add all 2020 transcripts

#GoVirCon_transcript.txt

@@ -0,0 +1,623 @@
+**Mat Ryer:** Hello, and welcome to Go Time. I'm Mat Ryer. Today we're talking about GopherCon. GopherCon was the first Go conference which I was lucky enough to go to in 2014, and now it's the biggest in the world, and has inspired lots of other GopherCons around the globe; or if you prefer, on top of the flat Earth, for those believers.
+
+Today we're joined by one of the creators of GopherCon, and the king of the lightning talks, or Thor's jester, as I like to call him - it's Erik St. Martin and Mark Bates. Hello, gentlemen!
+
+**Mark Bates:** I'm trying to figure out who's who.
+
+**Mat Ryer:** You're Mark. You should know who you are. That's one of the stipulations for being on the show.
+
+**Mark Bates:** Hang on, let me check my script.
+
+**Mat Ryer:** \[laughs\]
+
+**Erik St. Martin:** So we're trying something a little fun here, too. I'm actually pushing this thing to Twitch. I don't know who's listening, but... We thought we'll get some viewers there, too. I'll try to read in chat as I see it, too.
+
+**Mat Ryer:** Yeah, great. So this is multi-platform.
+
+**Mark Bates:** Is it concurrent and scalable?
+
+**Mat Ryer:** Probably, knowing us...
+
+**Erik St. Martin:** I'm gonna drop the link in the Go Time chat too, so if people wanna see our mugs while we do this...
+
+**Mark Bates:** Oh, whoa, whoa, whoa! Really?!
+
+**Mat Ryer:** Well, it's Twitch, isn't it?
+
+**Mark Bates:** Let me go put some clothes on. Hang on.
+
+**Mat Ryer:** No, please. I've already sold seats for people to see you without them, Mark. \[laughter\]
+
+**Mark Bates:** All they're gonna see is my bright, shining mug, with my nice Go Community T-shirt.
+
+**Mat Ryer:** Yeah. You look good though, mate.
+
+**Mark Bates:** Thank you. Well, you know, we should be talking about GopherCon, but I'll tell you why I look good; GopherCon this year has a really great -- because it's a virtual conference; we're gonna get into that. But one of the really great things they've decided to do is instead of giving their speakers some sort of GIF, they're actually working one-on-one with their speakers to look at their entire environment. Lighting, sound, camera, and everything, just to make their visual presentation and audio presentation so much better on the conference.
+
+**Mat Ryer:** \[04:26\] Wow.
+
+**Mark Bates:** And then they're shipping that gear to each speaker around the world, so that they can have just wonderful conditions. So these are my GopherCon lights that they've sent me. I've just set them up for this call, so I'm very excited. I think it's gonna be really fun, just based off of that alone. But anyway...
+
+**Mat Ryer:** That's amazing.
+
+**Mark Bates:** Yeah, it's a really great idea. It's a virtual conference, why not make everybody look and sound good? And they're literally doing it for every single person; they're not just sending one solution to everybody.
+
+**Erik St. Martin:** I think that we make assumptions sometimes about people's knowledge about things... So just because you're speaking doesn't necessarily mean you're familiar with lighting, and all of these things. I've got key lights and everything here, and all of that stuff... So I think it's important to try to help our speakers adapt to new climate, too; try to help them learn a little bit about audio, and lighting, and cameras... So our producer has been kind of meeting with each one, making recommendations based on -- you know, if they already have a good mic, then let's talk about your lighting. So yeah, we've been making recommendations for a year, and sending gift cards to all of our speakers to be able to do that.
+
+**Mat Ryer:** That's amazing.
+
+**Mark Bates:** Really cool, isn't it?
+
+**Mat Ryer:** Yeah. I mean, that speaks to the kind of attention to detail that you get with GopherCon. And when you're an attendee of a conference like GopherCon, everything kind of just feels effortless. And I suppose that's the point - it's meant to be a good show for people. But there's a tremendous amount of work that goes on behind the scenes, which we'll get to... But I'd love to first of all hear a little bit about the history of GopherCon, and where the idea came from originally, and what it was like that first time. Because it was back in 2014, wasn't it, Erik?
+
+**Erik St. Martin:** May of 2013 is when Brian and I first came up with the idea.
+
+**Mark Bates:** I still have my GopherCon pint glass from that conference. It's all faded, it no longer has any color, but it does say 2013.
+
+**Mat Ryer:** I don't think so, mate. It was 2014 the first one.
+
+**Mark Bates:** No. 2013, wasn't it? Erik just said 2013.
+
+**Erik St. Martin:** No, so 2013 is when we started planning. 2014 was the very first event.
+
+**Mat Ryer:** It's rubbed off his pint glass, and that's where he stored a lot of his information.
+
+**Mark Bates:** \[unintelligible 00:06:40.25\]
+
+**Erik St. Martin:** Yeah, so Brian and I worked together at a company, and 2011 was roughly when I started using Go. For a couple of years we had both been kind of going to tech conferences for other technologies we used, and kind of complaining about the fact that there wasn't one for Go... And then it was sometime April or May of 2013 where -- I think it was one of the members of Denver Gophers meetup group was talking about like "We really need a Go conference", and Brian and I are both like "We've been saying that..."
+
+So that tweet went out, and they were basically like "You guys should do it." And Brian and I had a conversation in his office, and we were like "I mean, how hard could it be", right? \[laughter\] It turns out very hard... But yeah, we decided to put up some money for it, and sign some contracts that would not end well if people didn't show up...
+
+It was really interesting, because we had no idea how many people would show. There hadn't been an in-person event yet for something like that... So if we got 200 or 300 people in a room, we were gonna be mind-blown.
+
+**Mark Bates:** And the tickets were really cheap, if I remember. They were like $200-$250, weren't they?
+
+**Erik St. Martin:** Yeah, that goes back to the "How hard could it be" thing... Like, "How expensive could it really be", right?
+
+**Mark Bates:** Very...
+
+**Erik St. Martin:** And then it turns out coffee is $75 a gallon...
+
+**Mat Ryer:** Wow.
+
+**Mark Bates:** I remember them being just dirt cheap, comparatively to conferences in general back then. It was a fun conference.
+
+**Erik St. Martin:** \[08:13\] Yeah, it ended up being 700+ people, almost 750, the first year.
+
+**Mark Bates:** Yeah. All in one big room. You remember - just one single room, one track conference...
+
+**Erik St. Martin:** Yeah. I was really surprised with how fast it grew. And then year two was right around 1,200 people... And then it was right around 1,500...
+
+**Mat Ryer:** We have the numbers here, actually, Erik... Because you're getting them wrong. \[laughter\] So in 2014 - 778. Then up to 1015, 1100... Yeah, so it is kind of growing. And this year would have been, according to Heather, this would have been over 2,000 in-person people.
+
+**Mark Bates:** Yeah, it was gearing up to be huge this year in Florida. We had a lot of just amazing speakers, and content, and the facilities... It was gonna be definitely the largest. And most of the gophers were going to be located on-site together, in one hotel, too... Which would have been really fun. I mean, it would have been a super-spreader event, but boy, it would have been fun!
+
+**Mat Ryer:** \[laughs\] So when we talk about the work that goes in, what were the big surprises that first year when you did the organizing of this, Erik?
+
+**Erik St. Martin:** There's a lot of coordination with speakers that takes place, especially if you want to be involved in all of that. And sponsors is a lot of work too, not just in the trying to acquire sponsors, but trying to facilitate the perks, and things like that. There was a lot of work in there. There's things you don't really understand, like banquet event orders, and all of these things that go in, and kind of considering dietary restrictions, and sometimes you might have to facilitate using external kitchens, and things like that... So there's a lot of stuff there that I don't think we realized in the beginning.
+
+I don't think we realized even contractual stuff sometimes too, about the fact that if you don't sell the hotel rooms that you blocked, you pay for them. So if we say 1,000 people are coming and we book rooms for 1,000 people, and 500 people show up - they take their rooms back, but we still pay for those 500.
+
+**Mat Ryer:** Oh...
+
+**Mark Bates:** Yeah, that's one of the biggest gotchas or overheads in conference planning. It's those rooms. That's where they get you.
+
+**Erik St. Martin:** Yeah.
+
+**Mark Bates:** Because it's such a big liability, just sitting there... You're hoping and relying on the community to fill that, so you don't lose tens of thousands of dollars on these rooms.
+
+**Erik St. Martin:** Yeah, so all of us are terrified, because all of you like to book your hotel rooms at the last minute.
+
+**Mark Bates:** \[laughs\]
+
+**Erik St. Martin:** So we look at it like this - several hundred thousand dollars that we are going to owe up into the last minute. Thank you very much. \[laughter\]
+
+**Mark Bates:** And while we're at it, get your CFPs earlier than the last day! \[laughs\]
+
+**Mat Ryer:** Yeah. There are like a series of things that you hear conferences asking people to do, and you sort of never really think about it, but yeah - get your talks in early, don't just submit them on the last day. But of course, if you give people a deadline... That's what happens.
+
+**Mark Bates:** And a lot of people don't think or consider what does registration mean for the conference organizers. When you get there, they often offer off-peak times to do registration sometimes; usually the day before, or whatever. Try and take advantage of those, because that not only makes it quicker for you to get in and out, but also it kind of flattens that curve of a rush of people at 8 AM coming in and trying to do it. And try to get there early. Try not to wait till five minutes before the keynote, rushing in... I mean, this huge wave of people assaulting the staff. It's hard to take care of that many people, especially 2,000 people. Imagine checking in 2,000 people. That's a lot.
+
+**Mat Ryer:** \[12:24\] Yeah, we don't want the attendees to be DDOS-ing the conference. So spread it out. \[laughter\]
+
+**Erik St. Martin:** Yeah, there's a lot of stuff like that. The CFP staff - we had about 300 proposals a year, and 290 of them are submitted in the last 24 hours. \[laughter\]
+
+**Mark Bates:** That's typical of every single conference. And we've definitely talked about that and the benefits of submitting early a lot... But let's get back to the conference. So I have been fortunate to have been to every single GopherCon.
+
+**Erik St. Martin:** Me too!
+
+**Mark Bates:** I should hope so. \[laughter\] And since year two, I've been fortunate to be a part of GopherCon, which has been so wonderful. We wanna talk about the organization of GopherCon, and I think Erik and Brian did a great job the first year for two very inexperienced people. It was this wonderful conference, we were all very excited about it... I could believe how many Ruby people I knew as I was walking through the door. I literally pulled up in a cab, got out of the cab, and there was Bryan Liles just standing outside the hotel... It was just like "Wow, I know all these people!" It was just this really fun event. But starting year two, like I said, I got to be part of it. But year two is when Heather came in, right Erik?
+
+**Erik St. Martin:** Yeah, so actually her team was involved year one. We used them for all of the off-site events - speaker dinner, after party, things like that. And then year two, they took over some of the logistical components of the venue, and all of that stuff. So it's been kind of like a progressive relationship, where each year they've taken on more and more for Brian and I, which has been insanely helpful... Because I think if we had to put in the amount of work that we did the first year or two, I don't think that we would still be doing it. Or at least also being engineers at the same time.
+
+**Mark Bates:** And not losing your mind either... Because it's such a crazy job. We work with Heather and Convention Designs, and they do all of this work... And you watch Heather at a facility, and it's amazing. So you walk around and your Wi-Fi is working at GopherCon - that's not an easy task. Anybody who's ever organized a conference knows that the conference Wi-Fi is a desperate \[unintelligible 00:14:50.07\] She makes sure. She wants that to be good.
+
+You talked about the food earlier? What was it last year - seven different options? Something like that. It was like seven or eight different food options. There's so much involved in all of this. And when you get there and it looks so gorgeous and slick and beautiful and just seamless - you've gotta know that there's people like Heather and all at Convention Designs, and Erik, and Brian, and Mat, and me, and all the CFP people...
+
+**Erik St. Martin:** It takes a lot more people than you'd think it would...
+
+**Mark Bates:** ...there's dozens of people involved in this conference every single year. The workshops -- you know, that's not including the workshop instructors or the speakers. That's just the people who are doing all the legwork in the background... And it's been different this year, I would say. Wouldn't you, Erik?
+
+**Erik St. Martin:** \[15:50\] Yeah, it's definitely been a learning experience, and trying to adapt, and a lot of scrambling... Because in the beginning we had contractual obligations, and everything. We had planned to hold an in-person event... So kind of the first stage was figuring out how do we move past all of that, and postpone it; all of this is when Covid first started coming out, and we're kind of hanging out, seeing what's happening... You're getting conflicting things about what's gonna happen - this is March - so there's a lot of struggle there, and just legalities of, you know, contracts were in, and deposits we've made, and things like that... And then we managed to shift it. Obviously, that didn't happen, so then we're going through the same process again... And then it's like "Okay, how do we do it completely online?", which is a whole new world too, right? We all do online things, but trying to do an online event is a whole other challenge.
+
+So it's been interesting in trying to figure out how to recreate what we have in-person... And I think GopherCon - my opinion is biased; I think GopherCon is a very special event, and how do we recreate that? I think the answer is probably that we can't. We can do the best we can, but it can't be this same exact thing, and we have to kind of rethink what an online version of it would look like, and not "How do we take GopherCon and put it online?"
+
+**Mat Ryer:** Yeah. It's definitely gonna be different. And it's got a different name, hasn't it?
+
+**Erik St. Martin:** It's just sort of like a hashtag, yeah.
+
+**Mat Ryer:** Yeah, \#GoVirCon.
+
+**Erik St. Martin:** It's something I said one time, and Heather loves it, so she uses the branding everywhere. \[laughter\]
+
+**Mat Ryer:** How is it gonna be different then? Obviously, we're not all gonna be collocating and being in the same place... Which is a big part of conferences, isn't it? So we do lose something by taking this online.
+
+**Erik St. Martin:** Yeah. And that's the difficulty. When we look at online content, it's never gonna be the same. When we go to an in-person event, all of us are essentially clocking out from work, from life, and we're moving somewhere and then we're kind of just fully immersed in it. But when we think about taking that and moving it to online context, it's not the same. We're all gonna be distracted. None of us is probably going to get -- well, I guess I wouldn't say I'd have the day off either; I'm gonna have a lot to do. But nobody is really gonna have the day off, right? They're gonna have it on a second monitor, they're gonna be kind of half in, half out... And that's just sort of the nature of this type of world.
+
+So I think it's just kind of accepting that that's the way it is, that people won't be able to be 100% in... And trying to create as many opportunities as we can for interactions between people... So we're still exploring how to do all of that. There's obviously going to be all of the online content that's part of the ticketed attendance, that people will see... And of course, all that content will be released to YouTube later... But we're trying to create areas where people can interact and discuss the things that somebody was speaking about, or do kind of like voice channels, and Discord, or something, for like birds of a feather, where you might be able to meet new people... You just drop into a voice channel that's discussing Buffalo, or something. And then in addition to that...
+
+**Mark Bates:** Why would we be discussing some city in New York? \[laughter\]
+
+**Erik St. Martin:** Because they invented Buffalo wings.
+
+**Mark Bates:** Oh, okay then. Never mind. Absolutely. Continue.
+
+**Erik St. Martin:** Yeah, and so there's some other things we're doing, too... We're gonna take the lightning talks and push those to Twitch, so we're gonna open those up to basically everybody... And then I've been kind of chatting with a bunch of regular Twitch streamers that do Go stuff too, so we're gonna hopefully try to create a hallway track, where if you wanna just go hang out in the chat, and watch people code, and kind of pop in and pop out, and "What are you working on?" "Explain this bit of code for me", and things like that...
+
+\[20:06\] They're not the same interactions, but they are new types of interactions that we're kind of learning. And I don't just mean we as GopherCon, but even we as people... I never realized how much I relied on interactions that I had while out at conferences, and stuff like that, until you're stuck at home for months in a room by yourself. This room looks really cool, but I'm sitting here by myself.
+
+So some of these things - you look for some of those interactions, and sitting in chat and Slack... You know, it's interaction, but it's not quite the same. But in recent months I've actually found relief in going and hanging out in Twitch chats. Sitting there, watching somebody code, having a conversation with them while they're doing it, and stuff like that. It's not the same thing, but it kind of fills a hole that's there.
+
+So how do we create new interactions, how do we create some interactions to get some of that back? Obviously, it's never gonna be 100%, right?
+
+**Mat Ryer:** Yeah... It's funny, because a virtual conference is gonna be more like your day job, whereas, like you say, when you go to a conference, it's a completely different thing. But could that be advice to the attendees? Could it be that you recommend clocking off, and properly take the day off and be out of touch, like you might be if you were at a conference?
+
+**Erik St. Martin:** I guess you could... But the likelihood that any of us are going to do that - lots of us have families, and things like that, so even being clocked in isn't being clocked in. We're all trying to find a new normal... So I think we have to accept that even if you got the day off for the conference, the likelihood that you're not gonna have to break away to help kids with school work, or meals, or things like that - it's going to be different, and we adapt. But yeah, I think the clocking out, and this just being my thing is probably not going to be...
+
+**Mark Bates:** The way people consume the conference?
+
+**Erik St. Martin:** Even just sitting in front of the computer is exhausting. Often we're energized by the people around us. Any type of event. You can be dead tired and go to a sporting event, or a concert, and then just find energy from the people around you. Me sitting here in this room in front of a computer for eight hours, watching people talk to me virtually - I'm not gonna get the same energy. It's going to get tiring. Eventually, I'm gonna walk away...
+
+**Mark Bates:** We're speaking of ways for the community to interact... Erik, do you mind if I make a little bit of an announcement about what we're doing at GopherCon?
+
+**Erik St. Martin:** Yeah, yeah.
+
+**Mat Ryer:** Oh, breaking news...
+
+**Mark Bates:** Breaking news, right here on GopherCon -- Gopher Time, or whatever the show is called...
+
+**Mat Ryer:** Yeah, Gopher Time, yeah.
+
+**Mark Bates:** Gopher Time.
+
+**Mat Ryer:** Mark doesn't listen to the show. \[laughter\] It's not his cup of tea.
+
+**Mark Bates:** I thought we were on the Rust Podcast for the first 15 minutes, to be honest... So who am I to judge? I thought I came in late to that podcast. So every year we do the GopherCon band, which is for me, one of my highlights; I just love it. We get all these fun people together, and we all get up on stage, we're playing music, and usually there's a big party... And again, talk about Heather going above and beyond last year in San Diego - we had the GopherCon band and the party on the deck of a battleship. It was a -- what do you call that?
+
+**Erik St. Martin:** An aircraft carrier.
+
+**Mark Bates:** An aircraft carrier, thank you. How amazing was that?! And it had the Gopher projected behind the band, on the ship, and stuff...
+
+**Erik St. Martin:** We always try to find ways to one-up ourselves... And eventually, you're gonna run out, right? \[laughter\]
+
+**Mark Bates:** \[24:00\] Yeah. We had some fun plans too for this year in Florida, but unfortunately, everything being the same... So we were talking about what we can do this year; we are doing, like we said, an hour of lightning talks each day, and like Erik said, we're gonna stream them live to everyone. So even if you don't have a conference ticket, you are welcome to join us for that hour and watch the live lightning talks, and Erik can talk a little bit about where you can find those, and how that's gonna work...
+
+As for sign-up, what we're really hoping to do - because that's a big question people have over here, "How do I sign up for lightning talks?", we're gonna try to give it more of an in-person feeling. What we're gonna do -- one of the biggest problems we have with lightning talks is the technical logistics; getting laptops, and microphones, and all that stuff to work. And I can't tell you how many hours I've spent tap-dancing on stage, while I fix somebody's computer to work with the projector...
+
+**Erik St. Martin:** I think Mark breaks them on purpose, just to--
+
+**Mark Bates:** Just to get more stage time?
+
+**Mat Ryer:** \[unintelligible 00:25:10.24\] "Oh, the laptop's not working again. What a surprise..." Oh, good. He's got his tap shoes with him.
+
+**Mark Bates:** \[laughs\] Yeah, now I'm gonna have to learn how to tap.
+
+**Mat Ryer:** Yeah, you should. If you can't already, you should.
+
+**Mark Bates:** I think that's clearly the next evolution of Mark Bates, is the tap dancing. What we're gonna do is we're gonna try and do a kind of first-come/first-serve on lightning talks. You show up, you get into the room, and your AV hopefully should be connected... And we'll just build up a queue. So you've gotta get there early for each of the days, and we're gonna just go through them as fast as we can, just like regular lightning talks; we're cutting people off at seven minutes, just like we did before. We're gonna give warnings at one minute... All the rules that we normally have for our regular, in-person lightning talks, we're gonna do the same thing.
+
+Like I said, we're trying to use that Go line-up on the side of the stage, by the stairs approach for the lightning talks themselves. So come with your laptop ready, open your slides, ready to go... And virtually stand in line by the side of the stage, and we'll call everybody up seven minutes at a time. It should be fun.
+
+**Mat Ryer:** Yeah, so that's very different to the main content or the main talks that we see at a conference, and I'm always quite interested in this - how do you select the talks? Because the thing is the conferences, particularly GopherCon--
+
+**Mark Bates:** We use a chicken system. We lay the numbers out on a field, and a chicken kind of goes along and just picks at them, and that's the speaker.
+
+**Mat Ryer:** The speaker selector.
+
+**Mark Bates:** Yeah.
+
+**Erik St. Martin:** Yeah, so since the very beginning of GopherCon, Brian and I have always wanted to ensure that our speaker line-up was composed of -- we didn't wanna play favoritism and things like that, so we created a system where basically we do blind CFP. So there's anywhere between -- some years I think we've had eight people, I think some years we've had as many as close to 20 reviewers...
+
+**Mat Ryer:** Is that where people have to submit the CFP in Braille?
+
+**Erik St. Martin:** \[laughs\] Submit it in Braille?
+
+**Mat Ryer:** Yeah. Blind process. No?
+
+**Erik St. Martin:** No, no... More like statistical blinding of data...
+
+**Mark Bates:** We're gonna get a lot of calls on that one, Mat...
+
+**Mat Ryer:** \[laughs\] Don't have a phone.
+
+**Erik St. Martin:** Basically, you submit the proposal, and it's wiped of any personal data. We don't know who you are, or anything about you... And there's a series of reviewers who will review it based on its technical merit... So if you are submitting a proposal, you have to remember that. Somebody - the only thing they know about you and your content and your knowledge on that subject is what you give us. If you give us three sentences that says "I'm going to talk about Buffalo", or whatever... I'm gonna just keep picking on Buffalo today.
+
+**Mark Bates:** \[28:08\] Um, excuse me, that was an exact submission from this year. Almost verbatim. I don't appreciate your breaking my anonymity here.
+
+**Erik St. Martin:** So you're gonna end up with a bunch of people who are like "Well, I don't know. I don't know whether they're gonna be able to write good content about this, I don't know how well they know the subject matter... And it's gonna get rated poorly." So I think it's important to make sure that somebody who knows nothing about you or anything would be able to read that and gauge how well you're going to be able to present this topic.
+
+So after that, we take roughly the top third of submissions that are rated, and then we unblind it, and then we have a panel of people - the program chair, and Brian, and myself, and stuff like that. Then we'll sit down with that and try to formulate that into a diverse kind of line-up. And when I say diverse, I also mean subject matter. The top three talks could be all one thing; there could be five highly-rated gRPC talks, and you're like "Well, this isn't a gRPC conference, so..."
+
+**Mark Bates:** It's quite often the case, too. It's usually hot technology of the day, and there's five or six all bunched together.
+
+**Erik St. Martin:** Yeah. So we'll do stuff like that, and make sure that it's fair, in the sense that we don't have eight speakers all from the same company, or we have content that's targeted both towards beginners to the language, as well as more advanced content... So we try to finagle a program out of that.
+
+**Mat Ryer:** Whatever the process is, it really works, because I always find that every talk I've been to, even on subjects where I don't have a particular interest or knowledge or anything, I find the talks to just always be of a really high quality. So there's a testament to that process, I would say.
+
+**Erik St. Martin:** Yeah. And I think it's also important to do a couple of things... So the other thing we agreed upon to help with this is that all speaker costs would be paid for. I think that's important, because if companies are paying for you to travel, you only get people submitting who work for companies that will send them, or have the money to send themselves... So that reduces it. Plus, you also get a lot of companies -- this has gotten better in recent years, but it user to be really bad. Companies want ROI, so they send you to speak at a conference, and... Yeah, I think it opens it up to a broader community, I think it allows them to control their own messaging and not be beholden to ROI on the company that sent them, and all of those things.
+
+**Mat Ryer:** Yeah.
+
+**Mark Bates:** And from a personal note, I've attended quite a few conferences where they have provided no speaker support at all. Just none. No flights, no hotels, nothing. I went to one conference that made me pay to be part of the speaker dinner.
+
+**Mat Ryer:** I think I went to that one as well.
+
+**Mark Bates:** You did? You were also at that one...
+
+**Mat Ryer:** It was really expensive, as well.
+
+**Mark Bates:** It was 25 Euros, yes... Anyway, we're not gonna get into that. But GopherCon has always treated their speakers just wonderfully, and really have made sure that they don't need to worry about anything like that, like getting there, and being there, and any of the special circumstances that sometimes arise with speakers, too.
+
+**Erik St. Martin:** And we try to leave stuff in their hotel room for them, little care packages and notes...
+
+**Mark Bates:** Oh, there's always a nice little gift bag, and...
+
+**Erik St. Martin:** I think it's important, especially for first-time speakers. You show up somewhere, you're very nervous, you don't know how things are gonna go... Sometimes you don't even know anybody. So I think it's kind of nice; you get back to your hotel room, you're tired...
+
+**Mat Ryer:** Yeah.
+
+**Mark Bates:** It's wonderful. And like I said, this year they're taking care of all the speakers by giving them equipment, to actually make their virtual presence even better. We talked about it right at the beginning of the show here.
+
+**Erik St. Martin:** It's theirs to keep.
+
+**Mark Bates:** \[32:10\] It's theirs to keep. They're not sending it to make them return it; they're sending them lights, they're sending them microphones, they're sending them whatever they need to make their 45 minutes, or 20 minutes, whatever amount of time they get, the best it can be. And that benefits both the speaker, and it benefits the conference, and it benefits the whole community. When you put the level of detail into the conference that they do, it really makes a difference. And anybody who's even been at GopherCon can definitely attest to that.
+
+**Mat Ryer:** Well, aren't you running the risk of ruining other conferences there by making it too good?
+
+**Mark Bates:** That's why I'm here.
+
+**Mat Ryer:** Yeah, you balance it out.
+
+**Mark Bates:** I balance it out, I bring it down a little bit.
+
+**Erik St. Martin:** This is kind of a hot take, but aren't they kind of ruining themselves...? \[laughter\]
+
+**Mark Bates:** Wow... I'm not engaging in that.
+
+**Erik St. Martin:** And there are some conferences that I'm not very happy with the way they're run, but... Yeah.
+
+**Break:** \[33:11\]
+
+**Mat Ryer:** So I asked a question on Twitter earlier... Twitter is a website where you post little tiny messages, and people can reply, correcting you and/or giving you hate.
+
+**Mark Bates:** Isn't that The Telegraph?
+
+**Mat Ryer:** \[laughs\] I thought we weren't having any political commentary...
+
+**Mark Bates:** Yeah, what a terrible British joke...
+
+**Mat Ryer:** Oh, yeah... Well, I had a message from -- I think his name is Gram Orsinium, which does sound like a precious metal from the periodic table... And he was saying that he came from Python to Go, and the community there has initiatives like PyLadies, and Django Girls, and things that focus on diversity and making sure that opportunities are open to everyone... This is something that I remember hearing from that very first GopherCon. There was a big emphasis on making sure that people that otherwise wouldn't be represented at conferences did get represented. So where did that come from, Erik, and how effective has it been to have that sort of focus?
+
+**Erik St. Martin:** I think that came from just basic human decency... We all need to do a better job at this, right? So I actually kind of would like to take this opportunity -- I know that we do a lot for diversity initiatives, and it's something that's very important to us... But I also don't wanna take this opportunity to pat ourselves on the back either, because I think as a conference, we still have a lot more that we can do, and I think even as a tech community we have even more that we still need to do. So I think that we need to keep iterating, we need to keep getting better about this, and I think we need to keep making it a priority.
+
+**Mark Bates:** \[35:58\] Yeah. And if you haven't already, go and check out things like GoBridge.
+
+**Erik St. Martin:** Yeah. And Women Who Go.
+
+**Mark Bates:** If you're looking to make a difference -- well, they've merged. So yeah, check out both, but - they've merged. One thing - I am notorious for touting these T-shirts, but I'm gonna tout them again. If you could see me, you could see that I'm wearing a Go community T-shirt. An Ashley Simpson original. If you go to womenwhogo.threadless.com, you can buy really nice, high-quality T-shirts like this, that have all these amazing Ashley McNamara designs all over them. They're $20, $25 and the money all goes to Women Who Go. It's a really fun way to add to your wardrobe...
+
+I got comments on my T-shirts all the time. People just all the time go "I love your shirt. Where did you get it?" And of course, they have no idea what the Go community is, but Ashely did such an amazing job with logos, and the Gopher... So go and get some. It's a great way to stock up, add a little bit to your wardrobe. I wear them at conferences all the time; your not gonna -- not only you're not gonna upset anybody, but hopefully you're going to actually encourage people to get more of these shirts, and help these great initiatives. That stuff helps, but comes back and helps all these people go to GopherCon, and all sorts of great stuff.
+
+**Erik St. Martin:** Did you see the eSports jerseys?
+
+**Mark Bates:** Yes...
+
+**Mat Ryer:** They look very cool, actually.
+
+**Mark Bates:** I'm not sure they suit me, but they look very cool.
+
+**Erik St. Martin:** I'm gonna be wearing one every day. \[laughter\]
+
+**Mat Ryer:** Yeah, I do like your T-shirt actually, Mark... Just for the viewers, we don't know what's on the bottom. But you know, that focus of GopherCon having that diversity focus in the beginning... Because GopherCon has a big influence on the Go community... And I think it can take some credit for the fact that we do have a lot of diversity in the Go community. Of course, as we say, there's more that we can do there, but I hear a lot of people telling me that particularly the Go community just feels different; it feels better, because it makes that effort. And I do think GopherCon gets to take a bit of that credit. It was very early, and it was mentioned a lot in that first conference. I feel like it planted a lot of seeds there, that are flourishing now into beautiful trees of all different types...
+
+**Mark Bates:** Well, that analogy broke down really hard, didn't it?
+
+**Mat Ryer:** It did, didn't it? Yeah...
+
+**Mark Bates:** Yeah, pretty hard.
+
+**Mat Ryer:** The editors will put a better analogy in for me. \[laughter\] Yeah, they're good like that.
+
+**Erik St. Martin:** I think it's important to do this stuff... And yeah, we continue to iterate. It's important. Some of this comes from, I guess, the Ruby community, too. Mark, you know this, too - everybody used to say the Ruby community is nice because Matz is nice... So I think it's important that if you put people first, you put the attendees first, you put the speakers first, and you make sure that they are all taken care of, I think they in turn treat other people the same way, and you kind of kick-start this whole process of everybody learning how we interact with each other.
+
+If the people at the top, whether that be the people on the language team itself, the core team, whether that be the people who run conferences - if they are hostile, or they just don't care, I think that triggers the way we treat each other.
+
+**Mat Ryer:** Yeah. Well, thank you for all that, because I think it does make a big difference. I hear people tell me that, too. Another response that we had on Twitter is from Paul Bert Pizza Emoji. Paul mentions that you're doing a lot of Rust lately, Erik... So although this is Go Time, I'd love to hear your views on Rust and what you've been doing with it, and also how it's better than Go in your eyes, and where it's not...
+
+**Mark Bates:** And does it tie into GopherCon in any way?
+
+**Erik St. Martin:** Yes, to all of those questions.
+
+**Mat Ryer:** \[40:14\] That last question though, it's -- so obviously he knows something... Do you know what I mean? It's like when a really bad TV interviewer says "So, did anything happen to you on the way to the studio?" \[laughter\]
+
+**Erik St. Martin:** You all can -- because we're on a Zoom call, and anybody who's watching this on Twitch can see behind me; I've got like an Electronics desk back here... So Rust is something I've been wanting to learn, to do some embedded stuff on here... But we've had some projects at work kick up, that use Rust. One of those is a Kubernetes node agent that basically pretends to be a node in the cluster, and orchestrates WebAssembly to host, and stuff like that.
+
+**Mat Ryer:** Sorry, you've broken my brain.
+
+**Mark Bates:** He lost me at WebAssembly...
+
+**Erik St. Martin:** WebAssembly can have some really, really unique...
+
+**Mat Ryer:** That was at the end, Mark...
+
+**Mark Bates:** Then you lost me at the beginning...
+
+**Erik St. Martin:** So if we think about a container, we have to secure it after the fact... Whereas WebAssembly - we kind of get to rethink that, where it automatically has no permissions, and you have to allow every call, so we can kind of have like this actor thing. And each of those abilities can be cryptographically signed. Mat, you could have permissions to give applications this role, to run this thing. And Mark, you could have this... So each person would have to sign that for that to happen.
+
+Rust has really good WebAssembly support, so that's why we were doing that. I can't really take credit for that project; we jumped in and helped with a couple of things, built a really funny demo where we basically recreated the scene from The Martian, where Matt Damon is stuck on Mars and he can't communicate, and I built this robotic arm that would point the flags, and stuff like that. It was really funny... So that's probably what he's referring to.
+
+**Mat Ryer:** You didn't need to plant potatoes in poo for that, Erik, but you did that anyway.
+
+**Erik St. Martin:** We did not. We skipped that, yeah.
+
+**Mat Ryer:** Alright, that's good.
+
+**Mark Bates:** I do that because it's the best thing to do with your leftover -- well, when you've got leftover potatoes and poo, what are you supposed to do with them?
+
+**Erik St. Martin:** So that's probably what he's referring to. I don't know whether any of the other stuff I've been doing - unless you follow me on Twitch - has been public. I'm really bad about advertising myself, so I don't really post on Twitter or something like that when I do stuff on Twitch. But that's what I've been doing for the last month, and it's to solve two problems. One is Microsoft has like a Learn TV, which is basically just like linear television with scheduled content; a mixture between videos of a certain theme for the week, and live broadcasts. That's implemented in this really interesting way; it was put together as a prototype very quickly, but we're trying to productionalize that and do it differently, and trying to solve for doing online events.
+
+There's a lot of great -- in all of this I've found that 1) people like to charge a crap ton of money for online broadcast things... And it blows $20,000, $50,000 and you're like "This is a pretty UI over NGINX, with an RTMP module. I don't understand..." \[laughter\]
+
+It was actually pretty funny... We were talking with individual companies about what we were gonna use for our platform, and one of them was talking about something - I think it was Heather, or our production person, was like "You know, we're just evaluating... The owners of this business will probably just write their own", and the guy is just like "Wait, what?" \[laughter\]
+
+**Mark Bates:** So fast-forward 30 days later and Erik is building his own.
+
+**Erik St. Martin:** \[43:48\] Yeah, so there's a lot of places where you can take content, and this gets into a longer discussion about how broadcast works over the internet... But essentially, we use one protocol that is very low latency to get our stream somewhere, and then on the other side that's converted/transmuxed and transcoded into different resolutions, that are delivered over HTTP in a CDN, and you have kind of a smart player that fetches that. if you're interested in how that stuff works, ping me on Twitter, or find me on Twitch, because that's what we're streaming.
+
+So that part is solved. How do I take my content and get lots of viewers to view it all over the world? That's a solved problem. But at the broadcast point you're stuck. So I can send there, but if I go offline to let Mark take over the channel, my stream goes offline. There's a lot of weird things like that. So our options for how do I switch between people and their streams tend to be like use Zoom, or Skype, and then route through somebody's house... And it's just awkward. I don't like that, and I have sort of a knack for taking on projects I'm totally unqualified to build, so I decided to tackle that challenge, to solve it both for GopherCon, like how do we switch between, you know, going to the MC, going to the speaker, and all of that stuff, while maintaining a linear feed.
+
+And there's also things I wanna do, like measure bit rate. And if you drop out, then fall back to a pre-recorded version, roughly in the same time. There's a lot of cool stuff we can do. So that's sort of what I'm tackling... But there's two main libraries that we use for doing that type of stuff - Libav, which comes from FFmpeg, and GStreamer. Both are written in C. I don't want to write C. All of us have also seen enough talks about cgo for me to know that I probably don't wanna use cgo for this.
+
+**Mark Bates:** Cgo is not real Go...
+
+**Erik St. Martin:** And the bindings our out of date, and stuff... So it turned out that GStreamer, the library I chose, has really good Rust bindings written by core developers on the GStreamer team; they're actually writing some of the new elements in it in Rust... It seemed like the perfect opportunity for that. So yeah, that's why I'm using that.
+
+But it's been interesting, because there's some things I really love about Rust, too. I don't love writing HTTP APIs in Rust. It's way more time consuming, and I will totally write them in Go. But it has some interesting language features that I actually really, really like. I like the borrow checker...
+
+**Mat Ryer:** What is that?
+
+**Erik St. Martin:** So the borrow checker basically tracks when memory is used. So if you pass it into a function, you either let the function borrow it for the time that the function executes, or you're transferring ownership. So only one piece of code can basically own any given piece of data at any moment in time.
+
+**Mat Ryer:** So is that about making it thread-safe then?
+
+**Erik St. Martin:** Yeah, and so all of it is basically checked at compile-time for you. You can't write race conditions, because it won't let you give access to this memory to two places without wrapping it in an automatic reference counted type, and things like that.
+
+But looking at what they do for error handling, I really like that. Obviously, generics are a thing; people have been talking about it. I forgot how much I missed generics... But they have enum types where each variant of it can actually hold its own data. So one variant can just be its thing, one could hold a string, one could hold a struct, one could hold a tuple... And so when you match on these things, you can get the data contained in that variant. That's important, because the way they handle things - you can't have null. There's actually an option -- it's a generic type, but there's an option enum type where there's a variant that says none, and there's some that holds whatever your type is. So when you get the type back, you check to see "Hey, is this none or is it sum?" and get the thing out.
+
+\[47:53\] And then for error handling, there's one called result, and it's okay or error. Okay might actually contain some generic type, whatever you stored in there, and then error would return your error. It's a really, really interesting way you can do that. And then most people will end up using enums for the error variants that you could match on those, too. But yeah, so there's a lot of cool stuff in Rust that I actually really, really like.
+
+**Mat Ryer:** Are there other ideas that you think Go should borrow? I know that Rust has borrowed some ideas from Go, hasn't it?
+
+**Erik St. Martin:** Yeah, so they have channels, and things like that. They're implemented in an interesting way; I'm trying to remember exactly... I think it uses a lot of macro stuff, but I'm obviously not a Rust expert. But yeah, so they exist there; they're just not implemented into the language. But a lot of this is because Rust's whole design is based on zero cost abstractions. So anything that you use should be just as efficient as if you handwrote code, and they have the requirement where you should be able to run this without a runtime. So it just creates some unique challenges... But that's where you get into kind of like why Rust does things like monomorphism, which is basically you use generic types, and the compiler generates the code.
+
+**Mat Ryer:** I see. Oh, cool. Well, if you blog about that or anything, I would be very interested, Erik.
+
+**Break:** \[49:24\]
+
+**Mat Ryer:** We were already kind of talking about Rust, we were already at risk of treading upon our next regular section, because it's time for... Unpopular Opinions!
+
+**Erik St. Martin:** Unpopular Opinions. Okay.
+
+**Jingle:** \[51:27\]
+
+**Mat Ryer:** So do we have any unpopular opinions?
+
+**Erik St. Martin:** I probably have a lot of them... Which ones are appropriate to say on...
+
+**Mark Bates:** Yeah, I'm in the exact same boat.
+
+**Erik St. Martin:** ...on a medium that's going to exist forever? Yeah, I've got a few of them... Tech-wise, full-stack developers are a myth. I think that you can find people who can do all of this stuff, but -- I used to be a senior web developer, and a junior backend developer. Now I'm a principal backend developer and a junior web developer. It's really hard to stay up on all of these topics; so as you evolve in one, you're naturally gonna fall off in others. So I think it's important to -- you may be able to operate on all levels of the stack, but you're not the same level at all levels of the stack; and if you are, you're an amazing human being, and you deserve all of the money.
+
+**Mark Bates:** \[52:41\] Thank you.
+
+**Mat Ryer:** Yeah, thanks. \[laughter\]
+
+**Mark Bates:** Sorry, I tuned in at the end, and I just assumed you were discussing me.
+
+**Mat Ryer:** Yeah, well - that is an interesting one. It's that thing of specializing, the specializing point, which I think is a good point. Okay... Mark, have you got one?
+
+**Mark Bates:** I do, actually...
+
+**Mat Ryer:** Do you?
+
+**Mark Bates:** I do... I think as developers, we tend to prioritize build over buy.
+
+**Mat Ryer:** Oh, really?
+
+**Mark Bates:** And we need to do a lot less of that. It's so easy sometimes for us to get caught in fixing a problem, or coding it ourselves... And that's great, but sometimes it's also just like "Let's just buy." I realize that may seem like I was just hitting Erik, but that was not at all what I was thinking. I was actually thinking of something completely different. But yeah, for the most cases we do tend to do that, and I think we need to try to resist that urge a little bit as a developer. Not everything needs to be coded by hand. It's not artisanal code, it's not farm-to-table Wi-Fi... So get out of the business of writing a new text editor. \[laughs\] We don't need one. Just write some code.
+
+**Erik St. Martin:** Is that another shot at Microsoft? \[laughter\]
+
+**Mark Bates:** Well, these are Unpopular Opinions, Erik... While we're at it, let's talk about your hair. I don't like your haircut, grow it out!
+
+**Erik St. Martin:** I didn't think they need to be unpopular with me. \[laughter\]
+
+**Mark Bates:** No, that's it... Recently, in the last few weeks, I've seen a lot of different things, and I've been like "Oh man, we really just need to get better at it." Mostly from Reddit. \[laughs\] I'll leave it there. How's that? I also don't care for bacon.
+
+**Mat Ryer:** Right. When you're babysitting bacon, you're negligent. Is that what you mean?
+
+**Mark Bates:** I mean, you've met my kids. Clearly, I'm negligent. \[laughter\]
+
+**Erik St. Martin:** Shots fired at yourself...
+
+**Mat Ryer:** It's true. They're like they've been in Jumanji.
+
+**Mark Bates:** They do. At some point I'm gonna have to teach them how to speak... But I'll tell you, you should see them pick the nits out of each other's hair. It's quite fascinating.
+
+**Mat Ryer:** It's beautiful.
+
+**Mark Bates:** It really is. It's something to behold... Yeah, I just don't see what the fuss is with bacon; it's okay.
+
+**Erik St. Martin:** I'm pretty sure that's unamerican.
+
+**Mat Ryer:** Yeah, it might be treason.
+
+**Mark Bates:** It's alright... I'll take sausage over bacon any day.
+
+**Mat Ryer:** Okay, that's the marketing --
+
+**Mark Bates:** That's a real unpopular opinion, how's that.
+
+**Mat Ryer:** Well, we actually test these now... So the @GoTimeFM Twitter will be putting out a poll on these to find out once and for all if they are indeed unpopular. And if they're not unpopular, then the punishment is you have to come back on and do another episode of Go Time with us.
+
+**Erik St. Martin:** And apologize for the unpopular opinion...
+
+**Mark Bates:** That's why I'm on all the time...
+
+**Mat Ryer:** Yes, yes... Okay, well this has been great. Can people still get tickets for GopherCon?
+
+**Erik St. Martin:** Yup, they are still available on GopherCon.com. It's $100 for access to everything. Again, videos are gonna be released shortly after the conference, so really all you're doing there is helping us seed deposits for the next in-person conference, and everything like that.
+
+**Mark Bates:** \[56:07\] And you get access to tons of live content and exclusive content around the conference which you don't get on YouTube.
+
+**Mat Ryer:** Yes. And we're all gonna be hanging out... Go Time are doing some special shows, some special episodes...
+
+**Mark Bates:** Right, I'm excited to do this.
+
+**Erik St. Martin:** Yeah, we didn't talk about that, yeah.
+
+**Mark Bates:** Should we mention those really quickly?
+
+**Mat Ryer:** You can...
+
+**Mark Bates:** Well, why don't you talk about them? You're the Go Time person, Matthew...
+
+**Mat Ryer:** Oh. Well, don't give me work...
+
+**Mark Bates:** Well, sorry. We can't just sit here and be content-generating monkeys for you all episode. You have to put a little effort in...
+
+**Mat Ryer:** You haven't generated a single monkey.
+
+**Mark Bates:** That is true.
+
+**Mat Ryer:** At the lunchtime sessions we're gonna be having live Go Time recordings, and then on the 13th we're gonna do a special game show, which I think is gonna be a lot of fun. It's nice, because you know, the talks are gonna be quite serious, and deep, and we wanna try and bring some levity in the lunchtime sessions.
+
+**Mark Bates:** So who's gonna be on the game show? Come on, I wanna know. Who are the contestants?
+
+**Mat Ryer:** We'll have to wait and find out. It's a surprise.
+
+**Mark Bates:** Do they get a prize?
+
+**Mat Ryer:** We'll see. You'll have to wait and find out.
+
+**Mark Bates:** A new car? A years' worth of Turtle wax, something like that?
+
+**Mat Ryer:** Yeah, they sound like great gifts.
+
+**Mark Bates:** I'll trade it in for door number one.
+
+**Mat Ryer:** \[laughs\] Behind door number one, "Go Programming Blueprint", by Mat Ryer. Congratulations, Mark! I know you've already bought six of them...
+
+**Mark Bates:** Arrrgh... Well, I guess if there's another toilet paper crisis in the winter, I'll be all set.
+
+**Mat Ryer:** \[laughs\] You can learn about Go as you go.
+
+**Mark Bates:** As I go. Go on the go! It's a new series of Mat Ryer-based toilet paper, ladies and gentlemen. I'm gonna start selling it. Go on the go. It's got a picture of Mat and a little piece of Go code next to it.
+
+**Erik St. Martin:** Little programming idioms on each...
+
+**Mark Bates:** Exactly. \[laughter\] Go on the go, by Mat Ryer. We might have broken Mat, by the way, on that one.
+
+**Erik St. Martin:** Yeah, so we're gonna be doing AMAs too, and all of that stuff will actually be open to the general public.
+
+**Mat Ryer:** Okay, great. Yeah, so even if you don't have a ticket... There's plenty, but we won't be able to get tickets.
+
+**Erik St. Martin:** Turn it into a Week of Go. Let's all get together, celebrate as best as we can at safe distances... But let's just turn it into a whole week. Not just for the people who are attending, but for the broader Go community. Because I know everybody looks forward to the hype around the conference, and everything...
+
+**Mark Bates:** Well, there's usually lots of great announcements, and releases, and all sorts of great stuff at GopherCon, so it's always a fun time to be a part of that live, to hear the Go team do their keynotes, and stuff like that live, versus waiting a month, or just hearing about it on Twitter. If you're really curious as to what's gonna happen... You watched the Apple event today; why not watch Go Time? Come on.
+
+**Mat Ryer:** But the Apple event is pre-recorded, from what I could tell... And this is not. GopherCon is live.
+
+**Erik St. Martin:** The talks are pre-recorded. They kind of have to be, right?
+
+**Mark Bates:** Well, they don't have to be...
+
+**Erik St. Martin:** Some people can do live. However, not everybody is in places with stable bandwidth connections... There's some challenges.
+
+**Mat Ryer:** GopherCon, when it's an in-person conference - is that live, or are they pre-records as well? The people on stage. Because they look real. \[laughter\]
+
+**Mark Bates:** They're holograms.
+
+**Erik St. Martin:** Yeah, it's the hologram. It's revolutionary new television. We record everybody in holographic.
+
+**Mat Ryer:** That's brilliant.
+
+**Mark Bates:** Dave Cheney hasn't been there since year two. It's just the hologram walking around. It's all AI-generated Dave Cheney content.
+
+**Mat Ryer:** Yeah. On the blockcheney.
+
+**Mark Bates:** That would be the Chave Deney, by the way, I think.
+
+**Mat Ryer:** The Chave Deney. \[laughter\]
+
+**Erik St. Martin:** Yeah, so all of these things are challenges we have to work through... If we rely on people having dependable internet and all of these things, then we further reduce the number of people that we can have, and the opinions that can be shared, and things like that. Yeah, it's easier to do that. And some people actually prefer it that way. They can do multiple takes, and things like that.
+
+**Mat Ryer:** Yeah, the quality could be--
+
+**Erik St. Martin:** \[01:00:13.04\] Some of us love doing it live.
+
+**Mark Bates:** I love doing it live. I do not like pre-recording it. I want that energy, that feeling that at least there are hopefully somebody -- people are sitting at home, listening or watching, even if it's just one or two people, but I like that energy of knowing that there's that. But others don't, I guess. To each their own.
+
+**Mat Ryer:** But the lightning talks are gonna be live.
+
+**Mark Bates:** The lightning talks are gonna be 100% live, yeah. Streamed to everybody at the time. So yeah, that will be really fun.
+
+**Mat Ryer:** Yeah, so anything could happen. Because you want a bit of danger, don't you?
+
+**Mark Bates:** Exactly, yeah. That's why you like the live setting, it's that danger. You never know what's gonna happen.
+
+**Mat Ryer:** But the talks are gonna be of the highest quality then for sure. There's not gonna be any live demos that fail, or anything like that...
+
+**Mark Bates:** No demo gods gone bad?
+
+**Erik St. Martin:** This is true. How are we gonna get our fix of broken demos? \[laughter\] I guess we can hope that somebody just leaves it in, right?
+
+**Mat Ryer:** Yeah... So if you're going to GopherCon, try and take some time off, try and dedicate yourself to it... I don't know if you're allowed to tie your kids up in your country where you're listening to this from. If you are, maybe do that. Give yourself a break.
+
+**Erik St. Martin:** Tie your kids up... \[laughs\]
+
+**Mark Bates:** I might need to check into the laws around Massachusetts.
+
+**Mat Ryer:** Yeah, check with your local--
+
+**Mark Bates:** Is the trunk of the car okay?
+
+**Erik St. Martin:** Mark, are we now accessories? \[laughter\]
+
+**Mark Bates:** Maybe.
+
+**Erik St. Martin:** Are there any lawyers listening right now? Are we accessories?
+
+**Mark Bates:** "No, my kids are in the trunk of the car, they're not tied up. C'mon..."
+
+**Mat Ryer:** \[laughs\] There you go. You can always guarantee Mark will say something worse, and then lets me off the hook. That's my defense in court.
+
+**Mark Bates:** That's your defense? "He's worse"?
+
+**Mat Ryer:** Yeah, someone was way worse.
+
+**Mark Bates:** \[laughs\] Well, okay. Thank you.
+
+**Mat Ryer:** Thank you. And thanks so much for joining us today. Unfortunately, that's all the time we have. I could just hang out all day here, but you know, it's not appropriate. But thank you so much for joining us, and Erik, thanks for all your work. Mark, I can't wait to watch those live lightning talks.
+
+Thank you very much, and we'll see you next time.

All about that infra(structure)_transcript.txt

@@ -0,0 +1,435 @@
+**Mat Ryer:** Hello, and welcome to Go Time. I'm Mat Ryer, and today we're talking about infrastructure. We're joined by some very special guests and hosts... Jaana B. Dogan. Hello, Jaana.
+
+**Jaana Dogan:** Hey! How are you?
+
+**Mat Ryer:** Welcome back! How's it going?
+
+**Jaana Dogan:** Good.
+
+**Mat Ryer:** We're also joined by Johnny Boursiquot. Hello, Johnny.
+
+**Johnny Boursiquot:** Hello, Mat.
+
+**Mat Ryer:** How are you, mate? How's your week going so far?
+
+**Johnny Boursiquot:** Well, just before coming on this call I was just handling an incident... So yeah, this is actually a good time, compared to what I've just been doing...
+
+**Mat Ryer:** Oh...
+
+**Johnny Boursiquot:** Yeah, yeah.
+
+**Mat Ryer:** Well, hopefully they're always good, aren't they?
+
+**Johnny Boursiquot:** Yes, yes... \[laughter\]
+
+**Mat Ryer:** Fair enough.
+
+**Johnny Boursiquot:** Yeah, it's always good hanging out with you all.
+
+**Mat Ryer:** And we have a very special guest as well, joining us from London, actually... It's Shubheksha Jalan. Hello, Shubheksha.
+
+**Shubheksha Jalan:** Hello! It's really nice to be here.
+
+**Mat Ryer:** Thank you. It's good to have you. Yeah, so let's talk about infra then... Because this is such a big space, and I must admit, this is the first time I really feel out of my depth on this show... Because this isn't a world I'm very familiar with at all, so I'm very excited to learn as much as I can here. So forgive any silly questions... But yeah, let's get started. First of all, what do we mean? What is infra? What are we talking about here?
+
+**Shubheksha Jalan:** I think a huge issue is that we don't have standardized definitions for any of this as an industry. Everyone has their own. But when I say "infrastructure", what I mean by it is like "How do you run your software? Where does your software run, and what does it take to actually run it?" So not just writing the feature, shipping the code, but how it actually works and becomes useful.
+
+**Jaana Dogan:** That's such an interesting definition... I was just having a conversation with my partner the other day; they are slightly a smaller company, and their infra team is mainly people who are actually configuring things. They are provisioning resources, they are configuring things, they are giving the rest of the organization basically resources, and helping them manage.
+
+For me, working for a core infrastructure team at a large company to me is like building the core infrastructure from ground up; building storage systems, building databases, building network infrastructure. For each domain, and I think for each different size and area the company is operating there's always a different definition... And it was very interesting to hear your own definition too, which I just realize that it doesn't necessarily translate to me, or some of the other companies that I've been asking the question of what does infrastructure mean to them.
+
+**Shubheksha Jalan:** Yeah.
+
+**Johnny Boursiquot:** For those of us who have been around for a while, infra -- we've had to evolve our default interpretation of what infra means... So it used to be that when you said "infra", I would think you have some sort of data center somewhere, where you've got some space allocated for your rack, and you're gonna get some customized hardware in from HP, or Dell, and something like that, and you're gonna mount those things, your one or two U's, or whatever it is, and you're actually running wires and doing all these things. That used to be what infra used to mean to me.
+
+Then we go into the cloud age, where infra is like "Well, infrastructure as a service, so you no longer are handling your hardware anymore. We're gonna do that for you." It's all virtualized. So now you're still trying to figure out how to get all these things to network together and whatnot, but not for a developer, for a software engineer, for an operations, SRE, or somebody that far up the stack; infra is gonna mean something completely different to them as well. So now it's all about configuration of what you've got. You're not worried about the networking as much, you're not worried about the storage as much... Unless you've got some very specialized use cases and needs, that way you know to any degree what's going on under the hood. All of that stuff has been abstracted from you... So infra means different things to different people, and depending on the time scale you're looking at, it's gonna mean different things to different people.
+
+**Shubheksha Jalan:** Yeah, I think that captures it really well. This sort of issue I've been running into as well, because it's so hard to talk about these things. You have to first define it, and bring people on the same page, and then start the conversation. Otherwise, it just goes in completely different directions, because everyone has their own interpretation.
+
+**Mat Ryer:** Is that because everyone's needs are so different then, that things look differently in the way that they have to handle these problems are just totally different?
+
+**Shubheksha Jalan:** I guess yeah, that's definitely part of it. And probably another part of it is that we just never came together as an industry to actually assign meaningful definitions to it... So everyone sort of ran away with whatever they came up with.
+
+**Mat Ryer:** It's funny that that happens, isn't it? But I guess in a way a lot of the world has changed... As you were saying, Johnny, the world is kind of very different. So in a way, I suppose, that does make sense... But yeah, I liked your definition though, because it's about your code has to run somewhere; and that's the thing about -- sometimes as developers, as long as our unit tests pass, and I've probably been guilty of this in the past... Don't write in, anybody that I've worked with in the past. \[laughter\] We don't need confirmation about this... So as long as the unit tests pass, that's it; it gets thrown over the wall and it's someone else's problem. But a few times that's come back to bite me, where decisions I've made at dev time have not played well once it gets into the wild, where it's really gonna be running... So how important is the mechanical sympathy, if you like, of where your code is really gonna run? How important is that to devs? Or is it okay that they can just -- as long as it's working for them and the tests pass on their machine, they just push the code and they can go home.
+
+**Shubheksha Jalan:** I think this is going to be a very divisive topic, because everyone has very strong opinions about it... But for me personally, I do think that it is important. As developers building just features, you should be aware how your code is running, and where your code is running... Not all of the details of it, because that's not really possible, depending on the size of the organization, but some basic-level details of it definitely. Because it just helps you write better code.
+
+**Mat Ryer:** Yeah, that's it. It does, doesn't it?
+
+**Shubheksha Jalan:** Yeah.
+
+**Johnny Boursiquot:** So if you've ever had to rethink your approach as a developer, if you've ever had to rethink your approach of maybe how you're fetching data from the database in your application, and how much of it you're fetching... A typical use case - you're tasked with building a CRUD web app; a blog, or whatever. On your local machine, everything is fast, because you don't have a lot of travel; the database is right there... Even if you're running in a containerized version of your database, it's right there. The latency is virtually non-existent. You don't have to worry about these things. Everything is just moving fast. Your integration tests are just as fast almost as your unit tests, and everything is just fluid. It's like "Yeah, I'm getting my stuff done. It's working, it's working."
+
+And then it goes into maybe a staging or QA environment where they throw a heck of a lot more data than you have locally. These days you can't realistically have the same sort of production workloads or set of data that -- you can't realistically have that locally to test against. Or if you do, it's really just a small subset, depending on the scale of what you're dealing with.
+
+Now, your code that was super-fast when you were working locally is now in an environment where it's getting tons more data thrown at it, and things start to degrade. Performance is in the toilet. All of a sudden you're like "How am I as a developer supposed to build for these environments if I can't replicate them locally?" And I'm interested in what Jaana has to say about this.
+
+**Jaana Dogan:** Yeah, I have actually a lot of different opinions on this. Sometimes, like you specifically mentioned, it's much faster in your local environment, but it's actually not always true. Sometimes it's actually slower in your environment. I think what is the core is that our development environment is so different than our production environments... And even as a rule of thumb, we try to keep them very similar to each other in terms of the operating system, the toolchain and everything, so we have some sort of similarity in the production environment... It's just such a big, complex thing to bring all the data in, all the external dependencies, let's try to use the actual services... It's almost impossible to build that entire complexity. And one of the interesting things about this breakdown between product teams versus infra is infra teams have been kind of making things more as a service; I was giving you the example of databases... If you're working for a company as large as mine, they're also responsible for databases as a service. So sometimes they're allowing you and giving you all these staging environments, also sometimes taking care of just bringing similar data, so you can actually evaluate things before you start even developing... So it's a completely different approach. I think back in the old days, at least for all the smaller companies that I've worked for, everything was starting at the development environment. Now we see it as more of like an environment and evaluation of environments, and then if we think that things are feasible, we're kind of like collaborating with the development teams.
+
+So I think the work style and everything has changed, and I think this breakdown between product teams versus infrastructure is contributing to that... And I think it's a positive change.
+
+**Mat Ryer:** Maybe we should bust some of this jargon, because you hear infra, I hear systems engineering, and I hear DevOps... Do we agree at least on what these terms mean?
+
+**Johnny Boursiquot:** No. \[laughter\] It's kind of funny... Let's take for example one of the hot, new titles out there - SRE. You'd think that would carry some sort of consistency from organization to organization...
+
+**Jaana Dogan:** Yeah.
+
+**Johnny Boursiquot:** ...that is entirely not the case, at all. An SRE at Google is gonna be very different from an SRE at Salesforce, and it's gonna be very different from an SRE at Microsoft. Yes, there is a throughline between these things - and that goes for the other titles as well; systems engineer, DevOps engineer - I cringe a little bit when I say that, but... Basically, these terms are gonna mean different things in different organizations... And even over the lifespan of an engineering team, the definition of that role may also change. So I don't think there's one solid definition of what software engineer, database engineer, infrastructure engineer, ops, whatever it is -- I think it's gonna be different everywhere you go.
+
+**Shubheksha Jalan:** Yeah.
+
+**Mat Ryer:** I thought SRE was text speak for "sorry". That's how I always thought about it. \[laughter\]
+
+**Shubheksha Jalan:** I'm sure a lot of SREs will agree with you on that... \[laughter\]
+
+**Mat Ryer:** But that's interesting, because I think that sort of makes a lot of sense to me. Often, the problems that you're dealing with -- that's gonna depend on what you're building, isn't it? And probably even some of the design decisions taken at the architectural level of code; as you've pointed out, that will necessarily impact how that then runs in production... And vice-versa.
+
+And that's the thing, I was always interested in what I could do as a developer to make it easier and to make it simpler for things to be then deployed on the other side. I was always very interested in that, and I had a few tellings off in the past, I'll admit it. We can talk a bit about that too, but - yeah, that's interesting. Do you think it makes sense then that -- will we ever agree on these things, or necessarily are they always going to be different and versatile?
+
+**Shubheksha Jalan:** My main gripe with not having some sort of common understanding is that it makes looking for a job in this domain really hard... Because you have absolutely no idea what you're getting yourself into until you actually are on the job... And by then it might just be too late.
+
+So as someone who's spent the last couple of months looking for a job, I avoided all jobs with the title SRE, because of this reason. Because 9 times out of 10 they actually turn out to be just like sysadmin or a pure ops role with a different title, and probably a higher salary. And that's not obvious, because we don't have common definitions or a common understanding of what it actually means to have SRE as a culture. We have Google SRE, but that doesn't apply to 99.99% of the companies out there. So that just makes that super-duper hard.
+
+The other thing which is sort of related to that is that because of that, it's really hard to break into this field. I used to be a frontend engineer, and I started doing frontend when I first got into tech. Then I quickly realized that I don't actually like frontend as much; I enjoy distributed systems and backend a lot more. So while I was in college, I took up a distributed systems course in one of my later semesters, and it was like a whole new world opened to me. Before that, everyone around me was just doing JavaScript and web development, so I was never aware there's this whole domain of problems and jobs out there which is also an option, and not just web development.
+
+I have nothing against web development; I have lots of respect for frontend engineers. I cannot do what they do... But yeah, that's just something I really liked, and I wanted to eventually switch to... So I started on my own, I started learning about distributed systems in general by reading papers, blog posts, and consuming whatever material was already out there... And then very quickly I ran up a wall and I realized that you can only do so much on a single machine, unless you have lots of money and access to a cloud provider. That's when I started looking for a job in this domain. That was right after I got out of college; I ended up at Microsoft, and I was doing weird things with databases and stored procedures, which was not a lot of fun, so I wanted to move on and do other things... And I was getting more and more interested in infra and distributed systems and related domains.
+
+So I did as much as I could on my own, and when I started looking for a job, there were absolutely no job listings for junior engineers anywhere. It was impossible to find anything at all... Because everyone is looking for platform engineering experience for five years, but nobody wants to actually hire someone, invest in them and mentor them to get them there, and that's a huge problem that I need to address. At one point I tweeted something along the lines of "How do you get started in this domain?" because I was actually curious. I was like "I can't be the only person running into this." And they were just like "Yeah, we moved laterally" or "I was thrown into this role at my job because someone else left, and it was a complete coincidence, and I stuck with it", and stories like that.
+
+So people don't really know that this is actually an option from the get-go... And we actively exclude people who are curious and want to learn and have that drive to learn, and can level up really quickly, because we don't give them opportunity.
+
+**Johnny Boursiquot:** You've just shed a light on this that -- I didn't quite know how to articulate it, and you've done an excellent job of that. It's beyond being a developer who's building a line of business, or features, and putting together these straightforward, end-user-facing applications, perhaps with a backend that services these things, where there is a clear path, basically to go from junior, to intermediate, to advanced... You don't see that a lot for the operational side of things.
+
+I've never seen a job post for junior platform engineer, or junior operations engineer, or junior SRE. Maybe they're out there, but I've never seen one. I literally have never seen one. So how do you mature in such a role and in such a field if nobody is hiring for somebody who perhaps doesn't have that experience, and is looking to get that experience. Where else are they gonna get that kind of experience? They need the environment that provides those scenarios and circumstances for them to learn within.
+
+Sometimes you can't spin up your own -- well, I guess you could spin up your own infrastructure in the cloud to do that; it would cost you a lot to learn... But this is the kind of job where really an apprenticeship model is really needed. You need a starting point in this field, and like you said, people are moving laterally, and they've been around the company long enough, and perhaps they started as a software engineer, and maybe be exposed incrementally to the operational side, and they can finally make that jump... But if you don't have that, where do you begin?
+
+**Shubheksha Jalan:** Yeah, that's exactly right. That's exactly what I ran into as well. It's a chicken and egg problem. You need experience to get a job, but to get a job, you need experience. So where do you actually start? Because there is a very hard limit to how much you can do and learn on your own when it comes to a job of this type, especially if you wanna work at scale. You simply cannot replicate the kind of things you'll be doing every day at home.
+
+**Mat Ryer:** So why is that though?
+
+**Jaana Dogan:** And it's such a big missed opportunity. I'm seeing a lot of people without a lot of experience coming into this field and looking at things from a very new perspective. In my experience, they've been really much more effective in terms of pointing out the core friction points than the experienced engineers. I think experienced people have a lot of -- they're accepting the current status, and they always assume that there's all these layers of layers of things that you have to satisfy in order to provide. But as a new, inexperienced engineer, just coming in and questioning some of these things more carefully - that's the perspective that we are missing in infrastructure in general... Because there's no good connectivity and we're just doing a bad job in terms of hiring people into this area.
+
+**Shubheksha Jalan:** Yeah, I completely agree.
+
+**Mat Ryer:** Is it because people think that this is the kind of serious thing -- it can't go wrong, and therefore you need lots of experience? Does it fall into that kind of category in people's minds? Because I agree, I've never seen junior -- in fact, whenever I think of SRE or DevOps or the people that know this usually, they've been doing it a long time. So it's a very good observation, but I wonder why it is... And it could be that - that we feel like you can't let these things fail. It's got to stay running, so... Maybe they think they can't trust somebody that's junior to do that.
+
+**Shubheksha Jalan:** I think that's definitely true, and that's on a lot of people's minds... But I don't think that has anything to do with being junior. When you start in a new role, even if you've been doing something for four years and you start in a new role, on your first day you will be scared of screwing things up. So I don't get how that correlates that with experience, because you might be a backend dev, for example, but you might not have done infra... And you can still screw things up, and people do this.
+
+There's actually a really good doc on exactly this topic from SREcon a couple of years ago by Kate Taggart - sorry if I mispronounced her name - which basically makes the case for hiring junior folks into SRE roles, and what experienced folks on the team can learn from them.
+
+**Mat Ryer:** We'll put the link to that SREcon talk that you mentioned in our show notes.
+
+**Shubheksha Jalan:** Yeah. And the other thing that I run into and I keep hearing from other people about it as well is just the lack of resources in this domain, as well... Because a lot of stuff is very Academy-heavy; a lot of the research was done maybe 2-3 decades ago, and it's catered towards more experienced folks who have a lot of experience in the area. It has a lot of jargon that is not very accessible... So I also wanted to see what we can do to help break down those barriers and make the field in general more accessible.
+
+Something I really appreciated was this series of blog posts by Vaidehi Joshi called Baseds. She basically published an article about distributed systems, from scratch, every week, for an entire year. So it's kind of like a whole distributed systems curriculum... And we don't have a lot of stuff like that. For example the Amazon Builders library - you cannot expect someone who has just a little bit of experience to go and read about leader election and consensus algorithms there and actually make any sense of it. And that content has its own audience, I completely agree. There are people who definitely need to consume that and they learn from it. But at the same time, I feel like we need to cater to all levels of audience, not just people who've been doing this for a while and are looking to hone their craft, in a certain sense... And without breaking barriers for people who are coming behind them.
+
+My philosophy of just being in tech in general is leave doors wide open behind you as you go through them, and I feel like we don't do a very good job of that in this community, and we need to do better.
+
+**Jaana Dogan:** That's really interesting that I completely agree with this. A couple of years ago I actually tweeted something like "Distributed systems are not as intimidating as what people make them sound like", and there are a lot of pieces in the larger spectrum that you can learn, and learn some of the basics; you don't have to learn about any election algorithm to do anything. You can just go for the stuff that you have an understanding or you can relate to, and then go from there and enlarge your knowledge and your background in the field.
+
+But it's funny, most of the people were telling me that I'm actually trivializing distributed systems by saying that... And I think given how big of an area it is, and how complicated some of the problems are, I think people are always trying to be realistic about misrepresenting the fields, but at the same time it creates this -- it's super-intimidating, because the field is very large, and the work is not very accessible, and it's just hard to read, and there's not an easy way to begin, and there's no easy way to progress. It's just like a very large spectrum of things.
+
+When we were talking about databases, for example on Go Time, we also had that struggle. It was hard to create a mental model to think about these systems, and different trade-offs, and the larger spectrum of things that you have to learn... And it kind of is intimidating to people. But we have to do a better job in terms of surfacing out the entry points, and just making them more visible and accessible.
+
+**Shubheksha Jalan:** I completely agree. I dealt with exactly that. It's such a vast and such an intimidating field to get into... And there aren't clearly defined entry points at all. You sort of just fumble your way through ten different things, and then you'll come across something, and you'll be like "Fine, this makes sense. I'll stick with it, and I'll go through that, and then switch to something else."
+
+It's such a shame, because it's such an interesting field, with so many really interesting problems to solve, and we're not gonna run out of them anytime soon, at least not in our lifetimes, unless (I don't know) the world ends, or something...
+
+**Mat Ryer:** Let's not test it. \[laughter\]
+
+**Shubheksha Jalan:** Yeah, especially right now... \[laughter\]
+
+**Mat Ryer:** Yes. But at least then all of our infra-worries will go away.
+
+**Shubheksha Jalan:** Yeah. That's the plus side.
+
+**Break:** \[26:29\]
+
+**Mat Ryer:** I'm interested in Go's relationship in this world... Because I remember from the very first GopherCon - I remember that was the time when I started getting exposed to a lot more talks and ideas and things from this world. Docker obviously is a big, famous Go project... So I think Docker gave Go quite a lot of credibility in those early days, because we think of this world as very grown-up; this might be linked to what we were talking about, why it's difficult for juniors... But I certainly have this idea that infrastructure is very grown-up, it's very serious. Maybe you can't be as creative as you can - another perception I had, that it's not so creative, it's just very serious; it takes very grown-up attitudes and a very grown-up approach to it.
+
+I remember being exposed to more and more of that because of Go's part that it started to play in this world. Why is Go such a big name in this space?
+
+**Shubheksha Jalan:** Yeah, that's a really interesting question... And my story actually started at GopherCon as well. When I graduated, I was mostly doing Java and some C++ in college...
+
+**Mat Ryer:** I'm sorry...
+
+**Shubheksha Jalan:** Yes...
+
+**Mat Ryer:** \[laughs\]
+
+**Shubheksha Jalan:** I deserve that, thank you... So yeah, I was learning about Go, and I was seeing it pop up, and I was not sure what it was all about. I looked into it a little bit, and then I decided to apply for the GopherCon scholarship. I got that, and I attended a bunch of talks, went around talking to people in the community. I think that's the first time that I met Jaana as well. It was a really, really good time. People in the community were really, really nice. The language seemed really accessible. So that's how I started getting into it and actually actively learning it. It was already exploding in the infra distributed systems community, because more and more people were using it for these kinds of problems. That's why I made the relations that if I learn Go, then I'll be able to get a job in this domain, which in hindsight was probably not the best way to go about things, but it sort of worked out; so I guess it's okay.
+
+So yeah, that's one of the main reasons I actually got into the entire ecosystem, because Go was and is still being used heavily in the cloud-native ecosystem, which has sort of become synonymous with infrastructure, in a certain sense, in the last couple of years... And I think that has really lifted it up in the developer community as well.
+
+**Mat Ryer:** I love that. Actually, we should do a shout-out, because the Go community and GopherCon in particular, Brian and Erik, and all the team there - from the very beginning, they had a very big focus on making sure it was an inclusive community, and making sure that it was trying to open doors for people where the doors might not be open. So I think that's great to hear a success story on that. So how was learning Go for you? Did you find it tricky, coming from Java and C++? Was it too easy?
+
+**Shubheksha Jalan:** Not too easy, because it has its own gotchas, but it was definitely easy. I did not have trouble grasping the basics of the language. They were very straightforward. It was mostly just the syntax bit, which takes a while to practice. It takes a while to get your head around that. But yeah, it was fairly straightforward. One of the favorite things for me about Go is how easy and accessible it is to people just starting out... Which also in turn makes all of the domains \[unintelligible 00:31:27.08\] accessible to people. And the fact that it's used so widely in open source infrastructure projects is just a bonus, because it's so hard to get started anyway... If it's written in Lua, or Scala, or something like that - those are not languages that people always learn in college, for example. Go is becoming a language that people are actually actively learning in colleges, so it just reduces that barrier as well, and it's easy to pick up on your own too, which is great.
+
+**Johnny Boursiquot:** One thing I'll mention, and I definitely appreciate your experience and your journey with the Go language and the Go community... There was a time there that I was a little worried, because -- I'm not sure if you all can recall, but there was a time where Go was an elitist sort of language; it created this perceived barrier, like "Oh, you only do Go if you are the top echelon of people who are gonna be doing these ops systems, distributed systems." They were creating this aura, this mythos around the language itself, which created a barrier for people who might otherwise be interested. Because of that, you could really shut out somebody even before they even tried the language itself. Just the aura around the community. And I'd like to think that we got past that, and through effort, I'd say.
+
+You saw a lot of community members who care about the community sort of go out there and make Go accessible to beginners, teaching and writing blog posts, and in their videos, talking about the inclusive nature of the community... It was a really concerted effort. This spell, this notion that Go was really reserved for a select few... First of all, no technology holds that high a regard. Anybody can learn to program, I truly believe so; you need to put in the effort. For some it's gonna be harder than for others, and that's just the way life is. But creating these artificial barriers for people to learn things - that has no place in this community... And I'm glad your experience was a little different, but I do wanna acknowledge that we did have some of our own stumbling blocks within the community, that phase where artificial barriers were being talked about for a little while.
+
+**Jaana Dogan:** It was very painful in the beginning, to be honest. Right now I'm not sure if anybody can remember, but lots of the initial criticism maybe Go received was also related to that. It looked like a preserved place for elitist people. And if you take a look at the earlier posts on the mailing lists and so on, it was very different. And it took a long time, many years, and effort, and lots of people felt very burned out actually, and left the community and the project altogether... But I'm actually glad that we got to the point, and we still have a lot to do.
+
+**Mat Ryer:** Yeah, I remember that first GopherCon... I remember people actually actively talking about having diversity, and things like that... So you're right, it was a proactive thing, and it remains one. It's something that we do have to keep working at. It doesn't happen by default, does it?
+
+**Shubheksha Jalan:** Yeah, definitely. It has to be a conscious effort. I remember the year I went to GopherCon - that was 2017 - they actually had a diversity and inclusion room on the last day of the conference. It was mostly people like Women Who Go, there were quite a few number of men as well, and I remember Russ Cox handing out mics for folks to ask questions, and I was just like "Whoa... I did not expect that, but it's great to see."
+
+Something that has personally attracted me to this community has been - Go, as well as like Kubernetes and JSON communities - the number of women who are well respected and in positions of leadership and influence. I have not seen that in any other programming community to that extent so far. It's bad, yes, but Go seems to be slightly better at it than a lot of other communities out there.
+
+**Mat Ryer:** Great. Yeah. But again, more to do.
+
+**Shubheksha Jalan:** Yeah.
+
+**Mat Ryer:** Okay, so one of the things that I was told off by by my DevOps guy was just a specific example... I didn't set the timeout on an HTTP client; so by default, when you use that default client, there's no timeout. So it basically means that your code can hang if there's problems on the network. That was kind of an early sign that told me "I need to understand the actual environment in which my code is gonna run." Are there other things that you wish developers knew, that any of us wish that developers knew, even if we don't have the scope to become experts in this world?
+
+**Shubheksha Jalan:** I have one question for you... Did you think about all of this before the show, or do you just loop it in your head every night, before you go to sleep?
+
+**Mat Ryer:** What, the questions?
+
+**Shubheksha Jalan:** Yeah, all of the times people have told you off.
+
+**Mat Ryer:** Yeah, they come back to me just as I'm drifting off, yeah... \[laughter\]
+
+**Johnny Boursiquot:** I did not expect that. That was good. \[laughter\]
+
+**Mat Ryer:** Well, that's how I learn, actually... You raise an interesting point, because I think a lot of the learning comes from when things go bad, and then you think "Oh, I'm not gonna do that again...", especially in this kind of space.
+
+**Shubheksha Jalan:** Yeah, it sticks with you.
+
+**Mat Ryer:** Yeah.
+
+**Shubheksha Jalan:** Something that I've seen people get caught by is variable shadowing.
+
+**Mat Ryer:** Hm. Tell us more. What is that?
+
+**Shubheksha Jalan:** There was an interesting incident that was caused by it. Basically, you were looping through a bunch of machines, but because the variable was already assigned, it was only looping through one single machine, and not doing everything that it was supposed to do for all of the machines. That's something that's very easy to forget, it gets missed in code review all the time, but it has caused very bad things.
+
+**Mat Ryer:** Oh... Yeah, so you don't think that; it just seems like a simple little thing that you might do, and of course, it has a knock-on effect.
+
+**Johnny Boursiquot:** This is an example where I think the Go language is sort of forcing developers to come face-to-face with some of the things happening at a lower level within the stack. If you take a typical framework user, say maybe like a Rails developer or a Django developer, or Laravel - whatever, those popular frameworks that provide so much abstraction on top of what's going on under the hood; even your interactions with your databases, that's abstracted from you... All you have to do is use a singleton, call a method, and all of a sudden data comes back, and the mapping is already done for you... You're removed so far away from the interaction that there is another system somewhere that you have to talk to, that you start to simply be like "Yeah, this is like doing RPC, all the things. I don't have to worry about the networking, or setting timeouts..." That's framework-level concerns, right?
+
+So in Go, unless you choose to go and use one of those frameworks, you do have to be aware of and start to actually educate yourself about these things. When you get a value, when you initialize an HTTP client - yeah, you should read the documentation to see all the different fields that you can set, because if you don't realize that "Okay, yeah, there's a timeout there", your code could hang indefinitely if the server on the other end chooses to hang on to your connection, all of a sudden you're doing yourself a disservice by not knowing what's happening under the hood.
+
+Now, I'm not saying you have to read through every line of the standard library to figure out what's going on. I'm not saying go learn the inner workings of networking, understand all the different OSI levels... I'm not saying just get that deep, but actually understand what it is to have one system talk to another system over the network. Understand what it means to have one system talk to a number of replicas within a network. Understand what it means to spin off multiple goroutines, where each of them could be doing something. When do their lifetimes end? When do they rejoin the main goroutine?
+
+Understanding these things that are happening behind the scenes, that way you're not just saying "Oh yeah, I'm just gonna go to Stack Overflow, copy and paste what I see", and then not really know what it's doing and why it's doing it. So it kind of forces you to educate yourself a little bit... But I think that's a good thing.
+
+The assumption was that developers don't want to learn these things, that they just want frameworks to do all the things for them. Not necessarily. Some do want to know what's going on under the hood. My days when I was first writing HTML, we had this thing called WYSIWYG, "What you see is what you get", and you'd go to websites and you're like "Oh yeah, cool." You could use tools, drag-and-drop things, whatever it is... And then you need to solve something, and you're like "Oh, crap. I can't drag and drop my way out of this thing. I actually have to View Source. I have to actually see what's going on under the hood and figure out "Oh yeah, there's an errant tag here missing, or there's an unclosed something." So you have to just look at Go as the View Source behind all the abstracted network, the interaction between your different systems (your database, and your application), you have to look at Go as the View Source, peeking under the hood as to what's happening there. That's gonna make you a better developer, and if you're not already in the operational space, that's gonna get you closer to understanding what it takes to actually be somebody in that role, that has to deal with systems, that has to deal with network, and that has to deal with latency and all these terms \[unintelligible 00:41:36.29\] for those new roles that we have floating.
+
+**Shubheksha Jalan:** I completely agree with that.
+
+**Jaana Dogan:** And it is very interesting that, especially with a new language, the assumptions are very different, because you have some baggage from other languages... One example is the context package. A lot of people, when they look at it, they don't necessarily actually question what it does. They automatically assume that it's also propagating things over HTTP, or gRPC, even though it's just a package that provides context propagation inside the same process.
+
+So I think we have all these different assumptions because we know that other languages or library spaces are just more powerful and trying to bring all this convenience into things. I think one of the difficulties of Go was people were coming with that baggage and were expecting more from the standard library than it actually was doing... And then they would be surprised, all of a sudden, that it was actually not doing that much. But then would go and learn, and it's not a big deal once you learn; there's always an easy way to get whatever you wanna get done... But even learning what is there and what is not there because of the assumptions and the baggage was a difficult thing for Go, because it was trying to be as small as possible, trying to be as simple as possible, compared to the other languages or library ecosystems.
+
+**Mat Ryer:** Yeah, and so then lots of teams started to build their own tooling, didn't they, to solve these problems... Mihai in the Slack channel was talking about this point, that orgs end up with their own flavor of tools to address similar problems, and that can be difficult for this industry (if you like) to move, when you move between companies even; there can be a lot of new things that you have to learn, because we don't really have these established frameworks.
+
+**Jaana Dogan:** Yeah, it's true. It's almost like you're living in a microcosm, based on which subset of language, and which additional tools that you inherited... So you have different assumptions, and different expectations. It's a completely different culture almost. Large companies, especially with a very strong culture - they do things in one specific way sometimes, and you automatically assume everything is very similar outside... But there's a larger world outside.
+
+**Mat Ryer:** Yeah, I think so. One example - I know a particular company that have basically a kind of microservice package in Go, and it might even have a lot of codegen, in fact... And then if you want to add a service, you use this tool to generate basically the scaffold of that service, and you sort of then do the implementation. And presumably, all the ins and outs are standardized somewhat, and a lot of other common problems are solved, like monitoring and traceability and all these sorts of infra problems. Shubheksha, have you ever used something like that, and what do you think of that as an approach?
+
+**Shubheksha Jalan:** Yeah, that's kind of what I have used at one of my previous jobs. We had a package which generated like a skeleton for service, and a lot of things were standardized. I really liked that approach personally, because it was really easy to get onboarded as a new engineer to the platform and the ecosystem, and just get started. Having standard conventions for how you write a service, how you write a handler, or how to do RPCs and stuff - it was really, really useful. And just having that skeleton right off the bat to get started with, without having to learn every single thing was really, really useful.
+
+**Mat Ryer:** Yeah, I think that is a good lesson. But the danger, of course, as with any abstraction, is teams might try and preempt that by designing that upfront, and I suspect -- I know the case you're talking about, and I suspect these things have to kind of emerge in some way, don't they? Or could there be a standard package -- there are some Go packages that address this... Could there be a standard that we all just use?
+
+**Shubheksha Jalan:** I think what you say about it evolving - you basically notice that you're solving the same problem again and again and again, and then you try to get an abstraction out of it... But then - yeah, that abstraction can go wrong. There are no guarantees that you will get it right, and depending on how you design it, it can be very hard to change, especially if it's used as widely as something like a new service being generated.
+
+So yeah, there's always that, and I think the other danger is that it sort of sets expectation that you don't need to learn about what's actually going on. It's like a magic wand that you wave, and you get stuff out of it, and then you can start building on top of it, but you don't really understand what's going on underneath... And yeah, that's again something that you need to be careful about. Like Johnny said, there will always be people who are interested in figuring out what's going on underneath, but there will also be people who just wanna build on top of it. So I think getting that trade-off right between catering to both audiences is very hard, but it's also super-important to keep that in mind when designing abstractions.
+
+**Mat Ryer:** Brilliant.
+
+**Break:** \[46:57\]
+
+**Mat Ryer:** Well, it's that time again... It's time for our Unpopular Opinions!
+
+**Jingle:** \[48:14\]
+
+**Mat Ryer:** ...and the music plays... \[laughter\]
+
+**Johnny Boursiquot:** You paused there, kind of waiting for it to play...
+
+**Mat Ryer:** Yeah... That's another thing that I just hear over and over in my sleep, that thing with Unpopular Opinions...
+
+**Johnny Boursiquot:** \[laughs\] There's a pill for that, mate.
+
+**Mat Ryer:** Yeah, thank you. I'd like some.
+
+**Johnny Boursiquot:** \[laughs\]
+
+**Mat Ryer:** Do we have any unpopular opinions on this? You must have one, Johnny...
+
+**Johnny Boursiquot:** I mean, we have a guest on the show, so I'll allow them to...
+
+**Mat Ryer:** That sounds polite, but it's actually rude. \[laughter\] Okay, if we don't have one, then it's just a waste of theme tune, isn't it?
+
+**Shubheksha Jalan:** I mean, I can probably come up with one... One tech and one non-tech.
+
+**Mat Ryer:** Brilliant.
+
+**Shubheksha Jalan:** What Johnny said earlier about DevOps and juniors - I just cry a little bit every time I see someone hiring for that... Because I don't know what they expect or what the candidate should expect when applying for that role... Yeah, I just wish that people would stop doing that. And a non-tech one is that I think pizza is overrated.
+
+**Mat Ryer:** Oh, here we go...!
+
+**Johnny Boursiquot:** Ooooh...! \[laughter\] Oh, no...!
+
+**Shubheksha Jalan:** Nobody saw that coming.
+
+**Johnny Boursiquot:** Oh, man... \[laughter\] Oh, man...
+
+**Mat Ryer:** By the way, that isn't a non-tech one. We love pizza.
+
+**Shubheksha Jalan:** Oh, yeah. Yeah, I'm just really sick and tired of eating pizzas in all the meetups... \[laughter\]
+
+**Johnny Boursiquot:** Oh, man... \[unintelligible 00:49:52.06\] Oh, my goodness... Oh, my goodness. Okay, so I do have an unpopular opinion, because Shubheksha gave me the courage with her first one, so...
+
+**Shubheksha Jalan:** Yes...!
+
+**Johnny Boursiquot:** ...I'm gonna lean into this one a little bit, so...
+
+**Mat Ryer:** You don't like lasagna, do you? \[laughter\] Okay, go on.
+
+**Johnny Boursiquot:** I've been sort of noticing this trend on the interwebs, of people changing the names of their main branches or their master branches on their repositories, and things...
+
+**Shubheksha Jalan:** Oh, yeah...
+
+**Johnny Boursiquot:** Well, that's all well and good, but other than renaming your master branch, I'd settle for you rethinking your hiring practices, to be less laced with implicit bias. I think that's gonna be way more beneficial in the long-term than superficial gestures like that. It's fine if you think it helps, if you think it helps culturally within your organization; that's totally fine. A master branch really never bothered me much, because I choose to think beyond that, but again, that's not gonna carry as much weight for me within an organization if you do that, but the rest of your practices and the rest of your culture is uninclusive and uninviting for people who look like me.
+
+**Mat Ryer:** That's very well said. That's a shame, because I was hoping we could just change that and then we fix racism. \[laughter\]
+
+**Shubheksha Jalan:** \[unintelligible 00:51:23.12\]
+
+**Mat Ryer:** Actually, I saw a heartbreaking report where somebody - I think it was Oxford University in England... They applied for lots of jobs and just used different names. The CVs were exactly the same, they just changed the names, and the results were staggering. And honestly, I thought we were doing better than this report showed. I'll try and find it and put it in the show notes. It was genuinely heartbreaking that just by having a different name, or a name that's difficult to pronounce perhaps... Not for me, obviously. I nailed it. \[laughter\] Then it gives you a disadvantage. Just a name. Everything else is the same.
+
+**Shubheksha Jalan:** In the \[unintelligible 00:52:10.08\]
+
+**Mat Ryer:** Yeah, but with editing though. Don't forget. \[laughter\]
+
+**Johnny Boursiquot:** It's gotta be just once.
+
+**Mat Ryer:** It's gotta be the first time. \[laughter\] Yeah, that's heartbreaking, ain't' it?
+
+**Shubheksha Jalan:** Yeah... But that's the way of the world. Jaana, you need to have unpopular opinions. I'm sure you do.
+
+**Johnny Boursiquot:** Yeah, Jaana. Bring the heat. You've been missing in action for a while. Bring the heat.
+
+**Shubheksha Jalan:** Go for it!
+
+**Jaana Dogan:** Oh, my god... Tech or non-tech?
+
+**Shubheksha Jalan:** Both.
+
+**Johnny Boursiquot:** Whichever one you wanna--
+
+**Jaana Dogan:** Okay... First - tech. I think I shared my unpopular opinion earlier, when we were talking about how intimidating distributed systems is... And I do believe that distributed systems are not as intimidating as they look like. The main reason it looks intimidating is it's still seen as elitist. There's a bunch of people controlling the entire field. Or it hasn't been able to go out, other than that small circle, and that's why it's hard, and it's kind of like -- those people are not contributing to the state of the world by making things accessible or having more of like a broader... I mean, this is what other fields that -- I understand the challenge so well, because it's so big, but at the same time, this is one of the fields that I haven't seen any progress in in the last ten years, in order to categorize things, or in order to come up with ways to think about problems. That's why it keeps being at the hands of a small group of people... And that smaller group of people - they're good in their job, and they have experience and so on, but that's not enough. I think that's a slightly unpopular opinion in my circles, that distributed systems is not that hard... But that's my unpopular opinion.
+
+The other unpopular opinion I have - it's tech, actually; it's not non-tech. And it's about SRE. One thing that I don't understand about SRE is - to me, it's true that we don't know what it means to be a junior; so you're like "Who's hiring junior SREs?" and whatever. But you know, I also don't know what it means to be the top SRE in the world. What does their job look like?
+
+I have a really good understanding of what software engineering can evolve into, but what is a principle SRE though? I don't have a lot of insights, and I know some people at that level at my company, and I've worked with them, and I still don't have a good understanding. And it's partially because this field is still less known, and there's less conversation about the role... But also, at the same time, it might be the case that people at those levels are not necessarily representing themselves as SREs, because there's always some sort of issue with being an SRE versus being a software engineer.
+
+At Google, for example, software engineering is considered as a better role, even though these are completely orthogonal things. So people who are at the principal levels are probably finding it harder to identify with that role, and that contributes to the problem of us not having any insights about their work, which is really sad.
+
+It's probably not a very unpopular opinion, but I believe at the higher levels we also don't know what an SRE is doing, and we don't have a lot of insights.
+
+**Shubheksha Jalan:** That's a really, really good point.
+
+**Johnny Boursiquot:** Yeah. Like you're saying, I don't think -- it hasn't been around long enough. I mean, you take the general software engineering field - we've been at this for a little while, so you can get a sense... Even at the junior level, you kind of know who seems skilled, and as you mature yourself and you go through the levels, you get a sense of "Okay, what does a top software engineer -- what skills do they need to have?" Because you're using that as sort of a roadmap for yourself. Like "I want to become a top engineer. Oh, I need to spend more time learning about different architectures, I need to spend more time learning about different patterns, and different things..." Because in your mind you have an idea of "Okay, one of the best engineers exhibits these qualities, or they know these things", because we've been around long enough to get that sense. Even if the role looks different at different organizations, there's this ideal representation of what a top software engineer looks like. But SRE hasn't been around long enough for us to know what that is, especially if the experience varies so much from org to org; we don't yet know what that should look like.
+
+**Jaana Dogan:** That's also a barrier against people who are trying to get into this field, because they don't have a clear mindset of what they can achieve. Or they don't actually know what it actually translates into in the higher roles, to be able to feel confident enough that "This is what I wanna do in the long-term myself." Because all the roles, all the types -- the tech work is a long investment; it's not like you can just go in and make a lot of impact in a very short amount of time. You make impact in a very long amount of periods; you invest 5-10 years into roles, and grow up to really high levels... And people just wanna understand what the role ends up being after 5 years or 10 years in order to invest their very precious time into that role.
+
+**Shubheksha Jalan:** Yeah, this is something I've had a conversation with multiple people about, who are SREs, as well as people who've been in the software industry for a while... And the conclusion that I got was that it's -- for me, for example, I've just completed three years in the industry, so it's too early for me to pigeon-hole myself into an SRE role... Because there is bias against that title, no matter what your actual job is. People have a certain impression of you when you say you're an SRE, and they will interpret it according to what their definition is.
+
+It's much harder to shift from being a software engineer to being an SRE, than vice versa. We need to work on that as an industry, but that's just something that's there, and we need to make people aware of it and acknowledge that that bias exists. More often than not, the hiring pipeline is also very different for both of those roles, which doesn't help either.
+
+**Mat Ryer:** I think this has been a very important conversation. I started the show by saying this is an area that I really feel like I didn't have a good grasp on, and you've really opened my eyes as to why that might be... So one thing I will say - anybody out there that has a story or experience or something, I could imagine them thinking "No one's interested in this", but we should encourage people, if you've got these stories and if you've got some experience, please share it, because that could be a way that we could start to illuminate some of these dark corners that we are operating in. Yeah, very, very good stuff.
+
+**Shubheksha Jalan:** I am brainstorming on a bunch of projects in this area to help break down some of the barriers. So if anyone is interested in collaborating, sharing their story etc. feel free to hit me up. I'll be more than happy to chat.
+
+**Mat Ryer:** How should they get in touch with you?
+
+**Shubheksha Jalan:** DM me on Twitter. My handle is @scribblingon. Or email me at hi@Shubheksha.com.
+
+**Mat Ryer:** Brilliant. I wanted to ask about those things... We are definitely pushing it for time, but I wanna find out what those things are... What are they?
+
+**Shubheksha Jalan:** They're not super-concrete right now, but something I really want to do is have a beginner's version of, for example, the Amazon Builders library, which is accessible to folks who are new to this field. You can pick up an article, you can progress through it and make sense of things, and then go on to read those kind of articles.
+
+Another thing that I just thought of - it literally just popped into my head, based on what Mat said a few seconds ago - is that it'd be really cool to actually highlight some of these stories and talk to people about it either in a podcast format, or in interview format, about how they got into this field... Especially folks from under-represented people, because this is something I've struggled with as well, like to Jaana's point. It's very hard to find role models in this area, or people you can talk to and who can guide you, who also look like you. So it'd be really cool to highlight and share stories, so that that is some sort of a resource for folks to just go to and find like-minded people who are interested in the same things when it comes to technology as well... So yeah, that would be cool to do.
+
+**Mat Ryer:** Well, that's all the time, unfortunately... And I genuinely mean unfortunately this time; usually, I just say it to be nice... \[laughter\] But it's been such a great conversation. Thank you so much Jaana, Johnny and Shubheksha. Thank you very much, and we'll see you next time.
+
+**Outro:** \[01:01:31.13\]
+
+**Johnny Boursiquot:** It's close enough... Do you wanna get us going, Mat?
+
+**Mat Ryer:** Um, yes, I do... I just want to make sure I don't mispronounce your name...
+
+**Johnny Boursiquot:** Are you writing down the pronunciation?
+
+**Mat Ryer:** Yeah, but when I write it down, I kind of try to write it phonetically, and then I realize I don't know how to read that. \[laughter\] I just don't wanna be a typical English person, that's what's really going on.
+
+**Shubheksha Jalan:** I appreciate that.
+
+**Mat Ryer:** Thank you, yeah. So go ahead, one more time.
+
+**Shubheksha Jalan:** Shubheksha.
+
+**Mat Ryer:** And then your last name?
+
+**Shubheksha Jalan:** Jalan.
+
+**Mat Ryer:** Jalan.
+
+**Shubheksha Jalan:** But can I get this part of the clip, so that I can forward it to other people who mispronounce my name? \[laughter\] It would be so useful.
+
+**Mat Ryer:** Yeah, we'll release it as an episode on its own. \[laughter\]
+
+**Jaana Dogan:** Oh, gosh...
+
+**Mat Ryer:** Okay, Shubhepsha Jalan.
+
+**Shubheksha Jalan:** Shubheksha. There's no P. Where is the P coming from? It's a K.
+
+**Mat Ryer:** Yeah, that's true.
+
+**Jaana Dogan:** It took them like five years to learn how to pronounce my name, so don't take it personal.
+
+**Shubheksha Jalan:** Seriously? Then I have no shot in this lifetime.
+
+**Jaana Dogan:** I've changed my name five times in order to get it right.
+
+**Shubheksha Jalan:** Oh, wow.
+
+**Jaana Dogan:** I still can't use my full name. Nobody can pronounce it.
+
+**Mat Ryer:** So is Jaana, so we can hear it.
+
+**Jaana Dogan:** Like the full thing? Nah... Too much information.
+
+**Mat Ryer:** Even you can't say it. \[laughter\]
+
+**Johnny Boursiquot:** It's like, "I ain't putting my government name out in the open."
+
+**Mat Ryer:** Yeah, fair enough. What's your mother's maiden name? I'm just trying to slowly get enough information to steal Jaana's identity. \[laughter\]
+
+**Johnny Boursiquot:** Oh, man...
+
+**Jaana Dogan:** My name is a globally unique name... Anyways, okay. Let's be serious.
+
+**Mat Ryer:** Mine is, actually. Mine is, I think.
+
+**Shubheksha Jalan:** Yeah, I guess...
+
+**Johnny Boursiquot:** Wait, yours is a UUID too, Mat?
+
+**Mat Ryer:** Yeah, it's G235119-... \[laughter\] No, but really, there isn't another - as far as I know - with my spelling. But it doesn't help me now. I need to keep this in my realm, so can we not talk about too many things, please? \[laughter\] Okay, Shubheksha Jalan.
+
+**Shubheksha Jalan:** Yes! That is good.
+
+**Johnny Boursiquot:** That was the closest you got. That's nice!
+
+**Mat Ryer:** What I'll do is after the show I'll just record me saying it 50 times, and just send it to you and we'll edit in the right one. \[laughter\]
+
+**Shubheksha Jalan:** \[unintelligible 01:05:52.26\]
+
+**Mat Ryer:** Okay, I'll do a little intro then, and then we'll get started.

Beginnings_transcript.txt

@@ -0,0 +1,491 @@
+**Mat Ryer:** Hello, and welcome to Go Time. I'm Mat Ryer. Today we've got a very special show... We've invited three people along who are at various points in their careers. We have Aditya Prerepa, who is a high school programmer, we've got Shaquille Que, who has actually just started his first day internship, and we've got Benjamin Bryant, who is in his first Go position. Welcome, everybody! It'd be great if we could start off - maybe Aditya, you could tell us a little bit about yourself and what you're up to.
+
+**Aditya Prerepa:** Sure. I'm Aditya, I'm currently 15 years old and I'm a sophomore in high school. I have been an intern at Life360, doing Android stuff, and I'm an incoming intern at TaoSecurity, doing backend and frontend security stuff. I'm currently trying to open source at Istio.
+
+**Mat Ryer:** Cool! Welcome to the show.
+
+**Aditya Prerepa:** Thank you.
+
+**Mat Ryer:** Shaquille.
+
+**Shaquille Que:** Yeah. I'm Shaquille Que, I'm 20 years old, I'm studying computer science and math double major. I will be in my third year this coming fall, and like Mat said, I'm doing an internship right now. Today is actually my first day. I'm currently interning with a Go team to improve package.go.dev. This is my second internship; I interned at a small startup also in Go last year.
+
+**Mat Ryer:** Great. Thank you, and welcome to the show. And Benjamin Bryant - hello, Ben.
+
+**Benjamin Bryant:** Hello. I'm Benjamin Bryant and I graduated university around a year and nine months ago; I've been in my first professional Go position for around nine months. I'm currently 24. When I first got invited to the show I thought I'd be the youngest here...
+
+**Mat Ryer:** \[03:57\] \[laughs\] Yes... Well, this is actually a really interesting subject, I think, because a lot of us take a lot of things for granted; we've also been doing this for many years, and it's difficult to remember what it was like in those early days. Also, the world's probably completely different. So I'm excited to hear from you what it's like and what your experience has been, and I think also our listeners - and hopefully we'll pick up some listeners that are also kind of in your position, or maybe a year or two behind each of you, and hopefully we can shed some light on things for them as well. That will be very interesting.
+
+First of all, let's find out - why Go, specifically? We're all here because we're dealing in Go, in some way... Ben, what was it about Go that attracted you?
+
+**Benjamin Bryant:** Well, I kind of got into Go by accident, in the sense that when I first graduated, the only job I could really find was a C\# position... So when I first got that job, a friend invited me to the GopherCon pre-drinks in the U.K. At the time I'd never heard of Go, and I thought GopherCon was like a conference for the fluffy animals... So I was really looking forward to checking it out. But when I got there and I went downstairs and \[unintelligible 00:05:20.19\] there was like a programming language, I've gotta say I was actually very disappointed... \[laughter\] After talking and listening to (I guess) everyone talk about what they were doing - the language Go at least has a bunch of other different words attached to it, that I had also interacted with in university. That kind of made me want to at least follow up on that and try and find out what it was about. That's how it happened.
+
+**Mat Ryer:** Yeah, that's really cool. So those other things - do you mean things like channels, and goroutines, and these kinds of concepts?
+
+**Benjamin Bryant:** Well, even that. But also going to different Go meetups - similar to how on Go Time it says you guys talk about microservices and Kubernetes, and words like that, those were words I had never encountered at university, at least while I was there... And I definitely wasn't self-studying or researching, because I don't think the word "Kubernetes" even existed to me until I went into -- I mean, I have no idea why would you ever Google that word at all, for any reason... But yeah, it's very ubiquitous. Even something like Docker is something I'd never even heard about before encountering this kind of community.
+
+**Mat Ryer:** Yeah, so I wonder then -- because a lot of the things you've talked about are quite new, or certainly were quite new... And I suppose the computer science courses and things take a bit of time to catch up with that... Shaquille, when you started with your internship - well, you're starting today - how did you find out about that? How did you get into doing that?
+
+**Shaquille Que:** Are you talking about the Google internship right now?
+
+**Mat Ryer:** Yes. Yeah.
+
+**Shaquille Que:** Well, my experience is that at college a lot of people are really into applying for internships, and getting more experience there. Like Ben said, there's a lot of things that you get out of internships and industry that you don't get in university courses. So at least in computer science the majority of students - or I guess a lot of students - apply for internships, especially their sophomore and junior years... And I was just kind of going through a bunch of common places where people applied to, and Google had one of the best internship programs out there. I just applied.
+
+\[07:56\] The way it worked - I actually didn't get in by just applying online; I was actually looking at a bunch of conferences in Google, and then a panel popped up and said "Hey, up for a challenge?" And that was the famed Foo bar challenge; I tried my hand at it, it was really fun. I was a little disappointed there was no option to use Go as a language, but I went through the challenges and then a recruiter reached out. That's how I got started with this internship.
+
+**Mat Ryer:** So you knew about Go before that then... How did you find out about Go?
+
+**Shaquille Que:** Yeah, I found out about Go in my first internship last year. When I was applying to a bunch of jobs, several of them mentioned Go. At the time it seemed interesting, it seemed pretty hip. I looked it up online, the website was pretty cool. I learned a little of the tutorial for Go, but I didn't really get anything outside of that, aside from the basic syntax. I learned more about Go during my first internship, where everything was really in Go. I got to do really cool things, building for example internal tools, and fixing bugs, learning about the software that they were building... And that's really my first experience with Go.
+
+**Mat Ryer:** Yeah, that sounds really interesting. It's funny that you mentioned that the challenge you did was fun. I think that's something that I remember being very important - and still is today - whenever I'm looking at a new technology; the technologies that I find the most fun and I have the most enjoyment with are the ones that tend to stick. Aditya, is that the same for you? I mean, you're in high school, so you're technically -- I mean, you don't have to be doing any programming at all, do you?
+
+**Aditya Prerepa:** No, I don't. But I guess the entire idea of making stuff has always enthralled me. Just to give you a bit of an example - the way I got introduced to Go was I was doing a lot of server programming, and I was doing it primarily in Java. I didn't like it much, but I didn't really know what alternatives were out there, so I was just doing some research and I came across Go. And it turns out that Go is really good for server programming, and I slowly learned it through example in projects, and stuff... And the one condition that I did these projects on was they were fun; I kept doing these, and I slowly got better, and I went from basic Hello, World to concurrency, and stuff. I appreciated it for what it was.
+
+**Mat Ryer:** So what sorts of things were you building?
+
+**Aditya Prerepa:** For example, I use gRPC a lot with Java, so I try to implement that with Go --
+
+**Mat Ryer:** What 15-year-old doesn't...?
+
+**Aditya Prerepa:** \[laughs\] Yeah... I tried to implement that with Go. I did a little server, and then I used the grpc-gateway, if you guys have heard of it. I tried doing the transcoding from JSON to protobuf... And just little things like those, that were in my mind really fun, because I could transform one thing into another... Yeah, just things like that.
+
+And then recently I learned a lot about concurrency through this project I did, this Cloudflare internship program, which was to provide a ping. You could choose from C, C++, Rust or Go. I chose Go, and it was a great choice.
+
+**Mat Ryer:** Yeah. Did you say that was Cloudflare?
+
+**Aditya Prerepa:** Yeah.
+
+**Mat Ryer:** I know that they do have a lot of Go at Cloudflare. Yeah, it's interesting we talk about the fun aspect... It is about kind of playing and creating, isn't it...? But are there any things that stand out so far as the bits that you're not as interested in? Are there any things that you've noticed that you think "Yeah, that kind of area isn't for me"? Aditya?
+
+**Aditya Prerepa:** So all the buzz is around AI and ML, and stuff. I have tried that... I guess I'm just not ready for it, but I didn't really get it that much. And then, low-level bit operations - I've tried to write C code, and I've tried to write low-level code stuff in Go, and I didn't really like it that much, because I guess I couldn't do much with it, if that makes sense. I could do more with building stuff like servers, and stuff like that.
+
+**Mat Ryer:** \[12:10\] Yeah, absolutely. Higher-level concepts. That's definitely true, and one of the nice things about the Go team - because that's where they spend all their time, is down in the guts of the language, and the compilers, and all that stuff. And they get to do all that work, and then we don't have to worry about it. We can kind of just use all their hard work to then build real things. How about you, Ben? Is there anything that stands out so far? You're already into your first position, right?
+
+**Benjamin Bryant:** Yeah. But by "stand out" - do you mean in terms of the language, or in terms of working with it?
+
+**Mat Ryer:** Yeah, sorry, I meant - does anything stand out in particular, an area that you're interested in? And then also, is there an area that you find not so interesting?
+
+**Benjamin Bryant:** Currently, at least on a professional level, I haven't had any experience to at least delve into areas in practice, if that makes sense... Right now I'm working within the payments area of tech, which is an area I didn't even know existed... But there's a whole area that is all dedicated to how users pay for things. I assume there are possibly a whole bunch of different areas for a whole bunch of other things that I don't even know right now, or haven't even touched.
+
+I suppose there is kind of a difference between trying something out and then seeing how it's used in reality. So I'm probably still open to just trying things out, if I've got the opportunity to do so. But right now there isn't really anything that I would say I wouldn't look into. I haven't looked into everything.
+
+**Mat Ryer:** Yeah. Well, I think that's a great attitude, actually, because especially when you're starting out - you're right, things are different; the different types of problems... You get to flex different muscles, and things. Can you tell us a little bit about what you are working on? Is it secret, or can you talk a little bit about some of the tech behind it?
+
+**Benjamin Bryant:** I don't know if it's secret, which means I don't know if I can talk about it. \[laughs\] Well, there's something that currently came out recently, and since it's out, I suppose I can talk about it... It was interacting with a different payment -- this is actually not that interesting...
+
+**Mat Ryer:** No, it is.
+
+**Benjamin Bryant:** I'll probably give an example...
+
+**Mat Ryer:** Oh, okay. Well, it sounds like we're talking about interfacing with a different system, and you're doing that a lot. I've definitely spent a lot of time in my career interfacing with another system... So it might be more interesting than you think.
+
+**Benjamin Bryant:** Well, it was just the idea of -- there are different payment service providers, and we have a little service that talks to these different payment service providers, and I worked on implementing another one.
+
+**Mat Ryer:** Oh, right.
+
+**Benjamin Bryant:** That's why I \[unintelligible 00:15:05.02\] too much attention... But I guess there was also this other project that is currently out, that I got to work on a tiny bit, and that was rounding up for charity, when someone tried to pay for something.
+
+**Mat Ryer:** So tell us how that works. What is that feature, first of all?
+
+**Benjamin Bryant:** Well, what that feature is is that whenever a customer wants to buy something, they now have the option to round up -- I'm in the U.K, so it's pennies to pounds; in the U.S. it's cents to dollars... So it's just rounding up to the nearest dollar or the nearest pound. And then we have to also keep track of these donations, obviously... Since if we just took someone's donations and didn't keep track of it properly, I guess that would be illegal.
+
+**Mat Ryer:** \[15:53\] Yeah, you can't just cast a float to an int and go home, yeah. Absolutely. You wanna keep track of that. Okay, cool. So what was the tech involved? What sort of things did you actually do as a team to make that happen?
+
+**Benjamin Bryant:** Well, the thing that I've found interesting around that project - at least what I've found in building something new - is that I got to experience the software architecture park... I think if you guys have done interviews, and stuff, there's always that software architecture interview, which you basically have to design how things are gonna go in put together... But a lot of the times it doesn't show up straight away. So being able to take part in the high-level part before that, which was scoping out how I would do this, scoping out changes to the database... We would do scoping out how things would link together and how they would talk to other services. I actually found that kind of interesting; I never really got a chance to do that before, and I'd mainly just previously been working on -- even in my previous position, I was working mainly on doing bug fixes... So being able to move on to the full designing a system - that's what I found quite interesting.
+
+**Mat Ryer:** Yeah. I mean, that is quite unusual, because often as a junior dev you do kind of end up getting just bug fixes. That is a very common thing to happen. So it's kind of awesome that you're getting exposed to different things. Shaquille, how about you? It's just your early days for you, but have you been working on anything so far, or have you worked on anything interesting you could tell us about?
+
+**Shaquille Que:** In terms of personal projects, usually my bar for working on a personal project is pretty high. It's really something that I have to be interested in, so I have to be invested in it. That way, I can see it through. The last thing I worked on was actually writing an interpreter in Go. I've just mainly followed along with a book - I think it was Thorsten Ball's book "Writing an interpreter in Go". I've found that really interesting.
+
+I guess \[unintelligible 00:18:03.04\] as Aditya here, where I kind of like low-level stuff. I wanna see how things work under the hood, as opposed to higher-level stuff.
+
+**Mat Ryer:** That's interesting - Thorsten Ball, his book about... He wrote a book of how to write an interpreter in Go, and there's a follow-up for compilers. And I always think for me the best way to learn something is actually by building something. Do you find that that is a good way for you also to learn? Since you're building and implementing a real thing, an interpreter, in this case...
+
+**Shaquille Que:** Absolutely. A lot of times I read up on things, and that gives me a higher-level understanding of how things work, in a way... But I don't really understand how they fit in together until we start going down and actually doing the things yourself. At first I was just browsing through the interpreter book, and going "Huh. Interesting. Cool, cool." And then I'd say "I'm not really learning much out of this", so then I started getting my hands dirty and actually following along with the book, writing the code, and then trying to write the code myself without looking at the book, and see how differently the implementation would come out... And that's when I began to understand really "Okay, so that's how an interpreter works."
+
+**Mat Ryer:** Yeah... I think that's great. That's always my one piece of advice when people ask "How do I learn more?" It's to build real things. Aditya, it sounds like that's similar to your approach, too; you spend a lot of time building real things, prototypes or actual things, right?
+
+**Aditya Prerepa:** Yeah. Do you want me to talk about some of the stuff/projects I've built?
+
+**Mat Ryer:** Please, yeah.
+
+**Aditya Prerepa:** \[19:50\] Okay. Well, they're not all in Go, but they're still pretty cool. One of them was a fingerprint-based attendance system for my school. Currently, students walk in and they all sit down, and then if one teacher messes up, they have to redo the entire thing. And so I thought "Teachers shouldn't have to be counting the number of students", and them saying "Here. Here. Here. Present. Present. Present." So I was thinking, everybody has a fingerprint; fingerprint sensors are like $5 now, and if you connect it to a Raspberry Pi and connect that to a server and a database, you have an attendance system. So I built that.
+
+I built a prototype and everything worked, but then Covid-19 happened, so then no school. There's two sides to that.
+
+**Mat Ryer:** Yeah.
+
+**Aditya Prerepa:** And currently, I'm working on this order management system for grocery stores to help with Covid-19. On the website you fill out the items you wanna get, and then when you send that, it sends an order to the backend, and then the backend sends an email, and then there's also an interface for this store to view the orders, and then mark orders as completed, and stuff. And the backend is completely in Go; Lambda, some pretty cool technology there... But yeah.
+
+**Mat Ryer:** They sound great. That thing, the Raspberry Pi and the fingerprint thing - whatever tech crosses into the real world, I think things can get very interesting, can't they? Obviously, we're not allowed to touch things now, so that's ruined that, but... Yeah, okay. Brilliant.
+
+What were the challenges around doing that? What was the hard bits when you were building those projects?
+
+**Aditya Prerepa:** So I guess for the fingerprint stuff I had to go through the bureaucratic hassle of talking with the district technology representative person, and then I had to get an affirmative from him. Only then I could start collecting fingerprints... Because you could extract these fingerprints, and we're all minors, so it's sort of an invasion of privacy if it's not dealt with right... So yeah, we had to deal with that; that was the bad part about it.
+
+Like I said before, I don't really like to interact with really low-level stuff, but this fingerprint sensor didn't have great interfaces, so I had to do some bit-twiddling, and stuff... And that wasn't fun. But yeah, you do what you have to do.
+
+**Mat Ryer:** That's really interesting.
+
+**Break:** \[22:30\]
+
+**Mat Ryer:** It's funny that the politics almost in that was the challenging bit... And I wonder, is there anything that stands out as kind of intimidating or concerning? Is there anything that stands out like that when you look at tech? Or is it all just kind of new and exciting, and you're not too worried about anything?
+
+**Aditya Prerepa:** I will try not to think about that... Because if I think about all the overhead that I'm gonna get, I probably won't do the project. So I try not to think about that stuff.
+
+**Mat Ryer:** What about you, Ben?
+
+**Benjamin Bryant:** What I find to be a challenge sometimes is -- I don't know if this is a word, but self-propellence.
+
+**Mat Ryer:** Hm. What do you mean?
+
+**Benjamin Bryant:** \[24:07\] I don't know if everyone else feels this as well, but sometimes when you need to teach yourself something - let's just take a random example, like if I needed to teach myself Docker. You go to the website and then you have to read -- you do need to read; I mean, I guess you could watch some videos as well... But you kind of need to get your hands dirty and you need to start working on stuff. And I find that sometimes without some kind of catalyst, it can be difficult to want to continue with a project, or to see it through to the end.
+
+One of my first Go projects that I worked on as a sort of joke is that I wanted to model the life of a million gophers (as in the animal) in Go, just as a way to learn how to use Go, because I'd never really used it before. And if I was just doing that on my own, if I was just trying to learn Go on my own, I don't know if I could have pushed through and gone and tried to learn through everything... But because I had volunteered to try and give a talk about me learning Go, that had a fake deadline involved in it. And having a deadline there was like "Okay, I need to keep moving myself forward." I think that could be something that you could kind of use; if you can somehow find a way to maybe just impose a restriction onto what you're working on, it might drive you to work for it, essentially.
+
+**Mat Ryer:** Yeah, that is interesting. There's another thing that happens as well, which is we tend to want everything to be perfect. So when you're building something, you just want it to be the best it can be... And sometimes a deadline helps with that too, because it forces you to prioritize. You mentioned giving a talk... I saw that talk, because you did it at the London Gophers, right?
+
+**Benjamin Bryant:** Yes.
+
+**Mat Ryer:** Yeah, and it was great. I remember the audience loved it. You were kind of talking as well about your story of learning this, and this way that you went about that. So did that talk - it was a catalyst for you to actually do and deliver... But was there also value in that, in giving the talk as well? And was that a kind of scary thing to be doing?
+
+**Benjamin Bryant:** I found at least doing the talk kind of helped me get more engaged with the community, since after the fact, since people had seen the talk and they were willing to -- I guess it felt easier walking up to people and talking to them, which could help with at least unlocking a new language, or unlocking different places that you might find and think about applying or to go to.
+
+Yes, I guess it helped with me wanting to program something, because I knew that at the end of it I would be presenting it... So that kind of drove me to at least want to do a good(ish) job on what I was doing.
+
+**Mat Ryer:** Yeah, absolutely. And Shaquille, how about you? Is there anything that stands out as a kind of intimidating or worrying area in particular, as you look forward in your career?
+
+**Shaquille Que:** As I look forward in my career, I would say the most concerning right now is graduating in a recession. I hear a lot of stories from back in 2008, or even back in the 1980's, when it was very difficult to find jobs and when a lot of people were losing their jobs... And I think it's also already kind of happening right now, where a lot of companies are laying off and a lot of companies are also going under.
+
+\[28:02\] I know that in the future the tech industry will still keep growing, and even right now, a lot of the big companies like Facebook and Amazon are still aggressively hiring... But it's still scary to think that right now it's gonna become harder to find jobs, and graduate in a \[unintelligible 00:28:19.29\]
+
+**Mat Ryer:** Yeah, I think a lot of people share those kinds of concerns. Tech isn't immune, of course, but in some ways -- look at remote work; that's kind of an example of something that I think the world has changed with remote work now. The prevailing attitude was that remote work wasn't right for every company, and people were very used to being collocated... And since we're now all forced to be working remotely, people have seen "This actually works", and there can be some benefits to it, as well.
+
+So yeah, I think, Shaquille, you're right - there's gonna be things changing, and hopefully there is still enough opportunity within that to be kind of getting stuck into. Shaquille, have you done any talks, meetups or anything? Or would you consider something like that?
+
+**Shaquille Que:** I have not given a talk, if that's what the question is... But I have attended some meetups; I've gone to a couple Austin Go meetups, and that was pretty interesting. This was around last year, last summer. At the time I was a freshman student, and I didn't really understand a lot of what people were saying, and a lot of tradeoffs they were discussing were just not things that I'd run into at all yet, at least in my career... And now I think I have a little more understanding of that, just from looking around in Reddit threads and Hacker News posts... Which are pretty interesting, by the way.
+
+But yeah, I'd say those meetups - a lot of those meetups can be intimidating at times, because if you come in and you don't know anyone, it's pretty hard to interact with people. Also, when people start talking about what they're doing, and all this really cool tech stuff and you don't really understand what they are... It's just kind of hard to "Hm. I see..." or just keep asking "Can you tell me more about that?", because then you're gonna take up all their time.
+
+**Mat Ryer:** Is that what you did though? Would you ask people questions?
+
+**Shaquille Que:** I'd ask them to a certain level, like maybe four or five questions in, and then I'd stop right there, because I'm sure they wanna talk to other people more at that point.
+
+**Benjamin Bryant:** From my experience, I think people like talking about stuff that they're interested in... Because when I went to these meetups as well, as I mentioned earlier, I didn't know what Docker was, or anything like that... So similar to you, a lot of this stuff went over my head. But I'd still go to a pub afterwards and I'd still ask questions... And especially if you can somehow find someone who's working on something - if you simply ask why a couple of times, or even just ask some questions for them, they will continue talking for probably longer than you'd want them to anyway, because they're just that interested in what they're doing. So I wouldn't worry too much about that.
+
+I also found that meetups were a great place to unlocking language... So much stuff that's mentioned there is just stuff that's kind of internally known, and is very difficult to find entirely on your own. But at least if you go there, you can pick up the words you need to google afterwards.
+
+**Mat Ryer:** Oh yeah, that's really interesting. Do you find the same, Aditya? Do you find yourself in these communities and there are concepts and things that you don't seem familiar with, and it gives you something to go and look into?
+
+**Aditya Prerepa:** \[32:05\] Oh, yeah. I can talk a little bit about my open source -- like, trying to open source... Because I kind of knew nothing about -- like Benjamin, I knew nothing about Kubernetes, Docker, any of this stuff, but I found out the best way I learn was through experience... So I decided to pick a repository on GitHub, and you go and just jump into it. My first attempt was gRPC, and I knew nothing about a lot of things... But I've found out that I couldn't do much there, so I decided to go to something else. I've found that in Istio, where I'm currently contributing, where I've sort of found my niche, everybody there is super-nice and supportive... And I sort of like bug people on Slack about certain things, and they're so nice about it, and they always answer. Yeah, just learning through people, and experience, and reading stuff. That's how I sort of get myself to where I wanna be.
+
+**Mat Ryer:** Hm... That's interesting you say "bug people", and it's a shame that it feels like that... But I think there is a bit of that. You have to sort of be a little bit like that. In fact, Aditya, you reached out to me to pitch coming on a podcast, for example; so it's that very proactive, quite a bold step. It's not without risk; those kinds of things can be quite embarrassing if they don't work, so it's a risk you take... But if you can, I think it's a risk worth taking. Now we get to meet you and hang out on the podcast, so in that case it worked, so that's great.
+
+Are there other people your age and in your school that are doing the same kinds of things in tech as you, or are you sort of unusual there?
+
+**Aditya Prerepa:** So I don't wanna be unusual, but... \[laughs\] I have a friend, he's sort of the opposite of me. He loves doing little bit twiddling, and he loves doing low-level stuff. He's super-smart. But yeah, he's the only other friend I know who actually is super-interested in this stuff... But it's not to say -- like, I was just lucky to have found something that I really love this early. People will find their things eventually; it's no rush.
+
+**Mat Ryer:** Yeah. Very wise words, annoyingly... \[laughs\] You're right. And I mean "unusual" in the best sense. I think you do sometimes - and I don't know what it's like now, but I know that when I was... Because I was kind of a programmer at school, and it was strange to be doing that; sometimes you kind of can get a bit outcasted, things like this can happen, too. But what's the technical learning there like? Do you have a computer science class, and what's that like?
+
+**Aditya Prerepa:** Okay, so I go to school in the Bay Area. The Bay Area's high schools and stuff are notoriously known for being really competitive... And yeah, so we have a computer science course, but it's only available to 11th and 12th graders. I actually met the computer science teacher at my school; his name is Dr. Brocker, and I met him through a math class. I was taking geometry in ninth grade, and he taught that class for one period. And I met him through there... He saw I was interested in this stuff, but he sort of took me under his wing and taught me a lot of that stuff. He's sort of like a guru at Java, and stuff... And he really helped me throughout that process. But I'm really lucky to have that sort of person; a lot of people don't. I think having a mentor is what stops a lot of younger people from going out there and doing things. We do have a curriculum - it's sort of like algorithmically-oriented. Not really application-oriented, but yeah.
+
+**Mat Ryer:** \[35:58\] Yeah, so speaking of mentors then... Shaquille - have you experienced that, where you've had some mentorship from somebody?
+
+**Shaquille Que:** So I started in high school. I'm actually an international student; I think I forgot to mention that. I'm actually just here on an F-1 visa. I'm from the Philippines. Back in high school I started learning programming by myself, using an online curriculum from Stanford... And at the time, I really wanted to expand more in it, but I really couldn't find anyone at my school who either enjoyed programming as much as I did -- well, there were people in robotics, but it's a little different, because they're programming robots to move and do stuff, and I was more interested in using programming to solve problems.
+
+Yeah, so we didn't really have a good CS curriculum or program at my school, and it was my understanding that a lot of other schools as well didn't have that, unless they're top of the top schools... So I'd say the closest I have to a mentor is really the manager I had from last year, from my internship last year. We're still in touch now; from time to time we check up on each other, just giving us updates on what's happening in my life, what I'm learning, books he's reading, and some suggestions for things to dive into. But yeah, a lot of it is really just me talking to my friends about cool tech, and then them sharing what they've found, and then we sort of dive into those together, or go into discussions about those together.
+
+**Mat Ryer:** Yeah, that's really great, that you were able to find somebody like that, I think. And it's also funny, I noticed this theme of being really hobbyist, that then we turn into job, and definitely my route was that. I just couldn't believe I could make the computer do things. That was just an amazing feeling, and that never went away.
+
+So Ben, did you ever have anybody mentoring you in this way
+
+**Benjamin Bryant:** I feel if I say no, people who did might get offended... But I don't know--
+
+**Mat Ryer:** I see. So less formal, but obviously there's been people that have helped along the way, of course.
+
+**Benjamin Bryant:** Yeah, as in I've talked to many people about my issues through life; not just about programming issues, just issues in general... What kind of milk do I wanna drink, that kind of stuff.
+
+**Mat Ryer:** Yeah. What sort of milk did you land on in the end?
+
+**Benjamin Bryant:** I don't drink milk anymore.
+
+**Mat Ryer:** Okay then.
+
+**Benjamin Bryant:** It's a sad tale... \[laughter\] I think whole is good, because it has fat in it, and that's good for muscle growth... Anyway.
+
+**Mat Ryer:** There's oat milk... Other kinds of milk are available.
+
+**Benjamin Bryant:** Oh, you mean like milk that doesn't come from cows?
+
+**Mat Ryer:** I don't know.
+
+**Benjamin Bryant:** I guess that doesn't make sense... I guess programming is a bit like milk... \[laughs\]
+
+**Mat Ryer:** Programming is a bit like milk.
+
+**Shaquille Que:** Save it for the after-party. \[laughter\]
+
+**Mat Ryer:** We know how to party on this show... I was gonna say, save it for our sister podcast, Milk Time... It's not a real podcast, so don't look it up.
+
+**Benjamin Bryant:** I guess to answer your question - because I noticed you guys were talking about how you guys were sort of hobbyists and kind of went into it... But my route was slightly different; although I did do computer science at university, I don't think I opened the terminal on my computer and... Well, after all I was using a Windows computer, instead of Linux or Mac or something like that, which I guess all the hackers use (I said hacker/programmer).
+
+I didn't open a terminal till I think after my first class at university... So I came into programming I guess a tiny bit late. And I also remember avoiding -- because it was Java at university, and I remember avoiding installing it on my computer because I found it too complicated to figure out how to install and then run it... So I just used the university's computers to do my coursework.
+
+But then eventually, when I needed to actually work at home, I kind of had to sit down and figure out "Okay, how do I install Java, so I could actually get to work?" So a slightly different path...
+
+**Mat Ryer:** \[40:18\] Was that something that attracted you about Go then? The fact that actually it is quite easy to install and to build things. You don't have the JVM, right? You don't have these virtual machines... It is a simpler thing, is it not?
+
+**Benjamin Bryant:** Well, at that point I had been in university for four years, so I kind of got over my fear of installing programming languages... I guess some people at university actually avoid installing programming languages all the way to the final year. \[unintelligible 00:40:46.00\] in the first year. But yeah, it's actually pretty easy to install Go. I think at the time when I started using Go, there was this GOPATH thing. That wasn't easy... *go mod* made it slightly easier \[unintelligible 00:40:54.16\] I was like "I've messed up."
+
+**Shaquille Que:** Go modules were the biggest struggle I had last year. I remember just being super-frustrated, trying to download dependencies for Go modules, and not finding a definitive resource out there that tells me "Oh, this is how you start a Go module. This is how you use it." But now, with the official support, it's a lot easier, and pretty much every new tutorial out there can tell you how Go modules work. But that last year, I had to ask people a lot of questions to really understand how do you use Go modules...
+
+**Mat Ryer:** Yeah, me too, mate. And actually, it's the transition from GOPATH... GOPATH in the beginning was, in a way, kind of this minimalist idea, which is just this one path, and your code goes in there, and it follows the folder structure that also follows the go get kind of structure.
+
+For a while, it really kind of was quite simple, and then it didn't quite scale, of course... So modules comes along, and we have to transition. So that's why - yeah, all the articles that just assume GOPATH -- it's just "Setup your GOPATH, and just link off to some thing." Suddenly, it isn't as simple as that... And I think that happens sometimes, doesn't it? In communities, when the tech evolves and things start to change, it does create some tension.
+
+What about you, Aditya? Was modules a tricky thing, or did it make sense to you?
+
+**Aditya Prerepa:** Coming from a very JVM-heavy ecosystem, it baffled me, but I sort of got used to it through exposure. Yeah, GOPATH was sort of -- I got used to it, and then Go modules came out and it streamlined everything else, and now I just do go mod in it and I'm done, yeah.
+
+**Mat Ryer:** Right. It does keep getting better. The Go modules implementation really does keep getting better and better... And I think in the next release we don't need to be doing go mod tidy, and things... I think it's sort of self-healing almost; it tidies up after itself. We'll probably end up doing an episode on that.
+
+What other features of Go then do we like? This is a Go podcast... What's your favorite Go feature? Aditya?
+
+**Aditya Prerepa:** I really like select and choosing from channels. I don't like saying stuff like "Code is beautiful", but that is beautiful. I just have multiple channels... Okay, so when I was writing the pinger for the Cloudflare internship, I had multiple channels. I had an interval time ticker, I would send the packet on a different goroutine, and then send the packet through the channel... And I could just select from all those options, and then I could just handle them in my own way... It just made things so easy, as opposed to Java, where you'd have a thread pulled in, and then you would do all this stuff.
+
+Another thing I like about Go is you don't have to do a factory factory, like you would do in Java. It's just really nice and really simple.
+
+**Mat Ryer:** I think that's a great choice. For concurrent programming, which is usually difficult to keep in sync and difficult to communicate and share memory and all this stuff... And yeah, I agree, that select block - only one of these cases can run at a time, so it gives you that synchronization point... Yeah, it can conceptually simplify things a lot. I think that's a great choice. What about you, Ben? What's your favorite bit of Go, mate?
+
+**Benjamin Bryant:** \[44:21\] I would say the standardization of style - I think that's what it's called - in the sense that I've been using Ruby at my job as well, and that definitely has a lot more ways to do a lot of the same thing, and that can make it difficult to track things down. But having everything look the same and feel the same, I kind of appreciate that. It also kind of avoids the situation where you have to figure out what the team's code style is gonna be, and stuff like that.
+
+I remember there was a university project I had to do, and there was a heated debate about how many spaces we were gonna do, and if we'd even dare to use tabs... And that was just one small part of the kind of code style we were trying to get going. I just like how just everything already has a style. It's a small thing, but it's very useful.
+
+**Mat Ryer:** No, I completely agree. The Go project shipped in the very beginning with the go fmt tool, which did all the formatting. And you know, if you think about working on open source projects, or actually collaborating in any way, it's quite important those things, especially when we're dealing with pull requests, and things... You don't want to see pull requests with just lots of formatting noise in there. What's nice is as long as everybody go fmts their code before they push it, you really only see what's actually meaningful that's changed; you don't see "Oh, this change, but also, look, my editors put all these tabs as well", so it's just red and green everywhere. So yeah, I think that is actually a pretty good choice as well, mate.
+
+Shaquille, have you got a particular favorite language feature?
+
+**Shaquille Que:** For me it's goroutines and channels, and sort of the idiomatic way to use concurrency in Go. I think the patterns that they want you to use for concurrency is very good in terms of how you can avoid a lot of the common pitfalls with parallel programming... In particular, I just finished a class last semester on parallel programming and different patterns, and I find myself translating a lot of those patterns into how they would work in Go, and thinking "Hm. That gives me a better model of how this pattern really works." And kind of contrasting it with Go, I can see why people encourage you to use channels, rather than for example giving new text everywhere.
+
+**Mat Ryer:** Yeah. I think that's a great choice. The thing that attracts a lot of people to Go is the way that it handles concurrency, and treats it as a first-class concern. In fact, it was designed for modern architectures. If you think about C, it's kind of an older language, designed back in the '70s, when machines were just single-processor... It wasn't really networked in the same way. And modern systems are very networked, and multi-core, lots of processing speed and concurrent processing capabilities, and things... So the language kind of knew that when they designed it, so you do get that into the language. So again, great choice.
+
+**Break:** \[47:43\]
+
+**Mat Ryer:** Do you have any questions that we might be able to answer? And by "we" I mean me and also anyone else on the panel that might have an answer?
+
+**Benjamin Bryant:** Yeah, I have a sort of different question now.
+
+**Mat Ryer:** Go for it. Who doesn't like a surprise...? Is this gonna be about milk?
+
+**Benjamin Bryant:** No.
+
+**Mat Ryer:** Okay.
+
+**Benjamin Bryant:** But it ages like it... No, that's a terrible joke. Anyway. \[laughter\] So we're on different levels, in the sense that Aditya - I could be assuming for you, but eventually you'll be heading towards university potentially, right? And Shaquille, you've currently completed university and you're like an intern - well, close to completing university, you're an intern, and then after this you're most likely gonna graduate and get a position.
+
+**Shaquille Que:** I'm actually halfway through university.
+
+**Benjamin Bryant:** Oh, well you're almost there.
+
+**Shaquille Que:** \[laughs\] It's a long way off...
+
+**Benjamin Bryant:** Oh, it's not that long, trust me... And I've graduated, I'm in my first Go position, and eventually I will most likely look for maybe a new place to go, like maybe trying to go towards a more senior level, or so on and so forth. So I guess my question is -- because we've kind of all been working towards something... I guess once you get past my stage, you kind of have to decide for yourself where you wanna go, and how you want to choose how to learn, or to keep learning, or something like that. So I guess my question for you, Mat, is "What makes you wanna keep going?"
+
+**Mat Ryer:** Yeah, that is a really good question, which is code for "difficult to answer."
+
+**Benjamin Bryant:** Apologies...
+
+**Mat Ryer:** It's alright. I think it's just the love of doing it. I think it is that, ultimately. And to be honest, I've done jobs -- I've taken jobs before for the money. I took a job I thought "It's not gonna be the best job, but it's really well paid, and it's a step up", and all this... I always regretted those decisions. I was never happy. You very quickly get used to the money. Whatever money you make, you very quickly get used to it, and it stops being the incentive that it seems like it's gonna be, and so therefore it has to be the love of the thing, the belief in the project that keeps you going.
+
+Essentially, as you've all kind of given examples of, when you see a real problem that you want to solve, that's the drive that keeps you. But yeah, I think everyone also is gonna have a different path. I don't think there is -- I mean, you're right, Ben, initially you've got maybe high school, and a lot of that early stuff is kind of laid out in some ways; there's lots of room to do other things. I don't know that most people go on a podcast when they're 15, for example.
+
+So there's lots of ways you can break the rules or bend the rules and do different and more interesting things, and take different paths... But yeah, I think there isn't a path -- I don't like the idea, actually, that there is a single path that you follow to progress. If there is one -- some companies do it; they have explicit little programs (or big programs), which you can flow through. But I feel like it has to be so tailor-made to you for it to be valuable to you, that those things may or may not fit very well. I've never found anything like that really work, but if there's benefit to it, then that's what you do.
+
+\[52:15\] But yeah, I would say there isn't really a right or wrong path that you go through. It has to be a kind of thing that you find your own way on. I wouldn't recommend necessarily the way I have gone through my career. It's something that has kind of worked to an extent, I guess, and it's been somewhat random when I look back on it; it's somewhat random, it feels. I didn't have a big five-year plan, for example. I never really did that. I was much more agile about it, if you like; I was much more immediate and looking, that sort of approach.
+
+I know a lot of people do do that - they have the big plan, and stick to it, and it really works for them. So I'd say there isn't an answer; there isn't a route that everyone should take. You have to find your own path. How does that sound?
+
+**Benjamin Bryant:** It sounds good.
+
+**Shaquille Que:** On that note, we did discuss a lot on this podcast about hobbies and doing things that are interesting... So all the times I look back to, when I first learned programming, or when I first learned a new language, that sort of childlike curiosity and excitement you get out of doing something simple... And then slowly, as you start to learn more languages, programs and patterns, that sort of childlike wonder seems to be more muted and fades down a little, because a lot of these are things that you've seen before, in just a different form... So I'm curious, Mat, do you still have that sense of childlike wonder and curiosity when you dive into different things now? And if so, how do you keep that alive throughout the years?
+
+**Mat Ryer:** That's a great question. Actually, the answer is yes, I do still have it. And what happened to me was the things that I learned, and that just became kind of tools that I could use - in the beginning they were exciting things I didn't understand. And then once you understand them and they do somewhat become mundane, actually they just become tools. There's always more things, and that's the nice message I can tell you from the future, if you imagine I'm in the future. I can't tell you any lottery numbers, it's against the time-traveling laws, but...
+
+**Benjamin Bryant:** You can't?!
+
+**Mat Ryer:** No, sorry... But I can tell you that there's always stuff, new things to learn, things that are hard, and therefore enough to get stuck into. And there's lots of other languages to learn... But really, once you've along the way collected up these tools that you can use, the exciting thing then is mixing all that knowledge in new ways.
+
+Somebody I read said that creativity actually is just making connections that other people haven't made yet, between things that already exist. So in that sense, you might know about a messaging system, and you might know HTML, so you could build a web frontend; you might have these various tools, which in their own right, if you've done them for long enough, become not very exciting or cool. Don't have that sparkle to them. But then you can use them to mix them together in new ways that other people haven't done before, and tackle problems -- and even tackle problems that have already been solved, in different ways, in new ways. So yeah, it never runs dry, there's always more to learn, and I think that's the good news I can give you from the future.
+
+**Shaquille Que:** That's very wise. It sounds good. It sounds like there's a lot to look forward to.
+
+**Mat Ryer:** Yes, there is. There definitely is. Any other questions?
+
+**Aditya Prerepa:*** \[56:01\] Yes. So last year, at my Life360 internship, I was like -- there was this one senior engineer who would sit perpendicular to me... And we were both working on the same sort of area, so he knew a lot of stuff... And I would go to him a lot for help; after a certain point, I felt like I was being intrusive, so I stopped going for help that much, and that sort of slowed me down. I'm just wondering, how much time should be spent trying to figure out a problem before bugging a senior dev with that problem?
+
+**Mat Ryer:** I think that is a good question, which again, isn't gonna have a very simple answer... Because it'll just depend so much on the situation. But the fact that you're even thinking like that I think is the good thing, because it's about -- you know, you do have the respect for other people's time, and that's important. But I think it just depends on the situation. And having that as a sort of open conversation and asking people about it - I think that's the way to do it. "Do you mind if I come and bug you? Would you prefer I just did it at certain times? What works for you?" Those kinds of things. I think you'll find most of the time people like helping.
+
+In some cultures there's a lot of time pressure, which can spoil these kinds of things... Because of course, if somebody's busy and their neck's on the line, if that's how they're feeling, then of course they're gonna just be focused on their work and they're not gonna have time for other things, and it might feel like you're distracting them from that mission that they have. I think the culture is what's wrong there; it's not a very healthy culture. It does happen sometimes, and I think having that open dialogue is how you'll find out.
+
+Someone will say "Oh, I'm sorry, I've got to get this thing done for Friday, so I'm just gonna knuckle down and get on with it." But I think generally, most of the time, assuming a nice, somewhat more relaxed culture, one where people aren't under too much pressure - I think people would be more than happy to help... And as Ben said, sometimes you have the opposite problem - you want them to shut up.
+
+**Benjamin Bryant:** I did want to add to that question as well, since I actually was in a similar situation where I worked... And it was kind of tricky, because there were two people next to me. One to the right, that you could basically bug any time, and they were fine with it, and one to the left that you couldn't. So I originally was using the philosophy for the right person on the left person, and then it came up in a performance review. \[laughs\]
+
+What's important is you can sort of be proactive in just figuring out how people like to be worked with... And similar to what Mat said, simply just asking or setting up a little one-to-one meeting and asking "How would you like to be interacted with?" Or even from your perspective, like "I'm new here. I have quite a few questions to ask. How often are you okay with me asking?", just seeing how they want to do it. And then it's kind of on them to say how they like to be worked with, and at least then you know.
+
+**Mat Ryer:** Yeah, I think that's nice. Keep revisiting that conversation as well, because often people won't know the answer, and they might say "Oh yeah, interrupt me any time", and then once you've done it for a few weeks, then they might know more about that. That happens all the time, too; everyone's always learning. So revisit it occasionally. And I actually quite like it when this is done in a somewhat informal setting. I don't know what that's gonna be like now, when we're gonna be a lot more remote... But I would go for a coffee with somebody, and at those times chat about things like that.
+
+I think that's right, it will change over time, and it'll be different at different times. So acknowledging that can only help, really. It doesn't become a problem; it never boils up into -- it certainly shouldn't boil up into a performance review... But it can happen, I guess. So yeah, definitely talking I think is just the key thing. Talking to people and being personable. And actually, I can tell -- I mean, Ben, I know you a bit more, because we're both in London... How many times have you spoken at the London Gophers Meetup, by the way?
+
+**Benjamin Bryant:** Only twice.
+
+**Mat Ryer:** Okay, but I've been to both of your talks, in fact, so I do feel like I know you better... But I can tell you're all personable, and I think you're thinking about these things in the right way. So yeah, I think you're gonna do great. Don't worry too much, but yeah, having that conversation with people is gonna be important. It's time for Unpopular Opinions!
+
+**Jingle:** \[01:00:49.01\]
+
+**Mat Ryer:** Okay, does anyone have an unpopular opinion?
+
+**Aditya Prerepa:** Yes. It's sort of similar -- I listen to Go Time a lot, so it's sort of similar to one of when the NATS.io dude came on. I think that REST is obsolete, and I'm not sure why we're still using it. I know there's legacy involved, and it's easy, but there's much better protocols out there that we could use, that are faster, better, having IDL... This is just my opinion, but -- this is actually sort of a question, too. For client-facing apps, where you have to integrate with lots of third-parties, do you always have to use REST? Because I was using gRPC for that a lot, and it sort of worked. I wish we could use gRPC for everything. What do you guys thing?
+
+**Mat Ryer:** Ben, what do you think?
+
+**Benjamin Bryant:** gRPC is a new word for me right now, but I'll definitely look it up after this.
+
+**Mat Ryer:** There we go. We learn. Shaquille, have you encountered gRPC?
+
+**Shaquille Que:** I have. I was researching on whether to use REST or gRPC for a server I was gonna make... And the idea I seemed to get from it is that gRPC is really fast, and -- I actually don't recall the exact arguments, but it seemed to be that REST still has a place today, and it works for simple applications.
+
+**Mat Ryer:** Yeah, it's funny - I think one of the big things is browser support. Now, you can do gRPC in the browser; it's not as easy as RESTful services. And I think yeah, the legacy, Aditya, that you talk about, of the fact that we've had REST for a long time - that is probably what carries it.
+
+If you sat down and looked at them both, and just did a sort of analysis and a comparison, gRPC is a binary protocol. You use protobuf to describe the messages... Essentially, you describe a struct, basically, if you think of it in Go terms. But one of the cool things it has is you put a number for each field, and once you've used that number, you never will change it. You can delete them and not use it, but you can't ever reuse that number. So that's sort of future-proofs API design; old messages with still work in the future.
+
+There's a few other benefits, and things... It's a binary protocol where REST tends to be text, whether it's JSON or something... And yeah, you can also get some very fast serialization down at gRPC. Typically, I've seen people use it in systems, if you've got microservices and things... They'll use something like gRPC and sometimes protobuf on its own for the messaging there... But when it comes to building a web experience, there isn't yet the bit that reaches into the browser enough, I think, for that adoption to happen. But there's no reason why that won't happen.
+
+\[01:04:07.23\] So yeah, we could well see something like gRPC -- or there are other kind of binary protocols... But we could easily see that take over and become the one. You'll notice, if you go to a Stripe API, or Amazon Web Services, or these kinds of APIs - they tend to be still following the patterns of the web.
+
+REST APIs essentially follow the web's design as well. Rules like get requests will get information, they'll never change information. A get request doesn't have a body, so in a sense you can't really send data with a get request... Versus a post request, where you actually post a body with some data. I think deletes also don't have a body... And REST kind of fits that pattern of how the web was originally designed. That's why it's about getting documents, putting documents, making updates to resources, and it's done like that..
+
+RPC things, which gRPC is one, of course - not of course, but it is - is more about thinking of like... There's actually a procedure or a function I'm gonna call over there. And that's a very different way of thinking about things.
+
+So I think it's interesting - I think web protocols and REST still is by far the biggest player, but it doesn't mean it's better technically. It is probably easier to interface with, so that's why it might be winning.
+
+**Aditya Prerepa:** One last thing... When I see microservices that use REST to communicate, it baffles me... Because let's say a company wants to move to a faster RPC model; one of the reasons they might not do that is you will have to migrate all of your microservices from REST to gRPC, and there's gonna be so many things that could go wrong there. But there are services that I'm not sure -- maybe I'm just not in the loop, but I'm not sure why more entities aren't utilizing it; there's stuff like Envoy, and gRPC Gateway, which transcribes REST into JSON, and JSON into REST. And that actually -- you know, it supports REST while you're getting a taste of what gRPC is. So that's an interesting topic - adoption, players in adoption, like Envoy and gRPC Gateway are super-useful when dealing with that stuff.
+
+**Mat Ryer:** Yeah, absolutely. I think it's just a case of inertia. Here's the thing... If you get to a project and you're responsible for choosing which technology you're gonna use, you could sit down and just compare the two technologies and very reasonably come to the conclusion that "Well, gRPC is the one." But if that team has never heard of gRPC before, then that is quite a big thing, that's quite a big consideration that you have to make when choosing tech.
+
+So yes, you might compare the tech and land on an obvious answer, but what do people -- if you expect them to consume your API, are you gonna provide a client that they can use? And if you are, then you have to provide a client for every kind of platform that you want to support. You'd need to do an iOS, you'd need to do one for JavaScript, you'd probably need to do a Go one... And maybe that's fine, maybe that's what you're gonna do... And actually, a lot of the Google services do that; when you write code for App Engine and you use the Google Cloud APIs, they are all gRPC. The g in gRPC is short for Google, so that's probably why...
+
+But if instead you want to just provide a surface and let people write to that surface, then it's what the world wants, it's what the customers are gonna be happy consuming. It plays a big part in that, and I think that's why we still do see that technology - REST technologies, and just JSON, HTTP, even though yeah, you could easily argue... I mean, it's a text protocol; you have to encode and decode JSON, which can be slow compared to other technologies.
+
+\[01:08:17.09\] And the other thing is you might just not need the speed of gRPC. It's very common, as you learn about the tech -- it's easy to sort of look at and optimize for a particular metric, like "Let's just go for the smallest package size, because we're gonna be doing messages and we want them to be tying in messages." Of course we do. But if the cost of that is that no one's gonna then consume the API because they don't know how to do it, or you have to write and maintain a client library for them - maybe you might just think "You know what - we'll just do a JSON thing, because in this particular case we may get 1,000 users doing it, and it's not gonna make much difference." But yeah, it probably will change. It almost certainly will change. And it does keep getting better all the time.
+
+The gRPC Gateway thing is interesting. As you mentioned, proxies essentially lets you do -- I think it spins up the gRPC server and it spins up a JSON server... So depending on how you're deploying code, that might not work, too. You might not be able to do things like that. I'm thinking of how App Engine used to work in Google Cloud - it was a locked down kind of environment, so you weren't able to just open up any ports you want; it was very controlled. It's less so the case now, but there are definitely considerations like that too about deployment, and things... But yeah, I think it's a great one, and we should probably have another episode on it at some point, and we'll invite you back.
+
+**Aditya Prerepa:** I completely agree. One last thing - gRPC, to actually sort of move it away from a Google product, they actually renamed it to gRPC Remote Procedural Call, so it's sort of a recursive acronym.
+
+**Mat Ryer:** Right, right. So the g stands for gRPC.
+
+**Aditya Prerepa:** Yeah.
+
+**Mat Ryer:** Okay. \[laughs\] That's great. That's a paradox. It's a bit like how Go is written in Go, the compiler is written in Go. It didn't use to be. Any other unpopular opinions?
+
+**Benjamin Bryant:** I guess I have a really short one, since this is -- I guess we're nearing the end of the podcast... And this part could also be cut out... \[laughter\]
+
+**Mat Ryer:** This is definitely going in now, Ben...
+
+**Benjamin Bryant:** Okay... Well, it's very short, and there's not much to debate. London Gophers is the best Go meetup.
+
+**Mat Ryer:** Oh, that is a good one. It is a great meetup. Are you an organizer, by the way, now?
+
+**Benjamin Bryant:** I might be, but that doesn't affect my opinion.
+
+**Mat Ryer:** Okay, interesting. Well, hopefully you have meetups in your local areas, too. The London Gophers one now is online, so you don't even actually have to be in London. Anyone from around the world can join in. I think more and more of them are going online too, but it is a great meetup; I have to admit that. It is one of my favorites.
+
+Shaquille, have you got anything unpopular to say? So far you've just been saying popular things.
+
+**Shaquille Que:** Yeah. I'd say updating Go versions is a terrible experience. It's not as terrible as other languages, for example, but the fact that you have to do a lot of things manually - go through a website, download it, delete the old one, and all of that, where we have a lot of Go tools that do other things... I think that's kind of sad.
+
+**Mat Ryer:** Interesting. There is an installer which will just do it for you. It probably depends on how you installed it in the first place, I guess.
+
+**Shaquille Que:** I was talking more about updating versions. You have to download the latest version's installer from the website first, and delete/blow away your old installation... Or is there a new way to do it now?
+
+**Mat Ryer:** There might be, yeah. There's an installer. You still have to download it, I'm sorry... I don't think we can solve that one for you. But yeah, it will do that work for you, so you might be in luck there. We might be able to solve that one for you.
+
+**Shaquille Que:** Cool. I need to look more into it.
+
+**Mat Ryer:** Well, it's been an absolute pleasure. Thank you so much for spending this time with us. I think you've all got really great futures; please stay in touch... We'll see you hopefully online, we'll see you in meetups, speaking, and we look forward to seeing how things progress. Thanks so much.
+
+How about we have you on again next year and see where you are? We can check in... It could be really cool. Thank you again - thank you to Aditya, Shaquille and Ben. We'll see you next time on Go Time.
+
+**Outro:** \[01:13:11.11\]
+
+**Mat Ryer:** We could easily end up in a kind of arms' race against other people, and I never find that to be particularly enjoyable... So I think being different is the way. I like to say "Think different." \[laughs\]
+
+**Jerod Santo:** Did you just make that up?
+
+**Shaquille Que:** I've never heard that one before...
+
+**Mat Ryer:** It just came to me, yeah.
+
+**Jerod Santo:** It's just you who made that up. Wow.
+
+**Mat Ryer:** Yeah. It's good, isn't it?
+
+**Jerod Santo:** Aditya, you're 15; I think with your skills, and I would assume your relatively low expenses, there's actually no better time to bootstrap one of your side-projects into an income-producing endeavor. That's what I would suggest...
+
+**Aditya Prerepa:** Really?
+
+**Jerod Santo:** ...versus trying to just beat the streets for an internship. Yeah, man. It's a great time.
+
+**Aditya Prerepa:** Well, the projects I'm working on -- actually, you know what? I'll share. It's actually pretty interesting. So you know, because of Covid, group addiction therapy is not happening. You can't do it on a Zoom call, because from what I've heard it feels simulated and artificial. So what I'm doing - me and a few other people - we're building a platform... It's not really income-generating, like what you said, Jerod, but it's for social good. It's sort of a platform where you can do group counseling and random one-on-one conversations with other addicts, so you can talk out your problem. That's how it happens in real life, and that's actually my current project that I'm working on right now.
+
+And yeah, I've been reaching out and it's going really great. But hopefully, in the future, once we move back from remote to in-person, it can still be something that can be used, and at that point it could be turned into something revenue-generating. Does that make sense?
+
+**Benjamin Bryant:** Yes.
+
+**Mat Ryer:** Yeah, but I love the solving a real problem. That's a key thing. We can imagine cool tech, and we wanna build it, but sometimes the tech is the boring bit, and the problem you're solving and solving that problem is the valuable bit. So I think solving a real problem is a great little thing, a great insight, and you're doing it already.
+
+**Aditya Prerepa:** Awesome.
+
+**Jerod Santo:** What's it called? Is it out there?
+
+**Aditya Prerepa:** No. We're prototyping right now. But I'm with an organization, it's called Coding Tomorrow. We teach fifth-graders how to write Python, and that's gotta be the most painful thing I've ever done in my life. We're doing that as a group now, to try to get that up and prototype it.
+
+**Jerod Santo:** If you think fifth-graders are bad, try fourth-graders, or third-graders.
+
+**Mat Ryer:** How old are you in fifth grade? We don't have the same--
+
+**Aditya Prerepa:** Ten? Or eleven? I don't know. My sister's in fifth grade, so I should know, but...
+
+**Jerod Santo:** How old is your sister?
+
+**Aditya Prerepa:** She's, I think, ten or eleven. I'm really bad at this.
+
+**Jerod Santo:** Ten or eleven?!
+
+**Mat Ryer:** It's an off-by-one error. That's also very common.
+
+**Aditya Prerepa:** Well, you know, round it up, like Benjamin does.
+
+**Jerod Santo:** \[unintelligible 01:17:50.22\]
+
+**Aditya Prerepa:** Yeah...
+
+**Mat Ryer:** That's great though.
+
+**Jerod Santo:** Yeah, absolutely.
+
+**Mat Ryer:** That story might go in there. Are you still recording your audio?
+
+**Aditya Prerepa:** Yeah.
+
+**Mat Ryer:** We may snip that in.
+
+**Aditya Prerepa:** Awesome.
+
+**Jerod Santo:** Yeah, but to snip that in, we're gonna have to snip out most of the stuff Mat said earlier, so...
+
+**Mat Ryer:** That's the way it should be.
+
+**Jerod Santo:** We'll just have to \[01:18:14.15\]

Black Hat Go_transcript.txt

@@ -0,0 +1,655 @@
+**Mat Ryer:** Hello, and welcome to Go Time. I'm Mat Ryer. Today we're talking about Black Hat Go. We're gonna teach you how to steal billions of dollars while ensuring that your code remains simple, clear and maintainable.
+
+Joining me today, Johnny Boursiquot. Hello, Johnny!
+
+**Johnny Boursiquot:** Hello there.
+
+**Mat Ryer:** How's it going?
+
+**Johnny Boursiquot:** It's going quite well. I'm actually feeling pretty peppy. You might have heard about the -- there's this virtual Go conference going on in a few weeks... I think you might have something to do with it; maybe you're MC-ing, or something like that. I'll probably be making an appearance and giving a keynote, or something.
+
+**Mat Ryer:** Awesome. That's gogetcommunity.com, right?
+
+**Johnny Boursiquot:** That is correct.
+
+**Mat Ryer:** On May 21st.
+
+**Johnny Boursiquot:** Yes, sir.
+
+**Mat Ryer:** Okay, great. Well, are you excited to learn about hacking in that?
+
+**Johnny Boursiquot:** Oh, boy, am I!
+
+**Mat Ryer:** Yeah, me too. I feel like this is gonna be quite an episode... Don't worry though, we don't have to do it alone, as usual. I'd like to welcome back Roberto Clapis. Roberto is an ex security consultant and penetration tester, now a Chocolate Factory Boffin at Google. Hello, Roberto!
+
+**Roberto Clapis:** Hello, hello. Thanks for correctly pronouncing my actual job title. That's good.
+
+**Mat Ryer:** Hm. Security engineer at Google is the boring version.
+
+**Roberto Clapis:** Yeah... Who cares about that.
+
+**Mat Ryer:** But what was that? The register called you a Chocolate Factory Boffin, didn't they?
+
+**Roberto Clapis:** Yes, yes... We found some things in Safari and they weren't happy about that, and they decided to make a funny blog post about it.
+
+**Mat Ryer:** \[03:57\] Okay, we'll have to put that in the show notes, and you can tell us more about it as well later... Because I need to welcome our final guest today. It's Tom Steele. Tom is one of the authors of the book Black Hat Go, and a research consulting directory at Atredis Partners. Did I say that right, Tom?
+
+**Tom Steele:** Yeah, you got it good enough.
+
+**Mat Ryer:** Close enough.
+
+**Tom Steele:** Yeah, that works.
+
+**Mat Ryer:** Welcome to the show!
+
+**Tom Steele:** Thank you. Thanks for having me. I'm excited to talk about all manner of things, I guess so far.
+
+**Mat Ryer:** Yeah, well let's find out what we're gonna talk about. It's exciting. First of all, it's nice to kick off, to get to know each other a little bit with any working from home tips. We're all working from home a lot more these days. Has anyone got any tips?
+
+**Roberto Clapis:** The best tip that I can give you so far is to always have something to drink on your desk, and a lot. So you drink, you stay hydrated, and after a while it automatically reminds you to stand up and go somewhere else, like in another room in your house, so that you also keep the blood flowing.
+
+**Mat Ryer:** That's quite a good little system.
+
+**Tom Steele:** You can always tell how much work I've been doing by the amount of cups on my desk. They just kind of accumulate over the day.
+
+**Roberto Clapis:** You should garbage-collect sometimes.
+
+**Mat Ryer:** You know, I'm excited to talk about the Black Hat Go book, because it's not often you get a lot of technical books that have titles like -- and this sounds a bit like I'm just doing clickbait, but chapter seven is gonna blow your socks off... Is it called -- is it "Databases and file systems pilfering and abusing"? I saw it twice, it's something like that...
+
+**Tom Steele:** Abusing databases and file systems.
+
+**Mat Ryer:** Okay, so chapter seven is called Abusing databases and file systems. That is a great title for a chapter.
+
+**Johnny Boursiquot:** I have the book right here, I can double-check for you if you want.
+
+**Mat Ryer:** Yes, please...
+
+**Roberto Clapis:** Oh, physical one?
+
+**Johnny Boursiquot:** I know, right? Dead tree..
+
+**Mat Ryer:** I didn't know they did those. They just printed out the internet. \[laughter\] You should get Tom to sign it for you.
+
+**Tom Steele:** Anytime. Later.
+
+**Johnny Boursiquot:** Yeah, definitely. Later, yeah.
+
+**Mat Ryer:** I don't mean like with a signature, I mean like add up all the characters in it and send a checksum.
+
+**Johnny Boursiquot:** Oh, boy... Yeah, okay.
+
+**Tom Steele:** There's something about that though...
+
+**Mat Ryer:** Where does it come from, Black Hat Go, the name? What does that actually mean?
+
+**Tom Steele:** You could probably ask a hundred different people and get a hundred different answers. We kicked around a bunch of titles for the book at first, and I personally wanted to stay away from Black Hat Go, from the name Black Hat... So I think our original title that Chris and I came up with was Adversarial Go, which is probably more descriptive of what we actually do on a daily basis... Because the term Black Hat - I wouldn't exactly describe what I do on a daily basis as Black Hat, because I'm obviously always doing it legally. People know I'm doing these things, they're paying me to do them on a daily basis... So I'm not doing them to illegally make money or anything like that, which is probably more what you would think of Black Hat.
+
+But the publisher wanted to sort of relate it to Black Hat Python, which was a Python book that came out a number of years ago. I think they came out with the second edition. It's a really good Python book, too.
+
+So in that No Starch ecosystem there's -- I don't think there's a White Hat, but there's a Grey Hat C\# who one of my co-workers actually wrote. That's a really good one, too. But that has a lot more parsing out maybe vulnerability management style stuff. It also has SQL injection, and those sorts of things.
+
+So I think more along the terms of what No Starch was thinking, was like these are more like attacking-based tools, versus what a defender would use. But typically with white hat/black hat, that historically has been more criminal versus authorized type of testing. However, we all go to Black Hat, the conference over here, and most of us are not criminals anymore... \[laughs\]
+
+**Johnny Boursiquot:** \[07:57\] Plus it'd probably boost sales too, right? \[laughs\] It's not dull and boring, it's more of "Yeah, you're going in the offensive. You're gonna poke holes in people's systems." That sounds way more fun to me.
+
+**Tom Steele:** Yeah. So the goal for the book, when we got asked who the audience was, the audience wasn't people that know Go already. And I don't even think it's people that know Go, from the Go community. Certainly, if they wanted to go pick the book up and expose themselves to some things they might not be exposed to, that'd probably be a good choice, too. But the audience was really like the people that I work with in the security industry, or maybe people who are just getting started, or they wanted to learn a new language, and then they want useful tools that they can go apply.
+
+If you pick up a programming book and it' just strictly "This is how you do X", like "This is how you handle arrays", or "This is how you do a for loop" - that's great, but if you give yourself projects that you can actively use in your work, those tend you help you learn the language quicker. So that was really what we were going for.
+
+So it just sort of spurred off of my history with Go. I started using Go -- I think 7-8 years ago might be the first commit to GitHub. We had a tool that was written in JavaScript using Node, and I basically used Go to rewrite that. That was my first introduction. After that, it sort of became the tool for everything. Obviously, it doesn't handle anything that we need, but I think it was most of the tasks. I became very productive with it, so after an extended period of time I felt like we could do a book like this. My co-authors Dan Kottmann and Chris Patten, they're also amazing Go developers as well, and they certainly -- the book wouldn't have come to a close without those guys. Certainly Dan pushed it to the end for us as well.
+
+**Johnny Boursiquot:** So if I may give a review of this book, I've been having fun with it, honestly. It has shown me different use cases for Go, that I typically don't think about... I'm like "Oh, you CAN do that with Go..." And now I have a whole new appreciation for the net package... I'm blown away by some of the things you can do, and I've been having fun with it, extracting bits and pieces there, and meetups, and teaching people about writing port scanners, and building proxies, and things... It's been incredibly fun having these kinds of projects, where you're not just learning syntax, you're not just learning the mechanics of Go, but you have things to build that are kind of fun.
+
+I've really been enjoying this book, so kudos to you and the others. This is a really great book, and I'm having fun with it.
+
+**Tom Steele:** Awesome. Thank you. I think that's probably the most useful aspect of Go for us, is the ability to just interface with TCP, and do it very quickly, and start doing interesting things, like writing odd proxies. The amount of times that we've been trying to get out of a network how to write some crazy proxy to get around things has been -- we've done a lot, we've done quite a bit... So yeah, it's awesome.
+
+**Roberto Clapis:** That has been basically my experience with Go. During a penetration test we needed to build a weird proxy. We needed to take a WebSocket on one side, and needed to downgrade it to a TCP connection on the other side, because that's what we got; because when you do a penetration test, you get going whatever you get going. So we needed to write a proxy, and I had just started learning Go... And I said "Well, one is a net.Conn, the other one is a io.ReadWriteCloser. I can just io.Copy them together and let's see if it works." And it worked.
+
+It was my first experience with Go in writing a tool that I actually needed at work... And it's a blast, because you just use Go, stuff that is in the standard library. Actually, in your book I think most of the imports you don't even mention them, because just running Go import on your snippet would import all the right packages, because you just use a lot of standard library and some external packages here and there, which I really appreciate.
+
+**Tom Steele:** \[12:06\] Yeah, you can get away with -- like you said, you implement a reader, you implement a writer, and all of a sudden you've got a tunnel.
+
+**Johnny Boursiquot:** Yeah...
+
+**Tom Steele:** And we've done that across various -- you can do it at any layer. So if you wanna do command and control over S3, and then write that back to a WebSocket, it's pretty much just a piping Legos together type of thing.
+
+**Roberto Clapis:** Yeah. We used to use Python, but for capture the flag competitions and at work... But at some point it would just become so complicated to handle the multiple connections, and concurrency, and all the threadings... Plus, you need Python to be installed on the system that you are talking to, when instead with Go you can cross-compile to whatever weird, baffling architecture you find, and you just push the executable... Which was a big plus.
+
+**Tom Steele:** Yeah, for sure. I think that's probably the most welcome feature, is obviously a lack of dependencies... You know, for people that are familiar with this stuff. So a lot of the stuff that we do, we work together as a team. The projects can range, different types of things, but one of the things that we do a lot is we're trying to attack a company; this is like a bank, or something like that, let's just say hypothetically... And we're not necessarily testing a specific application. We do a lot of that stuff too, but we're trying to get into this company by any means necessary.
+
+So once we gain code execution on an end user's workstation, like a Windows PC, that's when the fun really begins, because you have to figure out how to get out of these networks that sometimes are locked down. That's where Go comes in handy quite a bit, is when you need something really quick, and you need to compile it to Windows with no dependencies. That's a great feature... So that's just been great.
+
+Or maybe you're working with a co-worker and you have some little utility, and they need to run it right away. You're moving fast, sometimes having them install Python, and then install the dependencies, and to handle all that can be a nightmare... Because now you're not just saying "Here's my tool. Use it. Here's package management." Maybe they've never used whatever language you chose to write it in. I don't know, Ruby Gems seems to be the most historically awful one, but... \[laughs\]
+
+**Roberto Clapis:** I was about to say, how much time did you spend trying to make a gem work for your Metasploit. \[laughter\] That has been one of the worst experiences in my life. And yeah, Go doesn't betray, in that sense... Especially after a while I started building a library that I just would use for scripting, because Go is also good for scripting... And in that library you had the utility functions, like some of the tunnels that you explain in your book. I would never dream of building in some of the things that you wrote about, but I have this utility library, and you just pull it in, and you call it, and it just gets you running in a short time.
+
+It's also nice to see that it doesn't require that big machines, or beefy computers; it just runs from very low-hand hardware. I once had to do a penetration test from a Raspberry Pi, and I was grateful that Go was low footprint.
+
+**Tom Steele:** Also, low compile times helps a lot, too.
+
+**Roberto Clapis:** Yup.
+
+**Tom Steele:** I've been playing a lot with Rust; we won't get into that, but it's...
+
+**Johnny Boursiquot:** \[laughs\] This is a Go podcast, mate...
+
+**Tom Steele:** ...it's just sit and wait. Yeah, for sure. For sure.
+
+**Johnny Boursiquot:** So I kept thinking, when going through the book - and I'm still going through it - I've kind of been jumping around a little bit, and learning about the different projects, and whatnot... It kind of begs the question, how did you pick some of these projects? Because a lot of them are really fun. There's like credential harvesting, there's keylogging, there's building proxies to get around corporate networks, there's interacting with Metasploit over RPC, there's writing DNS tunneling... How do you pick those (to me) really fun projects? Do you pick the projects and say "Okay, what are we teaching here?", or do you say "Okay, we need to teach these things. What kind of projects would convey what we're trying to teach?"
+
+**Tom Steele:** \[16:13\] I would say the majority of the chapters came from thinking about how do we use Go during our testing, and what kind of utilities have we written and do we know a lot about? And how was it useful for us? So pretty much every single thing in that book, we've used that. There might be something small that we've used.
+
+Now, the hard part was taking perhaps a bigger project and then pairing that down to something that you can actually digest in the book. So the stuff in the book is really useful, but it's all basically getting you started on "What can I do in addition to this?" But everything in there is pretty much -- like, we had used that first; there's something in there that we had used for a test.
+
+And then there's a few stuff that obviously some of the guys - they have a ton of expertise. Chris Patten wrote the steganography chapter, and I'm not the guy to talk about that, but he has a long history with that sort of thing.
+
+So basically we built a huge list, and then we kind of paired it down to what we could actually fit in the book. There's some stuff that didn't quite make the cut, because it would have been a little bit too much background, a little bit too much lab setup type of thing... Like, some of the VoIP stuff you can do is really awesome, but you can't really set up a virtual environment for some of that stuff, right?
+
+**Johnny Boursiquot:** Right.
+
+**Tom Steele:** So yeah, I think that's how we went about it. For example the DNS proxy - we use a lot of tools to do what we call "command and control", which is basically like we have a payload that's running on a system that we don't control, and that's communicating back with us. You can do that over multiple protocols, but one of the great ones is DNS, because it gets out at most networks, and certain things... But a lot of the tools that are used there - they don't necessarily handle redundancy or proxying natively. You're basically client to server, that's it. But if you can use Go to write up a quick proxy that maybe fans those out, or at least gives you a layer in front of those, that's a great way to do it.
+
+**Mat Ryer:** So when you talk about proxies then, is the purpose to intercept and sort of have a look at and spy into the traffic? Or is it to control it or modify it in some way? Or does it sort of depend on what you're doing?
+
+**Tom Steele:** I think there's room for both with Go. I think there's been times when we've needed to -- say we're not doing this adversarial type of testing of an entire company; say we're just testing some software, and we need to get in front of the TLS stack and start doing interesting things there with inspection. Go is great at that, because you can actually -- it's so easy to copy the existing TLS package from the standard library, and modify that, and then use that in your own code. That's really easy to do with Go. It might be easy to do in other languages, but for me it was easy to do with Go. Very simple.
+
+So we've done that before, where maybe some client's implementing TLS in a really weird way, and we've had to do that. But for the most part, for my use case and the stuff that you'll see in the book, it's more like shaping traffic, like being able to get in and out of networks, and around, and move things. So think like, you know, interesting style of reverse proxies type of thing.
+
+**Roberto Clapis:** \[19:45\] Since you mentioned rewriting the TLS package, one thing that I wrote a proxy for was the Metasploit; when you use it -- Metasploit is a common suite that collects exploits, and you just take the exploit and you run it. It uses Ruby, and it uses a specific client hello. So when you connect via TLS, some firewall detected that is the hello that usually comes from Metasploit, and other things... And the firewall is going to drop it, regardless what you send .
+
+So with Go it's quite easy if you write your own reverse proxy to just take whatever comes from Metasploit, rewrite some headers, rewrite the client hello, because you can do that, specify the ciphers that you do... So you can clone perfectly the Chrome handshake, and the firewall is going to let you through. That is one kind of proxy that you do to yourself, to make the tools behave, which Go is great at.
+
+**Johnny Boursiquot:** I must say, that's scary, dude. \[laughs\] Honestly. I knew this was gonna be a fun episode, but really, some of the things you can do, and I've seen talked about in this book, and the stuff you're mentioning now - you can do that stuff with Go. It really blows my mind away.
+
+**Break:** \[21:00\]
+
+**Mat Ryer:** So it makes me wonder then, is this book potentially arming people with information that they could use to do some harm with?
+
+**Tom Steele:** Probably.
+
+**Mat Ryer:** And what's your ethical position on that, Tom?
+
+**Tom Steele:** I'll put it this way... Oftentimes the best people that get into hacking - but I'll just call it the security industry, the offensive security industry - the best people often come from development roles. One of our guys, Ryan Hanson, he lives in Idaho with me... And when I first met him, he was a developer working in Boise, but he has become one of the best security people that I know, because he understood the software at a level that you just don't get by just only doing security. So all hacking is is really just understanding the underlying mechanisms and then thinking about how you can subvert them.
+
+Showing someone how to do SQL injection, for example - that would be more like a script kiddie type of thing. You can go read about that stuff. But the best way to understand SQL injection is to go write your own SQL package. That's a great way to handle it, because now you know how to defend it and you know how to attack it. I don't have any qualms about ethically teaching \[unintelligible 00:24:04.28\]
+
+**Roberto Clapis:** \[24:08\] There's one other ethical standpoint, which is if you put the book out there, the attackers and the defenders have the same amount of knowledge. If instead this knowledge is just shared across attackers, maybe the defenders - which usually are software engineers - don't get access to this, so they might not know about this stuff. I've found that putting information out there is usually good for defense, even if it is aggressive, or black hat, or adversarial knowledge.
+
+**Mat Ryer:** Yeah, it sounds like that answer is legally watertight, Roberto, if you actually had a lawyer check over that sentence... Because I think you've nailed it. \[laughter\]
+
+**Tom Steele:** Also, one of the things that we deal with a lot is bypassing an antivirus or EDR controls. So you have something running on your system that's supposed to prevent code execution and payloads from escaping, and all those things. I don't actually have a problem teaching people how to bypass those things, because if we bypass them, it's a game of cat and mouse; those companies will improve their detections, and they'll actually come up with detections that maybe work. Or they won't advertise detections that don't work, or are easily bypassed.
+
+You can say "We have this machine learning algorithm that blocks all this stuff." And it's like, "Well, not really... Not really." It turns out if I just name my binary worldofwarcraft.exe, it just kind of works, right? Those sorts of things.
+
+**Mat Ryer:** People just immediately click it, yeah.
+
+**Tom Steele:** Yeah. I think the ability to identify and bypass these controls just makes the defenses stronger, and that's how we evolve.
+
+**Mat Ryer:** Yeah. That defense could be used to excuse any crime... But yes, I do get your point.
+
+**Tom Steele:** So true! \[laughter\] I have broken into physical banks before, and yeah, that's fun stuff, too. But you should always get approval prior to attacking a network or a physical location.
+
+**Johnny Boursiquot:** Yeah, I was waiting for that bit after you said that you've broken into banks before... \[laughs\]
+
+**Tom Steele:** Yeah, you don't wanna be doing this stuff unauthorized, set up virtual labs, and that sort of thing. That's definitely the way to go. I wouldn't say "Go attack random websites", but there are programs like HackerOne and Bugcrowd and stuff like that where you can go get approval to test those sorts of things, even if you don't have experience.
+
+**Mat Ryer:** We'll bleep out the names of those pieces of software, although this book -- apparently, you can just use the book... \[laughter\]
+
+**Tom Steele:** Well, those are sites where anyone can go sign up to do security testing of websites. They facilitate those types of things.
+
+**Mat Ryer:** Gotcha.
+
+**Roberto Clapis:** Yeah, those are collectors of bounty programs, in which you go there and you see which companies will pay you to do the tests on their systems. You attack their systems, and if you find a bug, you get paid for it. It's not usually that high of a pay, but it's nice when companies put it out there.
+
+**Johnny Boursiquot:** So is it fair to say that the things that do make it into the book, such as this one, are the things that are at this point well known? Is there a new ground being broken? And I'm gonna go and assume yes, to some degree, but please do elaborate. Is there a new ground being broken in -- I don't wanna say black hat, but... Or is it like a remix of different exploits to get to new things?
+
+**Tom Steele:** There's no sort of zero-days in this book. They're all well-known techniques. I did wanna put something in there that I was advised not to... It was how to bypass a certain product, and they were basically "Hm, you can't do that." I was like, "Okay, cool. I won't."
+
+**Johnny Boursiquot:** We should ask our folks at Changelog how airtight are our legal stances... \[laughs\]
+
+**Mat Ryer:** We didn't send Tom the link for this Zoom... So there's your answer.
+
+**Tom Steele:** \[27:56\] So the book talks about things in generalities... And yeah, they're known techniques, they're known flaws. And I think if you're on the defense side, if you're doing software development, knowing the attacks is definitely better than not knowing them. For example, you wanna know how to do SQL injections, so that when you're writing your application, you don't develop with that.
+
+I think if you're writing software, you always wanna be like "Well, what can happen if someone does this?" I think those are really good questions to ask.
+
+**Mat Ryer:** Yeah, that might be why you say that developers make good security engineers, because when it's your job to defend against it and you realize how hard it is, maybe you just think "Do you know what? I'm just gonna go on the other side of this." I don't know if that in any way helps that, but...
+
+**Tom Steele:** A long time ago, at a company I worked for, we were trying to basically bring up a bunch of people up to speed on web application security... And a typical way you do that is you basically just build a vulnerable web app, and then you teach them all the offensive techniques that you can. But I actually took it the other way, where I'd made them build a web app and then implement the defenses... Because I felt that was the best way to get an in-depth knowledge of how to identify and possibly bypass those defenses.
+
+**Johnny Boursiquot:** It does.
+
+**Roberto Clapis:** When you teach security knowledge to software engineers, and you see that a certain engineer maybe didn't have a certain concept, he just starts clicking into it and he goes like "Hm..."
+
+Once I was out at dinner with a friend of mine, who is an extremely good software engineer. And I was talking about cross-site request forgery, which is one kind of vulnerability... And after a while, he was like "Hm. I have to go." And just walked out. \[laughter\]
+
+**Mat Ryer:** I just have to go. Unrelated.
+
+**Roberto Clapis:** It was like four people, and he walked out, and he deployed a fix, and he was like "Thanks for that. I didn't know about that." That was the best reaction.
+
+**Johnny Boursiquot:** Okay, so I'm seeing a lot of parallels with operations. In my line of work - and I do SRE currently - I've found that developers and engineers tend to think of the operability of something towards when they start needing to ship it. It's like a concern, like "Oh yeah, I'll worry about that when I get to it." And then we sit down, and I'm like "Well, have you thought about this? Have you thought about that, before putting this thing into production?" They're like "Ohhh...! Yeah, let me go back and fix that."
+
+So it's almost like security, and making sure that your software isn't vulnerable to some obvious things. It sounds like this is also one of those things where, you know, as an engineer, folks are building features, and then you go sit down with a security review team, and they start poking holes in your thing, and you're like "Ugh... Now I've gotta go back and plug those holes." The art of engineering requires so much more than just writing features, and I'm seeing this parallel in this talk.
+
+**Tom Steele:** Yeah, I agree. I think that you obviously wanna try to design your feature with security in mind from the beginning. However, I am very sympathetic to that, to basically having to just get something running. That's another flaw that you see a lot in the security industry too, is the exact opposite, where they come in and they basically go "This has no security", but they're not thinking about the fact that it works, or it needs to work first. So there's a lot that goes into building and just making a product run, that sometimes you need to be able to give up a little bit of security as well while you're getting that going. So I think there's a happy medium, and that medium is constantly moving, depending on your risk.
+
+**Mat Ryer:** Yeah, I wrote an application once, and I did something wrong and basically couldn't sign in... It was the most secure application I can think of.
+
+**Tom Steele:** Exactly. \[laughter\]
+
+**Johnny Boursiquot:** So secure it's unusable.
+
+**Mat Ryer:** But it just didn't work at all, yeah. So it was just a bit too secure. And that was how I was pitching it... "If anything, we've made it a bit too secure..."
+
+**Roberto Clapis:** \[32:00\] Well-designed APIs are made in a way so that using them in a secure way is easier, and less effort than using them in a wrong way. That's usually what a library designer tries to do... Because otherwise people are going to go for the easy road, because they have to deploy.
+
+**Mat Ryer:** Yeah, but they of course have to know what those things are, and that's where I think a book like this comes into play. Is Go good as a foundation for writing secure code? Because I've heard different things said on this subject. How does Go stack up when it comes to security?
+
+**Tom Steele:** Yeah, again, you'll probably get a bunch of different opinions on this. My particular opinion - I think it does a lot of things correct, and I think the way people that are probably using it, I think it's probably a good thing to grab. The fact that you have to be pretty explicit about things helps quite a bit. The compiler will help you with a lot of bugs. And then the fact that it's statically typed, it just makes you be explicit. You have to explicitly handle lots of different things.
+
+I'm not gonna get into this, but the lack of generics probably helps in some of those things. It makes the code easier to trace, and that sort of thing.
+
+**Johnny Boursiquot:** Shots fired... \[laughter\]
+
+**Tom Steele:** But yeah, and I think the Go standard library does most things correctly, as well. For example, code execution, calling command exec, the fact that there isn't a version of that where you can just provide a string with arguments, basically - that's a secure design in and of itself.
+
+**Roberto Clapis:** Yeah, and also if you look at the SQL package and the HTML package - the SQL package makes you use prepared statements, or parameterized queries...
+
+**Tom Steele:** You have to try very hard to write -- you have to basically pass the format strings, or something like that. You have to know what you're doing to screw it up there.
+
+**Mat Ryer:** That is cool then.
+
+**Roberto Clapis:** Yeah, the HTML template does contextual escaping.
+
+**Mat Ryer:** Yes.
+
+**Roberto Clapis:** I think Go was one of the first languages in the open source world that had it. Maybe there was a Java library by Google, but...
+
+**Tom Steele:** If you go check out the documentation for that library, you'll basically see a reference to a paper about context-aware coding, and I don't think any other language read that paper and implemented it... I'll talk about it with one of \[unintelligible 00:34:27.05\]
+
+**Mat Ryer:** So for anyone not familiar, this is where when the template is rendered, it knows if it's putting it inside, say, an attribute, versus in the body as HTML to be seen, and things like that. It does different things, depending on the context.
+
+**Tom Steele:** Yeah. And interesting things can happen if you HTML-encode data into a URL parameter type of thing, or JavaScript.
+
+**Roberto Clapis:** Yeah. I maintain the package, and every time I have to send a CL to fix some things, I always shiver a bit, because it's so precise and so well done, and has to have no bugs... Because any bug there can become a vulnerability. It is so disquieting to add code to that.
+
+Also, the paper that you were talking about, Tom - I think it's like several tens of pages of schemes and very hard to read text, so...
+
+**Tom Steele:** It's not easy.
+
+**Roberto Clapis:** Yeah, it's not for everyone.
+
+**Mat Ryer:** Well, the Go team read that kind of stuff for breakfast, don't they? That's what I like, because it's solved for us, and as developers we get to just use it. So thinking about that then, are there any common gotchas? And actually, this was a question I got from Kamal on Twitter. He asks "What are the most common mistakes that developers make that lead to exploits and security vulnerabilities?"
+
+**Tom Steele:** \[35:59\] Yeah, so I've thought about this... At Atredis we've actually tested quite a number of Go applications. We do a lot of software security type of things, and so the things that I see -- I was gonna go into that, this is actually a nice feature of Go... For example, say you wanna take JSON and you wanna move that into a struct that you eventually wanna insert into a database; this is a very common thing... The nice thing about Go is that you're explicitly defining the fields that you want in that struct. So you're not gonna get random fields.
+
+For example, if you're using something like Node, we see this a lot, where they basically take request.body, if you're familiar with that, and they slam it into a database and now you've got real problems at hands.
+
+So what I see is that actually works quite well, because now you know the objects that you want; "objects" in quotes. And now you've also defined the field types; you know that what's coming in is gonna be a string. Sometimes it might cast an integer to a string depending on the library that you're using, and that sort of thing, but you are defining at least the data types that you want coming in.
+
+So the stuff that I see just from that end is sometimes you have a struct that represents something that's in the database, a user item. If you serialize that directly into the database, directly from JSON, you've got something called mass assignment, where you're basically not filtering the fields that you expect, because your form might take three fields to find the user; the first name, the last name and the email address. However, what if you have another field that's like a boolean, "isadmin", or something like this... Just because your form doesn't provide that value doesn't mean that the user won't provide that. And when you serialize that JSON or that form value even into the struct, and then you put that struct in the database, you've got a serious problem on your hands now, because now they're starting to modify objects, and things like this. So I see that quite a bit.
+
+SQL injection - we don't see that a lot. Cross-site scripting, we don't see that. We've seen some really nasty bugs when people are writing their own templating engines; we've seen some particular nasty ones. One actually resulted in code execution...
+
+This particular instance was basically they had a partial template that would load the user input from a database. Then that partial template would get rendered into a parent template. And because they took that value as trusted, you could put code in there, so you had this double eval type of situation.
+
+But I think those are few and far between. I think most of the Go things that we're testing, people aren't actually using templates to render server-side anymore. They're mostly using a JSON or other services. However, I would probably stick to the standard Go template, or at least something that wraps that pretty thinly. That's just my opinion. But we don't see a lot of that stuff.
+
+We see that mass assignment issue... For whatever reason, when you've got cross-site request forgery, we see the protections implemented, but often incorrectly.
+
+**Roberto Clapis:** Yeah. If I might add, I also saw that a lot... Because people don't have it in the standard library. It's an XSRF/CSRF token, so people don't see it, people don't know about it. So far, I've found that to be one of the most common vulnerabilities in Go applications.
+
+**Mat Ryer:** What should you do?
+
+**Tom Steele:** Be careful. \[laughter\] Yeah, go ahead...
+
+**Roberto Clapis:** Basically, whenever you take a request that is state-changing for your server, so that afterwards your server will do something different... For example, if you take a search query, it's probably not state-changing. But if you get a submission of a subscription, or some user data, you might find that to be state-changing. You need to use the XSRF token package to protect your form.
+
+Basically, there needs to be an extra parameter that you strip away, but it needs to be there, and it needs to be tied to the user and to the session. Otherwise, attackers can craft those requests. I will link the package in the podcast.
+
+**Tom Steele:** \[39:57\] Yeah, and my advice there is -- yeah, the CSRF packages themselves are actually good. It's when you try to get clever. So my suggestion is be explicit. If you're building an HTTP application, build a handler that accepts only a POST request. Don't process HEAD, GET and POST in the same handler because that's when you're going to end up with what I've seen is they protected the POST verb with CSRF, but they will process a HEAD with the same value. That can be a big issue as well.
+
+**Mat Ryer:** Hm. I'm just gonna go... \[laughter\]
+
+**Johnny Boursiquot:** You'll fix some things... \[laughter\]
+
+**Tom Steele:** Yeah, so some CSRF bypass stuff is one... You see a lot of -- like, they wanna implement core setters, and I've seen some really interesting bypasses there, where you're not using the correct regex... But a lot of that stuff has been resolved.
+
+**Roberto Clapis:** Okay, but to be fair, [CORS](https://en.wikipedia.org/wiki/Cross-origin_resource_sharing) is hard.
+
+**Tom Steele:** It's difficult, yeah.
+
+**Roberto Clapis:** It was not made to be understood by humans. CORS was a big mistake, and it should never have been there in the first place.
+
+**Tom Steele:** Yeah, every time I have to go explain one of these bugs, I go reread the Mozilla documentation, so that I don't say anything stupid.
+
+**Roberto Clapis:** I do the same.
+
+**Johnny Boursiquot:** I think we've already got Roberto's unpopular opinion... \[laughter\]
+
+**Tom Steele:** On the idea of being explicit - if you are building your API to only take in JSON, don't accept anything else other than JSON. There's no reason to sit there and parse text/plain, or URL form encoded input if you are only willing to use JSON. And I haven't seen a lot of that in the Go web frameworks, where they require that level of explicity. I've written something that I can link later that does this on a per-handler basis, which I think is the way to do it.
+
+I actually stole a lot of these ideas from the author of Hapi.js. If you wanna go look at a framework that implements a lot of things well, go look at that framework, Hapi.js. Eran Hammer has done an excellent job with that. But yeah, validate all your inputs, too. Don't just depend on the struct mapping. If you're expecting an email address, make sure that's an email address.
+
+I'm not super-happy with the way the validator libraries in Go work. There is one that's pretty good. At least it gives you good definitions, but then you have to go and write your own wrappers, and things like that... Which - I don't know, that might be just the story of Go web development in general though, right? It's like, you've gotta implement a bunch of stuff yourself.
+
+**Roberto Clapis:** One of the things that I hated the most is the content type. In Go if you don't set any content type in your responses, it will still work, probably... Because the Go server tries to guess, based on the response content, what is the content type, and will set it on the response for you... Which means that if an attacker can somehow make the server respond with something that looks like HTML but is instead JSON, the server will set an HTML content type, even if your endpoint only returns JSON. So as Tom was saying, you should only accept JSON if you expect JSON. Same thing the other way around - if you want to return JSON, set the content type.
+
+**Tom Steele:** Yeah. Being explicit is definitely helpful.
+
+**Johnny Boursiquot:** Okay, so when you say that, there's something that immediately comes up to mind, and that's the fact that most developers basically -- there's always this bias towards "Oh, easy to get started, easy to get going. Just grab this thing, drop it in your code, and if you run into any issues, go to Stack Overflow, copy-paste whatever you find, and go. Get the job done!" There's all this emphasis on "Easy, easy, easy", and not enough on "Do you know actually what it is that you're putting in your codebase right now?"
+
+So the frameworks tend to hide and make things easy, and I remember back when I was doing Rails work, there were articles about how Rails basically prevents you from shooting yourself in the foot from a security standpoint. You can't get a little Bobby Tables kind of situation in your system if you're using Rails, because you're hoping that the framework is gonna take care of these things for you.
+
+\[43:57\] At the same time, when I think about it I'm like "Well, shouldn't I know as a developer why something is working or said it is working?" In order to be explicit - which is the advice we're giving here - that means you have to dive deeper; you have to go lower. Basically, leave the abstractions of the framework behind, get deeper, in order to know exactly what's going on. There must be some sort of happy medium there. I'm not sure where it is in the context of this conversation, but is it your advice to default to the explicitness, versus relying on packages and frameworks?
+
+**Roberto Clapis:** Personally, I think that if a framework is well-designed -- for example, Rails gives you a hand in Ruby. If a framework is well-designed or a package is well-designed, like the HTML template package, you should not know. You should be able to use it and not have vulnerabilities. That's how a framework should be done.
+
+But then if you want to do an extra step... For example, in the HTML template package there are some types that tell you "These might cause a security issue." It's written in the type. "If you use this type, you're exposing yourself to a risk." That's the time when you need to go back and reread the entire thing and understand the threat, understand how those features combine, and everything else... Because at that point you're on your own.
+
+**Tom Steele:** I definitely think there's things that you shouldn't implement yourself. I think you should rely on packages that you trust, if you can, from trusted sources. But you've gotta remember, if you import something, you're now responsible for it.
+
+With Go we've seen some interested things where some packages were overriding the user agent and HTTP clients and things like this, and you don't know that it's doing that. Or they're opening files in the background and you don't even look, because you're just like "This does the thing that I wanted it to do." But I don't think that you should implement everything yourself. When I said "be explicit", I meant more "Be explicit in what you accept as valid input from an unknown source." Don't trust the input coming in, and interrogate it as much as reasonable before processing it.
+
+In relation to that, we see good encryption, good use of encryption. No one's writing their own version of AES. But then I also see people using it incorrectly. The difficult thing with Go is people aren't writing -- typically, if they're leveraging Go, they're doing something probably pretty interesting, pretty difficult, as far as like they're breaking new ground in their application... Because this isn't just like "Well, I can just use WordPress and write a little plugin that does it." They're breaking new ground and doing something tough, so you might have to use encryption there.
+
+So yeah, I do see quite a bit of basically using encryption without validation - yeah, stuff like this. I don't really see hashing used incorrectly. For whatever reason, people seem to be pretty good with that.
+
+And then I guess the individual flaw that you won't see in other languages for the most part is unbound concurrency. It's huge. What I mean by that - I'm gonna put this in context of HTTP, just because it's the easiest one for me to relate to... You have an HTTP handler and you accept an array, and then you launch a goroutine for every item in that array, and process that. We've seen that quite a bit, and that always leads to...
+
+**Johnny Boursiquot:** Trouble.
+
+**Tom Steele:** ...trouble. I think the request limit on Go is 10 MB, so you can imagine an array of 10 MB; you can generate quite a bit of goroutines that way.
+
+**Break:** \[47:43\]
+
+**Mat Ryer:** It's funny, because you think of this stuff often as like hacking, and it's all low-level... If you believe the movies, of course, it's like 3D cubes that have to fit together. When it all lines up and lights up, you know you've hacked in somehow...
+
+**Johnny Boursiquot:** \[unintelligible 00:49:10.22\]
+
+**Mat Ryer:** Yeah, you sort of think of it as being a very alien process... But when you talk about that, that's a very everyday thing. I'm sure most of our listeners will have dealt with JSON coming into some endpoint, and done exactly the kind of things that you're talking about.
+
+So it's interesting that it's worth thinking of just the very practical ways of abusing something, as well as things that are down at the TCP layers that you might not be exposed to too often.
+
+**Tom Steele:** Yeah. I think when you get down to the TCP layer, if you're talking about the in-the-movies type of thing, \[unintelligible 00:49:46.08\] buffer overflows and memory corruption vulnerabilities. You just don't have those in Go. Of course, you would want to audit and be careful when you're calling C libraries from Go. If you're using unsafe - it's called unsafe for a reason, right? If you're taking in input from an untrusted resource and then passing that to a C library that you're wrapping, that could be disastrous. I don't see a lot of that though. For whatever reason, we don't see a whole lot of the need for people to wrap C libraries, and I think that's because Go makes it so easy to work with--
+
+**Roberto Clapis:** Cgo is terrible.
+
+**Tom Steele:** Yeah, cgo is pretty terrible. I wasn't gonna say it... \[laughter\]
+
+**Mat Ryer:** What's that? It's more secure?
+
+**Tom Steele:** It's not too bad... I use it quite a bit. It's not too bad, but it's definitely not easy. But yeah, I just don't see a lot of that. I think most people if they find themselves reaching for a C library, they're just rewriting the bits and pieces that they need, and processing those binaries by themselves... Which is good to see.
+
+**Roberto Clapis:** Yeah, and one thing that I also like is that people are using the race detector when they test their code, which is a -- so Go is memory-safe, which is great. But one thing that people don't know is that it is memory-safe until a race condition happens. If you have race conditions in Go, you might get memory corruption and remote code execution. So build your code with race.
+
+**Mat Ryer:** Could that be abused, or is it just gonna be random?
+
+**Roberto Clapis:** It can be abused. Actually, in Google we host a CTF every year, and one of the challenges of the finals last year was to exploit a Go server by using a race. So you could exploit a race on the server and get remote code execution by just abusing this race. So it's doable, people did it. I think six teams solved it... So yeah, use a race detector.
+
+**Mat Ryer:** And what about fuzzing? I feel like fuzzing is also gonna be really useful when it comes to this.
+
+**Roberto Clapis:** Fuzzing is going to be really useful, and also I think that the second or third most voted proposal on the Go public repo is about fuzzing. I would love to see it in the standard library, because fuzzing also makes it very easy to spot easy bugs and panics. I mean, a denial of service is usually a bad bug. It's not a memory corruption, it's not an RCE, but you don't want your server to go down. And the fuzzer will try to feed complicated input to your APIs, so you will find it EOS.
+
+**Tom Steele:** \[52:25\] Yeah, that was gonna be my unpopular opinion, was that denial of service is worse than you think. It's more critical than people think it is.
+
+**Mat Ryer:** Well, we'll get on to that. Before we do though, Thomas, I wanted to ask you one last question. What was your mother's maiden name? \[laughter\] Okay, good. Now you've passed the test.
+
+**Johnny Boursiquot:** Date of birth.
+
+**Mat Ryer:** Nearly. Now you've passed -- it's too quick. It's too good. It is good. It's actually time, speaking on unpopular opinions... It's time for our regular slot, Unpopular Opinions.
+
+**Jingle:** \[53:07\] to \[53:26\]
+
+**Mat Ryer:** So Tom, you were saying your unpopular opinion...
+
+**Tom Steele:** My unpopular opinion is that denial of service is a big deal.
+
+**Mat Ryer:** Oh...
+
+**Tom Steele:** And that's because if you build this awesome service and no one can use it, you're probably gonna lose quite a bit of money if it's profit-generating. And sometimes that can impact companies more than code execution can.
+
+**Mat Ryer:** Yeah, of course. I think that's a fair one. I mean, do people think denial of service is just not something to worry about?
+
+**Roberto Clapis:** Yes.
+
+**Tom Steele:** Yeah, most of the time. Most of the time people's eyes glaze over when I start showing them bugs in their code with that.
+
+**Mat Ryer:** Oh...
+
+**Johnny Boursiquot:** Or they assume it's gonna be handled at the edge of the network or something, by the provider or something.
+
+**Roberto Clapis:** One of my worst moments - I was presenting a vulnerability to the client and I showed that with a single packet I could make it backend-generated several tens of thousands of packets, and take it down. And they said "Well, it only goes down..."
+
+**Tom Steele:** It comes right back up?
+
+**Roberto Clapis:** Yes... \[laughter\] Why did you build this? If it's fine if it goes down, why is it up anyways? Turn it off."
+
+**Mat Ryer:** I thought they were gonna say "Don't send that bit." Don't send that byte, or whatever it is.
+
+**Tom Steele:** Well, why would anyone do that?
+
+**Mat Ryer:** Yeah, no one's gonna do that. Tell us some other things that people say that's infuriating when it comes to security? Because honestly, it's good to hear... And it's actually quite useful for people to learn.
+
+**Roberto Clapis:** I think I'm blessed at Atredis... The customers we work with - they actually get it, and I don't get too much pushback anymore. I think we're evolving, and that's really a nice thing to see. I think we're evolving to where security and software development aren't as adversarial as they used to be. I think we all want the same goals, and at least for the most part the developers I work with and talk to a lot - they really do care, and they wanna fix things, and they do take things seriously... But yeah, so I don't really see much of this "Well, why would anyone do that?" type of thing anymore.
+
+**Johnny Boursiquot:** So it doesn't take much convincing then to tell people "Hey, because of that code you wrote this way, you have a problem." You don't have to prove to them that they have a problem.
+
+**Tom Steele:** Sometimes... That's another debate, too. We call it illustrating impact. If you're describing a problem, versus having an actual proof of concept, proof of concept will be better.
+
+**Johnny Boursiquot:** I'm gonna borrow that, "illustrative impact."
+
+**Tom Steele:** Illustrating impact, yeah.
+
+**Johnny Boursiquot:** Nice... Nice.
+
+**Roberto Clapis:** I've had different experiences, probably with different clients... Because the most questions that I got was "Yeah, and then what?" You just prove that you can run arbitrary code on the server, and the answer is "Yeah, but then what?" You're like, "No, that's it... It doesn't go further than this." \[laughter\]
+
+**Mat Ryer:** \[56:02\] Is that not bad enough...?
+
+**Roberto Clapis:** Yeah, exactly. It's like, "But what is it that you want? It makes you coffee, but it's made with milk. What do you want from me?!" \[laughter\]
+
+**Mat Ryer:** Yeah, I'm looking forward to -- I'm gonna get the book. I'm really looking forward to number seven, the abusing databases one. I don't know if I'm gonna be able to learn anything. I've spent lots of years abusing databases... I've screamed "Blue murder" to Postgres, for example.
+
+**Tom Steele:** Yeah, I'm working with the publisher for getting a discount code for everybody too, so...
+
+**Mat Ryer:** Oh, that would be sweet. If we pull that off, we'll put it in the show notes for everyone. Nice one.
+
+**Roberto Clapis:** Isn't it the other way around, that the database abused you, in that case?
+
+**Mat Ryer:** Yeah, actually that's a good point. That's why I want revenge.
+
+**Johnny Boursiquot:** Yeah, it's the database's fault.
+
+**Mat Ryer:** Yeah. But what sorts of things...? I mean, obviously, a SQL injection and stuff like that is kind of well known, I think... But are there other things that you have to look out for?
+
+**Tom Steele:** Look out for as far as databases and security?
+
+**Mat Ryer:** Yeah. Are there any common gotchas around databases that people--
+
+**Tom Steele:** Don't store your encryption keys alongside your cipher's text.
+
+**Johnny Boursiquot:** Nice.
+
+**Roberto Clapis:** And do not expose that to the internet. Those are two steps you don't have to make, especially together.
+
+**Tom Steele:** Yeah. Don't just do everything as root. Use permissions, and use the least amount of privileges you need. Don't store your encryption keys in the database... Yeah. Also, validate the signed ciphertext that you store in the database, too. Don't assume that just because it's in the database it hasn't been tainted with.
+
+**Mat Ryer:** That's a good point.
+
+**Tom Steele:** Yeah.
+
+**Mat Ryer:** Another tip I like to do is I always have a field called Password, that I just put fake stuff in and I just don't use it. So if anyone ever does look in the database... \[laughter\]
+
+**Johnny Boursiquot:** It's like a honeypot.
+
+**Tom Steele:** Yeah. Don't encrypt passwords, hash them. Don't use a secret key to encrypt the passwords.
+
+**Mat Ryer:** Yeah. I always get worried when a web service -- if I've forgotten my password and then they just email me my password. They must have been storing it in some way that they can --
+
+**Tom Steele:** Well, it's also great when you dump a database and then you see 500 of the hashes, of the passwords as the same thing. Then you know they've got a default password, right?
+
+**Mat Ryer:** Ooh... Monkey.
+
+**Roberto Clapis:** Mat, you should put "Nice try" in the password column, instead of the random stuff.
+
+**Mat Ryer:** I feel like that's just gonna make the attackers angry though...
+
+**Johnny Boursiquot:** Yeah, that's an invitation.
+
+**Tom Steele:** I'll give you guys a good security bug that everyone probably wants to go check their code out for...
+
+**Mat Ryer:** Oh, yes...
+
+**Tom Steele:** Maybe you guys can answer - so when you're gonna email a password reset email to a user, how do you get the domain name that you're gonna email the link from?
+
+**Mat Ryer:** \[58:49\] Oh... So I assume you don't just get it from the email address...
+
+**Tom Steele:** Let's say the password reset - you're gonna take in the user's email address, and then you're gonna generate a link to your domain for your application with a reset token. Let's say you're securely generating the token and everything's good there. How do you decide how to generate the URL?
+
+**Johnny Boursiquot:** That's some sort of configuration in your application, right?
+
+**Tom Steele:** Yeah, for sure.
+
+**Johnny Boursiquot:** Maybe you pull that from a database, or from some sort of YAML file that gets pulled, or at runtime you read it and then Boom!
+
+**Tom Steele:** Yeah, that's the correct way.
+
+**Mat Ryer:** Oh, you nailed it.
+
+**Tom Steele:** What we see a lot is people using the Host header.
+
+**Johnny Boursiquot:** Oh, bad move.
+
+**Tom Steele:** That way it works in Test and in Prod.
+
+**Johnny Boursiquot:** Yup.
+
+**Roberto Clapis:** What!? Ahh...
+
+**Mat Ryer:** And what's the problem with that?
+
+**Tom Steele:** That the attacker can control the Host header, and then it generates a link to their domain. So the user clicks that.
+
+**Johnny Boursiquot:** Hey Mat, what's the host name to your new app you're working on? I'd like to try out some of these things from the book...
+
+**Mat Ryer:** Yes, I'm not gonna--
+
+**Tom Steele:** The book mostly focuses on like once you've got access to a database, what do you do?
+
+**Johnny Boursiquot:** Pilfer.
+
+**Tom Steele:** Yeah, you pilfer. That's right. You find the ciphertexts, and you find the banking data, things like that.
+
+**Mat Ryer:** Yeah... It's good stuff in it I bet. This is it any final thoughts? We're approaching the end, but we've got a bit of time...
+
+**Roberto Clapis:** I have an unpopular opinion.
+
+**Johnny Boursiquot:** Another one?! Roberto... You're on fire!
+
+**Mat Ryer:** Roberto, let's hear yours.
+
+**Roberto Clapis:** Go would benefit from enums. That's it, I want enums.
+
+**Mat Ryer:** Okay. So you want to be able to specify a type where there's only a set number of allowed values compiler-checked.
+
+**Roberto Clapis:** And I want the switch to be exhaustive. Because in many Go applications I've found a way to bypass security by just specifying four, in an enum that only expected three values... So you end up in the default, and the default does other things. So enums would be a big security benefit, in my opinion... But that's my opinion.
+
+**Mat Ryer:** It's a good one.
+
+**Johnny Boursiquot:** Yeah, that's not unpopular. I think that's fair.
+
+**Mat Ryer:** Yeah, because you can do them; you create a type and then you use the Iota to const block. So it's possible to do, but it's not the same, is it?
+
+**Roberto Clapis:** Yeah, and then you parse the form, you take the integer, you convert it, and where has your type gone?
+
+**Johnny Boursiquot:** Right. So you want proper enum support.
+
+**Roberto Clapis:** Yup.
+
+**Mat Ryer:** Proper enums... Yeah, fair enough. That's a good one. That could be popular or unpopular, I don't know. I'm afraid we just don't know... But we'll find out for you, Roberto, and let you know... Shall we? \[laughter\]
+
+**Roberto Clapis:** Thanks. Put it in the links.
+
+**Mat Ryer:** Yeah, we'll put it in the links. Okay, well I think that's all the time we have today. Thank you so much to our guests, Roberto Clapis and Tom Steele... And Tom's book is available... Roberto, are you selling anything?
+
+**Roberto Clapis:** No...
+
+**Mat Ryer:** Okay, Roberto's not selling anything, so that's fine... No, but seriously, Tom, where can we get your book if we're interested?
+
+**Tom Steele:** I think the best is No Starch Press.
+
+**Mat Ryer:** Cool.
+
+**Tom Steele:** Or it's on Amazon too, if you wanna go that route.
+
+**Mat Ryer:** And there will be a link in the show notes.
+
+**Roberto Clapis:** NoStarch.com/blackhatgo.
+
+**Tom Steele:** And I should be getting a discount code for everyone, for the eBook at least.
+
+**Mat Ryer:** Well, that would be great if we could pull that off.
+
+**Johnny Boursiquot:** Nice.
+
+**Mat Ryer:** That'd be awesome, Tom. Thank you so much. Well, that's all the time we have today. Thanks for listening, and we'll see you next week.
+
+**Break:** \[01:02:46.28\]
+
+**Jerod Santo:** I went back and watched The Neverending Story. It doesn't hold up, guys. It doesn't hold up at all.
+
+**Mat Ryer:** What do you mean? Which bit?
+
+**Tom Steele:** I haven't seen that in a very long time...
+
+**Jerod Santo:** The movie. The movie. \[laughs\]
+
+**Mat Ryer:** I thought it was ad time...
+
+**Johnny Boursiquot:** The graphics, you mean?
+
+**Jerod Santo:** The movie doesn't hold up.
+
+**Mat Ryer:** I thought it was logically sound.
+
+**Roberto Clapis:** I think the main issue is that it ends... \[laughter\]
+
+**Jerod Santo:** You never want it to end...
+
+**Johnny Boursiquot:** You liked it that much?
+
+**Jerod Santo:** This is just false advertising, right?
+
+**Roberto Clapis:** No, it's just that it's fake.
+
+**Mat Ryer:** Fake news. It's clickbait, ain't it?
+
+**Jerod Santo:** My issue is that it didn't end fast enough.
+
+**Johnny Boursiquot:** A flying dog... What kind of nonsense is that?
+
+**Mat Ryer:** Yeah, it's a lovedragon, but...
+
+**Jerod Santo:** \[laughs\] Mat's personally offended at this point. \[laughter\] He's like "Come on, guys..."
+
+**Mat Ryer:** It's my childhood.
+
+**Jerod Santo:** I mean, it was awesome during my childhood, so I was excited to watch it with my kids... I'm like "Come on, kids, this is gonna be amazing." And then I sat in there and I was like "What is going on...?"
+
+**Mat Ryer:** Yeah, but it can't compare to the Avengers stuff.
+
+**Tom Steele:** No, not at all.
+
+**Mat Ryer:** If you have kids, you have to show it to them in real-time, as you experienced it. You have to rate-limit. We could just do an episode on the Neverending Story, but we're gonna need a bit more time. Dylan on Slack just said that Artax drowning in the swamp was the saddest moment in movie history.
+
+**Jerod Santo:** In movie history?
+
+**Mat Ryer:** It's almost as sad as when the Terminator slowly goes down into that lava...
+
+**Johnny Boursiquot:** With the thumbs up.
+
+**Jerod Santo:** The thumbs up is hopeful...
+
+**Mat Ryer:** Yeah, but it gets you, doesn't it? Because you think "Aww... It's really sad that that terminator is dead. That machine sent in the future to murder people. I feel really bad now that it's dead." \[laughter\] So it works.

Building desktop apps with Go + web tech_transcript.txt

@@ -0,0 +1,375 @@
+**Mat Ryer:** Hello, and welcome to Go Time! I'm Mat Ryer. About six months ago - or, I suppose, three months, if you listen to this on 2x speed - we chatted with Elias about Gio, which is a way to draw graphical interfaces and build desktop apps in Go... And there's also the Fyne project, which gives you tools to build native UIs. But they kind of have a different thing to learn in order to get to use them, so we're gonna look today at an alternative approach, which probably will use a lot of existing skills that are already around, which is where we can mix the desktop and web technologies using the Wails project. That's what we're gonna talk about today.
+
+Joining me on this quest, Jon Calhoun. Hello, Jon.
+
+**Jon Calhoun:** Hey, Mat. How are you?
+
+**Mat Ryer:** I'm good, mate. Have you gotten all your jobs done that you were telling me about?
+
+**Jon Calhoun:** I think I have most things done that I need to for today.
+
+**Mat Ryer:** Oh, you're so good. It makes me feel bad, because I've achieved very little today. And joining us also, only the creator of Wails, Lea Anthony. Hello, Lea.
+
+**Lea Anthony:** Good morning, afternoon and evening.
+
+**Mat Ryer:** Welcome to Go Time, thanks for joining us.
+
+**Lea Anthony:** Thanks for having me.
+
+**Mat Ryer:** It's a pleasure. So maybe we could just start at the beginning - when we talk about desktop apps in Go, what do we mean, and why is it such an interesting subject, compared to other types of things that you can build in Go?
+
+**Lea Anthony:** I think originally, when I first started with Go, I went on the same path as everybody else, making CLI apps, making servers... I'd come from a Node world at the time, so I was kind of doing similar things on the backend. One of the things that I thought about when I was learning Go was "How can I make visual things?" I love visual things, so the obvious thing to do there is to create a server and to hook up your browser to it and to see things visually that way.
+
+\[04:10\] I sort of went along with that approach, I thought "What should I do? What sort of application should I make as a test?" I was using restic at the time to get my backups, and I thought "That's quite a good idea." It's fairly static, there's some information that you can get from the application... It didn't really have a good library - I still don't think it does, actually - so there's a bit of shelling and running code, and trying to parse it up, and all that stuff... But what I discovered pretty quickly was that what you could actually do through a web browser was by its very nature fairly restricted, and I kind of wanted to still use all of that funky JavaScript stuff on the frontend, or the nice libraries and tools that you could use to do visual stuff, but to hook it up to my Go application.
+
+So like I said, I kind of hit that sort of sandbox restriction pretty quickly, and I thought "How can I marry the two things that I now really love? ...this amazing, traditionally backend language, with a very rich ecosystem and frontend technologies?" That's where it kind of started... And I didn't really get my first break on that until I found the WebView project. That allows you to present web frontend using \[unintelligible 00:05:30.28\] I started looking at that, and I thought "That's cool. That'd be really great if I could combine this with Go." They do allow you to do that, but there was a lot of technical stuff you had to deal with, of message parsing, and all that stuff... So that's really kind of like the birth of the project there - how do I make this easy for other people, and myself, obviously?
+
+**Mat Ryer:** Yeah, that's cool then. So the idea is you wanna be able to use web technologies - JavaScript, HTML - and even web frameworks... And I noticed that by default, I think Wails comes with React, right? Or is it Vue?
+
+**Lea Anthony:** It comes with both. One of the key things to start with -- and I noticed this at Rails, and it is a play on words on that... I actually called it that basically because I wanted this quick scaffolding way of building up an application, and Rails was the original to do that... So it was kind of a play on words - web view Rails... So it takes a kind of similar approach. It comes with a CLI, so you use that to scaffold out your project; it comes with a bunch of templates, so you can use Vue, React, Angular... There's a couple of Vue templates there that use Beautify to \[unintelligible 00:06:52.15\] and there's also a vanilla JavaScript one you can use as well.
+
+**Mat Ryer:** Yeah, that's cool then. Because the point is you're gonna serve those static assets, essentially, whatever the frontend technology is gonna do to build -- essentially, it's gonna build a JavaScript, a CSS maybe, and some markup. And then Wails will serve that, won't it, into its own kind of frontend, which is web-based. Does that sum it up?
+
+**Lea Anthony:** That's correct, yeah. Absolutely. So there's essentially two parts to your project - there's your Go part, which is your backend, and there's the web frontend. And they're sort of fairly loosely coupled. The concerns of the frontend are entirely yours; you can use whatever you want. At the end of the day, you basically compile down to your classic disk directory, and you have your bundles, and you essentially end up with 2 or 3 assets. And the way that's bound together is through basically that library that Wails provides.
+
+You create your application, and then you, as part of the creation of that object, of that struct, you then give it a link to your assets essentially, and it will serve them for you. The way that it does that - it actually uses a packer; originally, I used Mark Bates' packer, and then he deprecated it, and I was very upset... So I took on the challenge and wrote something very similar, because I really liked the way it worked... And they're essentially just strings. You just pass a string to the object and it takes those and serves those.
+
+**Mat Ryer:** \[08:26\] So the reason you do that then is so that when you ship the app, you don't have to ship alongside it those assets. They get bundled inside because of a build step inside the binary, don't they?
+
+**Lea Anthony:** That's absolutely right, yeah.
+
+**Mat Ryer:** So you still only have to ship one thing.
+
+**Lea Anthony:** That's correct. There's no extra thing that you have to supply with it. The web view framework uses the web renderer that sits on your system, so you don't need to supply that. And there's pros and cons with that.
+
+**Mat Ryer:** Yeah. Well, we can get into that. So what sort of apps can you build with Wails then?
+
+**Lea Anthony:** I mean, that's like saying "What kind of websites can you build?"
+
+**Mat Ryer:** What kind of websites can you build?
+
+**Lea Anthony:** \[laughs\] I guess one of the key targets for this application would be somebody who wants the power of Go, somebody who wants to perhaps do low-level stuff, USB stuff... I've got a couple of projects that I noticed that are working in the cryptocurrency space, so there's a lot of \[unintelligible 00:09:36.27\] that happens... And it's for people who want to just present that sort of stuff visually, where you perhaps otherwise maybe use a web browser to connect to, but that's doesn't feel quite right... And it's something that you wanna interact with in real time, but still have that power of go. That's really your target.
+
+There's a keyboard company in Japan called ergodox-ez. They create these amazing keyboards, the ones that have like a million different parts and they're super-ergonomic... One of their main developers contacted me a while back, and he was using WebView to do that firmware flasher. He's recently ported their application called Wally, which is their firmware flasher - he's ported it over to Wails. I've worked pretty closely with him; we didn't have to do much. He gave us some great feedback, and... Yeah, I was really happy about that, because it really shows the depth of application you could do.
+
+**Mat Ryer:** Yeah, it's interesting, because we do think of Go a lot running as server with HTTP interfaces and thing... So when you start to then think "Well, now this is gonna be running in its entirety on a single machine for a person who is running it in the desktop context", it's kind of a different way of thinking, in some way... But does it end up being quite similar, because you have this frontend that's communicating somehow with the backend through some kind of remote procedure calls. That becomes the way that the backend and the frontend communicate, is that right?
+
+**Lea Anthony:** Yeah, that's right. The difference is that that mechanism there - and this was a key part, the design of the project... That mechanism for communication is completely hidden from you. So instead of having to worry about messages, requests, parameters, all that sort of stuff, you actually just call functions. The IPC mechanism, the way that it works is that you essentially bind Go functions or struct methods to your application, and they get presented at the frontend through a JavaScript endpoint. So it'll be windows.backend. and then it'll be a package name, because you have to qualify the methods in your structs, and then just the name of the Go method that you bound. That receives the parameters that would be sent through to your backend code, your backend code will run, and the result of that would be sent back to JavaScript. That mapping was pretty interesting, and the way that it was done was through promises.
+
+\[12:24\] What I sort of figured out early out was that if you had a Go method that returned a value and an error, that's essentially the same as a promise. And so what happens is if you send a value back and the error is null, then the promise resolves. If you send an error back, then the promise rejects. So you're able to easily deal with the backend calling the backend code as if it's part of your application... Which it is, obviously, but part of your frontend application.
+
+**Mat Ryer:** Yeah. So does it preserve types as well then, if you're passing strings and integers? Can you call them in your JavaScript code using those types and they turn up as those types in the Go side?
+
+**Lea Anthony:** Yes. Yes, is the answer, pretty much. It mostly does type conversions. I use one of the low-level Go functions to do that type conversion. So on the whole, yes. Structs appear as JSON in JavaScript. When you send structs down to Go, they appear as a standard map string interface. It can be tricky, but it deals with the 95% case, so that's kind of handy.
+
+**Mat Ryer:** Yeah, that's so cool. How does it do that then? From the user's point of view, you're just calling JavaScript asynchronous functions, and there's a promise returned, and then they either resolve or they get rejected... So that makes sense from that point of view, but how does it actually communicate between the web view component and the Go code? What's the mechanism that's being used to make that work?
+
+**Lea Anthony:** So web views provide the ability to call native functions. You can set up an endpoint in webview it's window.external.invoke and whatever you pass to that ends up on the listening end. So that method actually -- what I had to do early on was to kind of wrap that, and supply what you say, an IPC mechanism... And I chose to use JSON messages to capture that information, "Am I making a function call or am I trying to log? Am I trying to use the runtime?" So I've got a bunch of standard messages, they get sent through that mechanism where I have a listener in Go, a sort of dispatcher; it receives the message, works out what kind it is, and then sends it to various subsystems, based on the need. Once that's processed, the same thing happens, but in reverse.
+
+**Jon Calhoun:** So when you're building things like this, what types of limitations do you see? A lot of times, when you see a big company and they're gonna build something for Windows or for macOS, they go out and look at .NET, or they look at Swift, and Xcode and all those tools... Presumably with this there are some limitations. Maybe they're not as far as what you can do with the system, but have you noticed any limitations or things like that, that would cause somebody to lean towards going with Xcode and Swift versus this? Or what types of projects, I guess, would you see going each way?
+
+**Lea Anthony:** Yeah, I guess it just depends on the original project. The difference between running an application in native UI versus an application in a browser versus an application in WebView, which is even more limited than a browser - you're gonna hit some limitations somewhere, based on the use case. The obvious one is complexity on a native UI perspective. "How do I make that work on different platforms?" There are some approaches that you can take to do that.
+
+\[16:03\] When you do a server app, obviously, you have that disconnect between the frontend and the backend... And in the browser as well you have things like local storage, which you don't actually have in WebView. So the limitation I'd say of writing a Wails app would be twofold. One is it isn't a full browser, which makes it good, in a way; it's slimline it's making good use of the resources... But you also don't necessarily know which libraries would necessarily work. There are so many libraries out there; if you use one that uses something that's native to a browser, then you're gonna hit that limitation. That's one.
+
+The other one would be that Wails uses the native renderer on your system. There's pros and cons with that. The con is that on Windows it's still basically IE11, so you'd have to deal with that. And there's a lot of libraries that kind of get around. I've not really hit that problem, but some people do.
+
+**Mat Ryer:** It probably would have been more of a problem in the past, like when IE6 was around, for example, when you had to basically write two versions of every website.
+
+**Jon Calhoun:** You have like two CSS files. It's like "Here's for you people in IE, and here's for everybody else."
+
+**Mat Ryer:** Yeah. Here's for all the cool people, and then 80% of the population of the planet is using IE6, so we still have to support that...
+
+**Lea Anthony:** Yeah, that's right.
+
+**Mat Ryer:** But Lea, you mentioned something interesting about multiple platforms... That's something we should actually highlight a little bit, because you literally can build one frontend, and you've got your Go code, and since Go code can be built for different targets, you can build Wails apps for different architectures - Windows, macOS and others - and it's the same code for all of them, right?
+
+**Lea Anthony:** That's absolutely correct, yeah.
+
+**Mat Ryer:** See, that's pretty massive.
+
+**Lea Anthony:** Yeah, that is. Until fairly recently, the limitation around that was that you could only build the platform that you were actually compiling it on... And there's a couple of reasons for that. It uses cgo, so you have that complication... But fairly recently we had an amazing contribution to the project which allowed you to cross-compile using Docker. A big thanks to Travis as well, who works on the project; he spent quite a lot of time in getting that working really well.
+
+**Mat Ryer:** Is that TravisCI?
+
+**Lea Anthony:** \[laughs\] No... Travis McLane, he works on the project.
+
+**Mat Ryer:** Alright.
+
+**Lea Anthony:** Yeah, he managed to iron it out a bit, polish it off, and it works really well. You have your initial XGo download, which is fairly large - it's around 7 gigs, which to some people is a lot, to some others it isn't... But once you have it, you can compile multiple-platform targets on the same computer.
+
+**Mat Ryer:** And speaking of TravisCI - you could probably easily build that into a continuous integration environment and have your build system build and deploy these apps, right? Using that image.
+
+**Lea Anthony:** Yeah. One of the things moving forward - I wouldn't mind having a look at GoReleaser and seeing how I could integrate with that. Because that seems like a really good fit.
+
+**Mat Ryer:** Yeah, I used that recently, actually, and it was really good. Basically, you'd tag your repo with a semver tag, and then you'd just run GoReleaser. It creates a release, uploads it to GitHub for you, it writes the changelog... And there's a lot more you can make it do, but by default it sort of does that. It would be great, something like that...
+
+But the nice thing, I think, that makes this a really cool project is that however you build it, it's one codebase that's running everywhere. So you can really get a big saving. And if you think about all the things you can do in Go code, like accessing the file system, presumably you can access the file system in Wails apps, right?
+
+**Lea Anthony:** \[19:57\] Absolutely. You can do whatever you would like to do on your Go side. There's no limitations on that. What I have done is provided a runtime which operates both in the JavaScript land and the Go end, so some of that stuff is wrapped up for you. Maybe you wanna show a file select dialogue, so you can call this runtime command in Go, and Wails version - which we'll talk about in a bit - you can also do that in the JavaScript land. So you can essentially call a function, open a dialogue, get the user to select something, and it just appears as a string in Go. So it's almost like a synchronous operation in your go where you're saying "I wanna get a file name", and it goes away and does it for you.
+
+**Mat Ryer:** Is it like a file -- do you get literally the file scheme on that when it turns up in Go then? And is it just a path to a local file?
+
+**Lea Anthony:** It's just a path to the local file, and then you can use open and just read in.
+
+**Mat Ryer:** So it doesn't upload through the WebView. You're just selecting the string. I see.
+
+**Lea Anthony:** Yeah, it's just a means of getting a string into Go.
+
+**Mat Ryer:** I bet there's a big range of different things that you can suddenly do, which you wouldn't do probably in the server context... That's why I mean it's quite exciting when you start to think of building desktop apps like this in Go. Are there any other cool use cases that you've come across?
+
+**Lea Anthony:** In terms of actually the stuff that you can build, or in terms of the tooling that's provided so that you can build them?
+
+**Mat Ryer:** Either.
+
+**Lea Anthony:** One of the things that I'm pretty proud of in the project is early on I realized that the IPC mechanism, the way that you communicate between the frontend and the backend - it's not dissimilar to WebSockets. So what I started thinking early on - because I wanted the project to be very developer-focused, like "How do I make it easy for the developer?" One of the things that I kind of hit early on was the integrated debugger in the tooling that comes with WebView is actually pretty limited compared to the things that you're used to. In Chrome, the DevTools are just incredible. And that's what people wanna use; they wanna use the things that they're used to. So what I did was to create a compile flag... Because you don't compile using go build, you use wails build. And the reason for that is because there's a bunch of stuff that it needs to do, including sorting dependencies, and packaging, and all kinds of stuff... So I've wrapped that all up, and you can pass a flag which essentially creates a headless version of your backend code, and sets up an actual web server inside your code.
+
+What you can do then is if you run -- it tells you what command to run, you run the command, and the frontend gets served up by its native tools. In Vue I think it uses webpack-serve, or one of these runners. The frontend actually has some code that's injected into it, which connects to your backend code. So now you're running in the browser, but you're still connected to your backend. So you can open up the DevTools and you can start just calling your Go code from Chrome... And that becomes incredibly powerful, because you can all of a sudden start using your Vue tools, your React dev tools, all that kind of stuff.
+
+The drawback of that is people get used to things working in the browser... So sometimes you get tickets raised that say "I don't understand - it works in serve mode, but it doesn't work in compile mode." And that's a tricky line to walk, because you do wanna provide the best tooling, but you also know that it's not entirely 100% like for like. It's pretty similar...
+
+**Mat Ryer:** Yeah... It's kind of a trade-off.
+
+**Lea Anthony:** Yes.
+
+**Mat Ryer:** Yeah, I see what you mean. That's exactly it. The web view is more limited. And like you say, it comes from the operating system, so I guess on the Mac it would be Safari, would it, essentially?
+
+**Lea Anthony:** It uses WebKit.
+
+**Mat Ryer:** Alright, yeah... Which Safari uses, doesn't it?
+
+**Lea Anthony:** That's right.
+
+**Mat Ryer:** \[24:01\] Hm... And you mentioned events then as well, and that's quite an interesting thing, when we think about Go, and we have channels, and we have different ways of running things concurrently and communicating events... Can you also get those events to fire in the frontend part of the app? Like, something happens in the backend; maybe if it's gonna walk the filesystem, it's gonna take a long time... Maybe there's an event to say "I've just completed." How does that get into the frontend?
+
+**Lea Anthony:** Good question. So as part of the runtime there's a unified events system. On the Go side you have emit and you have on, and it's the same in the frontend, too. And whenever you emit a message, you do it in the standard JavaScript way and it has a name, and it has that optional data along with it. If you fire that from Go, you can read it in Go and you can try to run it in JavaScript. And the reverse is true. So you could, say, have a JSON string, or whatever kind of data in the frontend, you could emit an event by a name, and receive that in Go. It was quite interesting early on, thinking about the differences between that and a call.
+
+So yeah, that's where unified eventing really opens things up, because early on I hooked up jmeter, a web load testing tool. It's a library, essentially; it's just a library. And I wanted to put a frontend to that as a test and work out "How is this gonna work? How do I visualize this that it's doing?" Because obviously, it's gonna do a whole bunch of stuff in the background, and I wanna try and view that.
+
+So you have the obvious thing of "Here's my URL. This is how long I want you to run, and these many concurrent threads..." And one of the things that the library did - and probably still does - is provide a callback so that every 1,000 events it just returns you some information. And in Go, that's a struct. It's a struct with all of that information in it... And what I did was I basically spun that off in a goroutine; so you have that running in the background, and every thousand requests or whatever you get this struct. And all I did was just emit it; so I just passed the struct into an emit call. On the frontend I had a listener; when it received that information -- initially, I just printed it out in the console. What was amazing was because the developers had put JSON tags in their struct, all of a sudden I had this really rich JSON object, with all of this information in it. And because these days components are very much driven by JSON state, you can just pop that straight into a state store and all of a sudden your frontend reacts to everything that's happening in your Go library... And that blew my mind, I've gotta say. It was just incredible to see.
+
+**Mat Ryer:** Yeah, because you haven't had to do much work at all there for that quite powerful thing, the event-driven live updates occurring in your desktop app.
+
+**Lea Anthony:** That's right.
+
+**Mat Ryer:** It is quite exciting.
+
+**Break:** \[27:24\]
+
+**Jon Calhoun:** So as you were developing all this, it sounds like you had to come up with a lot of unique solutions to allow the two to communicate, and allow things to work the way that it makes the most sense... With things like WebAssembly coming along, do you think they would eventually make it so that this type of stuff would be easier to do, in almost any language? Or are there still gonna be big limitations?
+
+**Lea Anthony:** Yeah, I don't really know too much about WebAssembly, I've gotta say... I've followed its journey in the early days, but I don't really know how it would impact things... I'm pretty excited that the technology exists. I think it's gonna be great. A part of my brain things we're sort of going back to the Flash days, as well... I don't really know why, but there's something in my head about that. But I do think it's a great opportunity.
+
+In terms of the impact, I genuinely don't know. I think we're likely to see more and more WebAssembly targets, assemblies \[unintelligible 00:29:40.25\] kind of make sense... And maybe this project is kind of a stepping stone towards that sort of concept of "You write once and run many", which is sometimes that we've always tried to achieve in different ways.
+
+**Jon Calhoun:** Yeah, I should definitely say I don't know a lot about WebAssembly, but the idea of "Write it, compile it into JavaScript and run it wherever" - that's appealing, in the sense that if I can just write regular code and it works, that's real nice. And we're all spoiled with Go, because it works pretty much everywhere... But when you have languages where that doesn't work, or anybody who's tried to ship a product that they wrote in Python or some other language, and you ship it to somebody and they're like "Well, it's not working", and it's because their macOS has a really old version of Python or something, and it's like "How do you handle these situations?"
+
+I think that was one of the things that always made delivering any sort of software to a desktop user that wasn't written in their native language, which just could be a nightmare.
+
+**Lea Anthony:** Yeah, that's right.
+
+**Jon Calhoun:** I'm assuming this is sort of similar to how Electron works, but I don't know enough about Electron to say for sure. Is it comparable, in some ways?
+
+**Lea Anthony:** Many people do draw that comparison, and I'm very reluctant to draw that comparison. I would say that the two projects are different; they've got different target audiences, and they kind of work a bit differently. The only overlap is that they essentially achieve a similar thing, which is a desktop application with web technologies. That's pretty much where I draw the comparison there.
+
+**Jon Calhoun:** Okay. So do you know enough about Electron that I can ask questions to sort of flesh out some of those differences?
+
+**Lea Anthony:** Yeah, sure. I know a little bit, but I'm definitely not the world's expert on it.
+
+**Jon Calhoun:** So is Electron using web views?
+
+**Lea Anthony:** No, it uses an embedded Chrome instance.
+
+**Jon Calhoun:** Okay, so it uses something entirely different. So that would be one of those cases where it works on Chrome, but it could solve that type of problem, but it probably comes with its own set of challenges.
+
+**Lea Anthony:** Resources, mainly.
+
+**Jon Calhoun:** Okay.
+
+**Lea Anthony:** It uses a lot resources.
+
+**Jon Calhoun:** Okay, so from that point if view we start to step away; it's not even using web views, it's just something built entirely differently.
+
+**Lea Anthony:** Yeah, it's almost like bundling Chrome with your application, where Chrome only executes your code, is my understanding. But it also gives you a very easy one single target problem... So as long as it works in Chrome, you get it.
+
+**Jon Calhoun:** \[32:10\] So when you're running in WebView then, I assume you have the JavaScript running there... Do you run into some other limitations, like with browsers? One of the ones that comes to mind is cross-origin resource sharing; if you use the JavaScript to actually talk with an API, for some reason, which might not make sense with Wails, but I guess might be possible - not really sure... Well, first off, is that possible?
+
+**Lea Anthony:** Yeah. It's kind of interesting, because the frontend and backend - they have similar ways of doing things. So you can make an HTTP request from JavaScript, but you could also do it from Go. So you'll likely hit -- whatever limitation you hit on one end, you'd hit on the other. The cross-origin - I don't think you have that problem; I'm pretty sure you don't, because it's not really running into a server, and there's not a browser that's gonna guard you from that kind of thing.
+
+**Jon Calhoun:** I was gonna say, usually that's something browsers would prevent, and I wasn't sure if that itself was close enough to a browser that it had it or not, so I wasn't quite sure.
+
+**Lea Anthony:** Yeah, I don't believe so. I think \[unintelligible 00:33:08.11\]
+
+**Mat Ryer:** Well, even if you couldn't though, you can always just make it on the Go side, and then call a function, and...
+
+**Jon Calhoun:** Yeah.
+
+**Mat Ryer:** Probably that's how you'd want to do it anyway, I think - or at least I would - because frontend/backend does give you quite a nice boundary for testing. So it'd be a nice surface to test. If all your tests that were calling those public methods all passed, you'd have a level of confidence that the app itself is gonna work too, I suppose. So I would probably do that anyway; I would put as much of the logic as I could into the Go side, just because that's where I'm more comfortable. But it does let you do things like that - you could just make HTTP calls, you could make other kinds of connections that would be impossible in browsers, too... You could make bespoke binary protocols -- maybe some bespoke API or something, some service that you wanna consume... You do kind of solve that by having this environment that is both Go running natively, and also this web concept running right there, too.
+
+**Lea Anthony:** You make a great distinction there, and it's one of the recommendations of the project when you build your application. Think about your state and your business logic really should be living in Go, and your web frontend really should just be a Vue on that... Which sometimes is a concept that's different to people who have used a lot of web technology, and kind of dealing with state on the frontend. The idea of the eventing system is really just to keep those in sync... More of a -- you should probably just take that approach of pushing the state, rather than trying to deal with it in two places. I know that some projects do try and keep those in sync, so you might end up in a kind of split-brain situation... And yeah, my recommendation is just don't bother. Just push your state to the frontend and react to your state on the frontend in a visual way.
+
+**Jon Calhoun:** I think part of that might come from -- like, when you're building something with the web server, you have to worry about latency, especially if you have users who are in a low latency area... So they try to think "How can I keep this state and minimize some of that latency?" Whereas with something like this, where you're both running on the same server, there realistically shouldn't be much (if any) latency at all.
+
+When I was thinking about people making JavaScript calls, I was sort of thinking somebody who maybe has the JavaScript app and they're like "Can we make something that -- a self-hosted type version, or something like that?" And I could see them making that mistake of just trying to port the same thing over, when the two approaches - while the UI might be similar, but some of the how you do it might be something you'd wanna change up.
+
+**Lea Anthony:** Yeah, that's right.
+
+**Mat Ryer:** So what about a desktop icon? Is that how this binary looks? Because normally in Go when you build a binary, you get a default terminal-looking application, and if you double-click it, nothing happens; you have to run it in the terminal. What's the asset you get after you've done the Wails build?
+
+**Lea Anthony:** \[36:06\] So there's basically two build modes. Well, it's for two targets. There's a desktop application target, so a packaged version, which has an icon, and is something you can double click and run... And the other one is kind of like your standard Go output, which is just a terminal app that you run. You'd mainly use that for development. Like a Windows, for instance - it's good to run that in that mode, just so you can see the debug output, and all that kind of stuff.
+
+But Wails does come with the ability to pack down to the package. So on Windows, that's essentially -- you know, it generates the manifest file, it generates a different range of icons, and you'd have different sizes into an icon file... And then it compiles that up into the .exe that you would expect from any other build tool.
+
+For Mac it generates a .app directory structure, it creates the plist, it generates the icons... So it puts it together in a way that works on a Mac. So you don't need to package \[unintelligible 00:37:07.07\]
+
+**Mat Ryer:** And of course, the assets are all embedded in the Go binary, so you don't have to even put the assets into that package either, do you?
+
+**Lea Anthony:** Absolutely.
+
+**Mat Ryer:** Yeah.
+
+**Jon Calhoun:** So you mentioned that you have the wails build command, and you mentioned Packer, which I know from using it you have to use the packer build command... So I assume the wails build is in part to replicate that behavior of "Grab all the assets, turn them into byte slices and stick them in the binary." Are there other things going on behind the scenes when you're running Wails build? Can you talk about that a little bit?
+
+**Lea Anthony:** Yeah, sure. Basically, all of the things that it does is essentially -- there's a project.json file, which basically tells Wails how to build your project. So what command do I run to install my frontend dependencies? What command do I use to compile my frontend to the assets? Once it does those two commands, it then gets those assets, it runs an external command which is something that I created - and now wish I haven't - which packs up those assets and puts them inside a Go file, pretty much the way Packer works. It then compiles everything in Go land, including your packed assets, into a single binary.
+
+If you've decided to package it at that time, it will then consume through platform specifc packing, on windows I think it uses windres to compile that down to a single binary. On the Mac side it's a little bit more manual from a compile perspective, in that it just creates directories and puts things in various places... And that's pretty much what goes on.
+
+It's fairly customizable, because the actual system assets that it generates - it comes with a default manifest file for Windows, it comes with a default icon, for instance... And it leaves those in your project directory once it's built. So if you change them, then that's your opportunity to customize your builds.
+
+**Jon Calhoun:** So when you're embedding those assets - I know recently there was like an embed proposal for Go... Have you had a chance to check that out?
+
+**Lea Anthony:** Yeah, it's definitely something that I wish had been there before I decided to try and make a packer. That would have been good. There's many solutions to this problem, so...
+
+**Jon Calhoun:** Does that solution look like it would solve the needs you have then?
+
+**Lea Anthony:** Yeah, I doubt that there's too much involved in packing a binary asset into your application. I mean, it only does one thing; so long as it does that, I think we're good.
+
+**Mat Ryer:** \[39:38\] Yeah, I quite like the similarities... And Mark Bates was actually - or is working, or was, before this proposal, because this proposal may supersede it... But yeah, he was working on a new API that mirrored the OS package. So you basically open files, and you can stat them, and you use the existing API, essentially a copy of it... Except that it goes through this layer where they may be embedded in the binary, and you sort of can't tell.. And I think the go:draft proposal - and we'll post a link to it in the show notes - has that same kind of idea, which is nice, that you get to kind of use the learning you have already from existing APIs... That may well be cool. And the nice thing is Wails could adopt that and upgrade that, and it almost wouldn't change anything for anybody, would it? It's almost an internal piece which happens transparently, right?
+
+**Lea Anthony:** That's right. It's one of those things as well -- pretty much the Go philosophy is that frameworks are frowned upon. And I tend to agree, even though I've created a framework. I kind of tend to agree that it's--
+
+**Mat Ryer:** You frown upon it, even though you made it.
+
+**Lea Anthony:** Exactly, yeah.
+
+**Mat Ryer:** Do you do that with your kids as well?
+
+**Lea Anthony:** \[laughs\]
+
+**Mat Ryer:** You made them, but you're like--
+
+**Lea Anthony:** No... Only with \[unintelligible 00:40:56.26\] But there is a place for it, right? So yes, I firmly believe in composition, and having choices, so long as you're okay with the cost of that... Which is you have to do the integration, you have to do the plumbing, you have to deal with the idiosyncrasies of all the various components.
+
+And so whilst I've created this framework and there is some stuff that's hidden, like for compiling of the assets and the bundling of assets, you do have to ask the question "Do I care? Do I really wanna deal with that stuff? Is that something that an individual has the desire in?"
+
+**Mat Ryer:** Yeah, absolutely. And I think you're right; sometimes, if it's the right fit... And what's been valuable about this conversation is talking about the sorts of problems that Wails is gonna be good for. I think there are a class of applications, and certainly -- I mean, I've built an Electron app; I bought extra RAM and really wanted to put it through its paces, so I just built a basic Electron app. It was opinionated, and it did these things, and it also wasn't quite right. It wasn't what I was used to, and Wails kind of fit that perfectly; so I think there is a sweet spot for Wails, and that's why I'm glad we talked about this, and we got to hopefully get some people's eyes on it... Because the things you can build are kind of limitless, really. It's web technologies. We can even create images in Go using the image/draw packages you can actually build images, and things; so you'd be able to do even that kind of visualizations and stuff for Wails apps.
+
+So in a way, it's extremely powerful, and it doesn't do too much, that's the other thing. It solves core problems, and then not much else. As a developer, you still have to solve those problems in a way that makes sense for your particular case, or for the particular problems you're solving. So I think it finds that nice sweet spot, personally; that's why I recommend people play with it and see what they can build. It's very exciting, because doing Go, we spend a lot of time in servers, or in web contexts, or building things that are quite abstract... And to have something that's just an actual application - I think it's gonna be quite nostalgic and quite fun for people that haven't done it before.
+
+**Lea Anthony:** Yeah.
+
+**Break:** \[43:25\]
+
+**Mat Ryer:** It's time for Unpopular Opinions!
+
+**Jingle:** \[45:27\]
+
+**Mat Ryer:** Okay, do we have any unpopular opinions?
+
+**Lea Anthony:** I've spent some time thinking about what the difference was between unpopular and controversial, and I wasn't really sure where that line was...
+
+**Mat Ryer:** Okay... Well, I haven't got time to re-record the theme tune, but give us a controversial one, if it's controversial. We've never had to cut one of these, so if you can give us one that we have to cut, even better.
+
+**Jon Calhoun:** Controversial means that at least maybe half the audience doesn't like it, or they disagree with it in some way...
+
+**Mat Ryer:** Right.
+
+**Jon Calhoun:** It's not, like, popular by majority, I guess...
+
+**Lea Anthony:** Well, okay, so I guess half the audience potentially does not think that--
+
+**Mat Ryer:** It can't be half. It's gotta be more than half, slightly... Just 51%, can we just say...?
+
+**Lea Anthony:** \[laughs\]
+
+**Mat Ryer:** The pedantic shoe is on the other foot... I mean, hand...
+
+**Lea Anthony:** Unpopular opinion... Well, it's sometimes okay to mix Go and JavaScript code. Some people may think that that's not a great thing. I actually think that -- you know, use things to their strengths. Like we said about frameworks - I don't really have a problem with frameworks, I have a problem with frameworks that have opinions different to me. And I think that's the thing, isn't it - we all have our way of doing things. I guess that's the thing I love about Go - it doesn't really give you too many hard rules that you have to adhere to. Yeah, so I guess my unpopular opinion is everything is okay in whatever context it's okay in.
+
+**Mat Ryer:** That's very nice.
+
+**Jon Calhoun:** I was gonna say, everybody loves Rails until it doesn't work for them. And then they hate it. It's that one time when you need to configure something special, and it's like "No, we aren't gonna let you do that."
+
+**Mat Ryer:** Yeah, that's right. And it is a trade-off, isn't it? It's almost the universal thing that we face all the time, of kind of optimizing for read, or optimizing for write. We choose a tool like Rails, and with scaffolding we can quickly build real things, and we're gonna put things in the database... But then, of course, as soon as you wanna do something different, or something more, or whatever - you then have to put in the time to do that.
+
+And yeah, the problem is when you're fighting the framework. That's what I like about Wails - it doesn't force you to use a particular web technology. It's somewhat agnostic, so your teams will get to use expertise they already have in order to then build desktop apps... And I think that is quite a valuable thing in itself. It seems like if it was more opinionated, of course, then there's gonna be more cases where it hits up against things that it doesn't do naturally, or it doesn't do very well. So that's what I was saying, I think it has a nice sweet spot.
+
+**Lea Anthony:** \[48:30\] I think you've actually made me realize something. I've completely mis-sold this thing, haven't I? It's actually a tool, it's not really a framework; because frameworks imply there's lots of opinions, and there isn't. It's really just a tool to compile things, to build things... Yeah, it's more like a developer tool. Interesting.
+
+**Jon Calhoun:** A build tool is probably more accurate... Because we talked about Fyne I think the last time we looked at GUIs, and it's a lot more opinionated, I think, in how you -- like buttons, and things... I think all those things kind of look the same in that, if I recall correctly, and it has some basic scaffolding used to build things in the UI. And it gives you the ability to quickly throw something together and make it work on multiple operating systems, but there are limitations to that based on what they think things should look like, and what their version of a checkbox is, or something... So it goes both ways, and I think yours is probably as unopinionated as possible; that's probably what I'd call Wails.
+
+**Mat Ryer:** And tell us about the name, because it happens to also sound like a country, doesn't it?
+
+**Lea Anthony:** It does, Mat. It does. And you know what - it's really funny... Do you know sometimes you have a concept in your head, and you think "Oh, my context for this thing is this", and at some point something shifts and you're like "Oh my God, how could I not think about that?" This is a classic example of that. It was actually called Wails, just WebView Rails. That's all it was. No part of my brain at that time really associated it with the actual country I'm from... No idea why. \[laughter\] And then afterwards I thought, "Okay, yeah... Of course... That was probably there somewhere when I was naming this."
+
+When I was looking for a logo, I kind of developed one which was a bit of a homage to the Wales flag... So yeah, there is a link there, albeit unintentional.
+
+**Mat Ryer:** And if anybody doesn't know what the Welsh flag looks like, definitely google it. It's definitely one of the best flags there is. It's essentially a red dragon, which I think is really cool... And yeah, the Wails logo does invoke that; I think that's nice. It's nice to see a personal link this in a project, I think it's quite a cool thing. You don't see it too often.
+
+**Jon Calhoun:** So how did this not lead to confusion in your real life, in your outside the computer world, where you're like "Yeah, I work on Wails" or "I created Wails" and they're like "You didn't create the country, bud..." \[laughter\] They didn't get confused?
+
+**Lea Anthony:** There's a lot of confusion in my life, Jon.
+
+**Mat Ryer:** They probably don't think he means he's created the country Wales. That probably isn't the--
+
+**Jon Calhoun:** I mean, if he's out at the bar or something, and he's like "Yeah, I made this Wails--" They're like "Oh, this guys has just had too many. Let him be."
+
+**Lea Anthony:** \[laughs\]
+
+**Mat Ryer:** But you don't live there now, do you? You moved.
+
+**Lea Anthony:** I moved a long time ago, yes. And honestly, there's a wonderful word in Welsh called Hiraeth which is pretty much untranslatable in English, but it kind of means this longing for -- this nostalgic longing for things that were... And any Welsh person listening to this will potentially understand what that is.
+
+**Mat Ryer:** That's awesome. I feel like we should get some more Welsh on the show, because I think to people listening it sounds like Elvish, from Lord of the Rings. That's how I've heard people describe when they hear Welsh being spoken. That's like the close reference.
+
+**Lea Anthony:** The irony of that is that a certain type of Elvish is actually based on Welsh, so... That's why.
+
+**Mat Ryer:** Oh, that's why. Okay. Yeah, that makes sense.
+
+**Jon Calhoun:** It makes way more sense now.
+
+**Lea Anthony:** \[52:07\] Tolkien was obsessed with the language, which is kind of interesting. He was obsessed with languages in general, so...
+
+**Mat Ryer:** Yeah, yeah. He made the whole language, didn't he? Didn't he make Elvish, the language, and actually did all the work to build it into a real language?
+
+**Lea Anthony:** Yeah, I think that was his primary thing. That's what he really enjoyed doing. I don't know too much about it though.
+
+**Mat Ryer:** He did Lord of the Rings on the side.
+
+**Lea Anthony:** Yeah...
+
+**Jon Calhoun:** He had to pay the bills somehow...
+
+**Mat Ryer:** "Oh, I'm making up these bonkers languages and no one's gonna use them. I've gotta find a reason for people to use them. I'm gonna have to write a six-part epic."
+
+**Lea Anthony:** That's it. This D&D genre is not gonna invest itself, so...
+
+**Mat Ryer:** \[laughs\] Yeah. Magic.
+
+**Jon Calhoun:** He didn't go the open source route and be like "We're gonna provide support for the language first." That didn't work. He had to go some other route.
+
+**Mat Ryer:** Yeah, he had to build it into something he could sell.
+
+**Jon Calhoun:** I like talking about tools like this, because I feel like there is a lot of potential in the learning space too for them, where when you build things and you're learning, the terminal is kind of bland for anybody who's just getting into programming... But if you can combine it with Wails so that they actually see visually what's happening in their code, it can actually help them out. I feel like there's a lot of ways that you could make learning a lot easier, or at least different for different types of people, who learn in different ways.
+
+**Lea Anthony:** Yeah, you hit on a really good point there - combining Go code into your documentation or your teaching material is definitely possible. We know that that's possible already with the GoDocs stuff... But what this does is it actually allows you to embed components that are aware of Go code.
+
+One of the things I wondered about early on was this concept of a package... And it's something I'm still kind of fleshing out in my mind - that components historically are just JavaScript; but imagine if you could have a package which bundled your backend, your Go code, as well as your frontend. So it's kind of Go-aware frontend components. So you're able to compose your Wails applications using people's different components, and potentially have this concept of a package manager where you can pull those in. I think it would lead to some really interesting projects.
+
+**Mat Ryer:** Hm... We'll have to save that one for another day, I think. Okay, well that's all the time we have for today. Thank you so much to Lea Anthony for joining us. If you wanna build desktop apps with Go and web-based technologies, take a look at Wails. It's a genuine option, ready to go, it's past the v1 release, it's production-ready, you can build real things with it... So do it, and then tell us what you've built, please.
+
+Thank you very much, Jon. Always a pleasure. Lea, thanks for coming.
+
+**Lea Anthony:** Thanks for having me, it's been great.
+
+**Mat Ryer:** See you next time.

Challenges of distributed messaging systems_transcript.txt

@@ -0,0 +1,353 @@
+**Johnny Boursiquot:** Hello, and welcome back to Go Time. To those of you tuning in week after week to support us, thank you. And thanks to those of you joining us for the first time. Be sure to go back and listen to our previous episodes; lots of goodies in there.
+
+Did you know you can listen to this show live, as it's recorded, every Tuesday, 3 PM Eastern? Yup, that's right. Head on over to GoTime.fm for details. So a shout-out to our live listeners currently, and on the Go Time channel on Gopher Slack. Your participation surely makes for a better show.
+
+I'm Johnny Boursiquot, and joining me today are Jon Calhoun and Mat Ryer, whom you've come to know and hopefully love, as regular hosts on the show. Jon, Mat, do you wish to be loved?
+
+**Mat Ryer:** It'd be nice.
+
+**Johnny Boursiquot:** Yeah...
+
+**Jon Calhoun:** Are you still working on that, Mat?
+
+**Mat Ryer:** \[laughs\]
+
+**Jon Calhoun:** Here's the key - you have kids, and they'll love you unconditionally. At least for a little while. When they get older, I can't guarantee anything.
+
+**Johnny Boursiquot:** Yeah, it is challenging, as somebody who has a teenager, and some little ones... I can definitely see the drift happening. \[laughter\] In real time... Yeah, it's a surreal experience. So our special guest today, not to be forgotten, is Derek Collison.
+
+**Derek Collison:** Glad to be here.
+
+**Johnny Boursiquot:** Good stuff. So you are best-known for your work on NATS, a well-liked distributed messaging system written in Go. So yeah, definitely good to have you here. The topic for today's show is going to be the challenges of building distributed messaging systems, and we're hoping that you will be able to shed a lot of light on the topic for us.
+
+Before we start, I must say that when I was first getting into Go, and really starting to sort of say "Okay, you know what - Go seems like the next best thing for me, for my career, the next technology I should jump into, the next thing I should learn", I went to the first GopherCon back in 2014... And yours was probably the first or second talk of GopherCon. You blew me away with that talk. You were talking about NATS, and how Go was a good fit for these kinds of high-performance systems... And from then on I was hooked. I was like "Okay, this is really cool stuff."
+
+\[04:12\] You were showing all these little benchmarks, and everything else... I'm like "Okay, there is a lot to this. There's a lot to the language itself; the technology you were working on was sort of demonstrating, because we know NATS - and you'll probably get into that - was originally written in Ruby, and you switched to Go for a number of reasons, performance probably being one of the top ones... But basically this showed the power of the language and the kinds of things you could do with it... So why don't you introduce us to NATS, and then we'll get into -- this is not a show purely about NATS, but it's certainly a vehicle for proving out some of the concepts and things you've learned... But why don't you introduce us to the concept of distributed messaging? What is it, why do you want it, why should I as a developer care about it?
+
+**Derek Collison:** Well, it's a great question... I've been doing this now over 25 years. My career started with being thrown in the distributed systems by accident. When I was coming out of university, we were still on the scale up, not scale out; scale out was actually kind of a bad word... And it happened to be for me that my first job out of school was at the applied physics lab of Johns Hopkins University on the East Coast - from where you are, Johnny, and I was, originally... And by happenstance I got selected by the second-best physicists at the lab, not the first.
+
+The way it worked back then was that top physicists got almost all of the super-computer time... So I was working on advanced visualization for large datasets, and that kind of was my passion in school. My bosses kind of came to me and said "Hey, we don't have a lot of super-computer time, but we have these 12 (Sun) Sparc pizza boxes, make those go as fast as the connection machine." I was like "What?!" So I'd go down this path of trying to wire these boxes together, and figure out how they can coordinate, and break the work up... All of the things that I think a lot of listeners probably today take for granted; it's just the way we do things. But back then, it was like "Oh, really? I just can't get a faster computer?"
+
+And then when I went to California in late '91, I ran into the exact same problem with a healthcare startup. We had so many doctors wanting to watch the federal trial data that I was generating, it was crashing the server non-stop... So I had to figure out how to scale that out, because I had no budget to buy a faster machine. And the little next step, even with the 68040 chip, if anyone remembers those, just couldn't handle it.
+
+So eventually, I think I got smart and said "Oh, maybe I'm supposed to be doing something like this", instead of the stuff I thought I was gonna do. So I joined a startup called Technicron Software Systems, that became TIBCO. They were trying to revolutionize the way finance was working in Wall Street, and specifically was around the notion that we had started to get to a point where the pattern was breaking down. What I mean by that - and it's a very long answer to your short question, but... The way Wall Street was working back then, before TIBCO came along, was stock distribution - I was gonna update you on the stock of IBM. It was like a telephone call. I would dial you up, tell you what it was, hang up; then I would go to the next person, dial them up...
+
+And when the number of updates was low and the number of people who cared was low - and by the way, this was people looking at terminals, not machines yet... It wasn't that big of a deal. But they already got the sense of "Uh-oh, this is gonna be a problem when Derek is the one last at the end of the line, and it takes three seconds to update everyone. His data is now older than someone else." So whoever is at the front of the line has an advantage, right?
+
+\[07:51\] So the way we pitched this to Wall Street wasn't low-level technology, and unicast, multicast, Pub/Sub type stuff, but we simply said "We wanna change the paradigm from a telephone call to a radio broadcast." Everyone just tunes in on the station and they all get the updates at the same time. So Pub/Sub was kind of born out of that driving problem space, that opportunity... It's existed for a long time, and finance verticals, other verticals were very into it. Of course, then we had multicast, and multicast was supposed to solve everyone's problems, but it didn't... And what's happening now is that people are starting to come back to the tech, but not necessarily the old ways of thinking about it, but to solve newer problems in modern distributed systems in cloud-native stuff.
+
+Messaging is simply -- think of it as a connecting technology to glue together distributed systems. There's lots of those today, and you usually have things that are very specific, I'm talking to a database (MySQL or Postgres or Redis, as a KV, or an object store) and then there's some that are generic. What's interesting though is some are generic and they've actually kind of layered on top of point-to-point communication, which defeats the whole purpose of looking at a technology that allows multiple patterns.
+
+So for me, most people are using the absolute wrong technology to connect the distributed systems. They're using one-to-one, point-to-point request/reply. And they build a whole bunch of stuff around it, to get around the fact that that's not a good pattern to use to build distributed systems. So messaging systems start coming into the picture again. NATS itself - I built it, I could care less if anyone used it; it was solving a problem when I was architecting Cloud Foundry at VMware, which is now Pivotal, now back to VMware... It was kind of a stab for me at trying to develop an enterprise platform as a service when Heroku and Google App Engine were kind of all of the rage... And I've just been building systems like that for decades, and that's just how I build them, with messaging systems. But I realize that most people don't. They're like HTTP, or today gRPC, or maybe you see Kafka, or Pulsar type stuff.
+
+I just build them always like that, and a lot of the systems I built in the late '90s are still running, 25 years later, in very mission-critical situations... So usually, I don't get yelled at or kicked out of a room if I walk back in and it's still running there... But I realize that most people didn't care or wanna worry about that. Scale out was still fairly new, it was just starting to take hold, but these days I think people are coming back to it on their own. They're saying "This just doesn't feel right", you know what I mean? So that's kind of where the history of why messaging systems I think popped up, how NATS came to be... But again, we never planned on anyone to ever use it, to be honest with you, when we first built it.
+
+**Johnny Boursiquot:** So it's a nice surprise that now you're in -- what, version 2? It's gone GA a little while ago?
+
+**Derek Collison:** Yeah, I don't know how many releases we've done... Obviously, what you were talking about, Johnny, with GopherCon - by the way, that was great; I got to speak right behind Rob Pike, and they asked me specifically to not just point out the things that I liked... To point out the things that I was struggling with.
+
+If you go back and look at that talk, it's mex talking about a lot of the performance struggles that I had early on... But to be honest with you, it is one of the best bets we ever made. We started on it when it was 0.56, and it actually wasn't performance, believe it or not. The reason that I wanted to move away from Ruby, which is a language I still like, was deploying production systems with the dependency management piece had become so painful for us, trying to do Cloud Foundry and VMforce, and later some other initiatives... That I said "I need something that does not do that."
+
+So the only priorities I had were produce the static binaries, and the fact that the biggest thing for me, believe it or not - again, very early on with Go - was that it had real stacks... Which means if you're a programmer and you know what you're doing, you have the ability to relieve the pressure off the GC... Because Go's GC was very primitive at the beginning. But it didn't matter; I saw it right away. I was like "I can use a stack, for real." And work that I had done at TIBCO took me three months in C to figure out how to transparently transition from a stack to a heap pointer automatically... We got that for free in Go, right? Stick it on the stack; if it outgrows it, you just auto-promote it. So that was the only two things that made my mind up, and the rest is history.
+
+**Johnny Boursiquot:** \[12:34\] Probably more familiarity with messaging systems on the call than perhaps Mat or Jon, so I'm trying to leave a little bit of room for them to sort of jump in and ask questions... So to me, from a developer's standpoint, which you haven't jumped into yet - so we've gotten a use case of why you created, went down that road and created this technology, so this fan-out approach, where you say "Broadcast this, and allow those that are interested to pick up on that broadcast." That way everybody has the same timings in terms of what they get notified of and when.
+
+From a developer standpoint, I hear that, and I'm like "Okay, well that's a way of decoupling. That's a way of basically saying "Hey, I don't have to have this component in my architecture and my system be so tightly bound to this other one that if something even happens to this other one - maybe it goes down, or maybe it's running in an impaired state of some kind, that I'm still in control of my own environment (destiny, if you will). So how much does messaging, or how much should messaging play a role, especially in this sort of \[unintelligible 00:13:36.13\] nanoservice, all bunch of distributed pieces running everywhere? How much of a role do you think that should be playing when you're considering architecture like this?
+
+**Derek Collison:** I think that's a great question... And like I said, I took a stance for probably almost 20 years, of "Oh no, just use whatever you're gonna use." I really wouldn't even engage with people, because it would be a two-hour conversation. I would still probably lose the argument at the end, so to speak.
+
+What's happening now is that things are kind of coming back to it. So just to level-set - one of the things that I care deeply about with messaging systems is a couple things... And it's not about messaging, especially not about the message broker from the '90s type stuff; we need to not think of it like that. We need to think of it as a ubiquitous technology, kind of like network elements.
+
+I was around when there was a single cable, and if you ever kicked the terminator cap off the end, it took the whole network down... And I saw the birth of hubs, and switches, and smart switches, and top of rack, and all the crazy network elements that we have... And just as well, a modern system needs to be doing that similarly. But to start out with - the first thing it does is it says "I am gonna do addressing and discovery not based on an IP import, but on something else." You can call it a channel, a topic, a subject - I really don't care what you call it.
+
+Now, again, for a while, people were like "Why? This doesn't make any sense." But in today's cloud-native world, I would argue what everyone is doing today makes no sense. We struggled so hard to change servers from pet to cattle, and yet we're still saying "Oh, I wanna talk to Johnny's service, so I need to know the notion of an IP port." Now, I know what the audience is probably thinking, and I'll walk through an example of what this really looks like in practice.
+
+The other thing too is that messaging systems, when they do that abstraction, a lot of people call it pub/sub. And we still call it pub/sub, but again, we've gone away from that, because it's kind of got a bad rep. But what I mean by pub/sub is that if the technology can understand multiple messaging patterns - one to one, one to N, m to N, and then one to one of N, meaning I can automatically say "Hey, I wanna send a message and only one of you in this set will receive it." That's kind of what NATS does at the very basic levels...
+
+And folks always ask and they go "Oh, I just don't need any of that stuff", and I say "Okay, a couple things... One, decoupling is good." Pets versus cattle was legit; let's make sure our connected technologies follow the same path, and don't say "No, no, this one's special", whether it's an individual server, or a load balancer, or a special host... It just makes no sense in a modern world. So we push down on those things and we say "Okay, got it."
+
+\[16:18\] The last piece of advice I always give people from the '90s is "Never assume what a message..." - and a message could be a request, it could be data events, all kinds of stuff. But "never assume what the message is gonna be used for tomorrow." So everyone kind of looks at me and says "What does that mean?" and I say "Okay, I'll give you a really simple example." When we talk about NATS these days, with a very fortunate, growing user base, and all kinds of crazy interest from customers and clients these days, is - modern architectures, distributed architectures are built using connected patterns, and there's really only two. There's lots of nuances to it, but there's two. It's either a stream or it's a service.
+
+The service is I ask a question, I get an answer, and a stream is I'm just sending a piece of data, and then I don't care. And to level-set, distributed systems, even up to a couple years ago, were dominated - 98%+ of everything was a service interaction. I'm not saying it has to be synchronous, but everything was "I'm asking a question and getting an answer." HTTP is "I ask a question and I get an answer" type stuff. So I said "Fine, on day one you know who's gonna be answering that question." So you coded up so that I'm gonna send a question to Mat, Mat's gonna respond back with an answer. And you're doing it on your laptop, and you use HTTP, or gRPC, or whatever, and you're like "That's all I need." I go "Great."
+
+Let's not even get to the point of anyone else being interested in the message. Just to start with "Okay, let's go to production." Well, we need more than one map. Oh, crap. Now we need a load balancer. Well, now we need to put stuff in, and do DNS... And that's fine; production can handle that. I don't have to worry about that. So then they do health checks, and then they have to figure out rerouting, and all this stuff... And all these big companies have playbooks on exactly how they do it, and they all look very similar, but they're all slightly different.
+
+Now let's say someone says "Hey, for compliance we need to be able to watch all requests and we need to record them." Now all of a sudden you're gonna have to put a logger, and you don't want the logger in place of a request-response, which is a massive anti-pattern I see being proliferated these days. It's like "Oh, no... Put a messaging system in between it, and store it on disk, and try to retry, and stuff like that, in line with the microservice, in line with the service interaction. I'm hyped up on Red Bull, but that's the dumbest thing I've ever heard. It's like Google saying "We're gonna write all the log entries before we return your search results." It's just foolish, no one would ever do that. But there's a need to say "Hey, I need to be able to write these things down, and someone else is gonna look for anomaly detection, or any type of policy enforcement, whatever it is.
+
+So look at that system when you're getting to the end of the day, and now let's say we actually wanna spread it out from East Coast to West Coast, to Europe, and you need Anycast, and DNS, and all kinds of crazy stuff and coordination on the backend of state. They become really complicated, and they're trying to get around the fact that everything is just point-to-point; it is naturally a one-to-one conversation. Whereas with a messaging system, you write it, you run it in that server, let's say, or whatever, but think of the NATS server is extremely lightweight. It can run on a Raspberry Pi, in your top of the rack switch, you can run it in your home router, you can plug and play and put these things together in any arbitrarily complex topology that spans the globe... And that's another discussion on distributed systems that aren't really distributed.
+
+So you set it up on your laptop and you run the NATS server. Its Docker image has been downloaded 150 million times... It just runs. For example, a subsystem of GE doing nuclear stuff runs our server for two years at a time, with no monitoring, no anything. And when they come in to change things inside of that nuclear reactor type stuff, they shut it down and figure out if they wanna upgrade it or not.
+
+\[19:59\] So it's lightweight, it's always there, it's ubiquitous, it just works. So now all of a sudden you write the same program, you're sending a request, getting a response, doing it on your laptop... I would argue you'll take the same amount of time, or possibly less, but it's on the same level. You do have to run the server.
+
+But now when you go to the next level of production - I need more mats. Well, just run more NATS -- I mean, maths, not NATS. "Oh, well do I have to put in deployment a framework like Kubernetes and a service mesh?" No. I don't care how you run them. Run them on bare metal, in a VM, in a container, in Kubernetes... It does not matter, run them anywhere. The system automatically reacts.
+
+By the way, you haven't configured anything on a NATS server yet, ever. Now, all of a sudden you're like "Okay, well what happens if I wanna do compliance and I need to watch all the requests coming in?" Just start a listener for it. It's got all of those patterns built in, so there's nothing changing between me who's asking the request and Mat who's giving the response; we have to change. As a matter of fact, we don't even have to be brought down and restarted; we're just running along and people can come up and bring up anomaly detection, logging... All kinds of stuff.
+
+So as you keep making these systems more production-ready and complex, you realize that messaging gives a huge win over the other ones. Now, the other ones are kind of known. People know how to put up load balancers, and know how to do logging, and know how to do all this stuff... But when you see something running on a service mesh and you haven't even sent a single request check and you're spending $6,000 a month... And I can show you, we can do 60,000 requests a second, with all of the servers latency tracking, all transparently, doing it for real, and it runs on a Raspberry Pi... That also translates to OpEx savings, which is a big deal.
+
+NATS has always been known for how fast it is, but most people tell me "But we don't need to go that fast. We don't need to have a server doing 80 to 100 million messages a second." And I go "I know", but if you think about it, if you take that same thing for your workload and put it in the cloud, you can save 80% on your OpEx budget.
+
+So do people need messaging systems to build stuff? Of course not, because everything for the most part is built on essentially HTTP, which again, is an interesting one to me, unpopular opinion... But I know why we did it that way, and we don't have a reason to do it that way anymore and get it stuck, right?
+
+The notion of client-server or request/response in the old days was the requester and the responder were usually inside the same network firewall... Not firewall specifically, but essentially we're inside the company. And everyone started to say "Hey, I want the requesters, whatever those things are, to be able to walk outside of the corporate network." So all of a sudden people started doing this, they go "We can't get through the firewall", and the firewall people, or the old DB people, they go "No." It doesn't matter what you ask them, they say no.
+
+So people kind of went "Wait a minute... Port 80 is always open. We can piggyback off that and circumvent this whole thing. So we can just do request-response on HTTP or HTTPS, and it works." And it's true, and I remember doing some of those tricks myself. We're not in that world anymore; that makes no sense whatsoever to build a modern distributed system off of technology that existed for something totally different and was a workaround around security and firewall systems.
+
+**Break**: \[23:23\]
+
+**Jon Calhoun:** I wanna ask you some questions from a slightly different perspective... So a lot of the things I build are very small, and when you talk about, for instance, logging something before you send a response, I am guilty of (in my early days) building a service that literally logged every request to a SQL database before responding... And now it seems ridiculous, but at the time the request load was so low that it really didn't actually matter that much. It was just a simple solution to fix it, and we could quickly browse requests and actually track things, and it made support way easier at the time.
+
+So if I'm somebody like that, where I'm starting off in a smaller project, or working on my own, or doing something like that, where do you see people get into messaging systems? What problems do you commonly see them tackle, that they would want to look at it? And what are your suggestions for ways to get introduced to them? Because obviously, a lot of people aren't gonna jump into these really complex scenarios.
+
+**Derek Collison:** Yeah, that's totally fair... The biggest one we've seen early on with people starting to get interested in it is simple load balancing. So even though the request might be handled by a single one, they wanted more than one; they were immediately thrown into utilizing load balancers from a cloud provider, or setting one up themselves... From there, if they are aware of NATS at all, or another messaging system, but NATS especially, in terms of it's just so drop-dead simple, there's no configuration... And you get load balancing for any number of groups that you want to create.
+
+So you can have production, dev, tests... The system essentially dynamically responds to the fact that you say "Hey, I'm interested in this request, and I wanna be a part of this group. I wanna be part of the Go Time group", and the system just automatically responds. So immediately, they have less moving pieces and less OpEx time in terms of time spent trying to make sure the system is up, monitoring all the other pieces...
+
+So we've seen that, but -- I mean, that's totally a fair question. It's interesting, I've shown people how easy it is to do request/response when they're doing HTTP, and we can do it just as quickly even running a server, because with Docker now it's just so easy to say "Docker run NATS", and then it's like "Oh, that was it?" and it's like "Yeah, you have a server now, so you can do anything you want on your platform."
+
+But you're right, until people start feeling pain, they're not gonna be looking. Or they wanna find a solution that they don't think exists, and if they see that it's enabled by a different technology, then that'll draw them to it as well.
+
+**Mat Ryer:** Does it have advantages beyond the production side? Does it have advantages for software design itself? Because of course, if you think we're gonna suddenly treat these things slightly differently, we're gonna be communicating through this message queue, that has some impact on how you then think about the design of your application.
+
+**Derek Collison:** \[27:58\] Yes, it can, but again, this is one where I think people still equate messaging systems with a single broker, and queuing, and back from the '90s type stuff... NATS is extremely good at routing and framing, and that's it. But what is interesting now is that NATS is extremely lightweight, so a server can run anywhere. It can run on your phone, or whatever... And you can Lego-brick these together into any topology that you want. So if you think of it that way and you don't think "Oh, I've gotta go through this queue", where it's "I'm just routing a request to the appropriate responder and getting the response back", even just with that simple model, you've removed a lot of moving pieces in the transition from dev to production.
+
+The other thing that's really interesting to us - and again, this is where microservices has driven a lot of this, I think... All of a sudden with microservices it's like "Wow, I'm doing point-to-point", which you can argue for or against; I'm against, but I understand why people say it... But what people are interestingly enough struggling with is addressing a discovery in security. Everyone has their own security. And when someone raises their hand in the org and says "We can create a one org security model type stuff...", and it's usually painful; it's hard to do that, and it affects the developers themselves, and you don't have that really clean decoupling. NATS tries to preserve that.
+
+The program that I write on day one, that goes into production, literally has one thing that changes, which is "What are my credentials?" So you say "NATS connect" and you give it a URL to a system. We have a global system that runs all over the world - every major cloud provider, every major geo. All you need is a single URL and it just works. We find the closest server, we do all the right stuff for you. But the only thing to go from dev to prod is "Hey, we need credentials to prove who we are." But now, all of a sudden you have a consistent identity authentication and authorization system that has no private keys or passwords ever shared with the system itself, and it just kind of works. In other words, that pain point of "Oh crap, now we've gotta get secure, and locked down, and stuff" also goes away.
+
+A lot of times, when you're playing around, hobbying out, doing stuff, you don't see that, especially for enterprise, or playing around with services that are usually exposed over the web. Where we have seen people immediately jump on something like this is IoT, and the IoT landscape, where people are like "Crap! We don't really have anything, and the thing we have really doesn't have a cohesive server back-end." In other words, there's a lot of decisions. If you've got an MQTT client on your device, but you don't know what to do.
+
+We haven't landed it yet, but we've committed to it, promised it, and we've been coding on it... This notion that "Hey, you can just take those apps and connect them to a NATS server, too." And this is a global topology, meaning your IoT stuff will work anywhere in the world. You can sell your gadget, your software to anyone, wherever they are in the world, and they'll get a good experience, so to speak.
+
+**Mat Ryer:** That is really cool, isn't it? If you think about that.
+
+**Derek Collison:** What's interesting is -- I mean, I'll be frank with you... We see a lot of folks -- we're starting to hit this weird bow wave which is, to be honest with you, making us a little uncomfortable as an ecosystem... Because we were always under the radar, didn't have a lot of attention, type stuff, which was nicer than we thought... But it's still good.
+
+But what we're seeing now is everyone coming to our front doors wither either of two mindsets. One is "Holy smokes! Kafka is too complex, too costly, too blah-blah-blah. Help me!" Or it's this pattern where they say "We've got centralized things that can do request/response; we can ask this central thing about certain things... But we have all these remotes. And we want the remotes to be somewhat autonomous. In other words, we want them to be able to communicate amongst themselves, seamlessly and securely communicate with a central service..." They might be generating telemetry data, so this notion of streaming, that getting collected...
+
+\[31:52\] And what's interesting is that these use cases are all over the board, from the person walking in the front door, saying "Hey, this is the problem we're trying to solve." But when you look at it as "Do you have a centralized thing and you have these things as remotes", well, remotes can be anything. It all of a sudden starts lighting up as the same pattern - they just want a consistent communication system, multiple operators... You can run your own server, I can run my own server, but the system still works. It's almost like putting a cell power in your backyard and having great cell service when you're at the house, but as soon as you leave, it just transparently works. It connects to Verizon, or AT&T, or whatever.
+
+So when we present that to them, whether they're talking about the end point is like a Bose headset, or a Peloton bike, or it's a factory, or a plant, or a telephone pole, or whatever, we're like "It's the same pattern." We have some centralized service that does services, and it collects data. It might actually throw off data as well; telemetry, or sensor-type data... Let's say it's an airport - it can operate on itself, but it's got its own servers; the airport staff monitors those servers and it runs them, but the security model is cohesive. All the backend service providers, the airlines, whatever that is.
+
+So those patterns we've seen appear quite a bit, and not in only people that know all the buzzword bingo of Kubernetes and service mesh and cloud-native. It's more traditional systems, that are trying to do more. They're trying to expose more data at these remotes, throwing off more data, doing things with it locally, uploading it or centralizing it... And we see that pattern repeat and present itself six, seven, eight, ten times a week, non-stop these days.
+
+**Mat Ryer:** Yeah, so it feels like you've actually identified a real good abstraction for a lot of this stuff... Because that's the big problem whenever we try and solve problems like this in a generic way - you build it for one case, and it's perfect, and it doesn't quite fit with the next case, but close enough; a few bits of configuration get you through it. And then the third case - it really doesn't fit at all, but it's too late...
+
+So yeah, that does sound right. Do you have lots of examples then? Are there lots of examples where NATS in particular is used in IoT?
+
+**Derek Collison:** The IoT stuff is new and up-and-coming. Everyone is looking heavily at it. Right now, most of them are bridging across MQTT directly, usually on-device or on-controller... So plants, factories - we're seeing a lot of that stuff. And they're desperately waiting for us to put together the native connectivity so that they can have a cohesive one type of system to manage. It's not a bunch of silos that they're trying to glue together.
+
+And what's needed -- there's one other thing that NATS did about two-and-a-half years ago, that we felt was kind of important, maybe, for certain use cases... And what's happened is it's exploded into one of the most targeted things. Well, there's two of them that we did. One was that we made it truly multi-tenant.
+
+Earlier in the podcast we were talking about distributed systems, and I would argue most distributed systems aren't distributed if you really wanna stretch them. So pick your favorite open source technology that you consider a distributed system, and tell me how well does it do if you have pieces on the West Coast, East Coast, Europe, and Asia-Pacific. Most of those projects will say "Yeah, don't do that. Just run it in one region." So it's a distributed system, but it's a one-region distributed system. Which is totally fine, by the way, because NATS was the same way about two years ago, before we dipped into "We need this ability to span the whole globe if we want to" type stuff. But we knew it had to be multi-tenanted, and it had to be -- I'm dating myself, but I call it Pepsi & Coke secure meeting. They'll use cell towers and cell service, but certain software, they're saying "If Coke is here, there's no way we're gonna use it" type of stuff. So it had to understand that.
+
+\[35:52\] So we created the notion of accounts. Think of it like a sandbox or container for messaging. All the users in that account can see each other, no matter where they connect in the world, but they cannot by default see anybody else's account. But we introduced the notion of "Isolated by default, but secure sharing."
+
+So remember, not talking Pub/Sub, but talking streams and services. You can say "I wanna export one of these", and then other people can import it. It's like a Facebook thing. Both of you have to agree that you wanna do it. And you can make it public, but most of the people say "No, you need permission from me. I will sign off on a token", which is public-private key cryptography, so no private key is ever in the system, meaning you don't have to trust the operator of the server for it to do the right thing, to allow them to cross these boundaries.
+
+So that was a huge win, and there's even a smaller example... A little off in the weeds, so we won't go too far down it, but this notion that people who do know messaging systems and have been using them, it's like "Wow, we used to have two-week design sessions on the subject space, or the topic space. How many tokens, and what's where, and who can step on each other's foot." And when we did this, everyone realized that they were so lightweight that they're like throw-aways. So organizations are putting a single account per every microservice that they do, and then their imports or their dependencies, and then their exporters, their API.
+
+One of the other things too is it's your sandbox, it's your world... So when you import something, no one else can tell you what to do with your subject space. So you tell the system where you want it to show up. So I could release a service that just says "Send me a request on request, and I'll send you a response", and you can import and say "I want it on derek.coolservice.request" or whatever, and the system transparently does that. But the kicker was - and we did this with our own system, and people have started really getting onto this... It's "Yes, you can put it wherever you want." So I could export something like -- NGS is our global system, but ngs.usage.star. And it's a service, meaning you send me a request - a service interaction - and I send you a response. And I'm expecting ngs.usage.something. Star is a wildcard in our terminology.
+
+So Johnny comes in and says "Yup, I wanna sign up." I go "Great. Here's a secure token that you can be allowed to import this." But what you can import is ngs.usage.johnny, and that's it. So you cannot send to ngs.usage.derek, or ngs.usage.mat, or whoever. But again, Johnny controls his own sandbox, so what he says is he goes "Great! I'm just gonna stick it at ngs.usage". So now all of a sudden what happens is that everybody's experience is "Hey, if you want usage, you just send a request that looks like 1h for one hour, or 24 hours, or whatever, to the same subject", and you get a response - JSON, all bytes and messages sent and received. But the backend knows that it's guaranteed to receive messages only from people it's authorized, and it's guaranteed that that last token is who you are. So they have a secure context built in.
+
+So now you can build a system where that secure, authoritative context - submitted context of who's doing the request - is something that you don't have to think about. You don't have to build lots and lots of stuff on top of to get it. I know that's a little geeky, but we've seen people who the light bulb goes off and they go "Holy smokes!" And then I can deploy these responders anywhere I want.
+
+The other big thing that we did - which again, is subtle and I haven't seen anyone else do it, but I wanted to talk more about the abstract - as we've seen for many years, and I've seen this for a lot of my career as cloud and SaaS took off, that it was an or-conversation. You could run your own servers, or you could use the cloud service. So we thought really hard about "Hey, how can we make this an and-conversation?"
+
+What we did was we said "Hey, we have all these different network topologies", which are just the way servers talk to each other; so they can form small clusters, and then you can put clusters of clusters together into super-clusters, and all kinds of fun stuff... And they all use different topologies. Why don't we create one that allows you to extend a super-cluster at will, like a hub-and-spoke.
+
+\[40:06\] So when we did that -- and by the way, you can mix and match operators and security models, meaning you can use a shared SaaS utility inside of a big Fortune 50 company or a global utility, but you can also run your own servers and have the best of both worlds. Those two things - that account isolation with secure sharing, and then the ability to mix and match a utility model, if you guys have dealt with enterprise companies (and I know you have), we see a lot of this. "Hey, we had a problem and we picked this technology and we did a POC. Then we did four POCs. Then someone raised the hand in one of the meetings and said "Hey, instead of having four silos or six silos, let's create a utility that everyone can use." But you either have to use it or you don't, but we're gonna mandate that you do it." And there's two things that happen.
+
+One is the effort fails spectacularly, because the code is not actually multi-tenant. It doesn't really understand it. And it can't stretch. It can't actually service people on the East Coast, West Coast, Europe etc. Or people go "No, I'm not gonna do that. I'm running my own servers anyway." We heard that so much. Even in the previous company, Apcera, I was like "Man, if we can give them the ability to have the best of both worlds, that could be interesting." And we did it because we knew we were gonna do MQTT, so we have to honor their security model, which I believe still username/password or client cert is the highest level they have... But it would have to mix and match into a global NATS system that might be using our forward private key/public key scenario. But those two have really resonated with a lot of folks trying to solve these problems. I wish I could have predicted it, but I'm not that smart; I didn't. But we're excited about it.
+
+**Mat Ryer:** Well, you talk about predicting things that people are gonna need... NATS 2.0 is generally available, as it says on your website. What was the change? Why did you need the breaking major version update? What changed is significantly, and how did that come about?
+
+**Derek Collison:** The way NATS 2.0 came about was I tried to think through the notion of "Is there an opportunity to create a company where NATS might play a role?" Prior 2.0, NATS was rock-solid, performant... Again, it was like a lot of distributed systems today - you couldn't stretch it. It liked to be close to its neighbor, have good throughput, good RTT and so on. And it wasn't a company, by the way. A lot of startups aren't companies, they're just features of their systems.
+
+So I really wanted to think hard about that problem, regardless of whether NATS was a fit. And what I came up with - and we'll see if I'm right or wrong; probably wrong, but that's okay - was the notion of the internet, that defining moment in '94-'95, and then in the early 2000's with the realization of the global cellular network, and what those two platforms with hyper connectivity provided and it actually ended up, we haven't had that event for digital system services or devices. We pick a different technology, and even if we use the same technology and the same company, it's siloed up Yin Yang, you've got 40,000 RabbitMQ servers running, or whatever that is... So I said "What if we try to create the first secure digital dial tone for digital systems services and devices?" That all they do is connect, and it's kind of like connecting to the cell tower; you're not even aware of it. It just kind of works, and it's all in the background. But we have the ability to connect anything that's out there.
+
+So that's where we started, and NATS was not up for that. It needed three major things that we identified right up the get-go, and then the fourth one there we just talked about, about that hub-and-spoke extending utility SaaS and private owner mix. But the first three were pretty simple. The security model had to really be forward-looking, and there's lots of fun, interesting math, and all kinds of cool stuff underneath the covers, but the easy version is the system should never have private keys or passwords, period. It's really simple. So if someone routes all of our net server devices and steals them all, they don't have anything. So we did that.
+
+\[44:08\] The second was multi-tenancy, which we talked about. That was a big one. Multi-tenancy is not something you can slap on top. By the way, security isn't either, but most people slap both of them on at the end. It just doesn't work. So multi-tenancy goes all the way down to the root of the codebase, and start from the ground up.
+
+And then the last one was global topology, so the ability to have a network topology span lossy systems, very high RTT, low signal to noise ratio types of things. You can't use the same topology to talk between servers that you expect to be really close and buddy-buddy with, as to one all the way in China, let's say. So those three main things had to be done.
+
+Now, what's nice is that as you get older, you remember lots of stuff, meaning we knew we could never break anyone that was using NATS before 2.0. So it's totally backward-compatible. Any config that works with NATS 1.0 will work with NATS 2.0. Everything works. Which was hard, but we thought it was important. But the major version was the signal; this is something different. This is not a message broker, a queue. It is a ubiquitous routing and framing technology that can run anywhere, and can do any type of pattern, but the major ones are services and streams.
+
+**Johnny Boursiquot:** So with this approach now, is it a fair comparison for somebody to be like "Well, I want some sort of a broker system, so I'm gonna consider maybe SQS, or RabbitMQ, and I'm gonna also toss NATS in there"? Are they even solving the same kinds of problems anymore, or do you think it's a fair comparison?
+
+**Derek Collison:** No, I think they're basically solving similar problems... But again, you run out of runway. So with silos - not true multi-tenant, the security angle, all that stuff. We've seen a lot of legacy messaging tech coming to us, with a pain point somewhere in that realm. "We're tired of managing all these silos. Everytime we wanna connect two things, we've gotta figure out how to glue these two separate systems together, just to take advantage of something that we should have just been able to flip a switch and it just works type stuff." But we don't see a lot of people come and say "We wanna mix and match SQS, or Google Pub/Sub." I mean, we've seen it a little bit, but usually they go all in on NATS after they talk with us. Now, where we have seen interrupts is with MQ series and Kafka. People want to run NATS, and they still wanna run the Kafka, whether it's an existing investment, or something new - they just want it there, and it's not going away. We wanna protect that investment, so we've done a lot of those interactions.
+
+**Mat Ryer:** That is really cool.
+
+**Break:** \[46:56\]
+
+**Mat Ryer:** So the company that you talked about was built around the technology... How does that work then? This is a tech podcast, but I know a lot of the listeners are also quite interested in the commercial aspects as well, things like this.
+
+**Derek Collison:** Yeah, that's a great question, and hopefully the listeners will enjoy the next four hours of dialogue on this subject... \[laughter\]
+
+**Mat Ryer:** If they listen to it two times, though. That is only two hours, so...
+
+**Derek Collison:** Yeah... Frankly, open source and smaller companies that are trying to make a viable business directly off of open source is a challenge. The whole industry is going through some really serious pains, because a lot of the open source is being funded through indirect revenue channels... Like Google. They don't have to charge you to license Kubernetes, or do anything like that. They have their reasons for doing it, but they're making their money elsewhere... Which means that it drives a consumer bias that it should be free. And that's a challenge. When I saw that with my last company, \[unintelligible 00:49:55.24\] it's a huge challenge.
+
+So part of that "What do we wanna build a company to do?" was equally weighted with "How are we gonna make it a viable business?" So for me - again, a very unpopular opinion, I'm sure - I personally don't believe in open core. I think it's freemium enterprise repackaged, and it's gonna fail, just like freemium enterprise did. I could be wrong, but my bet was that there's really only three ways to make money off of open source: run it as a service, bundle it with hardware, because a consumer bias of a physical thing totally changes their bias, and they have no problem paying for it... And then the last one is augment with a service.
+
+Some people kind of push back and say "Well, that's open core." But the distinction is it's kind of like your phone and an AT&T contract. You're augmenting your phone with that, and it makes it better, and it makes it actually -- you know, you need a telephone, or a cellular plan, or whatever... But I do draw that distinction. So what we did was we looked at "How do we take those three rules (which could be wrong, but that's my bet) and where do we go?" Because running NATS as a service, as a silo - this is a big deal - it was a no op. We tried it at Apcera, no one signed up. And they were so nice to us, and they said "Derek, we literally run this on a Docker container, and it's been running for three years now. We don't even monitor it. We don't even care. So why are we gonna pay you to run a singleton?"
+
+So we thought to ourselves, how do we make it such that "the sum is greater that the parts" type stuff? One is that we can create a global network, all cloud providers, all major geos... Which you can do yourself; there's nothing that's not open source to do that. But it's cost-prohibitive. If you just wanna use two sites, so one in Europe and one in U.S. type stuff. We did that, obviously.
+
+\[51:45\] There's always the notion of on-premise recurring support. That's kind of the marquee that you want. NRE consulting, training education - it's usually one-to-one, so you don't get a market multiplier there whatsoever. Recurring support - if it actually is clean, you can get your 10x+ kicker, for those who know all the market value accelerators, and things like that.
+
+So we care deeply about that. We do do NRE training and consulting, but we know it's not a huge source of revenue. It's a huge source of customer experience and satisfaction, but not revenue.
+
+So we have the SaaS model with NGS, we have on-premise recurring support for our stuff, and whether it's good or bad, a lot of people coming to NATS now are like "Wow, that's so cool! It took me two seconds to write the NATS app and run it against the demo servers", which have always been free and available and such like that. "I wanna do something hard." It's just natural, engineers are just like that; they're like "That was too easy, I wanna do something hard."
+
+So NATS now allows you to set up some crazy complex topology, with some crazy security rules... So they try to do that, and now all of a sudden because of that complexity people go "Oh, we wanna get support." I don't like that; I like things that are simple and just work, but we have noticed that.
+
+The other thing that's interesting from our perspective is that we don't believe NATS is just a connective technology. It is, but how do you value it as a user? You know, I've been doing this so long, and my bias is "Everything is just a message." So whether you're using a database driver, or whatever, you're just sending messages back and forth. So what happens if we say "Everything is just a NATS message"? What I mean by that is you get everything that NATS does, you connect, distributed queuing, load balancing, circuit-breaking, self-healing... It puts itself back together, by the way, without any help from any platform technology type stuff, really...
+
+But what happens if we said "Hey, you know that export and import, those streams and services, and you can export one and someone can import it?" What happens if the system just has a service that you can import, that says "It's a KV service", and now you can do zero-trust, secure key-value set and get from anywhere in the world, with any application. Hm. Okay, now all of a sudden NATS can do simple state storage and retrieval. It doesn't solve all the apps, it doesn't solve world hunger, but hey, okay... Now what happens if it can do object storage? Very large objects, very efficiently. The system dynamically moves things around, it understands where requests are coming from... And again, because we don't care, we can just move Mat and run them wherever, without anything special coordinating. That's very possible.
+
+And then going further, it's like, well, what happens if there's a GraphQL service, and I'm just sending requests over NATS to a GraphQL service, but all the security works, all the authorization/authentication is built in, it's what I know, it just kind of works? So those would be premium services that we could charge for. So you get -- I call it basic cable, the dial tone, and then you can get the premium channels a la carte if you want to.
+
+And then the last piece of the business model is -- some of those may be very compelling; let's say anomaly detection, or some advanced analytical statistics on traffic patterns, and stuff like that, that a company might say "We really want to use that service, but we can't use yours. We have to run it in our own data center, our own VPS", whatever that is. Then that's software license revenue.
+
+The way we envision the company succeeding is the first bow wave, the first 2-4 years will be recurring support as the major revenue driver. Then, as we land -- we landed web and mobile, we're about to land MQTT for IoT... Those two will drive more of the NGS stuff, direct. Or they create a leaf node that they connected with and use NGS to talk across the world, which we have a couple of folks doing already.
+
+That's kind of the whole business model in a nutshell. We'll see how it works out, but it's a challenge, for sure. The consumer bias that it should be free is the hardest thing for any OSS developer to fight against.
+
+**Jon Calhoun:** \[55:45\] I feel like you run into that sort of thing even beyond open source software. It's kind of a weird thing, I guess, in the software world... To give an example, people write tutorials and books that teach things, and in most environments people expect to go out and have to buy a book. But in the programming world, it's very common to assume "Well, somebody will just make this free." And that happens for all sorts of things.
+
+There is some upside to trying to help people access things, and trying to make it accessible to the world, but then there's also, like you said, that flipside of it's very hard to support, unless you get to like a Google scale, or some sort of scale where it's possible. Because prior to then it can be very challenging.
+
+**Derek Collison:** I think you raise a great point, and one of the things that I debate with folks quite a bit is, you know, the notion and how people frame support. So people go "Oh, I'm paying you because of me. And if something's wrong, I want you to help me", which I totally get. But if you look at like healthcare systems - the Western worlds mostly do the same thing. I pay for healthcare when I'm sick, type stuff; for when I'm sick. But if you look at Asia-Pacific countries, it's the opposite. You pay for insurance or you pay a doctor when you're well. You don't pay them when you're sick.
+
+I've had this debate - and to be honest with you, I lose most of the time - with customers saying "We've been running this for two years. We love it, it's been running for two years, we haven't had to touch it, we don't even monitor it... Why would we wanna pay for support?" And I say "You wanna pay for support so that as we keep making it better, it keeps doing this thing where you never have to have an issue to deal with, or they're very low." But to be honest with you, it falls on deaf ears, most of the time. It's very challenging. We have a ton of usage, production usage in the tens of thousands of users, and a minuscule percentage of people that want to pay support.
+
+**Johnny Boursiquot:** Stop writing such good software. \[laughs\]
+
+**Derek Collison:** Yeah, and you know what - I dealt with that earlier in my career, where people were like "You need to make the book longer." I didn't write a book, but it was like for a manual. "It has to be 200 or some pages so we can charge more." Or "It needs to be more complex; they need us to get it up and running." And I just resisted that. I said "That just doesn't feel right." It should be simple and approachable both from an application standpoint and an op ex standpoint. But the current state of the world is that if it actually nails all of those, you will plummet your voluntary support contracts, for sure.
+
+**Jon Calhoun:** Some of this is challenging, because -- you even said the open core model is... I think that one's hard to get right, because it doesn't work for a lot of software. An example in the Go world is a lot of people have probably used Caddy server, and I think when they tried to transition to a paid model, they struggled because the core thing that a lot of people really wanted was it was hard to charge for that... Because it was already there, and they were used to getting it free.
+
+But then, you come from the -- I think you said the Rails world. Did you mention Rails earlier? Or Ruby?
+
+**Derek Collison:** Yeah, more Ruby, but I definitely understood the Rails community, and all that.
+
+**Jon Calhoun:** Okay, so in the Rails world there was something called Sidekiq, which is like a background job processing type thing. Whenever that came out, it's something that -- the core of what a free user would need is actually there for the free users... They actually had a nice separation between what enterprise users would actually want and pay for, so as a result that has worked well for them. But I think that most open source it's very hard to find that distinction, and it causes issues where it's really hard to make that business model work, because either nobody pays you, or you basically make software that the free users can't get any value out of it.
+
+Going back to the metaphor you had, with the phone, and you can buy an AT&T plan - it's almost like you sell them the phone and say "You have to pay me for the battery though", and they're like "Well, that's not very useful now. That's not augmenting it. That's making it functional."
+
+**Derek Collison:** You're absolutely right. And what was interesting - and you could see this a little bit with the Kubernetes ecosystem - is a lot of people would knee-jerk around a new technology taking off, and approaching let's say an open core model saying "Well, we're gonna add monitoring and management tooling. That makes it easier to \_\_\_\_\_\_", whatever. And I had a friendly bet with a bunch of folks that I have known for years, to say "Here's why I don't think the open core model will work..." Because these technologies now, when they become very ubiquitous -- Kubernetes is not easy, but you can get through it, and there's a large ecosystem, so there's a ton of people working on it...
+
+\[01:00:20.24\] The first thing that ecosystem is gonna do today - which it used to not do, because the barrier to entry was too high for someone to sit down and write a monitoring and management system... As I said, these ecosystems spike so fast and so quickly that anything that looks like an easy, tangential business opportunity around an open core model gets sucked in.
+
+Everyone complained on the first versions of Kubernetes about a lack of monitoring, management tooling, dashboards etc. and I think it took them less than two releases to have a skeleton version of that... And then of course, no one now would start a business on a dashboard for Kubernetes type stuff.
+
+So you have to think "Hey, how do I make what I'm offering way more valuable as a service?" And again, that's even a challenge, because if you really do nail the experience for a single use case, you might struggle with "Where's the value with you running it for me, versus me just going 'docker run', or whatever type stuff?"
+
+But I think this notion of this macro trend, that I believe the opportunity to enable this hyper connectivity opportunity for all the digital system services and devices is huge, if it's done right... But it has to be very approachable, very easy, open source, good governance model, good OSS licensing... I've gone through all the phases of closed open source, and licenses and governance bodies.
+
+And you can mix and match models, meaning I could use in Google Cloud, Google Cloud's version, that runs great. But if I want to run my own server, I could, and it all just kind of works seamlessly. I do believe that has value that people would want to pay for.
+
+Now, the basic dial tone that we talked about, just for the listeners - that is gonna be a low-margin volume play. The premium services - we can get better margins, we think. And of course, on-premise, the recurring support - we've got multiple tiers there... But it's definitely something that folks who are thinking about starting a company - I don't wanna ever dissuade someone from starting a company, because I think it's great; I think it's amazing. But it's very hard, it's very lonely... But I do encourage you to think really hard about "Am I building a company, or a technology feature?" If you're building a company, do I have a really thought out vision for what the business model looks like, or do I just say "Oh, I'll figure it out once I get a whole bunch of eyeballs, and millions of users?" That doesn't work as much anymore.
+
+**Mat Ryer:** Yeah, that is so true, the perception problem. That is a challenge for people. We had a similar thing - we had this technology that was extracting metadata from video content using machine learning... So machine learning models would look at the video frames, and then actually be able to describe what's going on in a video; and of course, make that searchable, and all those things you can imagine once you've done that... So we were thinking maybe that would be charged per gigabyte, or something. We tested it, and people were saying "Well, to store on Amazon it's only 2 cents/gigabyte", or something. And we're like "Well, yeah, but that's just storing. This is using machines." And they're like "No, it's way more expensive than Amazon..." And it was like "Okay..." Common sense - we almost should just not assume that there's common sense around, from my point of view... Do you know what I mean?
+
+**Derek Collison:** Yeah, and I'm an angel investor and I consult with lots of smaller companies, and a lot of companies struggle with "How do I turn it into a business?" I said "Think about a way where my experience with your software becomes better." If you have a system that collects data from everyone, keeps the privacy concerns in place - that's a big, big deal, of course... But it keeps all of that stuff at bay. Essentially, it makes my use of the product better because of that.
+
+\[01:04:09.29\] I was fortunate enough to work at Google from 2003 to 2010 or so, and I remember some of the -- obviously, Google has a lot of extremely bright people that think of very, very elegant, complex solutions to very hard problems. But what I liked to see a lot within Google was extremely simple solutions to complex problems. So spam was a huge deal when Gmail came out. It was just awful. But as Gmail became so popular -- and we had lots of usage on it. Even in the early days. And if we just put a little button that said "Hey, I don't like this message. It's Spam", it would see all the signals and say "Wow, in the last five seconds a thousand people clicked on the same message and said it was Spam, then we can automatically mark it and move it off." So that power of collecting data and using it to optimize individuals' experiences is a model that I've talked to a lot of startups about, and said "Can you encapsulate what you're trying to do with your software where the service is augmenting - it's not an open core, you're augmenting with it - but your experience gets tremendously better because of it, and it's something they can't recreate?"
+
+For your case, Mat, I agree with you. They're just saying, "Wait a minute, I can store a gig in my own data for way cheaper than you're doing it." But if you said "Hey, can you collect all of the spam signals from 40 million people for Gmail?", they can't do that. They're just like "I can't do that." So then what happens is they go "I know I can't do that on my own. Does it really help me that much that I'm willing to pay for it?" And that's always the trade-off.
+
+**Mat Ryer:** It's funny, because that is almost exactly the way it went. So that was really funny you said that.
+
+**Johnny Boursiquot:** It is indeed time for our Unpopular Opinions...
+
+**Jingle:** \[01:06:17.12\] to \[01:06:33.18\]
+
+**Johnny Boursiquot:** I know you dropped a lot of gems throughout the show here, but I'm wondering if you have a solid, solid unpopular opinion for us...
+
+**Derek Collison:** Well, in terms of the gems - remember, it's advice and it's free, so you get what you pay for. So your mileage may vary. I'm usually never short on unpopular opinions, so my two probably that are applicable to this is that most systems that you think are distributed aren't. So you pick your favorite open source, and I'm telling you it's distributed as long as it's all close together. If you try to stretch it, it's not distributed anymore. And the other one is that I really do feel that us using HTTP to connect modern distributed systems, where you have to have sidecars, and proxies, and load balancers, and everything under the sun is just madness. I cannot believe people keep doing it and go "Yup, that's the way we should do it."
+
+I understand how we got here, I can't understand how we haven't got to a tipping point to go "Hey, we don't have to do it like that anymore, because everything's modernized, so we can actually do something real now" type stuff. So those are my two.
+
+**Mat Ryer:** Yeah, that is a pretty good one. There's examples in real life like that. I find real life as basically like someone's legacy code, and we're just born into it, and then we're like "Why would that be the way it is?" But yeah, I get what you mean. I like that. You're right, HTTP is kind of crazy... But it works, doesn't it? It \*just\* works. It mostly works... And so it wins.
+
+**Johnny Boursiquot:** Most of the time.
+
+**Derek Collison:** \[01:07:52.12\] It works if you have a large team that can watch everything and keep the lights on. So if you don't have to deal with it, then for you it's like "Well, this works the same." If all of a sudden you go "Yeah, it just works, but man, I don't wanna be spending $8,000/month on my little system, so now I'm gonna do it myself." Then all of a sudden you start to realize why do we have all the... And this whole notion of -- I guess a third, smaller one, but this whole notion of "Architect everything with side cars." That also drives me nuts. It's like, "I'll just add another side car to it."
+
+**Mat Ryer:** Yeah, it doesn't stay simple for very long, does it, when you have to tackle things like that, tackle problems like that.
+
+**Derek Collison:** No, it doesn't. And to be honest with you, we're in a weird global situation, as we all know, and my hope and thoughts are with all the listeners, and I hope you're safe, and healthy, and all that kind of stuff. But when you're looking at a company and you're trying to figure out how to drive revenue, we talked a lot about different pieces, but I'll tell you, at the end of the day you're either selling a vitamin or an Aspirin. And when times get tough, people stop buying vitamins.
+
+So if you can figure out a pain point and make it easier for folks, that always is easier than everything else. So for NATS, to be honest with you, it's two-fold. One is OpEx spend is too high; too many moving pieces, or just it's too expensive to put it on Google Pub/Sub if we're trying to do two million messages a second type stuff... So it's like "Great. We can cut your OpEx out. No big deal." Or it's that pattern we talked about early in the show, of "I've got lots of remote thingies that I all want to communicate, East/West, North/South, those central things, and I don't wanna have to worry so much about security", and it's just one ubiquitous communication stuff.
+
+So those two pain points are what's mostly driving us as a business. Maybe not NATS as a project and an open source technology, but us as a business; it's solving those pain points.
+
+**Mat Ryer:** Great. Very interesting. Thank you very much. Thanks for all the insights into the commercial side as well. We often don't explore that on this show.
+
+**Johnny Boursiquot:** That is true.
+
+**Derek Collison:** Happy to share it, although I'm mostly probably wrong, so...
+
+**Johnny Boursiquot:** \[laughs\]
+
+**Mat Ryer:** Fair enough. Like HTTP. \[laughter\]
+
+**Derek Collison:** Yeah, \[unintelligible 01:10:01.09\]
+
+**Jon Calhoun:** And the hard part there is I think almost every open source project has been mostly wrong when trying to figure out how to build a business around it. Even ones -- I'm thinking of like CoreOS... I don't know how well they did, but they had to be acquired, and I assume that if they had a better alternative, they wouldn't have done that... So you see ones like that, and I'm like "CoreOS seemed like it was doing very well, and... Unfortunately, no."
+
+**Derek Collison:** Yeah, and it's interesting... Building a software company - you can get a lot of mileage out of thinking of it as a psychology problem, than a technical, or go-to market strategy. I try to frame everything now as "What is the psychology of the consumer?"
+
+If you're building a kernel - except if you're Microsoft, which even there they don't have to let go of it shortly, is "Oh, those are always free." Even though there's probably multiple hundreds of millions of dollars of expertise and investment into these, the consumer bias is it should be free. So always ask yourself, "What does my consumer look like, and what is their bias around what I'm trying to offer them?" If you resentingly say "Oh crap, they're gonna think it has to be free", you might wanna rethink what you're trying to do.
+
+**Mat Ryer:** Can we just all start going into stores and thinking "This should all be free"? Will that work, if we all do it? \[laughter\]
+
+**Derek Collison:** That's where I said OSS bundling with hardware is one of my three models, because the consumer bias around physical things is you have to pay for them. I think we were talking earlier about a book. Well, I have to buy the book, right? Even if it's audible, I have to buy the book.
+
+By the way, with our IoT strategy, which is slowly developing and we're launching it, the notion of saying "Hey, there's some functionality, but it's built onto this little teeny thing that you buy", even if it only costs you like $8 or $30, the consumer bias is it's not zero, which is the biggest thing. So the hardest is to go from zero to non-zero with the purchasing and consumer bias.
+
+**Johnny Boursiquot:** \[01:12:00.04\] Indeed.
+
+**Derek Collison:** Thank you guys for the time. I really appreciate it and I enjoyed it. Hopefully, the listeners got something out of it, but... I appreciate the invite.
+
+**Johnny Boursiquot:** Yeah, it's been great having you. We've learned a ton about distributed messaging systems, and the excellent work you're doing with NATS... I definitely wanna go try it now. I'm used to some of the other ones we've mentioned earlier, and I'm sure some of our listeners are definitely gonna be trying it out as well. It sounds very, very cool. Thank you so much for being on the show.
+
+Before we wrap up, I do wanna mention that for those of you who are supposed to go to conferences or speak at conferences, meet with friends at conferences, and obviously the state of the world right now prevents that - we have a lot of our conferences moving online, and virtual, which is great, because it still allows us to maintain a tight-knit community.
+
+I believe the very next conference coming up might be the GoGet community. GoGetCommunity.com is where you wanna go check out the next virtual conference coming up. Mat is gonna MC as well, along with Mark Bates, and your truly is gonna play a small part/role in there... So yeah, definitely check that out.
+
+If you have suggestions for the show, if you have your own unpopular opinions you wanna throw at us, that's fine - we'll take them on in strides, and try to keep coming up with great show topics for you. Again, Derek, thank you so much for being on the show. So long, everybody. Have a good one.
+
+**Outro:** \[01:13:39.05\]
+
+**Derek Collison:** If you guys have two minutes, I'll give you one last story to top all of it, \[unintelligible 01:15:22.09\] We were doing TIBCO, so we have all of the large financial institutions; every single one - Goldman, Lehman, everybody. And we were partnered with Sun, so you always had to run our software on Sun in the financial \[unintelligible 01:15:34.28\] And the CEO came into me one day -- I'm in Palo Alto, and he's got a suit... The guy walks in, he hands me the suit, and I go...
+
+**Jon Calhoun:** We are live, just so you know.
+
+**Derek Collison:** That's fine.
+
+**Jon Calhoun:** Okay. I didn't want something that shouldn't be live going out...
+
+**Derek Collison:** That's fine. So I said "What's this for?" and he goes "You have to fly to New York. There's a problem." So I fly to New York and I come into an unknown, large financial institution, and they go "Your software sucks." And I go "Okay, how is that?" and they said "Well..." - and I can probably say this now, since it's a non-existent formal company... They go "We bought this multi-million-dollar Sunbox, and we run your stuff on it, and it's terrible. They can run faster on a desktop box." So I sat in the room and they were really not happy. They didn't like it that they had to wait six hours for me to arrive, but that's how fast planes travel... \[laughter\]
+
+So I'm literally having people coming into this old-school server room; it's freezing cold, you're sitting in there, typing, or whatever, with CDs, and stuff... And it took me probably four hours to figure out what's going on. And it wasn't us. But that never helps anyone, usually. So the person came in and I said "It's not us." They said "It is you. It's your software. It sucks." I go "It's not us." I said "It's the operating system." And they said "No way."
+
+Long story short, the CEO of that company called the CEO of Sun and said "Hey, I need someone here in six hours." So I get to sit around for six hours. They said "You can't leave", so I couldn't leave. I got to go to the bathroom and that was it... And this person comes in and just unloads on me. The same thing probably happened to him; they walk in with a suit... "Here, put this suit on, go to airport" type of stuff.
+
+I said "I promise it's you." Or the kernel, sorry. Not you. Blame the problem, not the person. And he was yelling at me and yelling at me. So in that six hours I had to wait for him to show up, I wrote a program. And all the program did was it said "Hey, find out where the interrupt handler is, and then on every other core run a busy loop, meaning you totally take out all the other cores." So our software was running, it was running really bad, and I said "Watch this" and I go "Click!" and all of a sudden our rates went up. Not big, but they went up pretty good. And he's like "Oh...!" And I control-C-ed my app, and then -- of course, he goes "What did you do?" I said "I just pegged all 20 CPUs except for" -- or I can't remember how many they had; they had a lot. It was the most expensive Sunbox you could buy.
+
+So long story short, they had this weird thing where they were finitizing network interrupts and scheduling us on the same ring, so we were just sitting there, waiting for each other non-stop all the time. But once you did that, the OS goes like "Crap, I can't do that" and you've gotta move him somewhere else.
+
+And we had a good laugh at the end, but it was a tense 12-14 hours. So I remembered that, and said "Hey, even if it's not your problem, show up, own it like it is your problem. And when it's not your problem, remember it could have been, so be nice."
+
+**Johnny Boursiquot:** Right. That's a good lesson.
+
+**Mat Ryer:** Yeah, that's great. Actually, it's a shame that one wasn't in the show.
+
+**Johnny Boursiquot:** I know, yeah. That would have been a good one. Maybe we can splice it back in.

Cloud Native Go_transcript.txt

@@ -0,0 +1,527 @@
+**Mat Ryer:** Hello, and welcome to Go Time! Today we're talking about cloud native, whatever that means... We're gonna find out. Joining us today, Johnny Boursiquot. Hello, Johnny.
+
+**Johnny Boursiquot:** Hello there.
+
+**Mat Ryer:** How are you doing?
+
+**Johnny Boursiquot:** Feeling a little cloudy...
+
+**Mat Ryer:** Oh, that's good. That's good today. \[laughter\]
+
+**Aaron Schlesinger:** I like it.
+
+**Mat Ryer:** And we're also joined by Jon Calhoun. Hello, Jon.
+
+**Jon Calhoun:** Hey, Mat.
+
+**Mat Ryer:** How's it going?
+
+**Jon Calhoun:** It's going well.
+
+**Mat Ryer:** Great. This conversation intro is really good. Let's see if I can lift the spirits...
+
+**Jon Calhoun:** I'm just ready to get past this and get to the cloud-native part. \[laughter\]
+
+**Mat Ryer:** I know, you're so excited.
+
+**Jon Calhoun:** I, like you, don't know what we're talking about today.
+
+**Mat Ryer:** Yeah, Jon doesn't know what it is either. Well, we're also -- I've got very special guests today, as well. We're also joined, believe it or not, by Aaron Schlesinger. Yes!! Hello, Aaron.
+
+**Aaron Schlesinger:** Hello, hello.
+
+**Mat Ryer:** Welcome back!
+
+**Aaron Schlesinger:** Thank you. I think you get better every time, with the last name there.
+
+**Mat Ryer:** Ah, yes. Thank you.
+
+**Johnny Boursiquot:** He's so excited too, when he gets it right.
+
+**Aaron Schlesinger:** Yeah.
+
+**Mat Ryer:** I introduced him at a conference once, and spent a lot of time learning how to do it... And I became the fastest Schlesinger swinger in the West. \[laughter\]
+
+**Aaron Schlesinger:** That was the talk, actually. It was just you learning...
+
+**Johnny Boursiquot:** Learning his name. \[laughter\]
+
+**Jon Calhoun:** He's like going over all the background of it, and how they started pronouncing it this way...
+
+**Mat Ryer:** I'm actually really interested in that.
+
+**Jon Calhoun:** Live chat with his grandma... \[laughter\]
+
+**Johnny Boursiquot:** Etymology of parents' name... Yeah, cool.
+
+**Mat Ryer:** \[03:54\] Well, speaking of etymology, cloud native - I genuinely don't really know what this means. Now, I looked on the CNCF website on GitHub actually, and they have a nice description for it. And they have it in lots of languages, so which one should I read? I could do the German one maybe, or... Oh, maybe the Korean one could be cool. Maybe Jap-- do you know what, I'm think I'm gonna take a stab at the English. Cloud native technologies empower organizations to build and run scalable applications in modern dynamic environments such as public, private and hybrid clouds. Containers, services meshes, microservices, immutable infrastructure, and declarative APIs exemplify this approach. Okay, does that help us?
+
+**Jon Calhoun:** Is that like the UK English version? Is there like a US English version? It's probably dumbed down a little bit.
+
+**Aaron Schlesinger:** Yeah, dumb it down.
+
+**Johnny Boursiquot:** Is there a five-year-old version? What does this mean?!
+
+**Aaron Schlesinger:** Yeah, explain it like I'm five.
+
+**Mat Ryer:** I need that... Well, what do we mean then by cloud native?
+
+**Aaron Schlesinger:** I'll take a stab at it...
+
+**Johnny Boursiquot:** Stab it please.
+
+**Mat Ryer:** Stab it to death.
+
+**Aaron Schlesinger:** I'll only do it in four paragraphs instead of the five that you got there.
+
+**Mat Ryer:** Okay.
+
+**Aaron Schlesinger:** So to me, cloud native is the new way - maybe the trendy way, but I'll just leave it at the new way - to build applications for production. We call it new because there are new technologies, like containers - Docker, runC, and all that good stuff... But also there are new techniques. And the one that I go to every time I talk about this is the declarative API that you mentioned, Mat, in that description. That's a really powerful idea, because you don't write any more scripts to deploy to cloud-native systems. You do write YAML... And the rest of this podcast is now gonna be about YAML. \[laughter\]
+
+**Mat Ryer:** In YAML, I thought you were gonna say. Now we have to speak in YAML, including somehow pronouncing the white space...
+
+**Aaron Schlesinger:** Yeah, yeah. Tab-tab... \[laughter\] Yeah, so you describe what you want to happen, instead of how it should happen, essentially.
+
+**Johnny Boursiquot:** Aaron, you're jumping into the implementation details of all a bit. Let me add some preceding layers; speaking of layers... Let's try to define cloud native as sort of a layered cake, if you will. So your first layer -- first of all, let me go before that. Back in the old days - I'm making air quotes here - you used to have your monolithic applications that run somewhere in a data center with your rack of 1U servers, disaster recovery maybe, you have multiple data centers, and all these things, and you're managing all this infrastructure. You have a team dedicated to swapping out hardware, and all these things.
+
+And then we're like, okay, the cloud providers, like AWS, and now we have GCP and Azure - these things exist, so we don't have to have our own data centers. Why are we spending so much money managing our own hardware? So we're like "Okay, let's take advantage of infrastructure as a service."
+
+And then, what started to happen is that, okay, now that we have people taking care of the hardware, and the virtualization and all this stuff that we don't have to take care of ourselves anymore, we started thinking "Well, how can we then make developers more productive?" We've got infrastructure relatively taken care of; how do we make developers productive? Because in the world of monolithic applications - so it is said - things take longer to change, you don't have as much flexibility in all these things, and people are kind of stepping on each other, release processes are long, and all these things that folks attribute to monolithic applications, which - some of it has merit, some of it is just incorrect, but we'll get to that...
+
+\[08:19\] And then now we're like "Okay, let's build ourselves a layer here of abstractions, so we can actually allow developers to be more productive." So you take the cloud infrastructure, the infrastructure as a service component, you add on top of that a nice smattering of schedule and orchestration - this is where your ECS, your EKS, your Kubernetes, your OpenShift, all these things, this is where they live. And you're like "Okay, now we need some services, some application and data services to lay on top of that." So we put a nice smattering on top of that. This is where your databases live, your storage layer, this is where you have provider-supported building blocks and application building blocks. "Hey, you don't wanna run your own Redis caching server? We'll run that for you." Or "Hey, you can use DynamoDB. You don't have to run your own database, we'll run that for you", these kinds of things.
+
+Now we're like "Okay, how do we abstract things even more?" And now we have application runtimes. We used to call these things middleware back in the day... But that layer -- now we start to standardize things like logging, and events, and tracing, because with cloud native comes the whole shift to break up your monolith, and instead of having one big ball of mud, you've got now 500 balls of mud, smaller balls of mud... And they all need to talk to each other...
+
+You kind of need to observe the system, you kind of need to know what each piece is doing when a customer makes a request... It hits one frontend, and then that talks to a dozen other backends, and if something happens, you don't know where to look. So you need instrumentation, you need observability, you need these things to be connected to each other. This is where you've got the service meshes, and all these new, fancy things; the Istios, and things like that - that's the layer where they live. And then you've got your application code. This is what we've basically trying to get to all along - the Nirvana, developer productivity, where the only thing you need to do is write your code, package it up in a container, and then ship it. This is where your Dockers and Rockets and things live. This is where your serverless is - even another layer of abstraction on top of that.
+
+So basically, we're just trying to get to the point where we can actually ship software faster. So that's the layered cake. Visualize it. And it's complicated.
+
+**Mat Ryer:** Wow. I could listen to that for hours.
+
+**Jon Calhoun:** While Johnny is explaining how this is the modern way to do it, I'm looking over at an old Linux box sitting on my shelf, running a server that I use for stuff... And I'm like "Yup, that's not modern... But okay."
+
+**Mat Ryer:** What's running on there, Jon?
+
+**Jon Calhoun:** It's some backup file storage and some other stuff like that, that I sometimes wanna access if we're off traveling and my wife wants to get some photos off of it, that sort of stuff.
+
+**Mat Ryer:** Oh, right. That's cool though, that we have these ideas. And they obviously have come out of people solving the same problems again and again... And once you've done that a few times, and actually sometimes before that as developers, you sort of see these patterns, these common things that we could then build abstractions on. And it is very useful from a developer's point of view, of course, because -- and this is how I actually use this technology personally; I write applications, and I let then something else (the cloud) do all the heavy-lifting of making things work, and making things scale.
+
+\[11:48\] In our project, if we get a sudden spike of traffic, then more instances are automatically spun up to deal with that. And they'll spin down as well after, when they're not being used. So things like that then become possible, because we have these abstractions. But what does it mean for application developers then? Do we do things differently now, now that our code is gonna run in this different way? ...it's running in this abstract environment, versus on Jon's computer in a corner of his office.
+
+**Johnny Boursiquot:** Jon, you need more reliability, man... My apps keep crushing, man. You need some redundancy. \[laughter\]
+
+**Jon Calhoun:** I mean, luckily, these are things that don't need really high reliability. My reliability is a back-up battery plugged into the wall in case the power goes out.
+
+**Mat Ryer:** Oh, that's fancy.
+
+**Jon Calhoun:** That's pretty much it.
+
+**Johnny Boursiquot:** That is fancy. I don't have that. \[laughter\]
+
+**Aaron Schlesinger:** I wanted to add something to what you said, Johnny... There was a part of that cake in there that made a transition from what I think of as the traditional ops role... I don't really know if it's DevOps, or ops, but that kind of role... And it made the transition into the developer role, the developer realm. And I think that's a really important one, because we have a different transition now than before we had cloud native stuff. Before, we were talking about IS, VMs, and VNets, and all of that stuff. But now we have things like Kubernetes, that abstract over a bunch of compute resources and network resources, and storage, and all that stuff.
+
+So now, whether you look at Nomad, or Kubernetes, or ECS, or even App Engine, Heroku, that kind of stuff - now there is an API to abstract over your entire cloud, or your entire subscription to the cloud, or whatever you've got... Even over your entire physical data center as well. So that is really interesting to me now, because we kind of have an operating system here that describes your entire system; whether it's 1,000 VMs, or two VMs, or Jon's Linux box, whatever it is...
+
+And Jon, if you had it, of course, you could add more Linux boxes, and the system would work the same. I think that's a really powerful abstraction that lets us pretty easily standardize on that transition from traditional ops to the development experience you were talking about, Johnny. And your experience too, Mat, because we've got now this API at the OS layer, and now we've got standardization on that, so we can move up the chain and get to things like serverless... And I know you use App Engine, Mat; so you get to stuff like that, and maybe beyond. I think we'll have to see.
+
+**Jon Calhoun:** So can I ask you two a question, Johnny and Aaron? What problems were you first facing when this type of solution really looked like it was going to save you a lot of headache? Do you remember, when you first started looking at it, what problems you thought it would solve? Or did you have something in mind?
+
+**Johnny Boursiquot:** Before I answer, when you start talking to folks about what does cloud native mean, inevitably the conversation drifts into "Well, you need to decouple components, you need to transition from your monoliths to microservices and nanoservices, and serverless", and all these things that are supposed to work with each other, and all that stuff... There's an inevitable push towards breaking apart. And very seldom are you told why you need to do that, which is why there's such a backlash around microservices and all these things. It's trendy to have microservices, two-pizza teams, and all this stuff, but there's a reason why you break up a monolith... One of the reasons being to be able to independently scale one piece that has higher demand on it than another piece.
+
+\[16:11\] Because without that, you're forced to now scale your one application vertically. You need to beef up the server, add more RAM, more CPU power... That's called vertical scaling; you need a bigger server to run your one thing. Whereby if you have smaller components - well, this component only requires two logical CPUs, and another requires eight logical CPUs, then these things will have to live in the same box. You can sort of distribute that. You can scale these things independently and horizontally. So that is one of the great benefits of having this break-up into small components.
+
+So that was my first true realization of what -- I guess even before cloud native... I think the term cloud native, if I'm remembering correctly, came either around the same time as cloud native, or it was before it; I can't remember exactly. But I think the idea, the promise of microservices in cloud native in connection to that is have enough flexibility. It's all about flexibility. Because at the end of the day, we're doing this not because it's cool to break things apart into smaller pieces; we're all engineers and supposedly we all like Lego pieces to play around with and assemble things together - that's all well and fun and all that, but at the end of the day we work for businesses, and businesses have the goal of making money, and one of the ways to make money is by being innovative and being agile... And I'm really using the lower-case word agile here, to be explicit... \[laughter\]
+
+So the innovation comes in terms of speed, get to market first, or go to market quickly, to respond to market demand, and that kind of thing... When you have smaller pieces that you can orchestrate and put the horsepower where you need it kind of thing - all these things feed into the innovation. And having developers who can work on different pieces of the platform - it's not one big thing, and then you have to worry about deploying that one big thing; you have release managers whose entire job is to just move things around, and just prep things for release... You can have that flexibility of teams being able to ship things independently, and still have some sort of interconnectivity within these things. That's the flexibility; that's really why we have cloud-native applications. Not because it's cool to break out of a monolith into microservices, not because it's cool to run infrastructure.
+
+**Jon Calhoun:** Okay, so making sure I got that right - and I know this paraphrasing is gonna be butchering it... But it sounds like a huge part of it was that your organization needed to be more efficient, and part of the ways you could do that was by breaking things up into smaller pieces that could be independently deployed, worked on, everything like that.
+
+So I ask that not because I think cloud native is a bad thing... Because I worked at Google before I went off into my own stuff, and while it wasn't Kubernetes and Docker, they had essentially their own internal version of getting resources for all this stuff... And Google wasn't just spinning up a monolith. I think we all pretty much know that wasn't what was happening. They don't make computers that beefy that I'm aware of.
+
+So you know, it's just one of those things where I'm trying to figure out from other perspectives what problems you were solving as you dove into them... Because from my perspective, I haven't seen an obvious need for these things, and I think part of it is just because of the things I'm working on... But I also want to know, are there problems that I'm not paying attention to, that these might be able to help me with?
+
+**Aaron Schlesinger:** \[19:44\] The last job I had, we were building a platform as a service... And it was all these containers. We started on CoreOS, with their Fleet system. I think that's deprecated now. It was sort of "We will take a container and we'll put it on X number of machines", and that was what Fleet did. And that was pretty powerful at the time, because you had these sort of beginnings of that abstraction, of "I don't need to care about VMs anymore. I can give an API a container name, and it'll do its thing."
+
+And once we got to that point, we then had to start breaking things apart, because a platform as a service has lots of different logical components that don't necessarily fit together. And this is right along the lines of what you said, Johnny. It has a Git SSH server, it's got a logging component, it's got an administrative interface, and a control plane... And the list goes on.
+
+So once we hit that point where we said "We just can't have a monolith with all of that stuff in it at once, because managing that thing, opening all the different ports and managing certificates and all that - that's just not feasible for us." So once we got all of our stuff running on Fleet, we then had to reinvent the wheel and figure out how to do secrets, and distributed locking, and all that stuff. And then Kubernetes came out and then we just adopted all the primitives that Kubernetes gave you... But stripping away the Kubernetes part, even though that was great, and stripping away Fleet as well - the idea that we could have implemented it ourselves would have been painful, but we probably could have - I'm not gonna say definitely; we probably could have... It's the fact that - yes, we had a technical requirement that stuff was split up, while at the same time stuff could interact with the other stuff.
+
+Service A could interact with services B and C in a way that was manageable and that didn't require two different operations and release management teams to manage services A, B and C. And for me, right at that moment - and I remember this - I was dreading having to build those systems, to manage all the things and route network traffic and all that stuff. And once we've found Fleet, that was when -- we went down this road of starting to think about an abstraction, and starting to think about independently scaling, and starting to think about how to organize the team around all these different services, and manage the sort of organizational aspect of this... I started thinking about a lot more things, too... But right then and there was the seed that got planted in my mind, that started me down this whole cloud native road.
+
+**Break:** \[22:52\]
+
+**Mat Ryer:** So when you're building something simple and little, would you recommend that people still build it in this way? Or should you start with a monolith, and then when you need to break it out later, do it then? How would you approach that? What would your advice would be?
+
+**Aaron Schlesinger:** I'm gonna use my Unpopular Opinion card right now... \[laughter\]
+
+**Johnny Boursiquot:** There you go.
+
+**Aaron Schlesinger:** Yeah, absolutely not. If you're one or two people, write all the code in one repo and deploy it all as one thing... Because the overhead of taking your two-person team and trying to manage all of the different microservices, and getting them to talk together and play well together - that effort is gonna outweigh the actual code you write. Let me share one more metric there...
+
+**Mat Ryer:** Go on, please.
+
+**Aaron Schlesinger:** I've worked on and seen apps that have more Kubernetes YAML than they do code... So that metric is disturbing.
+
+**Johnny Boursiquot:** That's a sign right there you're doing it wrong.
+
+**Aaron Schlesinger:** Yeah, yeah...
+
+**Jon Calhoun:** That's when you ask "Is this a project for school?" Because otherwise it's a little scary...
+
+**Aaron Schlesinger:** Yeah, yeah, for sure.
+
+**Mat Ryer:** I was thinking then - so it's not that you can just use Kubernetes and everything's easy. There are still challenges that come with deploying in this way. It is still a big trade-off, right?
+
+**Aaron Schlesinger:** Yeah, absolutely.
+
+**Johnny Boursiquot:** To follow up with Aaron's statement here - I totally agree you should definitely not jump into microservices and all these things, and doing orchestration, and all that stuff, especially if you don't have the staff for it... It's a different competency; you're required a different kind of mindset. Yes, you can be a developer and be operations-minded. I can define myself as that; I'm a software engineer who happens to do operations. So the thing to keep in mind is that if all you need is to ship an app and have it run, if you stick with the monolith, use a pass; use Heroku, because they're taking care of all that for you - I guess I'm unconsciously biased here, because I work at Heroku - or something like it, to run these apps, so that you don't have to worry about...
+
+Again, remember why we were leaning towards cloud native, or want to take advantage of cloud native applications? Keep the ultimate goal in mind - it is for developer productivity. Your job is not to run operations; that is not where the business makes money... Unless your business is to build a Heroku or a PaaS, you have no business really trying to run your own Kubernetes, or something like that. To me - and I'm gonna get some heat for this, but that's kind of nonsensical. You're not building a platform as a service for other people to run stuff. Maybe you're an enterprise, you have lots of different departments that have a lot of business software, and things like that, and maybe having your own orchestration tooling and all these things to allow different teams to develop/deploy things is the right approach... But you have to get to a certain scale, a certain level right before you can reach that. If you just wanna launch an app, just run the darn thing on a PaaS.
+
+One of the mistakes I see developers make all the time is they get lured by all the announcements coming out from re:Invent, and from Google's conference, or Azure's conference. They see all these nice, shiny bells and whistles... Listen, AWS released 2,000 products last year; that's an average of six a day. That's ridiculous. So all these things are just lures, and to "Oh, let me see if I can integrate some ML services into my app here. But before I can do that, I need to set up an AWS account, I need to run things on EKS, or ECS... I can't just deploy this thing on an EC2 instance. I've gotta break it up. I've gotta have microservices, I have to have a container, and now I need a CI/CD pipeline, I need to break this thing into different storage components, and I've gotta use DynamoDB, and I've gotta use S3..." Now you're bringing all these things so that you can actually take advantage and do things the way it is prescribed, the way it is being marketed to you as a developer.
+
+\[28:34\] Again, it's tempting, it is very tempting, because we're engineers, and when we see new and shiny, we're like "I wanna use that." Or you're thinking maybe -- and I think we've all seen resume-driven development here too, and said "Hey, I wanna use that tech. Let me find a reason. Let me convince my boss that we need tech so and so." Now you're bringing in that complexity into your world, and now you have to orchestrate around it... It's basically like a self-fulfilling prophecy almost. You're bringing all these things, and now rather than having one problem, which is ship this app and then get paid for it, now you have two problems - ship this app and all the small pieces that go along with it; now I have to manage, and orchestrate, and all these things, and then still trying to make money.
+
+**Mat Ryer:** So why is Kubernetes so popular then, if most people shouldn't really be paying any attention to it, for example?
+
+**Aaron Schlesinger:** I think a lot of people have really good use cases for Kubernetes.
+
+**Mat Ryer:** So are they running it themselves, or are they building platforms as a service? ...or platform as a services. I don't know how to pluralize that.
+
+**Aaron Schlesinger:** Is it like deer, where you just say deer?
+
+**Mat Ryer:** \[laughs\] I don't know. Do you just say deer? Are people building deer?
+
+**Jon Calhoun:** Do you just say PaaS, and since it's an acronym, then you can put an s, or whatever ending they need on there?
+
+**Mat Ryer:** Brilliant. Thank you. You've solved it.
+
+**Aaron Schlesinger:** So I think it depends. The amount of PaaS increases as the organization size increases... Because once you get to your 30, your 40, your 50-person engineering groups, now you have to really draw those lines between "Oh, hey, here's the DevOps team. Oh, hey, here's the ops team if we have that. Oh, hey, here's release management. Hey, here's development." And those roles - that's a Venn diagram; that's not separate and mutually-exclusive groups. So now you're gonna have people on Kubernetes who are touching it as YAML developers; and you're gonna have other people writing Flask apps, who don't care about YAML. They wanna get their app up, and they need a public IP, and they need an SSL certificate, and whatever else.
+
+So those are really, really different personas, and at that point - now you're looking at a group of probably SREs who are gonna be writing code to make it really easy for developers to deploy that Flask app, but also make it really hard for them to mess up the Kubernetes install, and make it really hard for them to do something that's gonna deploy an API that doesn't have an SSL cert, for example, or something that's bad practice, pretty much.
+
+**Mat Ryer:** I didn't realize there were YAML developers now. \[laughter\]
+
+**Johnny Boursiquot:** Yeah, that's us.
+
+**Aaron Schlesinger:** That's us, yeah... For better or worse.
+
+**Mat Ryer:** Right.
+
+**Jon Calhoun:** They didn't ask you to start YAML Time yet? \[laughter\]
+
+**Mat Ryer:** It's YAML Time!
+
+**Aaron Schlesinger:** Today is the day. This is it now. \[laughter\]
+
+**Johnny Boursiquot:** \[31:46\] Here's another thing I'll add to that here... I think sometimes teams are also lured by the perceived discipline that running something like a Kubernetes brings. And by that I mean the microservices approach to tie these things together, the microservices approach requires -- basically, you have a network boundary between these things. So now these components have to have clearly defined APIs to talk to each other, which kind of forces some discipline... Hey, team X, you promised that you'd have these end points for this thing; or you promised you'd implement this RPC service for that thing, whatever it is. So it creates some nice sort of compartmentalization and some discipline. You have different managers managing different teams, different timelines, different deliverables, all these things. So it bubbles all the way up.
+
+That discipline that we're sort of inherently seeking as part of adopting these orchestration tools - I think that lure is dangerous... Because if you're an undisciplined team before orchestration tooling, you're still gonna be an undisciplined team after orchestration tooling. The tool is not gonna make your team more disciplined. You can have discipline around a monolith, you can have discipline around established processes for delivering software, which has nothing to do with the actual tooling itself. Tools are enablers.
+
+Basically, if you're relying on tools to help you establish discipline, that's like saying "Okay, I'm gonna buy a piece of software and I'm gonna shape my business to fit into how this software is coded. Now you're following somebody else's business processes, however they've chosen to encode that. And businesses -- I'm pretty sure your business is probably not seeing things the way you are. So again, try to be disciplined about your practice, the engineering practice, the software delivery practice that you have internally, before you attempt to bring in new tools into your world, because that's just gonna add to your chaos.
+
+**Mat Ryer:** That is a really interesting thing, because I kind of felt like yes, if you have a framework - and I know some companies essentially do this; they have a prescribed way, including command-line tools that will generate the stubs for services, and things. So in a way, and because everything else is automatic, you do get a lot for free; and like you say, you can't make those kinds of mistakes. But it does seem a little bit too good to be true. So that is very interesting, to hear that you still have to build good software... Which is kind of nice, isn't it?
+
+**Aaron Schlesinger:** I'll really quickly riff on that concept of discipline... So this will probably make me sound like a curmudgeon, but I--
+
+**Johnny Boursiquot:** "Back in my day..." \[laughter\] Here it comes...
+
+**Mat Ryer:** Can you do it in an old voice? That'd be great.
+
+**Aaron Schlesinger:** Maybe next time... \[laughter\] I guess I do sort of the misery-driven development thing, personally... So with code, what that means is you copy and paste it three times, and then you figure out what the abstractions should be. For infrastructure, for me, I need to see the mistake maybe 2-3 times before I can justify finding a technology that fixes that. So with Kubernetes, we're talking about a mistake like "I deployed service A, and it broke everything, because I didn't know what was talking to service A..." Or it wasn't compliant, whatever it might be. You almost have to crash your app (almost) in order to make that jump... Because Kubernetes is such a big leap, and a big commitment also...
+
+**Mat Ryer:** Well, I think that should be quite good for people to hear that, because in a way, people that just don't have a clue about this stuff, in a way we're saying "Don't worry about it yet. Focus on the bit your care about, the bit that's important and unique to what you're doing, and you will then be able to solve these problems later." But how easy is that? Should we be building even our monoliths in a specific way, with an eye to the future? Or should we just build them however we have to build them to get it working?
+
+**Johnny Boursiquot:** \[36:16\] Nein, nein, nein, nein...! No, that's me trying to have a German accent. No. Like--
+
+**Mat Ryer:** Oh, I thought you were doing an SLA... \[laughter\]
+
+**Johnny Boursiquot:** No, no... I can see how that-- here's the thing... So the earlier part of my career as a software engineer, I'd read the books, read the blogs, watch the videos, watch the conference talks... I'd basically be trying to become the perfect engineer, who's building reusable software, reusable components, and trying to basically have well-orchestrated systems... All these things are still good. But over time, I've been around this industry long enough, going on 23 years now; I've been in this industry long enough, I've been around enough businesses long enough, built enough businesses and applications over and over and over again to see that this dogma that we have around building the perfect system, kind of like Clu, and -- what's that Disney...? The TRON movie. Basically, solely concerned with building the perfect system, that you're failing to see what surrounds you. You're failing to see that "Okay, well, this perfect application that you wanna build right now, and have all these abstractions and these reusable components that you think you're gonna reuse - nobody's asking you for these reusable components yet, but you wanna build them ahead of time. You wanna create a future that you may have... And then you find out next month that the business is going in a different direction on all the painstaking work you've spent trying to build these abstractions and layers upon layers upon things is now trash... Because you don't know what's going on above you, how many levels above you.
+
+Now, in a smaller company, in a startup or whatnot, that tends to happen a little less. You have a more -- the information radiates to you, so you kind of get a sense of "Okay, this is generally what's happening", or maybe you have one product, you have one line of business focus, and you know exactly what you're building... So you can kind of a little bit look ahead... But not too far ahead, because business is always changing. Business is what controls what your software is going to do tomorrow, not you.
+
+So trying to build abstractions early on in the process -- again, I take Aaron's approach; I need to see something showing up 2, 3, 4, 5 times, and become a pain to deal with before I create an abstraction around it... Because I don't know what the business is gonna want tomorrow.
+
+So this whole notion -- we ingest all these notions, these best practices, as we like to call them in our space, around how to build reusable software, and all these things... Yes, but also add a layer of realistic lenses to that. Look at your software as something that needs to evolve because the business side that it's serving also needs to evolve. Don't get way ahead of yourself; that's just recipe for disaster.
+
+**Mat Ryer:** That's very interesting. And I think even if the business doesn't change, still, the best time to design solutions isn't at the very beginning, because we know the least about the problem then, don't we? Even if the business doesn't end up changing, still, the learning that you get from building is so valuable. Ernest Hemingway said "The only kind of writing is rewriting." He was obviously talking about poetry, novels, books... But that applies to software too, doesn't it? When you rewrite things, you're doing it usually for a good reason, and because you've learned something significant... Which can only help.
+
+**Break:** \[39:51\]
+
+**Jon Calhoun:** Earlier we mentioned that, Mat, you run an App Engine... And I think one of the reasons people levitate towards this "I need to do this all now" is that in their mind they're like "If I don't do it now, it's gonna be a lot of work later, and we're better to just put in the time now." And like you said, usually you don't know what your pain points are gonna be and what's gonna work for you...
+
+But what I like about a Heroku or an App Engine or any of those is that I think they get the major ones out of the way. They generally suggest "You're not running your database on the same server as your code. Your code usually doesn't have a file system that it can rely on being on the same physical hard drive." That could go away; they could talk to a different server. And if you get those major parts right, I feel like you're in a good position to start refactoring towards that, and that covers your major bases...
+
+At least from my perspective, when Heroku and App Engine and those got big, that was probably my favorite part - that it kind of got you the major ones, the major wins, and then later when you get to a really high scale, you'll know at that point what to focus on, because it's gonna be specific to your business.
+
+**Mat Ryer:** Yeah. I mean, you're still designing APIs, aren't you? If you're building a monolith, you're still internally building APIs. You're gonna have objects, or structs, or services, or something; you're gonna have functions with inputs and outputs and things... So you are already designing APIs. Of course, there's something nice about when you do have microservices, because you have to communicate in a more official way. You're not gonna necessarily have compiler time help making sure that you're passing in the right types for arguments, and things... So it is a little bit different. But that kind of thinking can be quite useful, I think, even if you're building a monolith.
+
+For example in Pace we have different services inside, even though it is a monolith. We still break things up by -- we sort of group them up by functionality.
+
+**Johnny Boursiquot:** And that's the discipline that you want, right? Expecting a network boundary to help you create that discipline around your componentization efforts is the wrong approach. Again, infrastructure is not going to make your team more disciplined. You have to be disciplined, within a monolith. Again, we seem to somehow now associate monolithic applications with bad. Monolith - bad. Microservices - good. It's never that simple.
+
+I've seen beautifully architected monoliths, that do the job really well. They have clean boundaries and separations between different components within the monolith. When I see those, I smile inside, because I'm like "Okay, somebody is being disciplined about how they build a software."
+
+\[43:51\] Now, when you need to take something out of the monolith, to make it a standalone service, you already have a nice, clean separation. You don't have that tight coupling between all the other things within that monolith. You cannot easily take it out, and now just have an interface, some sort of RPC or HTTP interface, a REST API, whatever it is that you want; this thing you've ripped out, that is now standalone, that can be scaled independently. Again, that's the ideal scenario for me - you have a well-architected monolith where the components are loosely coupled, that you can easily take something out, put it in its own server or in its own container or wherever you need it to run (serverless, or whatever it may be), and now you have the rest of the monolith still able to communicate to that thing very cleanly. That's the perfect world, in my view.
+
+**Mat Ryer:** Yeah, I like that. Writing components that are easy to throw away also turns out to be a bit of a super-power. If you build a monolith and there's functionality that's just spread throughout the codebase, then it's very difficult to unpick that. So if it stops serving you in the way that you want it to, you get stuck with it. And I've worked on projects where people will say "Just don't touch that area. Just don't go near it, because we're scared of it", or whoever built it, the wizard that originally built it has now gone to work for Microsoft. I lost the metaphor there... So people then become scared to touch things.
+
+Another thing that is a handy thing to have is the ability to be able to rip those things out later, and throw them away. And to do that without feeling like you're really losing to much, or it's such an expensive thing to do, or a risky thing to do.
+
+**Jon Calhoun:** You describing that to me -- I've talked to a lot of different people about ORMs, and using them in your code... And I feel like a lot of the times when people have really big issues with them, a lot of the issues come down to the fact that they worked in a Rails codebase, where people basically accessed essentially the database anywhere in the code. And it led to these issues where you had code that really should be interacting with an interface of some sort to get the data, is instead writing code that's actually executing SQL queries... And like you said, that makes it so that refactoring the code, or changing it as you need to, or throwing something out is like next to impossible at that point, because you couldn't change your database implementation - at least not easily - without rewriting all of that code.
+
+But Go, for instance, is one that -- I feel like interfaces in Go make it so much easier to separate these things, without actually caring how you're communicating. Like, I don't care if this is a microservice, or if it's something that's running on the same server. I just need an interface here, and I'm gonna talk to it. That's one of the things I've really liked about Go, is that it just makes it so much easier to completely ignore that part of it and just write your code to be like "If you can get me this little set of functionality, I can do my job."
+
+**Aaron Schlesinger:** I think once you get to that point where in your Go codebase you really are taking advantage of well-defined interfaces, and have multiple implementations of them hopefully, you're in a good spot then to try to evaluate "Hey, do I need to start splitting out microservices?" But what I think a lot of people underestimate is how much technology you have to learn in order to split that stuff out, just to get one process talking to another process on local host over a network. That's hard enough. And this is all day zero, before you actually see this thing running in production. Just day zero, "I've decided I'm gonna build a third implementation of my interface that talks over the network to this thing that I've split out..." Well, you have to figure out a couple things right then and there. Am I gonna do HTTP? Am I gonna do JSON? Am I gonna do gRPC? Am I gonna wade into that 3,000-project list of things in the CNCF site to figure out if there's something else?
+
+**Johnny Boursiquot:** \[48:04\] \[laughs\] "I think I need Istio", right?
+
+**Aaron Schlesinger:** Yeah, you've gotta have Istio for that. \[laughter\] But that's a good point, because that talks a little bit about compliance, too. Am I in an organization that needs end-to-end SSL? Am I in an organization that needs to get a self-signed cert? The list really goes on and on... And like I said, that's day zero. When you push to production, you start feeling some pain about "Well, I can't debug as easily. I need a stack trace. 99.9% uptime... 99999", whatever it is. There's gonna be a long, long road to get from "It's a function call in memory" to "I have a thing reliably running over here, on another server, and I need my stuff to talk to that stuff."
+
+So like you said, Jon, you have a huge headstart if you've broken this stuff out into really well-defined interfaces... And at that point I really think -- you know, this comes down to good software design, and I really think once you have that, you have a pretty good leg up on getting it to where I can split out to a microservice. But you've got a lot of work ahead of you still, and I think a lot of people underestimate really how much work it is to get something production-ready out, microservices-based... And that's where we get caught up. That's where we end up writing more YAML than code. That's where we end up not getting to market in six weeks, or a year, or whatever it is.
+
+**Johnny Boursiquot:** Because you're still building YAML... \[laughter\] And nothing takes you faster from being a developer, to all of a sudden being a distributed systems engineer faster than trying to adopt microservices and all this orchestration stuff. Like you said, Aaron, now you have to learn a whole bunch of stuff, concerning yourself with a whole bunch of stuff... And really, in the beginning you just wanted to build this app, provide the business with a feature that they wanted, and now you're managing infrastructure, and trying to learn YAML, and Kubernetes, and ECS/EKS, whatever these things are.
+
+I chuckle, and it makes me sort of happy inside at the same time, and I think a part of me is like okay, I don't want -- and hopefully I can speak for you here, Aaron, as well... I don't think we want to be gatekeepers; I think we can sound like gatekeepers, as people who are in this space and are doing these things... We can sound like gatekeepers, and basically trying to warn everybody away from these things... We're not. I'm certainly not. If you wanna be a distributed systems engineer, if you wanna be an operations engineer, if you wanna deal with YAML all day, every day - please, come into the field. We'll tell you "Here's the resources to learn, here are the things you need to know about it." Heck, there's tons of conference talks... I mean, you know how to educate yourselves. You're in this space, most of you are autodidacts, and you can educate yourself. You will find the resources to learn and be that kind of engineer. But don't kid yourself. If what you wanna do is remain a web developer, and produce business value, and build application, whether they be monolith, or serverless, or whatever it is you wanna do - there's a track for that. But the moment you cross into this whole orchestration/cloud-native arena and concern yourself with these kinds of things, that's gonna require you to basically expand what you thought you needed to know to ship an application.
+
+**Jon Calhoun:** Do you think that's one of the things that causes confusion, is that people blend those tracks together as one big "A web developer has to know how to do all of this stuff"?
+
+**Johnny Boursiquot:** I blame the marketing. The marketing from these companies - that's exactly what they're trying to tell you.
+
+**Jon Calhoun:** And I think that's hard, because I run into people all the time who -- basically, in their mind, "I have to learn Kubernetes, and all these things", and I'm like "I've been working as a web developer for quite a while, and I don't know these things. So I can assure you you don't need to learn them."
+
+**Mat Ryer:** Yeah, but you've got it running on a Linux box in your corner, mate... \[laughter\]
+
+**Jon Calhoun:** \[52:07\] That's not my actual main stuff...
+
+**Johnny Boursiquot:** With no redundancy. You've got a backup power supplier, but you don't have redundancy in your servers... Come on, man. \[laughs\]
+
+**Jon Calhoun:** I mean, this isn't like my really insensitive stuff...
+
+**Aaron Schlesinger:** You put your app on a USB key and plug it all in out there...
+
+**Jon Calhoun:** I mail it to people and say "Alright, now that you've purchased, here's the login." \[laughter\]
+
+**Johnny Boursiquot:** Remember when we used to ship software via FTP? Those were the days, huh?
+
+**Mat Ryer:** Yeah, yeah...
+
+**Aaron Schlesinger:** I have shipped software by rsync more than once in the past six months. It's easier to go binary, right?
+
+**Johnny Boursiquot:** Yeah, that's right. You just ship it over there, yeah.
+
+**Jon Calhoun:** I've definitely done that...
+
+**Mat Ryer:** Is rsync a band? \[laughter\]
+
+**Johnny Boursiquot:** The follow-up band to NSYNC?
+
+**Mat Ryer:** Yeah, it was the one after. They were just copying, basically... They were just copying NSYNC.
+
+**Johnny Boursiquot:** Nice, nice...
+
+**Aaron Schlesinger:** Wow... \[applauses\]
+
+**Mat Ryer:** So back to Go then... This is Go Time... So Go and the cloud - are they a match made in heaven? Why is Go -- and by the way, for balance, I should say... Because I don't wanna exclude satanists... So is it a match made in heaven/crafted by the dark lord in the depth of hell? Is it kind of special?
+
+**Jon Calhoun:** I want the PG rating.
+
+**Mat Ryer:** Is there something special about Go for the cloud?
+
+**Aaron Schlesinger:** I wouldn't say special... I think Go is a very special language, don't get me wrong... But I think Go was in the right place, at the right time. I really think so.
+
+**Mat Ryer:** Because it was designed for modern ways of deploying things. It was certainly designed to make the use of multi-core processes, for example, where in previous languages that's quite a difficult thing to do. So Go was at least designed - just because it is a bit more modern - with all this context.
+
+**Aaron Schlesinger:** Yes. I should clarify - it's a great language for the cloud. I think it was at the right place, at the right time for cloud native, for this whole open source CNCF landscape. And actually, it's because of what you've just said. Because when you're a developer and you're looking to, say, build Kubernetes - go back to 2014 or whenever it was that the original folks were building Kubernetes. You've gotta build a system that does need to take care of multi-core, and it does need to be really good at networking, and databases, and more... And your choices then are - let's see... C++, C, Java... I'm probably missing a bunch of those languages that give you good access to lower-level primitives of the system. But when you're looking at Go, you do.
+
+**Johnny Boursiquot:** I see you skipping Node there... That was a sleight -- I saw it.
+
+**Aaron Schlesinger:** Throw Node in there... \[laughter\] Throw everything, I guess. Throw everything in there. But what language can you pick up that doesn't need a VM, that has high-level concurrencies/primitives built in the language, that can do networking in a couple lines of code, that's got really good support for built-in high-level networking protocols... You know there's not a lot of other stuff out there. And if you're three engineers - or however many engineers it was - at Google and you're trying to build Kubernetes, or... Shout-out to Nomad, because I saw some Nomad in the chat there... If you're HashiCorp and you're trying to build a distributed system abstraction layer, you kind of need those things, and you're gonna get up and running with something like that faster with Go than probably with C++, or Java... Not sure about Node; maybe Node will get you there just as fast... Although you don't have access to multi-core on Node, so there's a trade-off there as well.
+
+\[56:08\] So I think Go really caught on because it has most of this stuff right out of the box that you need... And it was one of the only languages that had it at the time when a lot of these things were coming up.
+
+**Johnny Boursiquot:** Right place, right time. The language of the cloud.
+
+**Jon Calhoun:** I'd say it's probably not a coincidence that a company that probably has to deal with all these types of issues was also backing a language like that -- that it came from a company like that. It would be hard for me at least to imagine Go coming from a 20-person startup that just probably doesn't have the same problems...
+
+**Mat Ryer:** Yeah, which is a good thing in a way, because it is borne out of these real environments. I think that's cool. Well, it's that time again, everybody... It's time for your Unpopular Opinion!
+
+**Jingle:** \[57:10\]
+
+**Aaron Schlesinger:** I already gave mine...
+
+**Mat Ryer:** Yeah, you did. You broke the format, mate, essentially... And you will be receiving a strongly-worded letter from me. \[laughter\]
+
+**Johnny Boursiquot:** Handwritten.
+
+**Aaron Schlesinger:** Handwritten, alright...
+
+**Mat Ryer:** Yeah, sure.
+
+**Jon Calhoun:** We'll be updating the guest doc to say "Save your unpopular opinions." \[laughter\]
+
+**Aaron Schlesinger:** I can give a second, if you all would like...
+
+**Johnny Boursiquot:** Yes.
+
+**Aaron Schlesinger:** I think I have plenty. I am a curmudgeon after all, so I can give plenty.
+
+**Mat Ryer:** Do the voice.
+
+**Aaron Schlesinger:** Back in my day...
+
+**Mat Ryer:** There we go... \[laughter\] That's what I wanted. Okay, what's your second unpopular opinion, Aaron?
+
+**Aaron Schlesinger:** Well, I would say that Go is not the only language of the cloud, or language of cloud native either.
+
+**Mat Ryer:** Interesting.
+
+**Aaron Schlesinger:** We've started to see some new applications, I guess, of languages in the cloud native space. I'll shout out to some of my old colleagues, who I wrote that PaaS with. They're doing a lot with Rust for Kubernetes right now... And Linkerd is also written in Rust. That's a service mesh. Half of Linkerd is written in Rust. And Rust is turning out to be a pretty good language because of the safety aspect. It's got some really great performance characteristics, including with concurrency... And all that stuff is memory-safe and concurrency-safe when you compile it. So it's a little bit more difficult to learn. Some might say a lot more difficult to learn, it depends on who you ask... But you get some of those same properties of Go while also getting that safety guarantee, so you don't have race conditions, you don't have null pointer exceptions, and that kind of thing. So definitely not the scale of Go yet, but I have some confidence that it's gonna get up there soon.
+
+**Mat Ryer:** Is that also one of the benefits of microservices? You mentioned something's half-written in Rust... Can you mix different technologies and pick something that maybe is better suited for solving particular problems?
+
+**Aaron Schlesinger:** I think so, yeah.
+
+**Mat Ryer:** Because that's harder to do in a monolith, isn't it?
+
+**Aaron Schlesinger:** Yeah, yeah. Of course, at a technological perspective - yeah, for sure. Because you can have Rust talking to Go over gRPC, or whatever... But I think when we're talking about the layer cake that Johnny mentioned at the beginning, you're app-level code can be - and probably should be - something like Node or Python... Because that's where you're writing business logic, and you're talking to databases, and rendering templates, and all that. I would do it with Buffalo, for example, because that's my most preferred web framework right now these days.
+
+\[01:00:11.15\] But then when you're going lower down, while you still might be writing Go, you might end up writing Rust because you've gotta integrate with some C stuff, and that's doable there... You might go and write some C, if you've gotta integrate with whatever is going on there. And this comes down to right tool for the job, but we're now applying it to the cloud native layer cake.
+
+**Mat Ryer:** Right. Hm... Good one. Any other unpopular opinions?
+
+**Aaron Schlesinger:** Thank you. I try... I try to bring the curmudgeon to the show.
+
+**Mat Ryer:** I say good one. What we're doing is taking the clips of all the unpopular opinions and then doing a poll on Twitter. So you can follow @GoTimeFM on Twitter and you can vote to decide whether that is indeed unpopular or not. So far, they've all been popular... And I don't know if it's just that the guests make such a compelling case, and we're just suggestible... Or if in fact people are just not taking the segment seriously enough. \[laughter\] Has anyone else got any unpopular opinions?
+
+**Johnny Boursiquot:** I do... But again, I have to, I guess -- well, let me not poison the well here; I'm just gonna say it, and if you think it's unpopular, that's your business. I think every startup at least, but even within the enterprise - just because you're in the enterprise and you have more operation layers doesn't mean everything needs to be microservices... But I think you should reason your way out of a monolith before you adopt microservices. Find all the reasons why you don't need to deploy an application as a monolith before you come up with the reasons, and the excuses and whatever it is you wanna tell yourself for adopting microservices and all the orchestration and baggage that comes along with it. Because if you cannot, realistically - and I don't mean just sit down by yourself, write yourself a list and then kind of weigh pros and cons all by yourself, alone... No. Get some feedback. Have your entire team weigh in. Heck, create some language around it and have your business leadership team weigh in on it if you can. I bet they'll ask you "Hey, what's gonna take the longest to do? What's gonna take longest to manage/maintain? What's gonna cost more in terms of people?"
+
+All these things factor into that decision. So basically, you need to reason your way out of not choosing a monolith before you adopt microservices... Because sometimes I think we try to look for reasons why we should use the tech we think is shiny, and new, and cool, and we shy away from the stuff that might be old, but it works. The boring tech. There's a reason why SREs like boring tech... Because excitement in infrastructure is never a fun thing. \[laughter\] So yeah, that's my piece of advice, I guess.
+
+**Mat Ryer:** Great.
+
+**Jon Calhoun:** I did YCombinator a while back...
+
+**Mat Ryer:** Show off... What did you do? \[laughter\] On-site Linux...
+
+**Jon Calhoun:** I think it was Paul Graham that had said in his opinion every startup should be using the language that whatever the founders are most comfortable writing with - that's the language they should be using. Because in most cases, as a startup, you're going to be way quicker even if that language is something that's gonna be harder to hire for, or you don't think it's the best language for the job... Chances are you can build the thing faster if that's the language you're most efficient with.
+
+**Mat Ryer:** Right.
+
+**Jon Calhoun:** I forget what the example was... There was one company that was written in -- not Haskell... Maybe that's Erlang. It was some language that I would not have expected a modern startup to come with... And they did, and they did very well for the longest time. That language did not hold them back.
+
+**Mat Ryer:** COBOL?
+
+**Jon Calhoun:** \[01:03:55.15\] Now, I don't know if it did later, and maybe they needed to change some stuff; I don't know. But I think, to Johnny's point, I think at times people think "Oh, I'm doing a startup. This is my opportunity to try these new things", and in reality, if you want this business - which is what it should be - to succeed, you need to focus on solving business problems, not learning.
+
+Now, if you do wanna build a project to learn, I think that's different, but I don't think that should be a startup necessarily. Maybe I'm wrong there, but...
+
+**Mat Ryer:** No, it really makes sense. I think so. It's really about being pragmatic, isn't it, and resisting these shiny objects... But yeah, it is tough.
+
+**Jon Calhoun:** I should also add, if there are investors who wanna pay me to go learn other languages, I'm all for investment. \[laughter\] I will take it.
+
+**Aaron Schlesinger:** Drop them an email...
+
+**Mat Ryer:** Great. Well, next week we are actually gonna be -- this is a subject that comes up a lot. We're gonna be talking about how you can introduce Go into your team... And we talk about people chasing those shiny objects; Go, for some people, is a kind of shiny object, and we're gonna be digging into that and exploring some ways we could do that... So that'll be a very good one. Definitely tune into that.
+
+And also, we don't do ads on Go Time, really. We don't advertise Go Time... So tell your friends that; just tell them about it. Obviously, if they don't have a computer, maybe just leave them alone. But if they're a developer, spread the word a bit for us, please, because we'd love to grow our audience a bit, and meet new people...
+
+**Jon Calhoun:** Carlisia is joining us, for any of our old listeners... Listeners who have been around a long time, not old as in old age.
+
+**Mat Ryer:** Yes.
+
+**Aaron Schlesinger:** Cool.
+
+**Jon Calhoun:** Anybody who's been around a long time, one of the original hosts, Carlisia, is going to be joining us next week.
+
+**Mat Ryer:** Yeah. So that's gonna be very exciting. And we are gonna actually be doing something with Brian and Erik also coming up; they're part of the original cast...
+
+**Johnny Boursiquot:** OG.
+
+**Mat Ryer:** Yeah. What does OG mean?
+
+**Johnny Boursiquot:** You don't know OG -- wow...
+
+**Aaron Schlesinger:** Original gopher...
+
+**Johnny Boursiquot:** Goodness...
+
+**Mat Ryer:** Does it mean original gopher?
+
+**Johnny Boursiquot:** Well, we kind of coopted it, but... Yeah.
+
+**Mat Ryer:** Yeah. It doesn't make sense.
+
+**Johnny Boursiquot:** This is a safe version, yes.
+
+**Aaron Schlesinger:** Yeah. And we're gonna go with that for this part... \[laughter\]
+
+**Mat Ryer:** Okay, now I'm even more in the dark. Can someone on Twitter just -- I mean, can someone just tell me what OG stands for, please?
+
+**Jon Calhoun:** How about this - we'll stop recording, and then at that point we can discuss. \[laughter\]
+
+**Aaron Schlesinger:** Organic grapes. We've got someone -- organic grapes.
+
+**Mat Ryer:** Oh, organic grapes... \[laughter\] Okay, yeah. That's good.
+
+**Aaron Schlesinger:** Erik and Brian both grow grapes, so that's where the term comes from... \[laughter\]
+
+**Mat Ryer:** I don't mind being treated like an idiot... No probs. \[laughter\] Well, unfortunately, it's been great hanging out, but that's all the time we've got today. Everyone who's listening - go away. See you next time. \[laughter\]
+
+**Johnny Boursiquot:** Wow...!
+
+**Jon Calhoun:** Alright...
+
+**Aaron Schlesinger:** Take care, everybody...
+
+**Break:** \[01:07:19.21\]
+
+**Mat Ryer:** Okay, so we'll just do 1-2-3 and then clap after.
+
+**Johnny Boursiquot:** Yes.
+
+**Mat Ryer:** Okay. Here we go.
+
+**Jon Calhoun:** Who's counting?
+
+**Mat Ryer:** We only get one of these, we've been told. Okay... 1-2-3. \[claps\] Brilliant.
+
+**Johnny Boursiquot:** That was like the closest we've ever gotten.
+
+**Mat Ryer:** Yeah, I think that was good.
+
+**Aaron Schlesinger:** Nice.
+
+**Mat Ryer:** We're in sync.
+
+**Johnny Boursiquot:** We're in sync. NSYNC. That's the same as the band, right?
+
+**Mat Ryer:** Not NSYNC...
+
+**Johnny Boursiquot:** Okay, okay.
+
+**Mat Ryer:** No, that's a different podcast, where we invite guests on and we have to do four-part harmonies...
+
+**Johnny Boursiquot:** \[singing\] Everybody... Yeah-yeah... \[laughter\]
+
+**Aaron Schlesinger:** \[singing\] Rock your body...
+
+**Mat Ryer:** Cheers, mate. Yeah, I didn't get in.
+
+**Jon Calhoun:** You could start your own.
+
+**Johnny Boursiquot:** That was actually the wrong band. We're singing Backstreet Boys...
+
+**Jon Calhoun:** That's probably why Mat didn't get in. \[laughter\]
+
+**Mat Ryer:** Yeah... You're singing all the wrong band songs.
+
+**Jon Calhoun:** He's like "I'm a really big NSYNC fan, I swear", and starts singing their own song... \[laughter\]
+
+**Johnny Boursiquot:** That was a setup... Oh, man.
+
+**Mat Ryer:** This is already going well, ain't it?
+
+**Johnny Boursiquot:** Oh, yeah. It gets better?
+
+**Mat Ryer:** It will.

Cloudy with a chance of Kelsey Hightower_transcript.txt

@@ -0,0 +1,343 @@
+**Johnny Boursiquot:** Hello, Go Time listeners, and welcome to the first episode of 2020. My name is Johnny Boursiquot and I'm joined by an exquisite panel today. Joining me are Mat Ryer, Carmen Andoh, Jaana Dogan, and our special guest, Kelsey Hightower. Hello, Kelsey, and welcome to Go Time.
+
+**Kelsey Hightower:** Oh, awesome to be here.
+
+**Johnny Boursiquot:** Good stuff. So it's 2020, and I don't know about y'all, but usually after the holidays I have this sort of haze, I'm just trying to come out of the little bit of time off, and trying to figure out how to compute again... I'm not sure if anybody experiences that... Mat, I'm sure you do, because you're always like that, no?
+
+**Mat Ryer:** Yeah... Basically, that's how it is for me. But no, it is hard sometimes to take a break, but if you can do it, it is worth it. And when you get back -- I really miss it, so I'm really keen to get going again once we start, so... I've had a good week so far, actually.
+
+**Kelsey Hightower:** A new decade is like a time for reflection, and I was able to reflect back... And I noticed that in 2020, the most CPU-intensive application is my web browser... I never thought that's what we would end up in 2020. \[laughter\]
+
+**Mat Ryer:** Yeah. They didn't predict that on Back to the Future, did they?
+
+**Kelsey Hightower:** No... Tabs. Chrome tabs.
+
+**Carmen Andoh:** We can't invent flying cars because we're too busy trying to optimize our browsers.
+
+**Mat Ryer:** We're too busy on Twitter, yeah. \[laughter\]
+
+**Johnny Boursiquot:** Jaana, welcome back to the show. You've been gone for a little while.
+
+**Jaana Dogan:** Yeah, I took a couple of weeks of breaks, I think, here and there... I'm still trying to come back. Last day was my first day at the office, and you know, I was questioning everything... Like, "Hey, am I really qualified to do this job?" \[laughs\]
+
+**Carmen Andoh:** Ooh...
+
+**Kelsey Hightower:** The important question though is does your badge still work? You come back from holidays and you have trouble badging in? That's a sign. \[laughter\]
+
+**Johnny Boursiquot:** It's like, your heart skips a beat, like "Oh, crap. What happened?" \[laughs\]
+
+**Mat Ryer:** Yeah, I've had that... I was using my bank card. I thought I'd been fired. \[laughter\] I couldn't get in.
+
+**Jaana Dogan:** I'm trying to work from home, to be honest. I have this suspicion that I can get fired anytime... I'm just working from home. It's better not knowing. \[laughter\]
+
+**Carmen Andoh:** Just be home anyway. You're like "Whatever..."
+
+**Johnny Boursiquot:** \[03:58\] Any day could be your last... That's rough. Carmen, how are you?
+
+**Carmen Andoh:** I'm well. I need a vacation from my vacation, I think... But it was nice to get time off.
+
+**Johnny Boursiquot:** That's good, that's good. Like Kelsey says, it's time for reflection, and I think we should do a little bit of that for 2019. It's been a big year for cloud, I think... And I don't think there's anybody here more qualified than Kelsey to talk about the impact that things like Kubernetes, and things coming out of AWS, and all the other cloud providers and things that they're doing... It seems like they're pushing the boundary in terms of what makes folks productive.
+
+So operators are impacted, developers are impacted, and when I try to wrap my head around everything going on in the industry right now, it's kind of hard to see -- I'm seeing blurred lines... And Kelsey, maybe this is something we can start with. I've been an operator and a developer, so I kind of get a sense on both sides of what lens to look at things with... But if you're a traditional operator, what does technologies like Kubernetes, for example - how are they changing life for you? And on the other side too, as a developer, what should you care about? Or is there even a line?
+
+**Kelsey Hightower:** Yeah, so the good news is I think 2020 the hype is now just dying off. This is good. We are finally getting to the point where most of the new projects are just remixes of the existing ones. Communities are merging; we saw this with the OpenCensus and OpenTracing communities, to OpenTelemetry...
+
+So now that we've done a lot of experimentation, I would probably say the last four years have been just trying ideas, from infrastructure all the way to developer tooling. And now what we're starting to see is that the buzz is kind of dying down; now people are more focused on production... And I think one good indicator is -- probably we can't extrapolate big, but when Docker canceled their conference coming up in 2020...
+
+I remember 2,5 years ago if you were at DockerCon, it felt like that thing was never going to stop growing. It was at five or six thousand people, and there was just no end in sight... And now they're moving to a virtual conference. Now, there's probably all kinds of reasons for it, but my guess is probably the attendance started to drop or shift to other events... But that's a good signal to me to indicate that now we're in this other phase. We have all of these tools, let's just put them to use.
+
+Now we're getting to that part I think I like... When infrastructure gets boring, something else will become exciting above it. Right now we're starting to be in that phase. So I think if you're a practitioner looking at all of these things, you've made it. We're on the other side of the hype curve now. You can stand back and just start to really listen to what people were doing in production, and you're gonna probably see a lot of good examples in the wild of what people actually are using, versus what's being said out loud, or inside of that hype cloud that we just came out of.
+
+**Johnny Boursiquot:** So are we still at the stage then where things like Kubernetes are -- we're still trying to figure out how we should use them, what we should use them for? Obviously, you can approach this from different angles - if you're trying to build a platform for folks to use, or if you're trying to run your own data center, for your own business purposes... I think there's still some confusion around who should really be adopting and using Kubernetes and for what. Are we still in that phase where there's some confusion around its use cases?
+
+**Kelsey Hightower:** Some people are, but luckily for me, I was there from the very beginning, and now I'm stepping back and reflecting. If I reflect back, Kubernetes is about the how. Everything underneath is still the same. The kernel is still there, the virtual machines were still there... What we did was change the how.
+
+Before Kubernetes, Puppet, Chef or Ansible was your tool of choice to do some of the same things with multiple machines. Kubernetes changes the how. Some people call it abstraction; we introduce things like pods, we leverage some distributed system principles around service discovery... But all of that is just the how. How do you find your applications, how do you configure them into the load balancer?
+
+\[08:06\] Now, the interesting thing here is a part if that journey - we decided to take the Linux machine and treat it like the hypervisor. So we moved networking, security - all of these things that you typically got from a cloud provider like IaaS, or if you're on VMware on-prem, VMware already had these things to integrate your machine to the rest of the environment. We had to go rebuild all of that in the Kubernetes world, and we got pretty far. We did a decent job of networking; it's still confusing for a lot of people to be managing bridges on Linux machines instead of having the underlying networking fabric... And then we also did another thing - we took the application packages, container images, and we decided to just let them run side by side on the kernel, and we spent the last 3-4 years trying to securely do that... And that's a very hard problem, because the Linux kernel isn't designed to give you that kind of security for untrusted workloads.
+
+Now we're coming back full circle, where now we're gonna take those container images, some of the things we got from the Kubernetes world aka a pod - one or more containers, their volumes and their dependencies - and we're gonna put them on micro VMs. So we just went five years to come back full circle. We made the VM smaller, but we're keeping the API from Kubernetes, we're keeping the container image, and now we're back to much better virtualization, almost the original purpose of cloud.
+
+**Johnny Boursiquot:** Interesting.
+
+**Jaana Dogan:** Does this mean that individual nodes at some point end up being a virtual machine? We are trying multitenancy... Do you see Kubernetes APIs just replacing all, and each pod probably ends up being a virtual machine in the end?
+
+**Kelsey Hightower:** I can almost guarantee. Guarantees are bad, because people come back and listen, like "Dude, you were so wrong..." \[laughter\] "You were overly-confident." I am overly-confident that we will go back into a world where there will be one pod/VM. But to think we call a virtual machine is dramatically changing. I think Amazon hinted to this with things like Firecracker. We saw Firecracker come out... And for those that know their history, Firecracker comes from the Chrome team, where they decided that they want to isolate things at this level.
+
+When you can isolate things at this level almost using some of the machine-level isolation, then you can actually isolate a web browser tab, Amazon is isolating serverless workloads... Now you're in a really interesting spot, because guess what, one thing we already have - every cloud provider has a really nice hypervisor layer that most people don't see. We already have things like VPCs, Direct Connect... So imagine saying "Let's strip the kernel down by 80%." Strip down the virtual machine; no floppy disks, no things we don't need... And then just pack the container or the pod in there.
+
+That means you could be using the Kubernetes API, but all you're really doing is launching these lightweight virtual machines connecting to the existing stuff... And now we can take the eight years of advancements in the cloud, or VMware on-prem, and pair it with all the beautiful things we like about Kubernetes.
+
+**Johnny Boursiquot:** One of the things I've heard you speak about on a number of occasions is getting to the point where the things that we're moving around as the folks who look after the infrastructure is really data. I think the terminology you used is "infrastructure as data." Can you talk a little bit more about that? What exactly does that mean to me if I'm moving around YAML files today, to try and set up clusters and manage infrastructure? What does that mean to me?
+
+**Kelsey Hightower:** It's funny, people complain about YAML, like "Oh my god, YAMLs... So terrible." And you have to step back for a second. I'm gonna just talk through my journey of this to get to where I get to the concept of infrastructure as data. If you have five Linux machines, the first thing you're gonna try to do is write some Bash scripts. You're gonna write some Bash scripts, some for loops, you're gonna SSH things around...
+
+\[12:09\] And if you've ever seen people write Bash scripts over time, you go into their home directory and you see files like "Do this stuff 01. Do this stuff 002, but don't use it anymore, because you should be using the other script." \[laughter\] So you have no version controls, you have no semantics, no abstractions. You're just writing Bash scripts.
+
+Fast-forward to configuration management. We get things like CFEngine - big shout-out to Promise Theory - and then we get Puppet, Chef and Ansible, and then they formalize. It's almost like the Ruby on Rails for shell scripting. So now we have this configuration management error and we all start to say "Infrastructure as code." The problem is now you have to test it, people can write any code they want, it's unbounded context, and it's the same problems we had with software - how do you secure it? You're gonna have bugs... it's just all over the place. But it is a better place to be than we were before.
+
+Now let's get to infrastructure as configuration. Now we're removing all of the abstractions into the runtime, so the how. How do you create a load balancer, what goes in the load balancer and how do you remove it? That implementation detail - we're gonna have a lot of discipline and we're gonna move it into these controllers. So if we're talking about Kubernetes, these are gonna be the controllers. If you've been in cloud, you've already done this. We always have control planes that do the heavy-lifting. Same is true for networking - we expose ports and protocols, not the control plane.
+
+So configuration as data - we get to something very similar. Now what you say is "I want a load balancer in this region, pointed to those services. No for loops, no if statements, no programming language concepts." So all you have is a data model. And that data model represents everything that the state machine on the other side can do. Why is this more powerful? Well, when you're working with data, then you can manipulate the data much easier than you can manipulate code.
+
+We've seen this before, in the Go world - there's 10,000 Hacker News posts, "Oh, the same thing, but written in Go." I prefer Go as my favorite language. So every time we do things in a language-specific way, we end up having to rewrite this thing to be compatible with those libraries, and so forth. But when you move to infrastructure as data, we can have these high-level APIs. You can write them in JSON, you can write them in YAML, or if you're an enterprise, you can go XML if that's your thing...
+
+**Johnny Boursiquot:** Oof...
+
+**Kelsey Hightower:** But either way, every tool that comes out from here going forward can translate all the data that your infrastructure is described in, to do something else. That's extremely powerful, and I don't think people have really comprehended what this means going forward.
+
+**Mat Ryer:** It reminds me of actually JavaScript frameworks. A lot of those are -- you essentially have data, and they're applied against a template. And then you just focus on changing that data and mutating the data in different ways. That's what they mean by reactive; the experience reacts. And when you actually work like that, it is sometimes really quite surprising what it does automatically.
+
+If I think about that as an idea applied to infrastructure, it really does, from my point of view as a user of this - that is massive. It is a very different way of thinking about it.
+
+I wonder, in the design then, through the life of Kubernetes - because you know, building abstractions is hard - were there any abstractions that didn't fit, or things that didn't quite work, that had to evolve differently? Or because of the nature of it, you were able to get a good design from the very beginning?
+
+**Kelsey Hightower:** \[15:46\] This is where I learned the most as a developer, from working in the Kubernetes community - so a big shout-out to Brendan Burns, Joe Beda, and people like Brian Grant, that was a little bit more behind the scenes. The work that they put into the API design -- also someone named Clayton Coleman from Red Hat... They spent so much time on the API design.
+
+When you open up a Kubernetes configuration, you see the API version, and then the API group. And maybe about halfway into the lifetime of Kubernetes we decided there should be a thing called core. What's required to run a container? You need things like the scheduler, you need things like what is the definition of the thing you wanna run, the pod, and you need something like a replication controller. These things represent core, including things like services and so forth. With those core primitives you can build everything else.
+
+Red Hat came along and said "You know what - we can't just give customers raw Kubernetes..." And they have this concept of OpenShift; it's like a PaaS in a box, very similar to things like Cloud Foundry... And they started adding things like deployments, and namespaces. So those two things come from Red Hat. The problem though is they're very opinionated things, like a deployment that does rolling updates a certain way. What if you wanna do something like a blue-green deployment or a canary rollout? The deployment object is very opinionated in how it rolls things out. So we decided to say "Hey, let's slow down for a moment. We can't bring all of OpenShift into Kubernetes. We should probably leave those things on the outside", and I think Brendan Burns came up with this idea of the custom resource definition - this idea that you could just give us your schema, we would generate the rest of the API for you, so all the tooling will work, kubectl will work... And then you can actually implement your own control loops that would then watch for this data to show up in the system, and then you could build your own platform.
+
+So in wrapping, the thing to take away here is that the first system we've built on Kubernetes was a thing to manage containers and services. And then people went on to build things like Kubeflow for machine learning, and Istio for networking and service mesh, all on the same principles - control loops do the implementation, and your data model represents what that control loop can do.
+
+**Carmen Andoh:** Kelsey, are there cloud problems that you can't solve with the CRD approach that you were trying to work around?
+
+**Kelsey Hightower:** This is a really good question, because I go to all of these companies and they're just making CRDs for everything. There are CRDs to take a shower. I'm like "Dude, no. You just literally go take a shower. You don't need to do kubectl apply. Take a shower! Like, come on, we're going a little bit too far..." \[laughter\] I think what people have to understand is that there are things that really work well in a declarative model, and there are things that are a little bit in the grey area.
+
+For example, let's think about a CI/CD build. Typically, with a CI/CD build you're triggered by some action. Someone checked in a bit of code, and then that kicks over or sends a notification to run the build. And if that build fails, then you have a decision tree. Do you retry the build, or do you just alert the user?
+
+Now, let's take that and try to make it a declarative model. If you try to make that a declarative statement, you may define the trigger; if this source repository sees the change on master, then run the build. And then that could be a declarative object that sits there, and the system will then try to behave and represent that. But here's the thing - how do you represent a build failure? That's a very imperative thing. So what Kubernetes has - and it's a very important part of the Kubernetes API - there's a status field at the bottom. This allows us to capture the imperative side effects from the declarative statements we make. So when that build fails, if you look at your status field, it will say "Hey, this build failed for these reasons, but you're never touching the declarative state that you want, the trigger, you just have the results of the trigger." And people tend to mix the two.
+
+\[19:47\] Some people will put the status as part of the API, and now it's no longer declarative, because I'm not declaring that I want a failed build, so why is it part of the API? It's just an output of the API. So that nuance is really key to getting that right.
+
+When I see people trying to roll a bunch of imperative things into Kubernetes, it gets really tricky on how to build that API without munging the two worlds.
+
+**Mat Ryer:** That's really interesting. So in the event of a failure then, what happens? How do you say what I'm gonna do in this case?
+
+**Kelsey Hightower:** I love this part of Kubernetes... It is on the implementation to eventually get to the state. This is the whole promise theory. The whole promise theory is that I want a load balancer to exist in region A. That's my declarative statement. Here's the thing - it may never happen, but the control loop may have a policy that it's gonna try every three minutes until it does. That's the only thing that you can get from Kubernetes.
+
+So when people say it's taking long, or it's too slow, or let's say there is a failure - well, the promise is I'm going to tell you about the failure, and I'm going to try again with no further input from you, because the declarative statement is already in place.
+
+**Mat Ryer:** And that declarative statement is valid, is validated. So it's not like you're gonna get a runtime error or anything back.
+
+**Kelsey Hightower:** Exactly. So with this set of inputs, I promise you that I can go and configure a load balancer, because I created that contract.
+
+**Mat Ryer:** That's how some managers work with developers as well. They just say "This is the thing. Just go and do it."
+
+**Johnny Boursiquot:** Keep trying.
+
+**Mat Ryer:** "I'll try every three minutes, I'm gonna check to see if you've done it..." \[laughter\]
+
+**Kelsey Hightower:** But you know what? So here's the problem, and you all can probably chime in - I've never seen a set of product requirements so clear that if I did exactly that, everyone's happy. It's like "Hey, I gave you a vague idea of what I want, you did what I asked, but that's not what I wanted. Could you change it?"
+
+**Mat Ryer:** Yes, yes... And of course, it's hard to do, because when we design things, we're doing it at the beginning, often. And I don't think we pay enough attention to the learning we get from doing it. I still think that's a big issue that we have, just on that point.
+
+**Kelsey Hightower:** I'm gonna pull on that thread, because it's very important. Kubernetes has an alpha/beta stable progression, and it's for what you've just said - we start with these alpha objects, so we can actually learn, "Is the API good enough? Is the status field good enough?" And what we often learn is that we get it wrong and we start adding annotations to be this kind of side-channel for the official API. And then we study those annotations, and typically, the right annotations that are applicable to a wide range of problems become part of the official API, and then you have less. So to me, in Kubernetes, the if statements are the annotations. "If this is for NGINX and not HAProxy, then these annotations apply."
+
+**Mat Ryer:** Yes, that's a good lesson for anyone writing tech - that thing that getting things wrong is okay; it's part of the process. I once had a manager who pulled me into his office because he'd look over my screen, and -- I was doing TDD red-green testing; he'd look over my shoulder and kept seeing failing tests, so he thought I was struggling. \[laughter\] So you know, getting things wrong is what we do, isn't it, really? I mean, if you write code that runs the first time, do you really trust it?
+
+**Johnny Boursiquot:** No...
+
+**Mat Ryer:** I go and break it to make sure it's real, or... Yeah.
+
+**Jaana Dogan:** Is there anything like being right, to be honest? We had this discussion at the API design show... You are always wrong. As you gain more experience, you always have more insights, so it's almost impossible to get to the point that you are very confident about the end result.
+
+**Kelsey Hightower:** I think that's why it's really critical to expose as little of the API service as possible in the beginning... Because you can always usually add something, but it's really hard to take away.
+
+**Mat Ryer:** \[24:02\] Yeah. And there is an art to it, and it's useful in Go - you can do little tricks like putting your tests in the test package, so it's an external package, and you import the main package. That's quite a nice way to see the footprint of the API as you're writing it, and keeps that in mind. And literally, in the autocomplete in the editors, if the list is too long, it gets a bit annoying. Well, that's what it's gonna be like for your users as well... So yeah, keeping everything small, and don't commit to anything too much. And it's about future-proofing things, but you do have to be honest. The whole team has to be really honest about the reality of writing code.
+
+And if you do have influences in the team that perhaps aren't technical or don't think about things in this way, that's when you can get a lot of tension between product, and tech, and these things... So yeah, a part of our mission as developers probably is to communicate and educate on that point. It can be very difficult to do in practice, but it's really valuable... So it's great to hear that Kubernetes as a massive, successful project, has this approach. Sometimes it's wrong about things... I think that's a great lesson for people, actually.
+
+**Jaana Dogan:** Kelsey, you mentioned that you learned a lot from Kubernetes as a developer... I think that Kubernetes has been playing a teaching role, as a teaching tool. Developers know more about the load balancers, networking, all the scaling implementation details, deployment... Some people think that this is an overhead, but I do think that it generally had a better impact to the overall community, and making people more aware of some of the critical design primitives. Do you agree with that?
+
+**Kelsey Hightower:** Yeah, so if you're gonna go build a 30-foot skyscraper, usually all of the elements - the bathroom, the siding, the cooling, the heating - is on one blueprint. Maybe there's layers to the blueprint, but if you open up the set of blueprints for the building, everything's there. It all has to be connected. You need so much power to power the building, but you can't figure out how much power you need unless you know how many rooms, and the requirements.
+
+Before Kubernetes, people were really thinking about software at different APIs. Go to the network switch, you program it there, you go to your load balancer, you program it there... There's no single place where you see everything. And the Kubernetes experiment was "What if you took everything that it meant to deploy an app? The secrets, the volumes, the load balancer, even the RBAC permissions to even deploy it, and you give it one common set of languages?" We're starting to add security to this - network security policy, Istio policies... And now for the first time as a developer - you're right, we're now putting that front and center.
+
+So you can say "Here's my application. It listens on this port. It needs these secrets. Oh, and I would like it exposed to the outside world with this DNS name. Oh, and if it fails, roll over to this other region." I can articulate all of that now, using a very clean API, and submit it to the system and have it converge to the proper state.
+
+**Jaana Dogan:** Yeah, in the sense that we've been always living in the skyscraper, and we were avoiding what's going on there... But as the architect, you have to realize the truth, and Kubernetes is providing this really accessible API. It's just like you said, it's the common language. Once you know it, you can also extend it, and it's a really good foundational--
+
+**Kelsey Hightower:** I'm looking at getting a new house built from the ground-up. I've always bought maybe a house brand new, right? And I was like "How long will it take to build the house?" And they were very optimistic; they were happy and smiling. They said "We can get it done in a year." I was like "A year? What the hell are you doing?" And they're like "Well, here's the thing... We have to make sure that the land is prepped, we have to go get permits every step of the way. We need a permit for this piece, and we need a permit for that piece... Someone needs to go and check that we do exactly the work that the permit said we were going to do, and we have to make sure that we have all of that... And then building the house - that's roughly the easy part. It's all the other stuff that we have to do..." Because you can't mess up the house and be like "Yeah, the roof is off slightly. Don't worry. We'll ship a patch for the roof." Like, "No, I don't want a patch for the roof." \[laughter\]
+
+**Jaana Dogan:** \[28:34\] I have a question about this... Does this mean that you need a cloud provider to run Kubernetes?
+
+**Kelsey Hightower:** I love that question. Here's the thing... I meet a lot of people that say "Hey, I have very little experience. I don't know how to manage a Linux server. Hell, I don't even know how to patch a Linux server. But what I wanna do now is I wanna leapfrog all of that and just start running a massive distributed system. I can do it in probably ten minutes, fifteen tops." And I'm sitting there like "Wow, man... That's a bit disrespectful to the industry", because this is not a point-and-shoot camera. There's a lot going on to this. We're talking about the difference between making a movie and filming your child's birthday party. These aren't the same things.
+
+I think a cloud provider just brings to you a promise and a knowledge base that says "Look, we're managing this stuff for thousands of customers. We've learned all of this stuff, we're obligated to make it work as a full-time job, versus finding 10% of my downtime to play administrator." So I don't think you need it; you don't need a cloud provider. But unless you're being honest with yourself -- I go back to some of these teams and they've got like 20 people on their Kubernetes team because that team is responsible for networking, load balancing, security, metrics, AND the application, but they try to put it into the Kubernetes box, when the problem is much bigger than that.
+
+**Carmen Andoh:** Yeah, Kelsey, I was thinking -- when you were using the 30-story high-rise analogy and Kubernetes, and having this one-API-fits-all approach... I think a counterpoint to be made is that because its one API to do all of the things, if you can't encapsulate it into small pieces the way you would for the plumber or the way that you would for the HVAC operator, it can feel overwhelming, and it presents a mountain of complexity.
+
+So when you said "Wait, our managed service provider can take away that complexity and do this for you", that was really a good analogy to extend for companies who think that they wanna roll their own, and saying "Well, if you do, then everyone has to be the Jack or Jane of all the trades." So that's always the counterpart that I hear when I hear "Well, Kubernetes is using a 20-wheeler to carry one package." We've seen that meme on Twitter a lot.
+
+So yeah, this managed service provider seems to be the way to go... How do you feel like once somebody uses a managed service provider, what is the work to do after that? Where should the focus be? And I know this is gonna be not one-size-fits-all...
+
+**Kelsey Hightower:** You know what's interesting? We are a collection of communities of people who can probably do it ourselves. It's one of the very hardest industries to service, because we are selling solutions to people who can actually visualize, or even have the skillset to just do it themselves. So it's very tempting and enticing to just go do it yourself. But think about power. Most people are getting power from a power grid, the equivalent of a managed service. There's some grid by a provider who knows what the hell they're doing, and most people could consider that overkill, that we have these massive grids... And we run a line into your home, so you can just flip on the light switch.
+
+That works at scale, because imagine if everyone tried to build their own grid. This would be a straight-up disaster; people would be getting electrocuted all over the place, they'd all come up with some group therapy, like DevOps, to talk about what happens when the grid doesn't -- you know, like "Hey, I've been burned..." Everyone's walking around with half an arm, because they don't know how to run power grids. And there'll be a bunch of conferences about how to run power grids, and there'd be 20k-30k people talking about how to best manage power grids.
+
+\[32:21\] This is what we're seeing in tech right now - lots of people are trying to be cloud providers part-time, and without the same level of budgets. So as the technology advances, eventually it starts to make sense. And maybe there's a trust thing that's missing. With power, we assume that it's always going to flow. And when it stops flowing, it becomes like a national emergency, if we have a blackout, or something like that.
+
+**Mat Ryer:** Yeah, it's like The Purge. \[laughter\]
+
+**Johnny Boursiquot:** That's a bit extreme...
+
+**Mat Ryer:** I'm surprised how quickly it becomes The Purge, once we have a power cut. \[laughter\]
+
+**Kelsey Hightower:** Well, okay... That's an extreme thing... You're right though, because people do feel like there's chaos. If a region goes down, the number of people who go to Twitter and just start going crazy... Like, you're right.
+
+**Johnny Boursiquot:** Yeah, so along the lines of building our own power grids - is it fair to say that there's a bit of FOMO with the people who could do it themselves? Is it FOMO that's driving them to say "Well, I should try this, because heck, if I'm being selfish, this is a career thing. I wanna know how to run Kubernetes clusters, and do this and that..." So now you have this sort of resume-driven development, as we tend to call it. Even in the back of your mind, even that little voice in the back of your mind saying "Yeah, why don't you push for it? You can do that, right?" \[laughter\] Is there a bit of that?
+
+**Kelsey Hightower:** I'm the world's biggest hypocrite right now... I'm starting this podcast, and I've got some recordings, I've got the dope mic and I was like "I need to host this podcast somewhere." And I've found all of these services; they're like $10, some of them were even free. They were like "You just come in, fill out this form, they'll host your RSS feed... You can even upload your MP3, and you're done." I was like "Nah, man... We ain't going out like that. What I'm gonna do is write XML by hand, read about all the tags..." I got my RSS XML, I put it on Google Cloud Storage, I've got my MP3, I did the ID3 tags for the MP3, I put a CDN in front... Just so I didn't have to use a managed service. Because you're right, I had this appetite for "I wanna know exactly how podcasts work. I wanna have full control, I wanna back up my data..." I don't want some limit telling me I can only stream 50 concurrent streams at a time. I was like "You know what - I know what to do here, I've got this." So I spent two days to save $12. \[laughter\]
+
+**Carmen Andoh:** Oh, my god... I think you've just explained our collective zeitgeist or pathos as an industry. Like, why our ecosystem is an embarrassment to low-level bridges. In this new decade we need to innovate, we need to make Kubernetes boring, we need to experiment, we need to move up in the abstraction chain, but yet we also need to fundamentally know how things work. And that's the paradox we're in.
+
+**Kelsey Hightower:** Yeah, and I think, honestly, if I were to recommend this to someone else, I think I value the knowledge of how podcasting works than the service... And yeah, you're right, our industry really appreciates the knowledge and the power that comes from the knowledge, versus what the business actually needs... And I think that the catch-22 - your business needs to survive, needs to grow, needs to be efficient, and sometimes you can leverage your cloud provider and make margins on top. That's just good business. But when we're at the helm, making decisions, we're like "Yo, forget good business. I'm about to deploy Kafka to process 25 messages a year." \[laughter\]
+
+**Carmen Andoh:** It's nerd pride, right?
+
+**Johnny Boursiquot:** Seriously, seriously...
+
+**Break:** \[36:17\]
+
+**Johnny Boursiquot:** We've been talking about the API for some time, and I think there's a reason for that... Is it fair to say that the learnings, the journey that we've gone through to have the Kubernetes API as it is today - is that really the crown jewel? And this is why I'm asking that... We have projects that are even looking to make Kubernetes simpler; one of the things you're gonna hear over and over is that "Well, Kubernetes is a complex system. There's a lot of moving parts. You've just gotta know what you're doing" kind of thing. And yeah, we're gonna have the people who are gonna geek out, and they're gonna go through the manual, they're gonna figure out "Okay, this knob goes here, and this one attaches here..." They're gonna figure it out, great. But you have projects that are trying to simplify that, using the same API, the same abstraction. I think of things like K3S, versus K8S. There's that common ground that is the API. Is it fair to say that the Kubernetes API itself, as we know it today, is really the crown jewel of this project?
+
+**Kelsey Hightower:** Yes. And the reason why it's important is that Kubernetes makes it easy to have hyper-specialization. The internet is one of the best examples of hyper-specialization. Some companies just make web browsers, because you can rely on the specific protocols to show and display web pages. Some people go to Best Buy and you buy a modem and you screw it into the wall, and now you're online... Because now everyone can just hyper-specialize, because the connectors, their protocols are so standard. But remember, the internet is just massively complex, but the things it exposes means we can specialize.
+
+So if you are an ISP, you view the internet differently than someone watching Netflix. It's just a totally different ball game. Kubernetes is meant to be the same thing - if you need to run at a global scale, meaning "I just need to be available everywhere. I need to be able to pick and choose my provider of choice", then that's gonna require a much more complex system to lay on top of that. It's infrastructure. Now, that's for people implementing that kind of capability. You can either buy it, or you can build it, but you don't have to start from scratch. You download Kubernetes and you start from there. And then that becomes implementation details.
+
+So when I lay it on to of my infrastructure, if I've gotta lay it on top of GCP, you're probably better off starting with GKE. If I have to lay it on top of Amazon, you're probably better off starting with EKS, because all of the integration work to make it all really run underneath - that's the hard part. It's like what Linux and distros do - the drivers, the default package managers etc. Linux is also complex.
+
+On the other side, as a developer, let's assume that either your team or whoever you bought Kubernetes from is doing a good job in terms of keeping that thing available. You get to then deal with a different API.
+
+\[40:07\] You get to say "Kubernetes, run this. Kubernetes, run that. Kubernetes, tell me when that is no longer true." And for some people, that's all they know about Kubernetes. They've never installed it before. All they know is that if you give it things, it runs them. And when things aren't working, here's how you leverage the API to troubleshoot them. We have to do a better job as an industry of separating the two.
+
+**Carmen Andoh:** Well, what about the future? I mean, we're getting better and better at having managed service providers to be able to help you get jump-start it, and provide value to the business... But I know you've also talked a lot about *Serverless*, and maybe that kind of form of compute. Can you talk about maybe what this decade should bring, and maybe *Operators* and why would they choose that instead?
+
+**Kelsey Hightower:** Yeah, so the thing about serverless - I think that hype cycle will be very short, by design, because they have the right focus. Their focus is to make something useful, without managing the server infrastructure underneath it. That means that there's no reason to have long hype cycles.
+
+Think about it - storage. I just upload some files. There's no need to have *StorageCon*. I upload the files. I don't really care how you're storing my files, I just wanna make sure that they're there when I need them. And when you look at the service philosophy, it's more about the APIs and getting things done than implementing the system underneath. So now we're asking ourselves, "In the future, what else is going to get the serverless treatment?" We've seen it done for email, we've seen it done for storage, we've seen it done for the internet... Most people use the internet in more of a serverless model.
+
+So right now the last thing that's really standing are databases and compute. These are the two things that just won't die, for the things we've talked about earlier - people believe that they can do it themselves. One day I believe that there will be a CDN-like database. CockroachDB hints at this, the Spanner team also tries to provide this for customers... But imagine a world where you go and you put your data in the thing, and it's just available everywhere. That's going to be a game-changer, and maybe people may not wanna deal with a database service individually anymore, because they just don't make sense.
+
+**Carmen Andoh:** Yeah, it feels like sooner or later the core competency of companies is not to run Kubernetes clusters, in the same way that we're not running our own data centers anymore... And that core competency is gonna shift elsewhere.
+
+**Kelsey Hightower:** Or just make sure you get a lot of value from doing that yourselves. There's lots of companies where running a data center does make sense. If you think about what I think may happen, someone's going to collapse some of the layers. I remember when Oracle -- I've worked at enterprise a lot of my life before going to the startup world, and... \[laughter\] Actually, you know what - what makes me the most valuable now is I had that context of reality. And having these AIX boxes, or Solaris boxes running Oracle, and everyone just praying at the server not to go down... And when that happens, it becomes a very complex thing where you start to have 50 Oracle DBAs to manage two database servers; tweak and tune indexes, deal with the foul systems and upgrades...
+
+And then Oracle did something amazing. Ten years latet from my initial exposure to Oracle, they came out with the Exadata. They're like "Alright, that's enough. We're gonna wheel in this refrigerator-sized thing. And it has the storage, the SSDs, all the RAM... And you're gonna close the door. Now, here's the thing - if you open the door, don't call us. Leave the door closed; don't mess with it, don't devops it, don't run Kubernetes on it, nothing."
+
+**Mat Ryer:** Don't even look at it.
+
+**Kelsey Hightower:** \[43:52\] Yeah, if it look at it, it costs extra. \[laughter\] They send you photos of it, and you pay the invoice, so you know it's there. But the key there though is that now people just start to focus on using the database, and not managing the database.
+
+**Mat Ryer:** Yeah, I mean, that's me. That's my whole approach. It resonates a lot, because when I'm building things, I wanna just focus on the bits that are important or that are unique, that I'm doing. I really don't want -- I mean, if I could just get away with just stitching some things together to build a product, then I'm very happy doing that.
+
+I use Firebase quite a lot. I'm working on a new project, and we've chosen Firebase; we're doing experiments with it... And technologies like that, alongside other things like App Engine and things - it takes away a lot of the things I have to worry about. And it is slightly abstracted, and there are times when it doesn't quite fit perfectly, of course... But yeah, I do think that's the right approach, really. I'm not gonna add much value by doing these things myself. I don't even get that thing, that nerd pride thing, where I think "Oh, I'm really smart because I made this thing..." So I'm definitely not gonna be doing that.
+
+**Kelsey Hightower:** If you're listening to this and struggling with this concept, imagine you work in the IT department and you hire ten new people a week. You give them their laptop, and you give them the desk assignment. They go sit down at the desk, and you say "Hey, before you can get online, we practice devops around here. Here's a crimper cable and here's some raw Ethernet... And here's two in clips and you only get two so if you crimp this thing wrong you don't get online." And you lay the kit at their desk and say "Hey, make a Cat 5 cable, remember the crossover in the middle, and then you can get online."
+
+And people will be sitting there like "What the hell are you doing? I have work to do." You're like "Mm-hm. You need to make a patch cable first, so you can get online." That is what we're asking developers to do every time we expose infrastructure to them at that level.
+
+**Carmen Andoh:** Yeah. They need to work at that highest level of abstraction, as Mat was talking about. Firebase does that for him.
+
+**Mat Ryer:** Yeah. It's really about the important bit. What's important for what I'm doing. Where's the value that I'm bringing. I think also we do a thing which I think might be necessary, which is kind of like ignorance-driven development as well. When people see the "finished" product, or whether it's a good user experience design or something - when they look at that, it just seems obvious, because it's matured, it's been through the pain, the process of figuring out the right thing... So it looks easy. That's the thing. We're kind of ignorant about things. I think if we knew everything that was involved in a product before we started it, we'd never start anything, because it always ends up being really hard and really complicated, always. Even very simple projects. So we kind of need that (don't we) naivety at the beginning, and that ignorance...
+
+**Kelsey Hightower:** Well, I think it's hard, because so far in the industry we've attached negative terms with specialization. "Silos are bad. Not being a wider range of skills are bad." I don't think that's necessarily true. If I hired someone to do development, I wanna get them to focus a lot on development. If I'm in ops, I'm hoping part of my job is so that they can focus. There's nothing wrong with focus. Maybe there's something wrong where I'm not able to help you focus, because we need to collaborate to get there... But once the collaboration is done, can we not serialize the results, so that tomorrow we don't have to go through this again?
+
+I think the goal is I want people to be able to hyper-focus and be the best person they can be, while not preventing them from learning other things if that's what they choose.
+
+**Carmen Andoh:** \[47:58\] Well, you said in your Oracle experience they collapsed layers so that people could focus on a specialty, rather than all the other layers. And that particular specialty was still very productive and fruitful, and we've had lots of developers still working on it. So even if we were to collapse layers, we're still gonna find more work to do.
+
+**Jaana Dogan:** I feel like maybe specialization is -- there's some stigma; people don't necessarily wanna specialize because tech is changing so fast. If you just know how to wire the cable, there's no guarantee that you'll be able to take it to the next level. In tech, we change the way we work every 2-3 years, so I think that's one of the main reasons people try to avoid to specialize.
+
+**Kelsey Hightower:** You're right, specialization is a risk. If you over-specialize, you may find nothing to do. So I think a lot of times we take turns being specialists, even at the same job. You may specialize in QA for three months, and then you may specialize in development for another three months, and then when you step back, you say "Hey, the last year I've grown in three areas." But I typically like to focus, when I'm working, on one thing. Like, "Hey, I don't wanna play infrastructure person while I'm trying to figure out how to do my podcast."
+
+**Mat Ryer:** Yeah. The same applies for limiting scope, as well. If you try and do too much in a product, you kind of can only touch the surface on everything... Versus if you strip it way back and limit the scope as much as possible, you can really do a great job on those things. It's kind of -- the way to be a 10x developer might be to divide the workload by ten. I'm not a mathematician, but I assume that works. \[laughter\] We'll edit that out if it doesn't.
+
+**Johnny Boursiquot:** Good luck. \[laughs\]
+
+**Carmen Andoh:** I think that is the premise of the book Essentialism, Mat, if you've read it... It's by George Mckeown. But it's true. I think that's kind of where I want infrastructure to go. I mean, I like that we've been so low-level, but when Kelsey started off the podcast saying that it's finally kind of getting boring... You know, I think that that means this - that we've collapsed some layers and we're finally letting people get on with solving real problems for their business.
+
+**Kelsey Hightower:** And we see this in the runtime space quite a bit... Like the Go project in general - we have a very massive standard library that most people don't really think about implementing the HTTP protocol from scratch, to import net/http. And there's still room for improvement for higher-level things, but I think the programming world has really come to the conclusion that you need good, standard libraries, and that you kind of need a second layer of good, standard common libraries that everyone just uses, before people start writing their own libraries for very common things.
+
+**Mat Ryer:** Yeah. I do think it applies all the way up, as well. There's Hugo, that static site generator that was started by Steve Francia, and the Buffalo project, which is a kind of hyper-example of that, where that is extremely high-level and abstracted away from a lot of the other things that are really going on in that application... And the evolution of that - I know that they are working towards a v1 release, and... It's that same kind of principle really, that you don't necessarily have to build all the things. The point is to get something good done, and then get some value from it. Get people using it. You learn so much by doing that. I never can stress it enough.
+
+**Break:** \[51:45\]
+
+**Johnny Boursiquot:** So we're coming up on our time with Kelsey, and it's been awesome having him... But I'm also interested in perhaps maybe something controversial from Kelsey.
+
+**Kelsey Hightower:** Let's do it.
+
+**Johnny Boursiquot:** We're trying to introduce a new segment to this show, that focuses on unpopular opinions.
+
+**Jingle:** \[54:00\]
+
+**Johnny Boursiquot:** Now, I know a lot of people like most (if not all) the things Kelsey says, but I'm interested in hearing something that perhaps folks may not like from Kelsey.
+
+**Kelsey Hightower:** This is where you have to bring in the unpopular opinion, and let people comment on it. One unpopular opinion that people disagree with me a bit is -- there's two. Don't run stateful things on Kubernetes. That's a very unpopular thing, because people wanna Kubernetes all the things, and I'm just saying "But why?"
+
+And then the second one is "Monoliths are the future." Because the problem people are trying to solve with microservices doesn't really line up with reality. Just to be honest - and I've done this before, gone from microservices to monoliths and back again; both directions. Most people say "Look, we lost all of our discipline in the monolith. We just started creating classes, this person went and bought the Gang of Four book, came back and started doing design patterns and then quit, so half our codebase is doing this thing over here..." So now it's a nightmare.
+
+Now the codebase it's so bad, and you say "You know what we should do? We should break it up. We're gonna break it up and somehow find the engineering discipline we never had in the first place." And then what they end up doing is creating 50 deployables, but it's really a distributed monolith. So it's actually the same thing, but instead of function calls and class instantiation, they're initiating things and throwing it over a network and hoping that it comes back. And since they can't reliably make it come back, they introduce things like Prometheus, OpenTracing, all of this stuff. I'm like "What are you doing?!" Now you went from writing bad code to building bad infrastructure... And what you deploy the bad code on top of.
+
+\[56:15\] There are reasons that you do a microservice, so to me a microservice makes sense in the context of you're a bank, and you have this big monolith that does everything. Then mobile comes out. You wanna do mobile banking, but it requires a different set of APIs. You don't have to add that to the monolith. You can go create a new application that handles most of the mobile concerns, and then connect back to the existing infrastructure to do its work. That makes sense to me. But this idea of microservices are a best practice - it seems to be unpopular with most people's initiatives.
+
+They're like "Oh, we're bringing in Kubernetes, so we can do microservices. We are going to rearchitect everything" because it drives a lot of new spend, it drives a lot of new hiring... So a lot of people get addicted to all the flourishment of money, and marketing, and it's just a lot of buzz that people are attaching their assignment to, when honestly it's not gonna necessarily solve their problem.
+
+**Mat Ryer:** It's so true that microservice architectures that I've used - the components are coupled anyway. One message from one thing, if it doesn't go into the other component, or the other service, then your thing doesn't work. It's coupled... They are. It really does make sense.
+
+What about the practical teamworking aspects? Because one of the benefits of having microservices is you can kind of break your teams up in that way, and have the teams working somewhat independently.
+
+**Kelsey Hightower:** Now we're talking about the problem with merge conflicts. At least my experience has been "How do we avoid merge conflicts? I know... Let's start another repository. Let's have a better API contract." We've been so bad at language-level API contracts, we decided to leverage things like JSON and RESTful interfaces to give us a much harder contract. They're very hard to violate, because they're so rigid. You can't reach behind a class and call a method, because you can't do that with REST. It's not easy.
+
+So I think what happens in the team aspect is -- I like the idea of modules. When I saw the way Go did modules - you can actually have separate teams building modules, but that's independent of how you compile the modules into the final deployable.
+
+At CoreOS I remember we used to do this a lot, we used to have a lot of individual modules, and then package main is where the collaboration happened. I would bring in my module and maybe add a route to it, or something. But once you touch that file, it's only because you're saying "Hey, I'm now part of the contract. Here's my route", but then I would just go do the rest of my work in the module, and allow the build system to take all of our work and combine it together. And if you're using tools like Bazel, it could be build one big binary, build three little binaries, with flags... But either way, that's a separate concern, the way you layout your source trees and how you develop code collaboratively, versus how you deploy the results of that effort.
+
+**Mat Ryer:** It's funny, because I'm just starting a new project now, so I get to use all of my past experience on a greenfield project, which tends to be quite a unique thing to do. It's rare.
+
+**Kelsey Hightower:** Your resume is about to get amazing. \[laughs\]
+
+**Mat Ryer:** Yeah. Our approach is always to just kind of -- we'll start simple, with the simplest thing, so that we can get it working, and then let patterns evolve or emerge, and let different problems present themselves, and then you can solve them along the way. And it turns out - in the beginning at least - this is gonna be a kind of a monolith, and there's so far no problem with that.
+
+Another way to solve some of the other problems we talked about is just not to have loads of people working on a project. It's not always possible, of course, but when it is, that's a great way to cut out a lot of problems. If there's just two or three of you building something, it's so much easier.
+
+**Carmen Andoh:** \[01:00:14.12\] Yeah. The communication overheads to me, in any distributed project, are always the biggest ones, the biggest things that you need to solve for.
+
+**Johnny Boursiquot:** Indeed.
+
+**Mat Ryer:** Yeah, I actually pair-program exclusively with my business partner, David Hernandez. We only pair-program, so we immediately both know everything about the whole thing and we're basically doing the whole stack... But it's small enough, at least at the moment, that that's okay. And if you can keep it small, I think you should. It's a goal that you should have. It pays dividends immediately; I can't advocate for it enough. It's not always possible, of course, but it fights a little bit this instinct, particularly if you do raise money and you're VC-backed, or something like this; you wanna hire, and you want to build the team, because that's what you do... But I think it's worth just taking a step back from that; maybe you don't need to do it that way.
+
+**Jaana Dogan:** To be fair, I worked for a company -- I was the only developer, and we were developing ten microservices.
+
+**Mat Ryer:** \[laughs\] YOU were developing ten microservices.
+
+**Jaana Dogan:** \[laughs\] Exactly, yeah... So just for the sake of following the microservices pattern, we were doing it. And I think this is a very common pattern in the industry.
+
+**Johnny Boursiquot:** Indeed.
+
+**Carmen Andoh:** Well, we look left, and we look right, and we say "Who's doing what?" I mean, really... That's also why we use common libraries. I remember working in a company and saying "Why are we doing testing this way?" and it's because another company did it. The real unspoken truth about our industry is we sometimes look left and right without any real sitting down and reckoning with what works for us.
+
+**Mat Ryer:** Yeah. Everything we build, if you spend long enough on a project, you end up hating it. So whatever you've done, you've got something you hate. Oh, we had a monolith; we hated it. We did microservices, we hated it. If you spend long enough on a project, you're gonna hate it... And I think that drives a lot of decisions. You just want something different, you just want something fresh sometimes...
+
+**Johnny Boursiquot:** Novelty. \[laughter\]
+
+**Carmen Andoh:** Humans crave novelty. Humans absolutely crave novelty.
+
+**Mat Ryer:** Yeah, indeed. That's why I'm wearing these glasses. They're not even real. \[laughter\]
+
+**Carmen Andoh:** Well, I took a screenshot. I'll share on the Twitters... \[laughter\]
+
+**Johnny Boursiquot:** Wow... Alright, so this has been an awesome episode. Thank you again, Kelsey, for coming on, and for regaling us your tales of putting together an entire infrastructure for your podcast to save $12... \[laughs\]
+
+**Mat Ryer:** Are you gonna open source that, by the way? \[laughter\]
+
+**Kelsey Hightower:** I am. No, I actually have a set of command line tools that I will be publishing... Because I also do some things like translate the audio, using some of the cloud APIs... You know, you've gotta move it to the next level. You'll see that coming soon to GitHub.
+
+**Carmen Andoh:** You're a nerd in the best way, Kelsey.
+
+**Mat Ryer:** What did you write it in, Kelsey?
+
+**Kelsey Hightower:** You know what, I just fell back to Java. Go just wasn't -- I'm joking. It's written in Golang. Come on, what are you talking about? \[laughter\] It's gonna be written in Go, of course.
+
+**Johnny Boursiquot:** Ohhh...! Troll of the year, I hear. \[laughter\] Awesome. Well, thank you all for being a part of the show, thank you all for listening, and we'll catch you on the next GoTime.

Community Q&A_transcript.txt

@@ -0,0 +1,733 @@
+**Mat Ryer:** Hello, and welcome to Go Time. I'm Mat Ryer. Today we're doing a very special live Q&A episode, because yesterday Jon Calhoun, who is with us today... Hello, Jon.
+
+**Jon Calhoun:** Hey, Mat.
+
+**Mat Ryer:** Yesterday, Jon, you posted on Reddit, asking for people's questions, didn't you?
+
+**Jon Calhoun:** Yes, I did.
+
+**Mat Ryer:** And what did you think? Quite a good response, isn't it?
+
+**Jon Calhoun:** Lots of questions.
+
+**Mat Ryer:** Lots of very good ones... And we're going to do our best to answer them. But don't worry, we're not doing it on our own. Joining us today, Peter Bourgon and Roberto Clapis. Hello!
+
+**Peter Bourgon:** Hey!
+
+**Roberto Clapis:** Hello.
+
+**Mat Ryer:** Welcome. Peter, how long have you been doing Go?
+
+**Peter Bourgon:** I guess since the beginning, depending on what "doing Go" means precisely. I remember the day it came out; I had been at that point working a lot in C++, and basically getting frustrated with how difficult it was to do concurrency stuff correctly... So it was very timely and appropriate, and I remember building \[unintelligible 00:02:35.03\] on day one and being very frustrated and confused as to why it wasn't fast. That was my start.
+
+**Mat Ryer:** Hm... And what about you, Roberto? How long have you been at it?
+
+**Roberto Clapis:** Five years.
+
+**Mat Ryer:** And what made you get into it?
+
+**Roberto Clapis:** Basically, I was using a piece of software that was extremely slow, and took way too many resources, in my opinion... So I just decided "Let's rewrite it in Go and see how it goes." And it was much better.
+
+**Mat Ryer:** That's a very interesting thing. We've got some questions that relate to that, and I think that's kind of an interesting clue... So that'll be interesting. How about you, Jon? How long have you been doing Go?
+
+**Jon Calhoun:** That's a good question...
+
+**Mat Ryer:** You didn't expect it? \[unintelligible 00:03:21.28\]
+
+**Jon Calhoun:** No, I've never had anybody ask me that one. This is a first.
+
+**Mat Ryer:** \[laughs\]
+
+**Jon Calhoun:** It was probably like 2012-2013... That one stemmed from - I was working on a project where we had to talk with a bunch of APIs concurrently, and it was written in Ruby... So it was kind of just annoying, because Ruby handles one request for a server you have up - if you're using Ruby on Rails at least - and while we could do all those different requests in threads and wait for the response, it was annoying that our server would basically sit there doing nothing while it waited for some APIs to respond... And I was like "If we could actually have a server that could handle multiple requests at once, there's no real reason why this one can't do it." It just wasn't set up to do it at the time.
+
+**Mat Ryer:** \[04:03\] Yeah... See, that's another thing - you get a few things that it does out of the box, that are very attractive. And actually, one of our first questions from the Reddit feed is from dlibian. Dlibian asks "What kinds of projects are best for learning Go?" For people that really want to start to learn Go, are there any types of projects that are more suited and more suitable, or can they just sort of pick any problem and go after it? What do we think?
+
+**Jon Calhoun:** I think generally speaking, diving in is more important than getting the perfect project. For me, I always like doing things on my local file system, like opening up files, appending to them maybe... Just simple stuff like that that I can actually go look at a file and verify that it worked as I expected. Not having to worry about servers and that sort of stuff, that doesn't necessarily make it super-complicated in Go, but it's harder to verify everything... So that's where I tend to start. But I think each person is gonna be different.
+
+**Peter Bourgon:** I totally agree with this impulse - simple, and in a domain you are familiar with. You don't wanna be learning both a new problem and the language at the same time. Pick the thing to spend your energy on... Ideally, something you've written before, and it won't take you like a month.
+
+I've heard Go described as like a DSL for writing network servers, which resonates with me. All else equal, a simple thing that deals in packets and requests and stuff I think is probably the type of project that's gonna expose you to the most and widest variety of parts of the language, I think.
+
+**Roberto Clapis:** For me, one of the first things that I wrote was a web crawler... And I think that in general, Go shines best in tooling. When you need to do something - I don't know, a repetitive task that you don't want to do by hand, so you go automate it with Go; no UI, no nothing, you just write a console program and you run it... I think that's, in general, a good thing to start with.
+
+**Mat Ryer:** Yeah, that has the nice benefit of also being able to solve a real problem that you have, and I think that's always an advantage. When you're learning something new, if you can solve a real problem with it, then you're in a much better position to focus your learning. You have to learn the bits you need for this thing. So it's quite a nice way to cut out a lot of what you could be learning and really focus on something and then at the end of it, hopefully you've got something useful.
+
+I know teams that have done that as a way of introducing Go into the team. They've sort of found a little problem that they've all got, and somebody's just taken it upon themselves to solve it... And usually, it's a tool, it's exactly that; some command line thing, something that the developers are using. So it's not gonna go straight into production or anything like that. Something that you can have a slower introduction to. I think that is kind of a good strategy, and then yeah, I echo what you say about the domain as well.
+
+**Roberto Clapis:** One thing I found really funny is that in a company a friend of mine works in, they started using Go to generate code for Java... Basically, to generate some repetitive code, they found that the text template package in Go was working pretty well for them, so they just generated some code with Go in another language, because that was just an option, and they started using Go with that... Which I found pretty funny, because many people say that Go has the downside of having to generate code if you need to write generic algorithms, and that was kind of a funny experience to have...
+
+**Peter Bourgon:** I also wouldn't expect that Go's templating language, which I think we could all agree is maybe not the best in the world, would be the best default choice for something like this.
+
+**Jon Calhoun:** I wonder if it had to do with the fact that it's built into the standard library, and in a lot of other languages - that I'm aware of, at least - it's not really built into the standard library.
+
+**Roberto Clapis:** \[08:04\] Yeah, that and the fact that it could very easily expose language functions that you wrote in the language, to the templating language. So it was very easy for them to just write most of the logic in Go and just expose the function that they needed to the template.
+
+**Mat Ryer:** Yeah. Another question that was asked by ninevolt is "Why should people try Go?" Some people are kind of skeptical, and I understand this; there's so many new things... You don't wanna have this shiny object syndrome where you're just chasing everything and you never really get good at anything... So some people therefore become automatically skeptical and hold back from learning a language. This is the case for this particular questioner... And they asked "Is there a kind of convincing sales pitch to convince people to try something with Go?"
+
+And by the way, they are putting together a summer sound bites thing... So if your answers could be awesome little sound bites, then that'd be great. \[unintelligible 00:09:08.15\]
+
+**Peter Bourgon:** Yeah, we need to drive all human conversation more towards the Twitter model and away from reasoned discourse. Yeah, understood.
+
+**Mat Ryer:** Yeah, yeah. Because it's what works, isn't it?
+
+**Peter Bourgon:** Yeah, clearly. We're all building our personal brands here, so this is also -- it's all part of the big picture, yeah.
+
+**Mat Ryer:** By the way, that's a great example. That's going in. \[laughter\] \[unintelligible 00:09:29.28\] No, it's perfect.
+
+**Jon Calhoun:** I'm waiting for Pace to pivot into Zoom plugins that count your characters while you're talking...
+
+**Mat Ryer:** \[laughs\]
+
+**Roberto Clapis:** Wow. Let's make that never a reality, please.
+
+**Peter Bourgon:** Too complicated. So this is an interesting question, because there's a period of time where it wasn't clear that Go was gonna stick around, and you did need to make the pitch to get people interested. It's like, "Well, I understand you're not convinced, but here's some use cases, and here's some things that it has worked at, and here's some success stories." But I think we're past that point. I think Go has carved out a reasonably well-defined area/context in which it is useful, and I think if you look even a little bit, you'll find the success stories... And often, if you don't look at all, you'll find the things that people dislike about it, and you can get a pretty good idea of what's good and what's bad, and if it's appropriate to your use case.
+
+So I think we're at a point - and this is my personal opinion, where if you're not convinced, then fine. There's nothing wrong with that. If you need someone to pitch the language to you, take some of your short life on this Earth and spend it in a way that brings you more joy. It's fine to say "Go isn't something that's gonna be useful for me right now" and move on to something else, in my opinion.
+
+**Mat Ryer:** Absolutely brutal, as expected. Thank you, Peter. Roberto?
+
+**Roberto Clapis:** My answer is a little bit less brutal... I think people should learn Go, because it makes their code better, even if they decide not to use Go anymore afterwards.
+
+**Mat Ryer:** So how does it do that?
+
+**Roberto Clapis:** I've found out that after years of using Go, my code in other languages which I still use is better, because I try to keep the line of sight, stuff indented to the left, code is simple, stick to the native types, do not create unnecessary types where they're not needed... I would also say that the lack of generics so far forced me to try to write the simplest code, and code that was closer to the data that it was touching, instead of it just being generic and maybe inefficient.
+
+Now I've found that even when I write TypeScript, which is the other language that I use the most, and Java, my code is more readable.
+
+**Peter Bourgon:** Yeah, Go biases you away from abstractions in general, and I think this impulse is probably extremely good in many different languages.
+
+**Roberto Clapis:** Yeah, I agree.
+
+**Peter Bourgon:** Or at least it moves you in that direction.
+
+**Jon Calhoun:** I would generally agree with Peter, that if you haven't been exposed to a scenario where you've already considered Go, there's probably not much at this point that's gonna sell you, and I think trying to make the sales pitch is just not gonna resonate.
+
+\[12:06\] The only thing that I would probably add to what Roberto added is that the community in Go is also something that I think other languages should sort of look at and get an example of at least -- it's one of the better ones, in my experience, so getting a feel for what it could be and making sure they set the bar at the right level, versus just sort of accepting a toxic community...
+
+**Mat Ryer:** Yeah, I think that's a great point. I think these are all great points. For me, I'd say that in particular if you're building web servers, web services, JSON APIs, those kinds of things, Go does a good job with those kinds of things. You get a lot out of the box for free. For example, each HTTP request runs in its own goroutine because of the standard library, the way that that works. So kind of automatically you get a level of quite safe concurrency to operate in... And you may get that also with other languages.
+
+Another point is that there shouldn't be a kind of language war thing going on as well, which just seems to very naturally happen quite a lot. People want to know why is Go the answer, and really, we don't have the question. So it depends on what that question is. Some languages are great at different things, and I think that's probably something that is worth avoiding in yourself early. I know that I used to think more like that, and a more relaxed, open-minded attitude - I think that helps. Because like you say, there's things in other languages which we can learn from as well.
+
+Pure functions is one example that Rust makes extensive use of; in fact, it's the only way you can do it, I think... And you can then write pure functions in Go and take some of those lessons if you want to. So that sort of stuff is nice. But yeah, anything that involves web and APIs...
+
+I think tooling is another great area. When you've got a task, something repetitive you wanna run through, Go is nice; it's not just a good way to learn, but it also does quite a good job, even if you haven't written it perfectly. I think it's quite cool.
+
+**Peter Bourgon:** I just wanted to extend that a tiny bit, because you touched on something that I've seen over and over in the Newbies Gophers Slack. I think it's a little bit like what Go is or isn't suited for, and you said it's really good for writing web servers - I totally agree, but I think something a lot of people don't understand is that the Go model of what a web server or HTTP request is substantially different from a lot of other languages. In Go, the central model is the request itself, and the path that it takes through the handler stack, in the same way that the execution path takes through a call stack. So this is what Go wants you to think about, in terms of this relatively mechanical bit of the thing... And in many other languages, the model is completely different - the model-view-controller idea, where it's like a route being matched to a controller that interfaces with domain models in a repository. A lot of people who are used to this latter one come to Go and find it very low-level and unproductive, and I think for them it actually probably is.
+
+To go back to the question, like "What is it a bad fit for?", if you wanna build in the consultancy style of like high throughput, very rapidly procured, kind of cruddy web services, I think Go is probably not a great fit, because that's not the model that it uses to talk about HTTP. I think this is a really interesting thing that comes up over and over again. I don't know, maybe that's just me.
+
+**Roberto Clapis:** \[15:46\] I would say that to address that, some people wrote frameworks. I know that there are some frameworks that when you use them, they don't feel like Go, but they allow you to write Go in other parts of the code, and in the entire HTTP kind of CRUD stack, or an MVCstack is handled by the framework. But I would agree with you... There are other things in which Go is not good for.
+
+For example, it happens all the time on Twitter when people discuss that some things should be made in Rust, or in Go, or in C, or stuff like that, and I would say that most of the time, one of the two is the better choice. It's like, if you need to write firmware for an embedded device, and there's a very limited amount of memory - yes, you could use TinyGo or TamaGo, which are two perfect options, but I would just say that maybe that's not what you're looking for. If you have to write a real-time application, Go still has a GC. You can write GC-free Go. It's not pleasant. So just use the right tool for the job.
+
+**Jon Calhoun:** A follow-up question to this is bestform had asked "Do you follow other programming languages? Do you compare their design choices, and the ones they're making, with what Go is making as design choices? If so, which ones are you following? How has that influenced you?"
+
+**Roberto Clapis:** I'm following TypeScript, JavaScript, Rust and Java, mostly... And I have to say, there are some things that I really like about those languages, there are some things that I have to use from those languages... I try not to do the comparison. I know what the languages are for. Sometimes I carry over some concepts, but it doesn't influence me too much to know other languages or to follow other languages.
+
+**Peter Bourgon:** Do you think about them in terms of their founding principles, or theories? Or at a sort of more abstract level, and sort of compare that with Go? I find this often quite interesting, although maybe ultimately non-productive...
+
+**Roberto Clapis:** You know, I never think of it in these regards; I just use them, and I try to click in the mindset. When I speak English or when I speak Italian, it's not like I try to think in the other language and see how I would put the sentence in the other language. I just click into it and try to think in it.
+
+I've tried to build UIs with Go. I still prefer the TypeScript approach, some things... And I've tried to build stuff with Rust, and at one point I just gave up and I said, "Okay, I'm just gonna use Go. It doesn't work for me." But most of the time, when I start with a language, I just stay in that mindset.
+
+**Jon Calhoun:** A follow-up to that - since you're switching mindsets, one of the big things in Go is you write readable code. Do you find in other languages that you don't prioritize that as much, just because that's kind of the norm in the language?
+
+**Roberto Clapis:** Yes. I find Java unreadable. It's a big limiter for me, because Java uses very long lines, and most of the time you have an auto-format there that will split the line automatically. So I'm used that in Go one line means one thing, and in Java sometimes you have like 24 lines that are saying a single statement. That kills me.
+
+**Peter Bourgon:** What I find super-interesting is that even what constitutes readable is completely different from ecosystem to ecosystem, or person-to-person also. This subjective definition difference is also super-interesting to me, because I think for a lot of people, for example in maybe the Rust sphere (I don't know), readable is like what is somewhat terse, and compact, and conveys the most semantic information; not objectively the most, but a lot of semantic information, in a few characters. But for Go, simplicity has a completely different meaning. What we mean when we say it is totally something else.
+
+**Jon Calhoun:** I think Rust is probably the one that I see this the most in, because people will show me -- I think it's \[unintelligible 00:19:52.08\] and they'll show me code that does something... Like, they'll compare it to error handling in Go, and they'll say "Look, looking at if err != nil is distracting", and then they'll show me the example in Rust, and I'm like "I don't know Rust, so I know that's my issue here, but I don't know what that code is doing." It just is not clear to me.
+
+\[20:11\] So for me, at least in the Go sense, readability is somebody who basically is just a junior programmer could look at this and probably get a pretty good idea of what it's doing. But in Rust, I think you're right, they're viewing it differently. What they consider readable is different.
+
+**Peter Bourgon:** And this error handling thing is such a great example, because in Go error handling is explicit. It's part of the philosophy. Go considers error handling to be programming, and in a lot of other languages it's just not modeled that way. That has ramifications.
+
+**Roberto Clapis:** Right. But I also think that it matters how much of your brain you're willing to dedicate to the language, instead of the code that you are writing, instead of the logic that you are writing, and how much you're willing to dedicate to the actual problem that you're trying to solve. Sometimes Go might exceed on the other side, but I like that when I read Go and when I write Go, my brain doesn't invest any energy in "How do I do this in Go?" I know how to do this in Go, and I just do it, which is not true in many other cases... So I like that simplicity and I like that I don't have to focus to write Go on Go, but I can focus on something else.
+
+**Break:** \[21:27\]
+
+**Mat Ryer:** There's some interesting questions actually around some of the specifics of how we actually do things as well. There's a lot on structuring, which I think we'll come to later, but there's one that I quite like here, which we can discuss, and we may have differing opinions... To constructor or not to constructor. Some languages make heavy use - in fact, sometimes they're compulsory; in order to create a class you have a constructor. In Go you don't really have classes, but you can still have this idea of constructors, where you just sort of have a function, usually pretty fixed with new, to create something, and then it returns the thing, and maybe sometimes an error, if there's some work to do to get that thing. How do we feel about constructors?
+
+**Roberto Clapis:** I even have a follow-up question - how do we feel about builders? Once we decide on constructors, what about builders?
+
+**Mat Ryer:** Okay, let's do constructors then first, and we'll do builders after.
+
+**Jon Calhoun:** I can start by saying that when I started writing Go, I wrote a constructor for everything. And I think part of that stemmed from -- my history is Ruby and Java. And Go in some ways just sort of felt a little bit more like Java, because it was typed, and some of that stuff... So I think I just jumped into that mindset of "I need to write these." And then - I don't know when, but at some point I kind of realized that a lot of these constructors were not useful. I didn't need them at all. So I started taking a step back and asking myself "Do I really need it for this type? Why don't I just expose these fields and let the developer set them?"
+
+\[24:24\] So I don't really have a specific "Yes, you should use them/No, you shouldn't" type thing. For me it was just stepping back and deciding, case by case, is it necessary? And if it wasn't necessary, I just tried not to do it, because it didn't seem worth the effort.
+
+The other thing I would add is that if you are using a constructor, I think the common approach that people would take at first is to write -- like, if you're writing a thing, you write a new thing function that creates it... And I think there's a lot of times where you can make your constructor function a little bit more clear as to what it's doing. The database SQL package is a good example, where you call sql.open and it returns a db instance. I think that makes way more sense than sql.newdb, which isn't really clear what it's doing.
+
+**Mat Ryer:** Hm... So if it could be like a verb...
+
+**Jon Calhoun:** Yes. If you can explain what it's doing, I think that makes more sense.
+
+**Roberto Clapis:** And you're allowed to have multiple names in that case, because you might construct something, but performing different actions.
+
+**Mat Ryer:** Hm. Peter?
+
+**Peter Bourgon:** Yeah, like in any language, there's not one answer that is generally applicable. Sometimes constructors make sense, sometimes they don't. The one thing that I think is important to understand when thinking about this question is this fact of Go that except for very few exceptional circumstances it's always possible to construct the zero value of a type. Basically, there's no way to avoid allowing your callers to do that. So the zero value of a type is always something that can exist.
+
+With that in mind, it would be cool if the zero value of a type were useful. That's also like a proverb. If you can create a type where that is true, and all is equal, then maybe that's a good idea. You don't need a constructor; you can just use it in a useful way, without needing to initialize it. Of course, that's not always possible. If it's not possible, then it's probably better to use a constructor, because that gives you, by convention, control over what the -- I said "by convention" instead of like... It's not enforceable by the compiler, but it gives you some control over what the state will be in a way that setting fields in the caller context doesn't... So it's pros and cons. No single answer.
+
+**Roberto Clapis:** Have you ever written a destructor?
+
+**Peter Bourgon:** It's not possible in Go. It's finalizers, but that's something else, I guess.
+
+**Roberto Clapis:** Yeah. Have you ever written a finalizer?
+
+**Mat Ryer:** No, I haven't.
+
+**Peter Bourgon:** I have, but it was almost always a mistake. \[laughter\]
+
+**Mat Ryer:** I'm glad I haven't now. But that's interesting about the zero value. There are some examples, like the bytes buffer. You can just kind of say "I want a bytes buffer" and you can start using it immediately.
+
+One of the things you get with constructors is you can take in required arguments into the constructor. So then of course it's not possible to get the type, according to the API, without providing those values. Of course, the zero value is always there.
+
+I think one of the other counter things I hear is that in a way you're hiding what's happening. You could be spinning up goroutines, you could be allocating all kinds of memory inside a constructor, and by avoiding a constructor, it makes it very clear what's happening. If you're asking your user to create the instances, they see exactly everything that's going on.
+
+But to be honest, I think for me the convenience -- I went through a similar evolution where I used to always create them, and then I went off it and I started to just try and get the zero value to be useful. And then almost all of those types evolved eventually on their own into basically needing a constructor.
+
+So yes, but for me it depends on the case, which doesn't help. For services and things like this, if you've got these sort of service-level objects, structs and things, then I do tend to do it, because usually they have dependencies and it's kind of a nice way to tell that story.
+
+\[28:16\] But yeah, it depends... If I was building some lower-level thing, or something that had more of a data structure component to it, I think I'd probably think maybe differently. What do you think about the options as well, a way of customizing things by passing in little option functions, which can then run and modify things?
+
+**Peter Bourgon:** Functional options is a configuration pattern, right?
+
+**Mat Ryer:** Yeah, right. What do you think of those?
+
+**Jon Calhoun:** I can start by saying that I don't like builders that chain. In Go they just don't feel right to me half the time. I don't know, it probably comes down to like I've been bitten by error cases where one of the ways that they'll handle errors in a builder pattern is they'll have the created object have an error field, and you're expected to check it once you've built it. And I think it's really easy to miss that. Whereas if you use functional options instead, it's much easier to make your constructor actually return the thing and an error, and you know clearly what happened.
+
+**Mat Ryer:** That's a good point, actually... If in creating or in constructing the thing, if something could error, that is probably a case for having a constructor, isn't it?
+
+**Roberto Clapis:** Yeah, I agree.
+
+**Jon Calhoun:** We're all nodding our heads, for people who can't see us.
+
+**Roberto Clapis:** Yeah. And even for the builders that you were talking about, what do you do if something in the middle of the chain errs? Does everything else become a null operation? But what if some of the parameters that were passed actually called a function, or stuff like that? In that case, will it just return the error at the end? And how do you know what was executed?
+
+**Mat Ryer:** The ones I've seen that have done well with those fluent APIs, they just change state; they're basically just changing internal state in a controlled way. They can't error. So that's why. And then usually there's an operation at the end, which is the big operation, and that one will return an error potentially. But they do make testing hard, or they can make testing hard if you wanna stub things out, or have some kind of abstraction on that stuff. They can get in the way a little bit, since they're returning the concrete type often. So you can't do the trick of creating your own test interface in order to provide different mechanisms for it, or a different implementation.
+
+**Roberto Clapis:** Right. The one good use of builders that I've seen - and recently I had to use this - was for security reasons. We wanted to make sure that a certain object was constructed in a precise way by passing arguments in a certain order, and nothing would ever happen out of order. Basically, in the middle of the chain you would call a certain function that would return a new type, and on that one you could specify more stuff. So basically, it returned types that were not the entire thing, but in every phase you had to specify some things, and then you had to move on to the next state.
+
+**Mat Ryer:** How did you name those types? That sounds like a nightmare.
+
+**Roberto Clapis:** We still have to decide. \[laughs\]
+
+**Mat Ryer:** Yeah, that doesn't sound like an easy problem... Because they're all the same thing.
+
+**Jon Calhoun:** He's like "We started this six months ago and we still haven't decided."
+
+**Mat Ryer:** It's just gonna be 1, 2, 3, ain't it? Step one...
+
+**Roberto Clapis:** This issue was opened a month ago, and we haven't decided on the names yet. So yeah, you got a problem with that.
+
+**Peter Bourgon:** Wait, I'm trying to understand... So you have a thing and a high level, which is like in the end a sequence of substeps, or something like that, right? And you're using the builder pattern -- you're modeling the substeps as different types, and you're using the builder pattern to move from one substep to the next. Is that right?
+
+**Roberto Clapis:** Yeah, basically I can give you the concrete example. We are building a framework to develop web applications, and when you set up your entire server, you register handlers. After you have registered handlers, you're supposed to register plugins. That is something that will intercept requests and responses, for security reasons. Most of the logic will just be in handlers.
+
+\[32:06\] And then after you're done, you start a server, and at that point you are not supposed to change nor the handlers, not the plugins anymore. And vice-versa, when you install a plugin, you still had to register a handler at that point. We could do this by panicking, if you did things out of order... And this is still on the plate. But we found out that checking these at compile time, making sure that if your code compiles and you did stuff in the right order felt like a good thing to provide.
+
+**Mat Ryer:** I've genuinely never seen that as an example in Go.
+
+**Peter Bourgon:** I would love to have a deeper conversation about this, because I have many, many more questions that are not appropriate to this call... But okay, we'll see.
+
+**Roberto Clapis:** I can discuss it later.
+
+**Mat Ryer:** Interesting one.
+
+**Peter Bourgon:** Yeah. Just maybe quickly on functional options - I really liked them for a long time, but again, think about the actual property of the language or the invariant that they encode... If you're using functional options, your user can always not pass any - that is a valid state - which means that with functional options you can always create a zero value for a type, in a sense, by not passing any options... Which means that they're good when it's rare that you need to change anything away from the default value. When you do, you only wanna specify one or two things that are different.
+
+So if your type has these semantics, then they can be good. But with that said, I rarely have types that work that way, and when I do, it's often equally easy to configure them in a different way. This has been sort of my experience. These days I don't really use them anymore. I think the way they pollute or change the GoDoc for a package, or make it slightly harder to understand outweighs the benefits that they give you.
+
+**Mat Ryer:** Okay. Well, moving on, we have another question here from datacharmer... And I'm gonna paraphrase this, but it's basically saying "You know how when you parse dates in Go - that's weird, ain't it? This magic date - it's Monday, and it's January... It's a really specific time - why is it like that, and do you think it's weird?"
+
+**Peter Bourgon:** I'm so confused by questions like this. I'm so confused.
+
+**Mat Ryer:** Why?
+
+**Peter Bourgon:** Because everything in a language is like -- they made a lot of decisions, right? Why is it func and not fun, or function? Why are declarations in the order they're in? Can you do var blocking with parentheses? There are so many things that just are...
+
+**Mat Ryer:** So you just accept it.
+
+**Peter Bourgon:** It's just something you have to learn.
+
+**Mat Ryer:** It's just a law of physics.
+
+**Roberto Clapis:** I would say that's beautiful, actually. That's one of the things that I like the most about the date stuff in Go... Because when you think about parsing a date, usually you have an example in mind. It's like, I've tried to use other libraries, especially Java ones, for date parsing. They're not intuitive at all. Yeah, the Go one is odd, but once you have understood it, it's very easy to use. It's like, you just write a date in the format you expect it to come, which is -- this is user-centric APIs.
+
+**Peter Bourgon:** It's not hard, yeah.
+
+**Mat Ryer:** Except it's not your date, is it? You have to use the right values in each slot, essentially. And they're numbered, aren't they? It basically counts up, if you notice...
+
+**Roberto Clapis:** Yeah, exactly.
+
+**Mat Ryer:** Yeah. But that is odd.
+
+**Roberto Clapis:** Well, you have to remember \[unintelligible 00:35:34.04\]
+
+**Jon Calhoun:** It doesn't help that the example they tend to give doesn't show it obviously that it's counting up.
+
+**Mat Ryer:** Yeah.
+
+**Jon Calhoun:** The one they give is like Monday, January 2nd, 15...
+
+**Mat Ryer:** Yeah. Three, 04, 05...
+
+**Jon Calhoun:** Yeah. And the way it's put there, you don't realize that it's really going 1, 2, 3, 4, 5... Because 15 - well, that's three, but people don't quite realize that it's 3 PM. That throws them off.
+
+**Mat Ryer:** Yeah. And MM/DD/YY - that style... I feel like that is used in more places. What do you think?
+
+**Jon Calhoun:** I think the hard part is most MM/DDs have capital Ms and lower-case Ms, and they mean different things...
+
+**Mat Ryer:** Yeah, that's true.
+
+**Jon Calhoun:** \[36:15\] And you still have to look something up. So I think at the end of the day, the short answer is that you're gonna have to look something up. Looking up a specific date is really not that big of a deal, compared to looking something else up. And there's tooling that makes that easier.
+
+**Peter Bourgon:** Yeah, I always have to look up the \[unintelligible 00:36:29.08\] time stuff personally. I never remember that.
+
+**Mat Ryer:** Okay, well I'm with the questioner now. I think that is a bit weird. But the answer is that's just the way it is. This actually could be quite easy; we could just hammer through these with that, if that's our answer. \[laughter\]
+
+**Jon Calhoun:** That's just the way it is...
+
+**Mat Ryer:** Yeah. The next one actually, which we got a couple of times, is "How come there's no enums/enumerators in Go?" So this is that thing in other languages where you get a compiler-time way of declaring a set of values that are only valid for a particular type... And you get kind of compile-time help there. It's useful if you've got status, and you've only got four different statuses that are possible. You can make types give you safety there. What do you think about that? Because there is the iota pattern, which is basically how we do it most of the time. What do you think about this?
+
+**Roberto Clapis:** It's the only feature that I miss in Go. That is the one thing that I would add to Go. I would not add generics, I would not add many other things. Enums is the one thing that I miss.
+
+**Mat Ryer:** Yeah.
+
+**Jon Calhoun:** Go ahead, Peter.
+
+**Mat Ryer:** Please disagree.
+
+**Peter Bourgon:** No, I don't... I just wanted to say it would be nice, definitely.
+
+**Mat Ryer:** Oh, you've not disagreed? I'm sorry, it's supposed to be controversial.
+
+**Jon Calhoun:** He didn't do it.
+
+**Peter Bourgon:** Next one. When we talk about frameworks we'll get in the weeds.
+
+**Mat Ryer:** Okay. But Jon, you were going to say something about enums.
+
+**Jon Calhoun:** I was gonna say, iotas terrify me, because I always worry that somebody's gonna insert a random one in the middle of the iota values, and all my code is just gonna break. Every time I go to use them, I get terrified of that, and I end up just writing strings for everything, because I'm like "I know these are gonna work if they're stored in a database, or something." Am I the only one that feels that way at times?
+
+**Peter Bourgon:** I'm also afraid of iota. I tend not to use it.
+
+**Mat Ryer:** To be honest, I use strings.
+
+**Jon Calhoun:** So the advice we give everybody, we don't use.
+
+**Mat Ryer:** Well, that's the Go way to do enums, as I understand it. But yeah, the alternative is just to have constants with strings, or constants with integer values. You can still document them, you can still describe it... And usually, if you use a prefix, that's a quite useful way. HTTP status is an example. They've got all the status codes as constants, and they've also done the HTTP methods, which - I don't know if we needed those, but we have them. So you can do http.methodget, methodpost, methodpatch.
+
+**Roberto Clapis:** But that is weird.
+
+**Mat Ryer:** Yeah. It's longer to type that than to type the string.
+
+**Roberto Clapis:** Right.
+
+**Jon Calhoun:** I don't know... I've found that it doesn't bother me using it, especially with autocomplete and everything.
+
+**Roberto Clapis:** Right. But does it mean that we expect the get verb to change from get to another get? That is here to stay.
+
+**Mat Ryer:** Yeah, yeah.
+
+**Roberto Clapis:** I don't know, that feels weird.
+
+**Mat Ryer:** Yeah. I probably wouldn't have done it, but it's completely fine. You could make the same arguments for the status codes. Frankly, I find those to be much more readable, especially because I don't know all the HTTP status codes.
+
+**Roberto Clapis:** Yeah, exactly. That makes sense, because you read words, instead of a number.
+
+**Mat Ryer:** Right.
+
+**Peter Bourgon:** I didn't actually know that the methods were defined in the HTTP package.
+
+**Mat Ryer:** Yeah, exactly. There you go. So you don't use them.
+
+**Jon Calhoun:** At times I feel weird not using them, because somebody's gonna be like "Why aren't you using constants? Why are you hardcoding strings in here?" And even with status codes - I'll occasionally write 200, because I'm like "If you're writing a web service and you don't know what 200 means, you probably need to go figure that out first."
+
+**Mat Ryer:** That's true. 404 is another one. Seeing status.notfound to me isn't as clear as 404. That's the big one.
+
+**Jon Calhoun:** Yeah. There's a couple that are like that. There's just a couple that are like "If you don't know these ones, then you should probably figure them out." But then there's other ones where "Okay, I get it. This one's not quite as clear." So I have mixed feelings on that at times, but I tend to just use the written out "status okay" or whatever, just because I want all my code to be consistent... But sometimes I'm on the fence there.
+
+**Mat Ryer:** \[40:21\] Yeah. If I wanna say, as a string, "We'll post you a T-shirt", I'll use the post const there. So I'll go the other way and use it way too much, where you shouldn't use it. That's a joke, by the way. Any beginners - don't do that. It'd be crazy.
+
+**Peter Bourgon:** \[unintelligible 00:40:35.20\]
+
+**Mat Ryer:** Yeah, don't tell jokes. It's risky.
+
+**Peter Bourgon:** It's a tricky game.
+
+**Mat Ryer:** It's a tricky game... Especially when your editors are not kind to you in the edit, as it turns to be the case for me. Okay, so shifting gears a little, when should we inject dependencies? Maybe we could just have a quick overview of what we mean by dependency injection, and then we could talk a little bit about that. Who wants to tell us what is a dependency injection.
+
+**Peter Bourgon:** If you mean "When should you pass dependencies to things that need them?", the answer is always. Never not.
+
+**Mat Ryer:** \[laughs\] Yes, but what about an abstracted -- do you believe in any sort of abstracted dependency injection mechanisms, or is it just a case of pass arguments, because then you get all the compiler help?
+
+**Roberto Clapis:** I like to have a struct when I build web servers especially, that keeps all the dependencies that need to be injected. So if I need clockwork just to have a static time that I can measure in my tests, it's fine to inject time. Or inject a random source; like, have a crypto safe random source at startup on the real server, and one that is unsafe and predictable in tests. But I despise with all my strength dependency injection frameworks. I do not like them.
+
+**Peter Bourgon:** That's what a lot of people mean when they ask about dependency injection. They mean the thing that figures out your dependency graph for you based on annotations, or something like this. I agree; I understand some circumstances where they might deliver value, but I think they are so rare that anyone who's asking the question shouldn't be using it, basically.
+
+**Mat Ryer:** Yeah, I think I agree with that. You can't go wrong with just having fields and setting them. And if you do have constructors and there's a service that relies on other services that you have as different structs, then you can take those as arguments, or have them in the struct on the type, or whatever. And that is very clear -- and sometimes when you're then adding something, when you want to add a new thing, there is then some plumbing to go and do. And that's the bit that -- because it's repetitive, we wanna dry that up naturally, as programmers.
+
+**Peter Bourgon:** That's such an interesting point, because you said when there's some plumbing to do, but I see that work as so virtuous... Because it's not just you doing this mechanical stuff, this is you expressing what is now a new truth in your application. There's this thing which is now being used in this dependency chain that is now visible. Someone new to the code is gonna see that that's there and it's needed for these things, and that's so important to make explicit.
+
+**Mat Ryer:** Hm. So it's very rewarding, and usually it's quite easy work, too. It's just a case of looking at another service, going and looking where that's wired up, and doing the same. So it's quite easy... But you're right, it is quite rewarding. There is something very satisfying when you get to the point of actually exposing that service. It just feels like a bit of a milestone. That is a good point.
+
+**Roberto Clapis:** Right. And it allows you to do more general things. In dependency injection frameworks one thing that really happens is that they compute the graph, and the server responds. They will need to build up the entire graph and construct most of the things, even if you're not gonna use them, because they don't know if you're going to use them.
+
+\[44:06\] One thing that I like to do in my web servers, especially if I use cloud services - I use sync once everywhere to construct my stuff, so I know that stuff will be constructed only if I actually need it... And it's very hard to do that if you just inject magical things that depend on other magical things. I like to do the things that I need to do, not more.
+
+**Mat Ryer:** Right. So by default try not to do anything, and then when it's needed... And that comes down to also the environment in which you're gonna deploy the code... Because if you put that into a situation where it scales down to zero instances automatically very often, you're probably gonna get a big saving with only doing things on-demand. Whereas if you've got services that run for a long time, or that are just permanently running, maybe it's easier to just get all the work done upfront, so you know it's gonna succeed, and then you can go from there. So I think yeah, those kinds of decisions matter more, for sure.
+
+**Jon Calhoun:** This type of question is hard for me, because I agree with you guys that most of the time when it's asked, it's somebody coming from a Java background or something, where they're expecting to inject things with a framework... But on the off-chance they're asking when is it better to just put an actual type embedded into a struct, versus - or not necessarily embedded, but as a field in a struct - or to put an interface instead, I think for me it just comes down to "Are you actually going to replace this with something else at some point, either in your code, or in your tests, or something?" And if so, then the interface makes more sense. But if you're not ever replacing it, just start with a struct. It's really not that hard to change later.
+
+**Mat Ryer:** Why is there this natural aversion to this? Do you think that there's something about the fact that just setting fields on a struct feels too simple and too easy, and therefore not professional, in some way? Especially if people were used to big enterprise frameworks in the past.
+
+**Jon Calhoun:** I think some people just don't like having a main file that has all this setup of like "Okay, now I have to initialize this thing, and I've gotta assign it to all these different structs that need it..." Whereas if you come from a framework, like a Rails or whatever, you don't do any of that. It does all of that. Whatever assignment needs to happen, everything happens automagically, and you just access the things you need. So I think sometimes people are looking for this -- essentially, they wanna have some global db instance that they can connect to, and they don't wanna have to inject it anywhere; they just wanna be able to access it. And I think they don't quite realize that down the road that could potentially introduce some problems.
+
+**Mat Ryer:** Yeah.
+
+**Roberto Clapis:** I feel like people feel like that until they need to debug it. They like the magic until they need to find what's wrong. Because for me, I think everything magically injected was beautiful until I had to find out who was constructing that.
+
+**Mat Ryer:** Yeah, that's true. Any time you use a pre-built framework or an abstraction or something, you then are kind of tied to its decisions. And if you then have to fight it, it can get very unpleasant. A lot of people I hear talking about avoiding frameworks are coming at it because they've been burned so many times... So that's why we talk so much about this avoiding early abstractions.
+
+I think it was Ben Johnson on Twitter who made a good point about this... And he actually made the counterpoint, which was "Sometimes the right abstraction basically is amazing, is so powerful that if you do find that abstraction, then you really can unlock a lot of potential, enable lots of people to do things." That's why I think they're so attractive. But they're a bit like the Holy Grail; they're difficult to find. I think the Holy Grail is difficult to find, if I understand it...
+
+Anything else on that? If not, we have a question about init, which I wanna put to Peter.
+
+**Peter Bourgon:** \[47:57\] Why are you doing this...?
+
+**Mat Ryer:** Well, I know we're gonna get a good answer. What about init? When should you use init, Peter?
+
+**Peter Bourgon:** Since we've done a lot in this episode, let's go back to what are the mechanics of this thing; what are the truths of this part of the language. Init is a function that runs when you import a package, runs as part of program startup, and it's run before main starts... And it's a product of how it exists in a package. All it has access to, by definition, are the package-level consts and variables, the package-level declarations. It doesn't have access to any state that isn't declared at a package-level. And by design, its only purpose is to manipulate package-level state. That's the only reason it exists.
+
+But you shouldn't have package-level state, right? In general, you should try to avoid it. I took it a step further, I say "You should actually never have any", and so therefore, for me every instance of func init is a huge red flag. You were doing something in there which ought to be done somewhere else.
+
+**Mat Ryer:** Yeah, I concur with that. I would say just avoid init. It's a bit too magic, and anything that's too magic like that becomes a problem later. It's so fun when you're first doing it, and it gets things working, when you're writing it's fun, but when it comes to debug code and figure out what's going on, that's just too magic and probably worth avoiding. Jon, what do you reckon, mate?
+
+**Jon Calhoun:** So I guess my only caveat to that would be -- let's say you're writing a SQL driver. Pretty much the convention at this point is to register you driver via init. And as much as I wish that would change, I think at this point you're better off just doing it because that's the convention... But I don't love it, it's just one of those "Because that's convention, at this point trying to change that is not worth the effort. Just do it."
+
+**Mat Ryer:** So this is that thing where just by importing a package, its init function will register itself with another package.
+
+**Jon Calhoun:** Yes.
+
+**Mat Ryer:** See what I mean about magic?
+
+**Peter Bourgon:** Right. And I hope it's uncontroversial to say, this is a design choice that was made in pre v1 of the language, that there would be this global registry and package db, and this would be basically the way to use it. And I hope it's uncontroversial to say that this pattern has not stood the test of time. This is not a good pattern. So to use the global registry and package db you have to do it this way, and that's true, and it's too bad. But if you're writing your own package, do not copy this pattern. This is not the way to do things.
+
+**Break:** \[50:50\]
+
+**Mat Ryer:** I think probably our final question is gonna be a big, meaty one. We get this a lot, I certainly get this a lot, and it actually turns out to be a very difficult one to answer, but we're gonna do our best... Or your money back; it's free. You get to forget the ads, that's the way we'll refund you; you can forget all about the ads from this podcast. The question is "How do you structure Go code?" This is something that people care a lot about, it's the kind of thing that for beginners is very intimidating.
+
+You can learn the mechanics of "Well, this is how packages work, this is how you import packages", but what about "When should we build a package? When should it be in main? How should we structure this?" Because of course, a package is a folder in Go, so now we're dealing with folder structures, and things. This is a question I get a lot, so I'm just gonna ask you now - how do you structure code in Go?
+
+**Jon Calhoun:** I think we can all agree on the general idea of creating a CMD or Command folder, and inside of there sticking your different main packages. And generally, those just sort of initialize some state and start the program. From there, I think that's where it gets more confusing for some people, trying to figure out where to put the rest of the code and how it gets organized. So I'm just gonna say let's assume we have that.
+
+My general advice is stuff that I don't think people like to hear, but I really do think starting with a flat structure and waiting until you see a good way to break things up is the best way to go. Now, I will say that I have shown people how to do MVC simply because I didn't want them to think about it any longer. I just wanted them to get their stuff up and going, and I'm like "Later we can come back and figure this out." But for now, your goal is just to write some code. I think it gets asked so much because I think it's something that blocks people from moving forward, and it shouldn't.
+
+**Peter Bourgon:** A lot of languages have this (I guess) convention of all of the project structures being roughly the same, for the same type of project... Like, if you're doing a web service in Ruby, you're gonna have this layout, and the packages are gonna be named after the architectural patterns you're using. MVC, for example; controllers etc. But in Go, that's not really what we do. We have package and project structures that are basically reflective of the domain of the thing we're implementing. Not of the patterns we use, not of the scaffolding, but of the specific types and entities in the domain of the project that we're working in.
+
+So it is always idiomatically different from project to project, by definition. What makes sense in one doesn't make sense in another. Not to say it's the only way to do things, but this is what we tend to do... And so yeah, there's no answer, and that kind of truth about what is idiomatic in the language is extremely confusing for a lot of people, and it may be the wrong choice as a result... I don't know, but I think that's the main point.
+
+**Mat Ryer:** I think it can be quite liberating though to say that there isn't really a way to do it, which also means you can't really do it wrong. It's what fits for your case. And if you're not sure yet, then just sort of defer it. Worry about it later, if you can. It doesn't always work like that, and of course, if you're talking teams at scale, I think that's a little bit different. It's sometimes then worth investing a bit of time and trying to do some kind of design based on what you know about it.
+
+I personally follow the approach, when I'm building something new, of I just have a single folder, it's got a main in there, and then all the types. I just use file name prefixes with underscores actually to group the functionality up together, and have it all together. And then later, if we have to deploy this other service in a different way, or something changes like that, then I've got a good reason to go and have a look and make some changes. But I'm doing it in an informed way.
+
+**Roberto Clapis:** \[56:07\] I will say that the sentence that you said - it seems there is no scheme that you will adhere to; you can't get it wrong. I would phrase it different; I would say that since there is no mandatory structure, you might have a chance to get it right... Because so many times in a project I had--
+
+**Mat Ryer:** Less optimistic...
+
+**Roberto Clapis:** Right. So many times in a project in another language I had to force my structure to adhere to the one of the framework or of the language, that it made no sense for me and it was impossible for everyone to browse it. Because if the structure is the same for every project, it gives no information. It's like, sometimes when you visit a package, the package structure already gives you some information, and in other languages you lose that chance.
+
+**Mat Ryer:** So do you think that within a project you might even have contradictory things? You might have some things laid out one way, and other parts laid out differently, because it happens to suit?
+
+**Roberto Clapis:** Yeah, I do, because I try -- to the question "How do you structure your packages?", my answer is "The best way I can for the user." If I think that a user will benefit from one type, for example, being the exception of a rule that is general in that framework, in the thing I'm building, I will do that. I will have that type stick out in another thing. It's like cookies in HTTP. They're a header, but they're not treated as a header in the HTTP package, and that makes it so much more usable. And I think there's a good reason for that.
+
+**Jon Calhoun:** I will say the one thing that I have seen is that people will try -- Mat says there's no wrong way, I guess, and I agree that you should go try things... Because truthfully, if something is (I'm gonna say) "wrong" in quotes (not really wrong), but if something doesn't work, it's easier to understand why it doesn't work when you actually build it and start to actually experience why it doesn't work... And one of the common ones that you'll see is they'll end up with cyclical dependencies, where they're trying to import multiple packages this way... And I'm curious to get your take on this, but what I've been trying to advise people lately when I talk to them about this is -- I know nested folders don't really matter for packages in Go, but I basically said to try to think about your code almost as if you can only import things going up the tree, in your packages. If you think about it that way, there's no way you can possibly have--
+
+**Mat Ryer:** What do you mean by that, Jon?
+
+**Jon Calhoun:** Okay, so let's say your root level - you have an app. You just call it your app package. Underneath that you might have a SQL package that you're writing your database stuff in, and then inside of that you might have something even more specific, like a Postgres package or something.
+
+**Mat Ryer:** Right. Specific to the database.
+
+**Jon Calhoun:** So the Postgres package can import the SQL package, and it can use whatever types in there, and it can import the app package, but those ones can't actually import the Postgres package. They have to define an interface that the Postgres implementation would be used to satisfy.
+
+**Mat Ryer:** It's funny, because I've done that the other way around, where you can only import packages that are inside, and you do it the other way. It's the same kind of idea, I think, but I wonder if we're saying opposite things.
+
+**Jon Calhoun:** I'm viewing it like it might be similar. I'm viewing this as like you have a central object and it defines all your -- like, if you've ever read Ben Johnson's Standard Package Layout, it's the same idea. You have a central app that defines interfaces, and then as you expand, you get more specific implementations... And you might still have interfaces there that get implemented by even more specifics. And it's a little bit tricky, because as you work your way out, there's always going to be people who come up with some weird edge case.
+
+For instance, if you're using an ORM, you'll be like "Well, my app might have a user type, but then my ORM has a user type that's representing what's in the database and which one's correct." And I tend to treat that, if you're gonna use the ORM at least to this point, my suggestion is generally to treat it like an API, where your application might have a Stripe Charge, or some sort of object representing that internally, but then the Stripe API has their own representation of it and you have to translate between the two to get it into your application. There are some downsides to that, I don't think it's perfect, but it tends to get you going in the right direction without cycles in your dependencies and things like that.
+
+**Peter Bourgon:** \[01:00:17.25\] Ben expressed this really well... I think at one point he said "Packages should stack, and not interrelate." So there should be a clear stacking effect relationship between packages, and this is another way of expressing in the clean code, or hexagonal architecture, or domain-driven design - they call it the inward-facing dependency role. Your HTTP package can import your business logic, but your business logic cannot import the HTTP package. It should only go in one direction.
+
+**Mat Ryer:** Yeah, and that is key. When we talk about things going wrong or not working, from a structure point of view it is usually cyclical dependencies, like you say, Jon... And those kinds of things are easier to avoid if you just have one folder. So that's another reason -- you sort of just avoid those problems early by doing that.
+
+Well, it's that time of the show where we do our unpopular opinions.
+
+**Jingle:** \[01:01:22.26\] to \[01:01:40.10\]
+
+**Mat Ryer:** We did have an unpopular opinion from Reddit... It said "Unpopular opinion." This is from bklimczak, and bklimczak says "The simplicity of Go makes it less useful in some use cases." Which I think probably has some truth in it. There probably is a trade-off there... What do you think? Or we don't have to debate that one. Do you have any other of your own unpopular opinions?
+
+**Roberto Clapis:** I think I can quote Brad Fitz on this, in which he said that Go does everything you need, 100% of what you need, 80% of the time...
+
+**Mat Ryer:** It sounds like something from Anchorman.
+
+**Roberto Clapis:** ...which is true. Sometimes Go doesn't cut it, and that's fine. The simplicity and the benefit of having that hard, very opinionated language - the benefit overweights the downside of not being able to write some code, sometimes.
+
+**Mat Ryer:** Cool. Any other unpopular opinions?
+
+**Jon Calhoun:** Well, apparently Chris James has one, but I don't remember it in the Reddit thread... But he says if we don't say it, he's gonna riot, so...
+
+**Mat Ryer:** Okay... Then we're not gonna say it, obviously. Because we do not negotiate with terrorists on this show. We never have. They might tolerate that kind of thing on JS Party, but not on this show, baby. Or we can just do it though, if you want as well... Peter, do you have one, mate? You don't strike me as the type...
+
+**Peter Bourgon:** Well, unpopular to who? What audience we're talking?
+
+**Mat Ryer:** Most people. I think it just has to be most people wouldn't agree--
+
+**Jon Calhoun:** 51% of people. You get to choose which people.
+
+**Peter Bourgon:** To Go people, or to programmers in general?
+
+**Mat Ryer:** I don't know. Any.
+
+**Peter Bourgon:** Who am I offending? \[laughter\]
+
+**Roberto Clapis:** You can go this way - you set the group, and then you say the opinion. You say "This is unpopular in group -", and then you say it.
+
+**Mat Ryer:** Yeah, it's like targeted.
+
+**Roberto Clapis:** Right.
+
+**Peter Bourgon:** I didn't know this was a tradition actually, so I wish I had prepared a little bit better, because I'm sure I have lots of really offensive beliefs in here somewhere...
+
+**Roberto Clapis:** \[laughs\] So maybe it's better that you didn't.
+
+**Peter Bourgon:** No, we'll drag up something, and I'll try to keep it pithy, in a tweet format. This is maybe a little bit easy, but that's fine, we'll go with it... I think that in almost every case, if you're choosing whether to bias in favor of more work for you as a programmer, if it makes the reader of your code do less work, in every case you should make that bias.
+
+**Mat Ryer:** \[01:04:12.18\] Optimize for read.
+
+**Peter Bourgon:** Yeah. And the metric of how long it takes you to type code into your editor is approximately never worth tracking. It's just completely irrelevant. Anything that is like an optimization for that metric is incorrect, 100% of the time.
+
+**Mat Ryer:** Great.
+
+**Jon Calhoun:** But what if I'm writing code that only I'll read, and I'm very sadistic?
+
+**Roberto Clapis:** \[laughs\]
+
+**Jon Calhoun:** Like if I'm a masochist, or something. I don't know if I used the right word, but...
+
+**Peter Bourgon:** Don't bring us into your \[unintelligible 01:04:38.04\]
+
+**Mat Ryer:** What Jon does in the privacy of his own dungeon is his own business; in his own code dungeon.
+
+**Jon Calhoun:** I write this code thinking "I'm gonna have a real treat for myself later, when I try to read this..." \[laughter\]
+
+**Mat Ryer:** Well, that's usually how we really learn these things, is because we have actually done that, and we hate ourselves for it. So then we think "Oh, let's not do that again."
+
+**Jon Calhoun:** And then you do it again.
+
+**Mat Ryer:** Yeah.
+
+**Roberto Clapis:** Once I tried to re-read some code that deeply used map filter reduce chains, and I decided that it would have been easier to rewrite, trying to remember what it was trying to do, rather than understand it. Don't do that to yourself, or to anyone. Just don't do that kind of stuff.
+
+**Mat Ryer:** Yeah, that's interesting, because actually, if it's small enough, I quite like that. I like that I can actually just rewrite it. Because usually, then it's better, because some time has passed since I wrote it; and I didn't know anything about it when I wrote it.
+
+**Roberto Clapis:** Right. You write the code right on the third time, right?
+
+**Mat Ryer:** Yeah, if you're lucky.
+
+**Roberto Clapis:** \[laughs\] Right. I'm always thinking about Radio Sunshine in this podcast.
+
+**Jon Calhoun:** I think one of the other things to keep in mind there is just the fact that if you're gonna rewrite it, at least if it's for yourself sometimes - I view it as "I'll rewrite it better later, but if I never actually get to rewriting it, I save myself some time, because it didn't really matter; I'm not looking at this code." So there are cases where I'll just be like "I'll refactor this later", but I think I consciously make that decision. It's not like "Oh, this took me less keystrokes." It was like "Look, this was just me getting the first version done."
+
+**Mat Ryer:** Yeah. Sometimes if it's only gonna live for a little while, you just want to do one thing on your machine, and you need to just crunch through some data, and then you're gonna save it somewhere, or something like that. It's a one-off, and it doesn't matter. I like to still do it as exercise, to write the cleanest code I can.
+
+**Roberto Clapis:** Yeah, also because these things grow, don't they? Let's say that you start with a small Perl script that processes some data in one line. Then after a while you're like "I need to make this scale, or use it for more than one thing", and you have to do it again.
+
+**Mat Ryer:** Yeah, it happens all the time. Someone wrote one script to scrape some university website, and that thing's grown...
+
+**Peter Bourgon:** And now it's Facebook.
+
+**Mat Ryer:** ... and now it's Facebook.
+
+**Peter Bourgon:** Yeah, exactly. This is an interesting thing that we probably don't talk about nearly enough, which is like -- we're saying all these things like "You should do this. We don't like that. This is a good idiom. This is a bad idiom." But the context of all this stuff is maybe unstated. It's code you're writing with other people in an organization. None of this really applies if you're just writing some code that generates a fractal, or some thing. Do whatever you want. We're talking in the context of like how you can be a good citizen in an ecosystem. So maybe it's worth making that point a bit more explicit.
+
+**Mat Ryer:** Yeah. I think it's a good point, because context actually applies -- it changes lots of things, too; lots of advice changes with context. if you're just a tiny team of two people, you might well behave very differently to even a team of eight, and certainly a team of 50 or 100 people. You're necessarily gonna have different attitudes, different problems, different ways of interacting.
+
+**Peter Bourgon:** \[01:07:58.09\] And this is interesting too, because Go itself is biased for programming in the large, at large organizations with lots of engineers who come and go, on big teams... And that is explicitly what it's targeting, right?
+
+**Mat Ryer:** Yeah, yeah. Because the readability -- the focus is that optimized for read. So yeah, I think that's why lots of people learn it quite quickly - it's because it's smaller, it's optimized for read, and it's kind of easy to get going with the tools. Once you download it and install it, it's kind of like everything you need and a bit more. So yeah, I think it's great for that. Roberto, what were you gonna say?
+
+**Roberto Clapis:** I'm just saying that this also influences the community. There are communities in which the smartest code that you can write is a beautiful piece of code... And I'm totally fine with that. And there are communities like the Go one, in which if your code is smarter, it's too smart; you might want to consider dumbing it down a bit, because it's going to be hard to debug, and hard for newcomers to start understanding whatever you're working with.
+
+So it's not just that the language influenced the libraries and the ecosystem, it's also that now the ecosystem influences the way we write the language, to keep it simple and maintain that tradition... Which is something I like.
+
+**Mat Ryer:** Yeah. Well, I'm afraid that's all the time we have. I'm very sorry if we didn't get around to your question... It probably just means it wasn't interesting enough... No, we just didn't have time. There were so many great questions, and thank you so much. We will do another one of these again, because I've learned loads, and I'm sure the people have, too.
+
+Thank you so much to our guests, Peter Bourgon, Roberto Clapis, and Jon Calhoun was also here... And I was here too, obviously. See you next time!
+
+**Outro:** \[01:09:51.10\]
+
+**Jon Calhoun:** So if anyone's wondering why Mat has a team with just one other person, it's probably comments like that... \[laughter\]
+
+**Mat Ryer:** I'm the nice one.
+
+**Peter Bourgon:** That, and many other reasons.
+
+**Mat Ryer:** \[laughs\]
+
+**Jon Calhoun:** You're the nice one...?
+
+**Peter Bourgon:** I just thought of a much better controversial opinion, unfortunately...
+
+**Mat Ryer:** No, no, you can do it. We can edit it in.
+
+**Jon Calhoun:** We're still live, so yeah, it can get edited in, or whatever.
+
+**Mat Ryer:** We're still recording.
+
+**Peter Bourgon:** Well, maybe not worth it, but... Yeah, I've spent a lot of time thinking about the semantic import versioning rule in modules, and I'm increasingly convinced it's just a complete design error, and models are fundamentally broken in this way... But no one seems to agree with me, so maybe that's a deeply unpopular opinion.
+
+**Mat Ryer:** We did have a question about which Go version you should choose in your go.mod file. What do you think about that?
+
+**Peter Bourgon:** Well, it doesn't matter. At the moment it doesn't really have an effect. In the future it might, but at the moment it just doesn't really matter. If the question is like "Which Go version we should use in general?", it's the latest stable release, always. Definitely.
+
+**Roberto Clapis:** Yes. And I've had a problem with that... Because if you choose the latest stable release, a library for example relies on the SQL package; in Go 1.15 a function was just added, and my library is a wrapper around the SQL package, so I had to write two versions, basically; one that was more backward-compatible, and one that works with the new one and exposes that function.
+
+So for programs it's easy. Just the latest. For libraries it's not that easy, in my opinion.
+
+**Mat Ryer:** Hm. Good stuff.
+
+**Roberto Clapis:** I wish I could say "The latest that compiles with this code."
+
+**Jon Calhoun:** It's hard for me to talk about the versioning stuff, because I'm just not in a situation where I run into the issues that some people have... My dependency tree is not that complicated, so as a result, it's like, what they have is fine. So my feedback on that is I don't have -- I can give you the just-somebody-throwing-something-together's feedback, but really, the complicated cases are gonna come from Kubernetes or some big project like that, not mine.
+
+**Peter Bourgon:** But do you have projects that have a reasonable rate of change?
+
+**Jon Calhoun:** Yes... But it's probably not library changes as much. It's whenever I decide to go back and change them, I guess. For the most part, once something is in there and I used it for the things I'm using it for, it's pretty good.
+
+**Peter Bourgon:** \[01:14:01.10\] Okay, so you don't tend to have things that evolve over time, lots of contributors, or...
+
+**Jon Calhoun:** No, right now I don't have many projects like that, no.
+
+**Peter Bourgon:** Yeah, okay.
+
+**Mat Ryer:** Yeah. That would be different, wouldn't it? Especially like when you release - you suddenly have to release the tag versions, and things...
+
+**Peter Bourgon:** Right... Which semantic import versioning makes a much bigger deal out of than in any other ecosystem, and that's like the entry door to my Pandora's box of complaints.
+
+**Mat Ryer:** Well, maybe we'll have to do another episode on that altogether, Peter.
+
+**Peter Bourgon:** I would love that so much... And if you could get Russ on, so I can understand his perspective a little bit better - even better. \[laughter\]
+
+**Roberto Clapis:** I think you should make episode on rants. You just bring people that have complaints, and you throw them against someone that caused that pain.
+
+**Mat Ryer:** Like face your -- yeah, it's grim... But yeah. Just this. You get to do just this. Yeah. I don't know, because we're quite nice, aren't we? We're all a bit too nice for that. It sounds like a roast. Maybe we should just do a Gopher roast, where we just get Dave Cheney and we just write loads of horrible jokes about him.
+
+**Peter Bourgon:** Sarah Silverman probably, too. Yeah.
+
+**Mat Ryer:** Yeah, get Sarah Silverman up to do a bit; Jeff Ross will be in there, obviously... Yeah? No? Okay, won't do it then. I didn't get the support I was hoping for on that one... Grill a Gopher. Leah Anthony on Slack said Grill a Gopher.
+
+**Roberto Clapis:** Grill a Gopher. If it is voluntary, like the person agrees to that, why not? We might end up learning something.
+
+**Mat Ryer:** Okay, yeah. Fair enough. Grill a Gopher. I don't know... I don't think it's right for our community, is it?
+
+**Roberto Clapis:** Also, a lot of people in the community are vegetarian. It doesn't sound good. A gopher is an animal still.
+
+**Mat Ryer:** You can still grow vegetables though. Gopher like a person though, don't worry. It's a person. It's not an actual gopher animal.
+
+**Roberto Clapis:** Right. Yeah, I got that.
+
+**Mat Ryer:** And if they're prepared --
+
+**Jon Calhoun:** If the community guidelines don't say anything now, all of a sudden there'd be an addition... Like, "We had to add no grilling gophers."
+
+**Mat Ryer:** I haven't seen it say you can't grill gophers in the code of conduct, personally. And I read them every day.
+
+**Roberto Clapis:** \[laughs\] Just to be sure...?
+
+**Mat Ryer:** Yeah, just to be sure.
+
+**Jon Calhoun:** In other news, we have an opening for a Go Time panelist...
+
+**Roberto Clapis:** \[laughs\]
+
+**Jon Calhoun:** ...because Mat is gonna lose his job.
+
+**Peter Bourgon:** Okay, all joke aside, Mat - have you ever hunted and killed a human being, the most dangerous game?
+
+**Mat Ryer:** Oh, no.
+
+**Peter Bourgon:** Oh, you really should. It's so thrilling!
+
+**Roberto Clapis:** How did it get so dark so soon?
+
+**Jon Calhoun:** Like, it's time to go off live...
+
+**Roberto Clapis:** It was so sudden!
+
+**Peter Bourgon:** Oh, are we still live?
+
+**Jon Calhoun:** We are. Sorry.
+
+**Mat Ryer:** It's alright. It's obviously a joke.
+
+**Jon Calhoun:** Obviously, it' s a joke.
+
+**Mat Ryer:** I don't think anyone listening would not figure that one was a joke. But the answer, Peter, is "No. Not yet."
+
+**Peter Bourgon:** Okay, very good. Very good. \[unintelligible 01:16:59.23\] It's probably more thrilling that way, too.
+
+**Mat Ryer:** \[laughs\] We could have a gopher purge, where we just delete all the libraries that we don't like. Just one day where we can forget the consequences and just go around and delete all the libraries.
+
+**Peter Bourgon:** There's a fun game - if you could delete one library from existence...
+
+**Mat Ryer:** That is quite a fun game... \[01:17:22.00\]
+
+**Roberto Clapis:** One library.
+
+**Mat Ryer:** A Go package.
+
+**Peter Bourgon:** Yeah. And force everyone who's using it to use something else.
+
+**Mat Ryer:** Hm...
+
+**Roberto Clapis:** \[unintelligible 01:17:29.28\]
+
+**Jon Calhoun:** I feel like Peter would just be like "We're gonna delete database SQL" just so we can fix that init function. \[laughter\]
+
+**Roberto Clapis:** Or that has an alternative, or would you build the alternative?
+
+**Peter Bourgon:** It doesn't matter.
+
+**Jon Calhoun:** I assumed they'd have to rebuild something...
+
+**Mat Ryer:** No one's got to.
+
+**Roberto Clapis:** Oh, it doesn't matter. HTTP. HTTP, every day.
+
+**Peter Bourgon:** What's wrong with it?
+
+**Roberto Clapis:** That is old. \[unintelligible 01:17:53.18\] done in the worst possible way, too many default things, recovering panics... Too many things. It's like, I have a document.
+
+**Peter Bourgon:** Right. Designers... The main thing I think for me is allocations, of course.
+
+**Roberto Clapis:** Yeah, also that, but -- I care about performance when it's the last concern... And I have a 24 pages document with the other concerns to be addressed before performance.
+
+**Mat Ryer:** That's too many pages, mate.
+
+**Peter Bourgon:** So Brad, that's a shot across the bow from Roberto.
+
+**Mat Ryer:** Brad Fitzpatrick is on next week's show, so we'll absolutely be playing a clip of this, and getting his reaction.
+
+**Roberto Clapis:** He would agree. He started redesigning the HTTP package a couple of years ago. Most of the stuff in my document comes from his.
+
+**Mat Ryer:** It was written and then it evolved at a time kind of before we'd been doing Go for very long... So you can see, in the standard library there are lots of examples of things that really don't look very Go-like at all, and I think there's probably just stuff like that going around, isn't there, as well?
+
+**Jon Calhoun:** I think this is a good thing for beginners too, to realize that even these people they look up to as amazing developers still look back on things they created and say "This could be improved drastically, now that I know more..."
+
+**Mat Ryer:** And it's just in use by maybe millions of people; certainly hundreds of thousands of people are using these things, and yet they still feel like that. That is quite a good lesson, because software is never really finished, is it? That must be quite encouraging for a junior developer to hear, I would hope.
+
+**Peter Bourgon:** Yeah, there's no hope. You're gonna feel awful about the things you write, in perpetuity. \[laughter\]

Enterprise Go_transcript.txt

@@ -0,0 +1,409 @@
+**Johnny Boursiquot:** Hello, and welcome to this episode of Go Time. I'm your host, Johnny Boursiquot. Joining me is Mr. Mat Ryer. How are you, Mat?
+
+**Mat Ryer:** Hello! I'm good, thanks. How are you?
+
+**Johnny Boursiquot:** I'm okay, I'm feeling quite chipper today. Joining us is a special guest; in some circles he doesn't need any introduction, but we're gonna give him one anyway.
+
+**Mat Ryer:** Not in this circle...
+
+**Johnny Boursiquot:** Not in this circle... \[laughs\] So our guest today is Mr. Bryan Liles. Bryan is currently a senior staff engineer at VMware, where he actually runs multiple projects, including Octant, which you might have heard about quite recently. He actually unveiled it on Twitter, for all to see. I think the project has been getting some popularity and getting some contributions from the broader Kubernetes community, so that's awesome; we'll probably touch on that too a little bit.
+
+Bryan is known to talk on a number of different things, from machine learning, to developer health, to programming techniques... When I first came across Bryan, back in our Ruby days, where he was talking about "Test all the effin' time"... \[laughter\] Yeah, so Bryan has been around quite a while, and I've had the pleasure of knowing him for a few years as well. For those of you who don't know about Bryan, do check him out on Twitter; he's hilarious, and he'll get you to think with some of his tweets as well. Bryan, welcome to the show!
+
+**Bryan Liles:** Well, that was a great introduction, Johnny. If I could blush... If I was not brown, I would be blushing right now.
+
+**Mat Ryer:** I can blush, but I've never needed to, because Johnny has never been nice to me. \[laughter\]
+
+**Johnny Boursiquot:** No, the way you should look at it is if I'm not in the act of being mean to you, that means I'm being nice to you.
+
+**Mat Ryer:** Oh, so that's just like a sensible default.
+
+**Johnny Boursiquot:** \[04:01\] Yeah, you should look at it like that. So today we have Bryan on the show because we've been kind of stumped on something. We're trying to figure out what Go in the enterprise means. What does that even mean, "Go in the enterprise"? If you google it, you'll find a bunch of different opinions, and things... Some would say Go is even built for the enterprise; a lot of people who are trying to solve enterprise "problems"... But I don't think I've ever written Go differently, whenever I've written Go, I'm like "Oh, I'm writing for the enterprise now. A bunch of different things that I'm not doing when I'm writing for the non-enterprise." Like, what does that even mean? Bryan, please, enlighten us.
+
+**Bryan Liles:** Alright, so... I don't know, actually, what Go in the enterprise means. Johnny said "Come up with a topic, Bryan", and I went over all the other topics and I said "Well, no one's ever talked about this. We'll just explore it together." But I will say that now I've worked in two enterprises where Go was not a foreign language. Where I work at VMware right now, we have multiple business units using Go. When I was at Capital One - I've been there away long enough, so I can say this now - things that your credit card transactions ran through are now based on Go. And I think that's interesting.
+
+The reason actually why I thought this was an interesting thing to talk about - whenever I fire up whatever I decide to write Ruby in in a day, and I write code, or I write for an open source project, you code in one way. But whenever you introduce this horrible thing called "people", and then "a lot of people" to your process, things change. So that's why when it's a -- and I knew I would have Johnny here, and I guessed I would have Mat here, so I just wanted to throw out some ideas that I've seen, and then we can talk about those types of things for a little bit.
+
+**Johnny Boursiquot:** Does enterprise mean big money, big corporations, big expenditures? Because personally, when I think of enterprise, that's what comes to mind; I'm thinking "Big corporations, big tech companies, maybe non-tech companies that have lots and lots of software, lots and lots of process, and people and getting anything done takes months or years... That's what comes to mind.
+
+**Bryan Liles:** Yeah, that's interesting - what is an enterprise? I don't know... I do think that once your company has multiple business units, because they're chasing revenue in multiple directions, they've reached out beyond that one thing that they were known for... So VMware started with VMs, now VMware does software-defined data centers - so it's networking, computes and storage... And now Kubernetes, and protecting office devices, and whatever hyperconverged infrastructure is, and a few other things - we're now an enterprise.
+
+And generally, what it is is that whenever we can totally separate our cloud-native revenue line from what we're doing in vSphere in your clusters. Maybe that's what it is. Or a more simpler explanation is that an enterprise is whenever your chain to the CEO goes through a director, a senior director, some kind of VP, and then their boss who's a VP, and then their boss who's a VP, and then maybe a C-level person. Then you're an enterprise, even if you only have one product.
+
+**Johnny Boursiquot:** Layers.
+
+**Mat Ryer:** I've seen five-team startups that has that kind of a hierarchy.
+
+**Johnny Boursiquot:** \[laughs\] That's unfortunate....
+
+**Bryan Liles:** And you know, it's funny, you laugh, but yeah... \[laughter\]
+
+**Mat Ryer:** \[08:00\] For me, security also comes up a lot with enterprises. It seems to be a thing that almost there's a suggestion that startups don't care about it being secure. It's only when it's an enterprise that they suddenly have all this concern.
+
+**Bryan Liles:** Well, here's a secret fact here - and I'm not speaking for my employer when I say this, I'm just speaking in general; Bryan, that guy - most companies don't care about security. They care about SOX compliance, and HIPAA compliance, and they care about regulators, and they care about people who can cost them money if they don't get it right. And this is not a negative one, all the great security people that I know and don't know. I think there's people out there doing a great job, but really, a lot of companies don't care about security as a thing, they care about it as a liability. And when it becomes a liability, then that's whenever they start investing lots of money and people into it.
+
+**Johnny Boursiquot:** Not to go down too deep of a security tangent here, but when I think of developing software, I like the boundaries that security provides. Once I know what the boundaries are, then I feel free to explore up to those boundaries. It's liberating in some way. If I know what the constraints are, if I know what security review is gonna entail... Because I know I can't just ship my software out there and assume that there are no holes in it, and assume that it's just gonna do the right thing all the time, and if somebody tries to crack it... Like, I'm not perfect, and I don't think any team out there is perfect in how they create their software.
+
+Security is not a switch you just flip. You can't buy it off the shelf and apply it to your product and you're done. It's a continuous process, it's something that mitigates... So I feel comfortable working within that boundary. Shouldn't that be how we view things like security, and compliance, and all these other things?
+
+**Bryan Liles:** Yes. And not to go super-deep into this - in a perfect world, yes, of course, it is liberating. But in the real world that we live in there's people and there's ulterior motives, and it is a little more complicated than that. But there are better parts of enterprise, like Go in the enterprise...
+
+I actually have a premise. My premise is this - we have lots of companies out there, whenever you say "enterprise development", what they think of first off is they think of Java. Really deep Java. And I say this as a company now where we own Spring Cloud platform. There are many enterprises that do literally everything on Spring Cloud platform. But I'm actually here to say that Go - what are we, almost 11 years in now? Go in many cases is a viable language and ecosystem for enterprises. And we actually just got here recently, but I think now is a good time, and that's why I wanted to actually think about and riff on with you all today.
+
+I have my first controversial item... And this is something we could do years ago, but I will throw it out there right now and see what you all think. It's actually the same concept that I was brought on Go Time to speak about a few years ago with Brian. Whenever you have a monorepo - I love that word - and you have the right amount of tooling for it, Go becomes very, very powerful... Because it's strongly-typed, and fast compile times, and it generates binaries; there's no need for a runtime. These three things right here are great for whenever you have to code with lots of other people. You can define not the soft APIs that we define in Python, JavaScript and Ruby, but we actually can get some firm, strongly-typed APIs.
+
+\[11:54\] Then, when you put this all in a monorepo and it's all in the same place, then what you get is that in a perfect world, of course, you get this thing where if I'm depending on a library that someone else is using, I can always make sure it's up to date. And yes, there are definitely arguments of why that might not happen or why it's hard, and I can do that.
+
+Then another thing which is actually super-important for many large enterprises now, especially when we're all trying to move to this cloud-native technology, so there's a lot of containers and things like Kubernetes, being able to generate a binary that I can actually run and not have a runtime - oh my gosh, this is amazing. So this is why I just wanna talk about Go in the enterprise.
+
+**Johnny Boursiquot:** That's a very good observation, because a lot of folks are gonna think "enterprise Java" when you say the word "enterprise" and "programming" and things like that. Back in my Java days even some of the product launches had the word "enterprise" in them. Like, you were doing enterprise JavaBeans, enterprise service bus, and this and that. The name alone implied that you were programming differently, or you had different concerns, or additional concerns that you had to worry about when you were doing enterprise versus non-enterprise Java development.
+
+Now, in Go, is there a material difference in how you program, say, your startup's business logic in Go, or your network layer, or whatever the case may be? Is there a material difference between how you do that, say, at a Google, at a Salesforce, at a VMware?
+
+**Bryan Liles:** You know, I don't think so. I think that there is this misconception that in large enterprises, large enterprise development teams aren't capable of producing code... So we give them Java, so we can give them all the pacifiers and all the hand-holding they need... But actually, after seeing this -- and I worked at a large bank, and after seeing what these teams were capable of... No, I don't see any difference. Most of them spent their day inside of IntelliJ. I mean, that's cool... I spend most of my day inside of Go land. Made by the same people. It's literally the same editing engine. It's the same thing.
+
+I think we need to remove those misconceptions, because that's all it is - I hate to use the term FUD (the whole fear, uncertainty and doubt), but that's all this is. We use Java because -- it's the old IBM saying, "No one got fired for buying IBM" and "Oh, we're using Java because everyone else is using Java." But now in 2020 - I think that's the year - there's multiple viable options out there. I'm talking about what Microsoft is doing, what .NET and this dude over there David Fowler - amazing stuff, what Go is doing...
+
+I only bring up this because a friend of mine now works on the Go team, Carlos Amedee. And if you've seen an email from him lately, he's shipping Go right now. And then there's other languages... I think it's time to not diversify, because diversification is bad whenever you're in a big place, but I think it's time that we can explore other big projects... Because look right now - is Kubernetes almost the largest project on GitHub right now? They're doing fine(ish)...
+
+**Mat Ryer:** Bryan, earlier you mentioned that now Go kind of got there; it's now at the enterprise. What had to happen? What was it lacking before that? And is it just a sort of maturity thing, and a community kind of adoption?
+
+**Bryan Liles:** Well, there's only one thing that I'm really thinking about, and that was sort of kind of fixed recently... So in 2015(ish) I wrote this blog post - it was pretty popular - when I was at DigitalOcean, about DigitalOcean moving to Go. I didn't do that work by myself (as a matter of fact, I tried to do none of it), but I was definitely the person who got everybody on board with moving Digital Ocean to a monorepo. And the reason we had to move to a monorepo the way that we did is because Go package management five years ago was horrible. Yeah, I could make stuff, but it was bad.
+
+\[16:12\] And then just until recently, with the introduction of vgo, and now that we have modules on Go - it took us a little bit of time to get used to it. And I won't say it's perfect, but I will say that it's more than good enough now that we can actually do dependency management in Go in a proper way.
+
+And then you think about the introduction of the GOPRIVATE environment variable, and local caching - okay, now my enterprise doesn't have to take down the world's infrastructure to actually be able to do modules properly. Even better yet, I can use [Athens](https://docs.gomods.io/) and now I can have a private proxy where I can actually do this. So the infrastructure is now getting there, and I really appreciate that. That's what I'm saying, that Go wasn't ready. I guess in the last year or so - we're just right there, right now; we're at the point where I'm like "Yeah, I could actually recommend this to you and not laugh and you turning your back."
+
+**Mat Ryer:** Do you think that the fact that Go is open source was something that held it back? In the past I've worked in some big companies, and this may not be as true - and hopefully isn't - but definitely there used to be this attitude of open source wasn't proper software, and if you want proper software, you have to pay for it. Do you think that held Go back at all, when it came to the enterprise?
+
+**Bryan Liles:** It did. And guess what - 50 years ago, black people had to sit at the back of the bus. Or no, it's more than that now; now I would say 60-70 years ago. Was it right then? No. Is it right now? No. This is the same thing. The reason why a lot of these things held true for so long is because someone influential said these things... And was it right? No. And actually, it's great now, because as we go into this new decade, we're seeing that there's a lot more diversity in development.
+
+Because of David Fowler that I mentioned earlier, from Microsoft, I actually follow .NET development. And just a few years ago, ".NET - wow, is that a great idea? I don't even wanna run Windows." And now people are successfully deploying it off of Windows, and there's some really cool tech in there, made by some really good people. We need to start looking in other directions.
+
+And also because, think about it -- are they ready to do Java 14? Look at what Java 14 is doing. They're getting to the point now where they're like "Uh-oh... The world is catching up. Maybe we need to do some great things" and some of the new language features coming in in the new version of Java are definitely -- you can definitely see that languages like Go had some impact on them. But you know, Java has had generics for forever, so I rest my case...
+
+**Mat Ryer:** Yeah, that's definitely true, the different kind of language features, and things... I've seen Swift also has a defer statement. It doesn't quite work the same, but the trouble is they tend to add onto what they already have with these other languages. One of the nice things about Go is that it is quite a tiny little language, and there aren't too many ways of doing one thing... Which turns out to be quite important, because taking choice away means "Oh, then it's kind of obvious what you're gonna use to solve a particular problem." What are the Java features -- do you know what sort of things they're adding to it?
+
+**Bryan Liles:** You know what, no. I didn't come prepared to talk about that.
+
+**Mat Ryer:** Yeah... Sorry, mate.
+
+**Bryan Liles:** But I do remember - and someone's gonna check me on this and tell me on Twitter, and I'm gonna tell you right now, I don't care... What we're thinking about is one of the biggest complaints with Java is the -xmx flag. You know, whenever you're actually saying how big this thing is going to be, and that Java needs a runtime. And I think actually as we move on in years, I think Java is gonna move past that. We're just gonna assume that the JRE is part of the linking, and "We're just gonna give you a binary that runs on your platform." I see them moving there, and I think that it's actually a pretty good idea. But years ago, I can definitely see why they had a JRE.
+
+\[20:18\] But you know what - it's Oracle, and I don't understand how they are doing their development, but I do know that there's definitely some super-smart people over there, so hopefully they're trending toward the right direction.
+
+**Johnny Boursiquot:** Speaking of enterprise... Oracle - you're almost at the pinnacle of what you can think of as enterprise... So with all these projects - and I'm talking about the projects that you don't have to go and pay for. I'm talking about what you were talking about earlier... For open source projects to succeed -- Go, for example, is an open source project. It still has a corporate benefactor, it still has a corporate entity behind it. From what I can see, be it Go, be it other languages with backing from corporations, it seems like those are the projects that are faring quite well nowadays, than the non-funded, the non-backed projects. Is that me, or is that something that you're noticing as well?
+
+**Bryan Liles:** Yeah. I'm gonna tell you all a secret - it takes money to develop stuff. And you're like "No, no, no... I developed this on my weekend." Alright, that's cool... But to have a large, sustainable project, it takes lots of money. I don't know how big the Go team is, but I know that from the first time I went to GopherCon till right now it's a huge difference. And \[unintelligible 00:21:36.11\] to have all these hugely popular open source software projects, the reason why they all come from companies is because they need a lot of money.
+
+This is also why you have groups like the Linux Foundation, CNCF under them, or to a lesser extent the Apache Foundation... Because it takes money to not only write the code - because if you think writing the code is the easy part, ha-ha-ha - but to take care of the governance, all the legal aspects... And who's gonna pay for your security review?
+
+People don't think about that. They think about "No, this is easy. I can just write this on the weekend." And you can, to a certain level. But with a project like Go - think about all the pieces of Go. There's the build infrastructure, there is low-level language things, there's people who are getting their PhD's to think about how to make a better defer (I'm just guessing, I don't really know). Or how do we actually do generics - you can't just code that up on a whim.
+
+And then there's all the other support things, like the module depth infrastructure that we have for Go. Someone has to take care of that. And then all the advocacy. Someone has to go out and talk about how we're using these things, and work with their internal partners at Google to make sure that they're using it in a certain way.
+
+When projects get to a certain size, we realize that it's not easy anymore. Kubernetes itself - just go look at their repository, the kubernetes/kubernetes repository on GitHub, and then look at all the work that goes into that. You can't do that without financial backing. And then I know that at one time Google was footing the bill for Kubernetes' backend. It was not cheap. I mean, Johnny, you and I aren't gonna chuck in a couple bucks and pay for that. No, no, no...
+
+**Break:** \[23:37\]
+
+**Johnny Boursiquot:** Maybe I'm jaded... I've been on this Earth for a few years, so I've seen a few things... So once you accept money, that means "they" (the proverbial they) can tell you what to do. Do you think that this model, based on what we're seeing out there right now, the examples of open source projects that are backed by some sort of corporate benefactor - do you think that model is in the long-term going to retain what works best for everybody, instead of just what works for the corporate benefactor? For example, I'm sure there are some who do not like the way the Go team does certain things with the Go language, right?
+
+**Bryan Liles:** They're all on Hacker News right now. \[laughter\]
+
+**Johnny Boursiquot:** Yeah, there's no shortage of opinions on things like Go modules, for example... And semantic import versioning is a hot topic these days, which makes a few people not happy with that. But to them, you may seem like "Okay, well, the Go team decided to do something that works against the broader community's interests. Something that may work well at Google, but not for everybody else." But you're always gonna have these things... So is it naive of me to think that the corporate benefactor is going to always look after the best interest of the entire community, and not just for things that work well for "it"?
+
+**Bryan Liles:** A corporate benefactor is \*never\* looking out for the community. No, no, no. Companies, CEOs have fiduciary responsibilities to make money for their shareholders... Or - you can go to jail, in the United States.
+
+**Mat Ryer:** What if you don't make enough money?
+
+**Bryan Liles:** Well, the board will get rid of you.
+
+**Mat Ryer:** But you mean there's like a legal duty though to look after shareholders?
+
+**Bryan Liles:** Yeah. Your job is to do things that make more money. This is why companies should not be considered people. They do not work on the same set of ethics. This is just how the law works in the United States. And that's fine... But think about it this way. Companies aren't looking at big open source projects where -- and this is just speculation, but Google looking at Go, Google and whoever else looking at Kubernetes... They're not doing this to be nice to the community; they're doing it because if you support this project properly, ultimately you have one of two things - you either have a better funnel of people who are coming to use things that you can charge money for, or you have better integrations with other parts of the industry. For example with Kubernetes.
+
+\[27:53\] It's actually like with Linux. Linux was a good operating system that was free ("free", air quotes) that allowed us to build and work on all sorts of different types of hardware. Think about that. We didn't have to pay Microsoft to do these things. Kubernetes is now taking that a level up, where now I can actually install Kubernetes - and I'm going to really gloss over this, because we're not selling Kubernetes right now... But I can install Kubernetes, and with that I can run workloads that span over multiple machines. It's making it easier. The industry is looking at that as actually a boon; it helps us sell this type of thing, it helps us do these things better. And I think that's great.
+
+Actually, last year at KubeCon I alluded to this in a talk - industry is a bunch of verticals. Think about this - your company, whatever they're doing, they're looking up and down at "How do we take from 0 to 100 to make as much money as we can?" But they realize that they're not by themselves, they're not islands, and the ecosystem is horizontal. Now, there's certain things that we need to worry about on the horizontal, whether it be our operating systems, or our platform or platforms like Kubernetes, or certain standards.
+
+Think about this - if we didn't have an SMTP standard, how in the world do we send mail? Or even better, if we didn't have our TCP, or any of the OSI stack; how would we actually communicate? So there are certain things that we realize as companies in the industry that we need to standardize on, and then we compete around the edges, or we bring another angle to it. And you know what - I don't think that's a problem.
+
+I think that if we try to judge companies and what they do as people - yeah, it looks like companies are jerks. But if you look at companies as doing what they're supposed to do to actually make money - yeah, I guess they're doing a good job. And here's the best part - we don't have to like any of it. That's the biggest thing that I think -- and here's a soapbox that I will stand on... Us as adults, we need to understand - and especially as a black dude, I can just tell you this - that a large part of the world sucks, and it wasn't made for me. And you know what I learned to do? I learned to maneuver within it. There's things that I cannot change without killing myself, literally. So guess what. Knowing that, now what do I do? And actually, you know what - it's crazy; all these developers out here should embrace that.
+
+We love constraints, and we actually work better in constraints. I've actually used these constraints society has given me to do better things in certain places. What we should do in our technologies, with our languages - we should be doing the same thing. And yeah, it sucks, but guess what - that's okay. Not everything is made for you. It's made for society, and we should look at how it's benefitting society as a whole. I know there's another whole huge capitalist conversation that needs to be had here, but guess what - that's not what Go Time is for.
+
+**Johnny Boursiquot:** \[laughs\] And we appreciate that. We do have a segment on this show called Unpopular Opinions, but I think you might have been dropping maybe a couple already, sprinkling a few...
+
+**Mat Ryer:** Yeah, maybe we should change it to Popular Opinions for Bryan. Maybe he can give us one of those at the end.
+
+**Johnny Boursiquot:** \[laughs\]
+
+**Bryan Liles:** I actually had another one too, so we can do that later.
+
+**Mat Ryer:** Okay, we'll do that.
+
+**Johnny Boursiquot:** Good, we'll reserve it.
+
+**Mat Ryer:** On that, about the enterprise playing nicely with communities, one of the biggest resistances I've seen in the past to people adopting open source, or contributing even to open source, is this kind of -- it's almost a blanket, default attitude that all of our software is our IP, and it's valuable, and we need to protect it. It's a kind of default position that companies take, often by people that don't necessarily understand the nuances in that. So that's part of that, I think, and I've seen this happen; it's definitely getting better.
+
+\[32:01\] Cory on the Slack channel mentioned Rails, and Ruby. And maybe Ruby did pave the way a little bit for Go. Ruby was this open source project, Ruby on Rails of course the framework that became popular kind of at grassroots level... And it took a long time to get into enterprises. Compare that to Go. Go has kind of gone almost -- well, it hasn't taken as long as it probably took Ruby... So I feel like maybe Rails did help us there. But I think the attitude shift that I've seen is suddenly people realize not everything's gonna be that valuable to the company. Some things, like you say, Bryan, you're better off collaborating with, because then you all get the benefit of that particular problem being solved... So I think it's a great one, but that's the attitude, I think, that we need to bear in mind.
+
+**Bryan Liles:** I have another metaphor for you. The other day I went and bought a sampler, an MPC ONE. It was not pricey. Go look it up; you're gonna be like "Bryan, you have too much money." And no... I just like having hobbies. I bought the sampler --
+
+**Mat Ryer:** Is it an audio sampler?
+
+**Bryan Liles:** An audio sampler. So I'm going to, in my spare time now, make hot beats. That's what I'm gonna do. You can call me Kanye Junior Junior. The less self-hating, more righteous beat maker.
+
+**Mat Ryer:** More righteous... Wow.
+
+**Bryan Liles:** More righteous. \[laughter\] So I bought this thing, this box. It's the size of 11x11, it has lots of buttons on it... And I went to go make a beat the other day, and realize I can't even figure out how to sample. Okay... So I go to YouTube, and I learn, and I'm watching this guy and I'm like "Holy crap. This is hard." But here's the thing - our software is this sampler. You make a software, and it's a sampler. What does it do? Well, until you can actually use it in a proper, novel way, it's not useful. I could make a song that you could hear on the radio; literally, I can make songs that you hear on the radio, off of this machine. But because I don't have the knowledge and the expertise and all the practice, I'm just over here, entertaining myself with loud noises.
+
+And I think it's the same thing with software. We approach software like "My software is my IP." I said "Unless it's encryption codes, or things that we can't export, or something that was really novel, like pagerank when it first came out, back in the '90, most of your software is not that serious, dude..." And I did say dude; I meant that. "And to tell you the truth, it's probably not very good." That's something that we need to realize. It's how we employ our software, how we are helping our users. That is just a tool. We just happen to be using software to help our users out, because they are using computers. It works that way.
+
+But just the ability to have source is not super-helpful. Or it's not a thing that should be a differentiator. And companies want to actually optimize for that. And the reason they do is because... It goes back to this whole thing -- I'm reading a book called "Good Strategy, Bad Strategy", and I've read a couple of strategy books... It's because we're not good at strategy. We don't understand how to actually get from here to there. We play goals "I should be stronger" and "I'm going to work out". Then we don't really follow through, and we don't have actionable plans to actually get stronger. And the same thing, companies do this all the time. Developers do this all the time. We're lazy. We're people.
+
+And I'm not going to try to argue any company's lawyer down, because I would lose, but I think that we can see by having something like Kubernetes or something like Go, or something like what Microsoft has done with .NET - isn't it all open source now? Think about that. It doesn't matter... Because guess what - we are architecting now at a level that is so high... Who cares what letters people are using? I want novel applications of this thing.
+
+\[36:08\] And then soon - maybe in 20 years - we might even not be thinking about software in the same way we think about software now, and this is all silly. But you know what - it keeps people entertained, and it definitely makes a lot of money, so that's why a lot of people are sticking to it. That's controversial...
+
+**Mat Ryer:** A little bit.
+
+**Bryan Liles:** And you know what, it's fine.
+
+**Mat Ryer:** I like it.
+
+**Johnny Boursiquot:** If you're a language geek -- admittedly, this is niche, right? If you're a language geek and you revel in the esoteric knowledge of how language does things, and constructs, and all these things, then yeah, maybe you're writing code, and maybe just the source alone is valuable to you. But again, very niche, and that's not gonna really be of value to anybody than yourself, right? So it's okay, it's a play, it's okay to experiment, but not everything we write as software developers is of value.
+
+We always think that by virtue of writing code we're giving this gift to humanity. Well, not exactly... It actually has to be useful beyond you. Just writing the code - that's just the start; that's where you begin. Can it be used and produce value for others?
+
+Every time I hear somebody say "Oh, Facebook and Twitter - that's just a stream. You just scroll down and auto-refresh. I can build that in a weekend." \[laughter\] Every time you hear that, you look at them with the side eye, like "A-ha... Sure." For some reason -- maybe it's programmer culture, coder culture; I don't know what it is... And usually from dudes. We have this sort of ego about us, about the ability to create. We call ourselves "makers" and "builders", because we think just the act of creating something makes it valuable. That's only the beginning.
+
+**Bryan Liles:** Yeah, and that's my unpopular opinion, by the way. I'll share that when it comes time.
+
+**Mat Ryer:** Okay. That's true also even down at the code level... I had a friend who was looking at Go, and they started to learn it and they sort of dropped it, because they were confused about arrays and slices... And I was kind of heartbroken, because actually you don't need to really know everything about how arrays and slices work in Go to be able to use them, to be useful.
+
+This sometimes sounds a little bit anti-intellectualist or something, and it's not that. I think the more you learn, the better, of course... But there is something about that getting useful and solving real problems for people. If that's your focus - not nothing else matters, but almost nothing else matters. I feel like that's a way to give yourself a best chance of doing something that's gonna stand the test of time, or be useful, or be used, or be successful, whatever it is you're doing.
+
+**Bryan Liles:** Yeah, definitely. You're exactly right. I think I have definitely indoctrinated you into my church of haters... \[laughter\] I was called a Kill-Joy... Now I'm a professional Joy-Stealer/Killer. And it's not that, it's just how I get through today. I realize that the world is not friendly, and our goal is to make it better for people that come behind us... And that's really all we can do. But we realize that you can't change the whole world. I tried boiling an ocean once, and guess what happened - nothing.
+
+**Mat Ryer:** Was it a digital ocean, or analog?
+
+**Bryan Liles:** Ha-ha... That was good.
+
+**Mat Ryer:** No, I mean, was it a DigitalOcean? Were you making a pun or not? That's what I'm asking.
+
+**Bryan Liles:** No, I wasn't. I mean, I definitely boiled things while I was there... \[laughter\] I am definitely not one of those seen and not heard people. You're gonna feel me if I'm here.
+
+**Mat Ryer:** \[laughter\] Well, good.
+
+**Break:** \[40:03\]
+
+**Johnny Boursiquot:** I'm curious if you've witnessed situations where Go didn't take in an enterprise/organization...
+
+**Bryan Liles:** You know, I have not seen that yet. I've seen places where Go didn't take over. At VMware I will guarantee that our ESXi - that's our hypervisor that runs your virtual machines on bare metal - is never gonna be written in Go... But I think in places where it has been applied at a higher level, where we are building APIs, and doing distributed computing, I've never seen it loose. But I've been hearing more rumbles about Rust lately, and I've been learning it... And I look at it like this - a programming language is a programming language is a programming language. And it's not what you say, it's how many people are listening.
+
+In my professional career I've touched over 20 languages. And we're doing Go now. We might not be doing it in the future... And guess what - that's fine.
+
+**Mat Ryer:** Yeah. This is a Go podcast though... That's like going on the Great British Baking Show and going "Yeah, an oven is an oven is an oven. Food's all food...", and you're gonna make a casserole today. \[laughter\]
+
+**Bryan Liles:** But. The best part about this being a podcast is that the majority of people listening to this are not live, so if they have ill feelings about this, I'll just be like "I don't know what you're talking about." \[laughter\] Or even better yet, "That wasn't me."
+
+**Mat Ryer:** Yes. Fake.
+
+**Johnny Boursiquot:** That's a different Bryan Liles than the one who's behind things like Octant... Speaking of which - let's talk about Octant. That is arguably an enterprise project, because it's solving problems that the enterprise have.
+
+**Bryan Liles:** No, I disagree.
+
+**Johnny Boursiquot:** You disagree...
+
+**Bryan Liles:** 100%.
+
+**Johnny Boursiquot:** Yes, tell me about that.
+
+**Mat Ryer:** It doesn't solve problems?
+
+**Bryan Liles:** No.
+
+**Johnny Boursiquot:** \[laughs\]
+
+**Bryan Liles:** First of all, shout-out to Johnny. I launched Octant to the world at Johnny's meetup in Baltimore last August. He was the first person who had seen that, outside of folks at VMware. But to go back even further, we started Octant at Heptio. We were way less than 100 people at that time. Or maybe around 100 people. And it wasn't enterprise software.
+
+The problem I have with software, or just tech in general - it's all too hard. Yeah, I can code in over 20 languages, and I have code in the Linux source -- well, probably not anymore; that's probably been dropped off. And I've done all these crazy things... But guess what - my kid hasn't. And I hate to say my mom, but I do say my mom, because she's particularly tech-adverse.
+
+We need to realize that if we are trying to bring up technologies - and in my case Kubernetes - we can't just say "Oh, go use the command line." \[laughter\] That doesn't work.
+
+**Mat Ryer:** But you're not saying that to your mother, right?
+
+**Bryan Liles:** No, my mom doesn't know what Kubernetes is, nor does she care. And she shouldn't care, because it's not in her sphere.
+
+**Mat Ryer:** I don't even care... I don't even care about it. \[laughter\]
+
+**Bryan Liles:** \[43:51\] And you shouldn't. If it's not in your sphere, you shouldn't care. And actually, it should get to the point where it doesn't really matter. But it was like, "How do I know what's going on with my workloads in Kubernetes?" That was the first premise. Then it kind of evolved into a dashboard, and people are like "There's a dashboard." I'm like, "Well, no. It's not a dashboard. It has a dashboard in it", and now we're evolving it in a couple different directions, where for a long time I was really resistant.
+
+So Octant runs as a Go app, and Angular, and TypeScript... But it's a Go app that runs on Windows, Mac and Linux just fine. We're moving it to two different directions. We're moving it to the cluster, running in the cluster, as a website, because I come to find out -- and here's a lesson for everyone out there... You have to build software people wanna use, and you have to meet them where they are. So we find, with our enterprise customer - and this is the enterprise feature - we wanna run it in cluster. That's cool, we'll make that happen.
+
+But for our other users, and for our small office, home office people, or people like me, who just want a tool that can talk to different clusters, we're also moving it to an Electron app as well. And I know people hate Electron, but go look at what else is out there, and then come back to me and say that you hate Electron. Electron is the best thing, if you look at everything else out there. And go look at what VS Code has done with Vim. It is possible to make a good app. If it's not great, guess what - that's someone else's fault; it's not Electron's fault every piece of the time.
+
+So we're moving this app, but we use Go, and the reason we use Go is first of all because the first client we could find for Kubernetes was in Go. But when you write an app that's small, write in whatever you want. But when the app gets bigger -- so Octant is super-complex. It probably has about 12-13 different domains, different things that it does... And we find that having a strongly-typed language in this case makes it easier for people to come step in. We have no weird duck typing; there's only three lines of reflect in the whole entire app, and I've figured out a way to get rid of those.
+
+It's easy for people to be able to hook up to something like IntelliJ or Go land. And now they can actually view this thing and they can navigate through it. There's lots of benefits there. But here's the crazy part - have you ever tried writing a web app in Go? Don't do that...
+
+**Johnny Boursiquot:** \[laughs\] It's painful.
+
+**Bryan Liles:** Don't do that. So Go just serves a website that actually is an Angular-TypeScript app, and we do it that way. So really it's go find the best tool for your problem. If you run around with the Go hammer for everything, you're just gonna make Go-size holes everywhere, and that's not gonna work. \[laughter\]
+
+**Mat Ryer:** That's brilliant. Now I want to buy a Go hammer. \[laughter\] But I actually remember the launch of Octant. I was following the Twitter storm when you announced it in the Baltimore meetup. I mean, I say storm... They had likes in the high ones, but that was --
+
+**Bryan Liles:** I hear what you are saying. You say that, but I look at it like this - it's success, to me. At VMware I have a team of people working full-time on software that I made. I won.
+
+**Johnny Boursiquot:** \[laughs\]
+
+**Mat Ryer:** Well, actually, what I love about it is it just solves a real problem. And the thing is that is a fundamental thing that a lot of projects miss, I think... Especially, we like to write packages; I love making packages and open sourcing them. I really love doing that, and I used to do it all the time. And the only reason I don't do it as often now is because that process can involve a lot of imagining things, and building hypothetical software, solving hypothetical problems.
+
+When you solve an actual real problem that you have, it's a whole different ball game, and everyone who's working on software should understand the why. Really make sure that you are solving a real problem for somebody and it hasn't just been imagined, I think.
+
+**Bryan Liles:** \[48:02\] Yeah, that's a real thing. Now, don't get me wrong, if you go look at my GitHub right now - I think I looked at it this morning - I have 255 projects that I've created over the life of GitHub... And that's not even everything I've created, because I think I have some GitLab stuff, too. I enjoy writing software. But every once in a while, you have to write software that is usable. It either solves your problem or solves someone else's problem. You can't just go out there and be like "I'm writing software" all the time... Unless you're independently wealthy or you just don't care. But if you wanna progress in this world, you need to actually write software that helps people.
+
+**Mat Ryer:** Yeah. I mean, that's what we all want to do really anyway. It's just easy to forget that. Or often it gets deferred, as well. It's like "Well, that's the product -- that's that team's responsibility... So I just take instructions, or something." And I just think it's never as good as when you understand the problem you're really solving yourself.
+
+**Bryan Liles:** Well, that's another problem that we have... You know, we're all over the map on this conversation, but - that attitude "Well, that's someone else's problem." No... No, no, no. I mean, it might be someone else's job to solve it, but if it's blocking you, it's your problem, too. And today especially, people I work with that I see - and this is like no complaint; it's actually great where I work.
+
+But I see this now, and I see it on social media, where people are like "Oh, it's not perfect for me", they throw their hands up. They throw the whole baby out. And you really can't do that. Villages are full of people, and if someone doesn't do their job, the village could fail, or we have to kill that person and find someone else.
+
+**Mat Ryer:** Whoa, whoa...
+
+**Johnny Boursiquot:** That's been dark... \[laughter\]
+
+**Mat Ryer:** That's quite a strong policy...
+
+**Bryan Liles:** It is, but we can't kill people anymore. And we don't wanna kill people at all, because that's wrong.
+
+**Mat Ryer:** Oh, alright...
+
+**Johnny Boursiquot:** Good, good, good.
+
+**Bryan Liles:** So we need to realize that no one's perfect, and we have to work with everyone around us to make that world we want to be in.
+
+**Mat Ryer:** Why did that sound so forced, Bryan? I feel like you've had a lawyer tell you before that you have to say "No, no, no. Remember, killing people is wrong."
+
+**Johnny Boursiquot:** \[laughs\]
+
+**Bryan Liles:** No, I mean -- if you sat down with me, I am anti a whole bunch of bad things. I don't like any of that stuff. So this is all metaphorical. The village doesn't even exist.
+
+**Mat Ryer:** Oh, you don't actually do any murders or anything.
+
+**Bryan Liles:** Oh, gosh, no.
+
+**Mat Ryer:** Okay, good. That's great.
+
+**Johnny Boursiquot:** Yeah, this would be a different podcast.
+
+**Bryan Liles:** Yeah...
+
+**Mat Ryer:** Or at least save it for the unpopular opinion bit.
+
+**Johnny Boursiquot:** Speaking of which - we're at time for unpopular opinions.
+
+**Bryan Liles:** Oh, my...
+
+**Mat Ryer:** This episode has absolutely flown, I think...
+
+**Johnny Boursiquot:** I know, right?
+
+**Bryan Liles:** Yeah. Welcome to my world of Random Bryan Thoughts... \[laughter\] Now you see what my Twitter is like.
+
+**Johnny Boursiquot:** \[laughs\] Alright, drop it like it's hot.
+
+**Jingle:** \[51:03\]
+
+**Bryan Liles:** Alright, my unpopular opinion. Here it comes. In my mind, the world owes you nothing. So if you go to a job and you're a beginning developer, and you say "Well, I've worked four years. I should be a senior." Nope. Nope, you should not be a senior. What the world owes you though is not blocking you from moving forward. And that's the difference.
+
+So my unpopular opinion is that you don't deserve anything. You should have to go work, and get, and earn everything you have. And it should be fair. Whenever I see people out there saying "Well, we deserve this..." No, you don't. People hate when I say that to them. But guess what - your life is great. You know why it is? Because you can see my tweets. \[laughter\] And you're like "What does that mean?"
+
+\[51:58\] Well, there's a whole part of this world that either can't because they're sick, can't because they can't afford to get here, or can't because they're just looking in other directions because they're so busy trying to actually get through life... And I think you're not one of those people.
+
+This is not basically saying that there's someone worse than us, but think about this - you don't deserve anything. Everything you have, you earned it. This is how you get rid of impostor syndrome. You know how you're here right now? Because you earned it. There is no impostor syndrome. What it is - you're here because you're supposed to be here. And that's it. Don't question it.
+
+**Johnny Boursiquot:** I can live with that.
+
+**Bryan Liles:** Hold on one second though...
+
+**Johnny Boursiquot:** Oh. Another one?
+
+**Bryan Liles:** No... In the peanut gallery somebody said that's not unpopular. Alright... So one more quick one. Here's what it is. Twitter fame, work fame - it's all crap. Maybe this is not unpopular either, but here's the thing. That famous person you know is only famous because they impressed another famous person. And people who are seeking that, trying to be famous, and trying to do this - basically, you're not even measuring up to your own levels, if that's what you're seeking. What you're actually seeking is someone who probably doesn't really care about your acceptance and you really have to stop that.
+
+So those are my unpopular opinions. And actually, here's the most unpopular opinion - and this is where we'll leave it... I hate to say this, but most likely I am smarter than you. And you might wanna debate it, and you might think "Oh, no, Bryan's wrong..." No, literally. I am probably smarter than you. So we can debate that, but you will be wrong, and I will leave it at that, definitely.
+
+**Johnny Boursiquot:** \[laughs\]
+
+**Mat Ryer:** That is a great ending... \[laughs\]
+
+**Johnny Boursiquot:** If there's rebuttals, they can come debate you on this show.
+
+**Mat Ryer:** Yeah.
+
+**Johnny Boursiquot:** It's like a mic drop, right there.
+
+**Bryan Liles:** Because I'm smarter than them. \[laughter\]
+
+**Johnny Boursiquot:** Oh, my goodness... Awesome, awesome. Well, Bryan, this has been an educational and entertaining show. Thank you so much for coming on the show. It's been a pleasure having you. Yeah, any parting words?
+
+**Bryan Liles:** Yes, I've got a parting word... Leave the world better than you've found it. There's always someone doing worse than you. You don't have to help everyone, but turn around and help that next person, and tell them the only way you can continue doing this is if they turn around and help that next person.
+
+So realize that the world gets better when we all work together to make it better. And it ain't about politics, it ain't about anything else. It's about people helping people be better people. That's it.
+
+**Mat Ryer:** I love that, that you don't have to help everybody, but just help the next person in the line. It's like some kind of kindness blockchain... Yeah, I like that one.
+
+**Johnny Boursiquot:** \[laughs\] Awesome. Well, listeners, we hope you've enjoyed the show. Please tune in next time on the next Go Time.
+
+**Break:** \[55:10\]
+
+**Mat Ryer:** Okay, yeah... I was thinking, actually, earlier - do you prefer to change code when you're refactoring something, to just change it? Or do you prefer to kind of rewrite it wholesale, that piece? Does anyone have any references?
+
+**Bryan Liles:** It depends... That's a really weird question, but it really depends. It depends on really how bad past Bryan was. I was looking at this code earlier - it was yesterday, or the day before - and I was trying to figure out how to refactor it, because I was like "We need to change this, because I need to add new features." And I was going through the call tree, and I usually just write it down; I don't even use the debug print stack trace. I just went through all the method calls to see what it was doing... And it was pretty much eight more calls than it needed to be. In that case, I am actually going to rewrite it. And just wholesale -- I actually said "No, this is done. I'm gonna rewrite the whole thing." But generally, if it's just a small piece and I'm either 1) confident I have test coverage, or 2) don't really care, I might just go in and say "You only live once."
+
+**Mat Ryer:** Just tweak it, hack it... Test coverage does enable that, doesn't it? With quite a lot of confidence you can sort of be quite bold. But I always find if I have to change something that already exists... You're right, it depends; if the design is changing a lot, it's different... But I like to just rewrite it. It's just like "Now I know what it's meant to be. I can do it again." But I wondered if that was just a preference thing, or if others do have a different thought.
+
+**Bryan Liles:** But I think it's a testament to 1) how well your tests are written, and 2) how well your abstraction is around what you're trying to do.
+
+Not to sound like an old guy, but really when it comes down to it, if it's hard to change, you've messed up. And that's fine. We're developers. They wouldn't give us Backspaces and Git history if we were gonna get it right the first time.
+
+**Mat Ryer:** Wait, there's a Backspace? I could have really use that. I've been just starting again when I make a mistake... \[laughter\] Yeah, Backspace. I like that it's called Backspace as well. It's like, it properly comes from old typewriters probably, doesn't it? Like underscore... Right? Backspace. Literally, it's just moving it back a space, I suppose.
+
+**Bryan Liles:** Yeah, we definitely dropped a good chance to give it a much better name...
+
+**Mat Ryer:** Yeah. And then Delete, of course... It's this strange deleting the other way. On Windows it was that, wasn't it?
+
+**Bryan Liles:** Yeah, I don't know how it works. I have a Delete key on my keyboard... I don't think I ever use it. No, I do. I use it because my keyboard can control Spotify, and I think that's the Stop command if i use it with the Fn key, so...
+
+**Mat Ryer:** I thought you were gonna say you could delete sounds, or something. Anything that you just don't like - gone. That'd be good, wouldn't it...?
+
+**Johnny Boursiquot:** Alright, I'm ready if you are...
+
+**Mat Ryer:** Yes, I believe so. This is just a pre-show. So if the pre-show is that good, then the rest of it...
+
+**Johnny Boursiquot:** It's gonna be, you know...
+
+**Mat Ryer:** It's gonna be great. Unless we've peaked.
+
+**Johnny Boursiquot:** No, not yet. We haven't peaked yet.
+
+**Mat Ryer:** Okay, good.
+
+**Johnny Boursiquot:** We haven't jumped the shark, as they say...
+
+**Bryan Liles:** I mean, this is my second time on Go Time, so... Yeah, maybe it has jumped the shark. If you're back to me... My gosh. There's so many interesting people out there. Yup. Maybe we're done. This is it, people.
+
+**Johnny Boursiquot:** \[laughs\] This is the last episode we didn't know about.
+
+**Bryan Liles:** Yeah, the last one. \[laughter\]
+
+**Johnny Boursiquot:** Alright, let me get in character here. Let me get my voice... I should have drunk some tea before this...
+
+**Mat Ryer:** You're not wearing your wig?
+
+**Johnny Boursiquot:** \[laughs\]
+
+**Mat Ryer:** He normally wears a podcasting wig.
+
+**Johnny Boursiquot:** Yeah, my podcast wig, and socks... \[laughs\]

Enterprise Go？_transcript.txt

@@ -0,0 +1,1075 @@
+[0.00 --> 7.12]  Or a more simpler explanation is that an enterprise is when whenever you're chained to the CEO,
+[7.50 --> 13.00]  goes through a director, senior director, some kind of VP, and then their boss who's a VP,
+[13.44 --> 18.36]  and then their boss who's a VP, and then maybe a C-level person, then you're enterprise.
+[18.84 --> 20.32]  Even if you only have one product.
+[21.18 --> 21.70]  Layers.
+[22.12 --> 25.66]  I've seen five team startups that has that kind of a hierarchy.
+[27.98 --> 28.88]  That's unfortunate.
+[28.88 --> 32.36]  And you know, it's funny, you laugh, but yeah.
+[36.40 --> 39.24]  Bandwidth for Changelog is provided by Fastly.
+[39.62 --> 41.50]  Learn more at Fastly.com.
+[41.74 --> 44.82]  We move fast and fix things here at Changelog because of Rollbar.
+[44.96 --> 46.64]  Check them out at Rollbar.com.
+[46.88 --> 49.06]  And we're hosted on Linode Cloud servers.
+[49.40 --> 51.40]  Head to Linode.com slash Changelog.
+[54.10 --> 56.70]  This episode is brought to you by DigitalOcean.
+[56.70 --> 61.80]  DigitalOcean's developer cloud makes it simple to launch in the cloud and scale up as you grow.
+[62.18 --> 69.18]  They have an intuitive control panel, predictable pricing, team accounts, worldwide availability with a 99.99 uptime SLA,
+[70.02 --> 73.48]  and 24-7, 365 world-class support to back that up.
+[73.74 --> 79.20]  DigitalOcean makes it easy to deploy, scale, store, secure, and monitor your cloud environments.
+[79.20 --> 83.04]  Head to do.co slash Changelog to get started with a $100 credit.
+[83.40 --> 85.50]  Again, do.co slash Changelog.
+[97.88 --> 98.80]  Let's do it.
+[99.38 --> 100.42]  It's go time.
+[100.42 --> 105.88]  Welcome to Go Time, your source for diverse discussions from around the Go community.
+[106.22 --> 107.74]  One quick note before we get started.
+[108.12 --> 113.82]  This conversation was recorded back on March 10th, which was less than a month ago, but feels like a lifetime now.
+[114.00 --> 118.62]  We rushed out the Working From Home episode since it was of the moment, and this one is more evergreen.
+[119.04 --> 123.28]  So if you're wondering why there's no talk of coronavirus and global pandemics, that's why.
+[123.58 --> 125.08]  Okay, here we go.
+[125.08 --> 135.92]  Hello, and welcome to this episode of Go Time.
+[136.14 --> 137.66]  I am your host, Johnny Borsico.
+[138.26 --> 140.12]  Joining me is Mr. Matt Ryer.
+[140.18 --> 140.66]  How are you, Matt?
+[141.06 --> 141.42]  Hello.
+[141.64 --> 142.28]  I'm good, thanks.
+[142.32 --> 142.70]  How are you?
+[143.58 --> 144.20]  I'm okay.
+[144.30 --> 146.48]  I'm feeling quite chipper today.
+[147.40 --> 149.28]  Joining us is a special guest.
+[149.86 --> 154.20]  In some circles, he doesn't need any introduction, but we're going to give him one anyway.
+[154.20 --> 155.76]  Not in this circle.
+[156.22 --> 157.48]  Not in this circle.
+[159.20 --> 162.32]  So our guest today is Mr. Brian Lyles.
+[162.60 --> 171.20]  Brian is currently a senior staff engineer at VMware, where he actually runs multiple projects, including Octent, which you might have heard about quite recently.
+[171.40 --> 173.96]  He actually unveiled it on Twitter for all to see.
+[174.44 --> 180.48]  And I think the project has been getting some popularity and getting some contributions from the broader Kubernetes community.
+[180.64 --> 181.30]  So that's awesome.
+[181.44 --> 182.96]  So we'll probably touch on that, too, a little bit.
+[182.96 --> 191.58]  But Brian is known to talk on a number of different things, from machine learning to developer health to programming techniques.
+[192.58 --> 199.62]  I first came across Brian back in our Ruby days, where he was talking about tests all the effing time.
+[199.62 --> 201.94]  So, yeah.
+[202.08 --> 206.32]  So Brian's been around quite a while, and I've had the pleasure of knowing him for a few years as well.
+[206.82 --> 209.64]  And for those of you who don't know about Brian, do check him out on Twitter.
+[210.40 --> 214.42]  He's hilarious, and he'll get you to think with some of the tweets as well.
+[214.74 --> 216.54]  So, Brian, thank you, and welcome to the show.
+[217.42 --> 217.98]  Oh, wow.
+[218.12 --> 220.52]  That was a great introduction, Johnny.
+[221.00 --> 221.28]  Yeah.
+[221.28 --> 225.24]  If I could blush, if I was not brown, I would be blushing right now.
+[226.58 --> 230.42]  I can blush, but I've never needed to because Johnny's never been nice to me.
+[231.80 --> 232.80]  That's not so far.
+[234.42 --> 239.38]  No, the way you should look at it is if I'm not in the act of being mean to you, that means I'm being nice to you.
+[239.64 --> 241.44]  Oh, just like a sensible default.
+[241.44 --> 243.40]  Yeah, yeah, I should look at that.
+[243.58 --> 244.26]  Yeah, yeah.
+[244.44 --> 251.98]  So today, we have Brian on the show because we've been kind of stomped on something.
+[252.22 --> 258.36]  We're trying to figure out what Go in the Enterprise means.
+[258.58 --> 262.16]  What does that even mean, Go in the Enterprise?
+[262.44 --> 266.86]  If you Google it, you'll find a bunch of different opinions and things.
+[266.86 --> 274.62]  Some would say Go was even built for the Enterprise, about the people who are trying to solve Enterprise, you know, quote unquote problems.
+[275.00 --> 277.42]  But I don't think I've ever written Go differently.
+[277.58 --> 280.02]  Whenever I've written Go, I'm like, oh, I'm writing for the Enterprise now.
+[280.14 --> 283.52]  A bunch of different things that I'm not doing when I'm writing for non-Enterprise.
+[283.72 --> 285.12]  Like, what does that even mean?
+[285.24 --> 286.88]  Brian, please enlighten us.
+[287.62 --> 287.90]  All right.
+[288.08 --> 291.96]  So, I don't know, actually, what Go in the Enterprise means.
+[292.38 --> 294.46]  Johnny said, come up with a topic, Brian.
+[294.46 --> 299.26]  And I went over all the other topics and I said, well, no one's ever talked about this.
+[299.68 --> 300.80]  We'll just explore it together.
+[301.42 --> 310.16]  But I will say that I've worked in, now, I've worked in two enterprises where Go was not a foreign language.
+[310.30 --> 314.92]  Where I work at VMware right now, we have multiple business units using Go.
+[315.16 --> 320.96]  And then when I was at Capital One, yeah, actually, yeah, I've been there way long enough so I can say this now.
+[320.96 --> 327.18]  Things that your credit card transactions run through are now based on Go.
+[327.66 --> 328.72]  And I think that's interesting.
+[329.50 --> 338.62]  And the reason why, actually, why I thought this was an interesting thing to talk about, because whenever I fire up whatever I decide to write Ruby in in a day,
+[338.62 --> 343.40]  and I write code or I write for an open source project, you code in one way.
+[343.40 --> 351.86]  But whenever you introduce this horrible thing called people, and then a lot of people to your process, things change.
+[352.82 --> 357.82]  So, that's why I wanted to, and I knew I would have Johnny here, and I guess I would have Matt here.
+[357.82 --> 365.58]  So, I just wanted to throw out some ideas that I've seen, and then we could talk about those types of things for a little bit.
+[366.08 --> 368.94]  Does enterprise mean big money or something, right?
+[368.98 --> 372.96]  Big money, big corporations, big expenditures.
+[373.46 --> 376.72]  Because personally, when I think of enterprise, that's what comes to mind.
+[376.72 --> 384.08]  I'm like thinking big corporations, big tech companies, maybe non-tech companies that have lots and lots of software, lots and lots of process and people,
+[384.28 --> 386.62]  and getting anything done takes months or years.
+[386.98 --> 387.74]  That's what comes to mind.
+[388.64 --> 391.20]  Yeah, you know, that's an interesting, what is an enterprise?
+[391.70 --> 392.20]  I don't know.
+[392.20 --> 402.64]  I do think that once your company has multiple business units, because they're chasing revenue in multiple directions,
+[402.96 --> 407.52]  they've reached out beyond that one thing that they were known for.
+[407.74 --> 409.64]  So, VMware started with VMs.
+[410.22 --> 413.74]  Now, VMware does software-defined data centers.
+[413.94 --> 417.42]  So, it's networking, compute, and storage.
+[417.42 --> 426.44]  And now, Kubernetes and protecting office devices and whatever hyper-converged infrastructure is and a few other things.
+[426.50 --> 427.48]  We're now our enterprise.
+[428.30 --> 433.04]  And generally, what it is is that whenever your revenue line is separate,
+[433.20 --> 443.18]  like we can totally separate our cloud native revenue line from what we're doing in vSphere and your clusters.
+[443.50 --> 445.26]  Maybe, but maybe that's what it is.
+[445.26 --> 454.08]  Or, a more simpler explanation is that an enterprise is when whenever your chain to the CEO goes through a director,
+[454.26 --> 460.18]  senior director, some kind of VP, and then their boss who's a VP, and then their boss who's a VP,
+[460.64 --> 465.62]  and then maybe a C-level person, then your enterprise, even if you only have one product.
+[466.48 --> 467.00]  Layers.
+[467.00 --> 470.98]  I've seen five team startups that has that kind of a hierarchy.
+[473.36 --> 474.18]  That's unfortunate.
+[474.46 --> 477.66]  You know, it's funny, you laugh, but yeah.
+[479.60 --> 480.00]  Yeah.
+[481.40 --> 486.92]  I also think, for me, security also comes up a lot, again, with enterprises.
+[486.92 --> 493.10]  It seems to be a thing that it's almost there's a suggestion that startups don't care about it being secure.
+[493.38 --> 497.48]  It's only when it's an enterprise do they suddenly have all this concern.
+[498.12 --> 500.42]  Oh, well, here's a secret fact here.
+[500.54 --> 502.90]  And I am not speaking for my employer when I say this.
+[503.14 --> 504.44]  I'm just speaking in general.
+[504.56 --> 505.32]  Brian, that guy.
+[506.58 --> 508.52]  Most companies don't care about security.
+[508.52 --> 514.28]  They care about SOX compliance and HIPAA compliance, and they care about regulators,
+[514.60 --> 517.94]  and they care about people who can cost them money if they don't get it right.
+[518.10 --> 522.72]  And this is not a negative on all the great security people that I know and don't know.
+[523.00 --> 525.32]  I think there's people out there doing a great job.
+[525.32 --> 530.58]  But really, a lot of companies, they don't care about security as a thing.
+[530.90 --> 532.42]  They care about it as a liability.
+[533.00 --> 538.00]  And when it becomes a liability, then that's whenever they start investing lots of money and people into it.
+[538.52 --> 545.94]  But wouldn't you say, not to go down too deep of a security tangent here, but when I think of developing software,
+[546.74 --> 551.08]  I like the boundaries that security provides.
+[551.56 --> 555.32]  Once I know what the boundaries are, then I feel free to explore up to that boundary.
+[555.50 --> 556.58]  It's liberating in some way.
+[556.72 --> 560.92]  If I know what the constraints are, if I know what security review is going to entail,
+[561.34 --> 566.66]  because I know I can't just ship my software out there and assume that there are no holes in it
+[566.66 --> 568.98]  and assume that it's just going to do the right thing all the time.
+[569.10 --> 571.60]  And if somebody tries to crack it, I'm not perfect.
+[571.78 --> 575.12]  I don't think any team out there is perfect in how they create their software.
+[575.42 --> 577.66]  Security is not a switch you just flip.
+[577.72 --> 580.20]  You can't buy it off the shelf and apply it to your product and you're done.
+[580.50 --> 582.02]  It's a continuous process.
+[582.24 --> 583.98]  It's something that mitigates.
+[584.24 --> 587.72]  So I feel comfortable working within that boundary.
+[587.72 --> 592.66]  Shouldn't that be how we view things like security and compliance and all these other things?
+[593.42 --> 593.66]  Yes.
+[593.86 --> 595.72]  And not to go super deep into this.
+[596.12 --> 597.92]  In a perfect world, yes, of course.
+[598.20 --> 599.28]  Actually, it is liberating.
+[599.80 --> 605.90]  But in the real world that we live in, there's people and there's ulterior motives.
+[606.52 --> 608.40]  And it is a little more complicated than that.
+[608.84 --> 613.40]  But there are better parts of enterprise, like go in the enterprise.
+[613.40 --> 616.50]  So I actually had a premise.
+[617.26 --> 618.96]  My premise is this.
+[619.58 --> 622.56]  We have lots of companies out there.
+[622.70 --> 631.04]  Whenever you say enterprise development, what they think of, first off, is they think of Java, like really deep Java.
+[631.44 --> 637.24]  And I say this as a company now where we own a Spring Cloud Platform.
+[637.24 --> 643.10]  There are many enterprises that do literally everything on Spring Cloud Platform.
+[643.96 --> 649.26]  But I'm actually here to say that Go, what are we, almost 11 years in now?
+[649.52 --> 655.70]  Go, in many cases, is a viable language and ecosystem for enterprises.
+[656.08 --> 658.48]  And I think we actually just got here recently.
+[658.96 --> 661.68]  But I think now is a good time.
+[661.76 --> 666.80]  And that's why I wanted to actually think about and riff on with you all today.
+[666.80 --> 669.54]  And I have my first controversial item.
+[670.34 --> 675.04]  And this is something we could do years ago, but I will throw it out there right now and see what you all think.
+[675.08 --> 681.20]  And it's actually the same concept that I was brought on GoTime to speak about a few years ago with Brian.
+[681.20 --> 696.76]  Whenever you have a monorepo, I love that word, and you have the right amount of tooling for it, Go becomes very, very powerful because it's strongly typed and fast compile times.
+[696.76 --> 698.76]  And it generates binaries.
+[698.76 --> 699.52]  And it generates binaries.
+[699.80 --> 700.92]  There's no need for a runtime.
+[701.40 --> 706.26]  These three things right here are great for whenever you have to code with lots of other people.
+[706.52 --> 715.40]  You can define not the soft APIs that we defined in Python, JavaScript, and Ruby, but we actually can get some firm, strongly typed APIs.
+[715.40 --> 731.96]  And then when you put this all in a monorepo and it's all in the same place, then what you get is that in a perfect world, of course, you get this thing where if I'm depending on a library that someone else is using, I can always make sure it's up to date.
+[732.92 --> 736.76]  And yes, there are definitely arguments of why that might not happen or why it's hard.
+[736.96 --> 737.90]  And I can do that.
+[737.90 --> 757.80]  And then another thing, which is actually super important, I think, for many large enterprises now, especially when we're all trying to move to this cloud native technology, so there's a lot of containers and things like Kubernetes, being able to generate a binary that I can actually run and not have a runtime, oh my gosh, this is amazing.
+[758.22 --> 761.30]  So this is why I just want to talk about Go in the enterprise.
+[761.30 --> 769.20]  That's a very good observation because a lot of folks are going to think enterprise Java, right?
+[769.26 --> 772.14]  When you say the word enterprise and programming and things like that.
+[772.44 --> 777.24]  Like back in my Java days, you know, even some of the product launches had the word enterprise in them.
+[777.32 --> 782.08]  Like you were doing, you know, enterprise Java beans, you know, enterprise service bus and this and that.
+[782.08 --> 788.48]  It's like, you know, the name alone, right, implied that you were programming differently, right?
+[788.48 --> 796.88]  Or you had different concerns or additional concerns that you had to worry about when you were doing enterprise versus non-enterprise sort of Java development.
+[797.48 --> 809.04]  Now, in Go, is there a material difference in how you program, say, your startup's business logic in Go or your network layer or whatever the case may be?
+[809.28 --> 814.26]  Is there a material difference between how you do that, say, at a Google, at a Salesforce, at a VMware?
+[815.44 --> 817.26]  You know, I don't think so.
+[817.26 --> 828.84]  I think that there is this misconception that in large enterprises, you have that large enterprise development teams aren't capable of producing code.
+[829.12 --> 836.00]  So we give them Java so we can give them all the pacifiers and all the handholding they need.
+[836.00 --> 840.10]  But actually, after seeing this, I mean, I worked at a large bank.
+[840.74 --> 845.72]  And after seeing what these teams were capable of, no, I don't see anything, any difference.
+[846.08 --> 848.52]  I mean, most of them spent their day inside of IntelliJ.
+[849.14 --> 849.90]  I mean, that's cool.
+[850.26 --> 854.36]  I spend most of my day inside of GoLand, made by the same people.
+[854.48 --> 855.94]  It's literally the same editing engine.
+[856.64 --> 857.78]  It's the same thing.
+[857.78 --> 862.86]  And I think we need to remove those misconceptions because that's all it is.
+[863.08 --> 866.46]  I hate to use the term FUD, the whole fear, uncertainty, and doubt.
+[866.96 --> 868.00]  But that's all this is.
+[868.36 --> 873.44]  We use Java because it's the old IBM saying, no one got fired for buying IBM.
+[873.94 --> 876.54]  And, oh, we're using Java because everyone else is using Java.
+[876.54 --> 882.96]  But now in 2020, I think that's the year, there's multiple viable options out there.
+[883.08 --> 887.48]  And I mean, I'm talking about what Microsoft is doing with .NET and this dude over there,
+[887.54 --> 889.86]  David Fowler, amazing stuff.
+[890.22 --> 891.22]  What Go is doing.
+[891.56 --> 896.26]  And I only bring up this because a friend of mine now works on the Go team, Carlos Amadee.
+[896.56 --> 901.42]  And if you've seen an email from him lately, he is actually, you know, he's shipping Go right now.
+[901.92 --> 902.98]  And then there's other languages.
+[902.98 --> 909.26]  I think it's time to not diversify because diversification is bad whenever you're in a big place.
+[909.38 --> 912.50]  But I think it's time that we can explore other big projects.
+[912.64 --> 918.80]  Because look right now, what do you think one of the, is Kubernetes almost the largest project on GitHub right now?
+[918.94 --> 920.52]  They're doing fine-ish.
+[921.48 --> 925.56]  Brian, earlier you mentioned that now Go's kind of got there.
+[925.66 --> 926.66]  It's now at the enterprise.
+[926.96 --> 928.10]  What had to happen?
+[928.28 --> 929.88]  What was it lacking before that?
+[929.88 --> 935.38]  And is it just a sort of maturity thing and community kind of adoption and things?
+[935.98 --> 936.16]  All right.
+[936.24 --> 939.10]  Well, I think there's only one thing that I'm really thinking about.
+[939.38 --> 941.88]  And that was sort of kind of fixed recently.
+[942.62 --> 946.50]  So in 2015-ish, I wrote this blog post.
+[946.56 --> 952.08]  It was pretty popular when I was at DigitalOcean about DigitalOcean moving to Go.
+[952.46 --> 954.70]  And I didn't do that work by myself.
+[954.92 --> 956.36]  Matter of fact, I tried to do none of it.
+[956.36 --> 962.28]  But I was definitely the person who got everybody on board with moving DigitalOcean to a monorepo.
+[962.64 --> 970.54]  And the reason we had to move to the monorepo in the way that we did is because Go package management, five years ago, it was horrible.
+[971.50 --> 973.90]  I mean, yeah, I could make stuff, but it was bad.
+[973.90 --> 984.06]  And then just until recently, now with the introduction of Vgo and now we have how we have modules in Go, it took us a little bit of time to get used to it.
+[984.26 --> 992.50]  And I won't say it's perfect, but I will say that it's more than good enough now that we can actually do dependency management in Go in a proper way.
+[992.50 --> 1001.08]  And then you think about what the introduction of the Go private environment variable and local caching.
+[1001.56 --> 1009.00]  OK, now my enterprise doesn't have to take down the world's infrastructure to actually be able to do modules properly.
+[1009.16 --> 1015.76]  Or even better yet, I can use that Athena thing and now I can have a private proxy where I can actually do this.
+[1016.06 --> 1018.12]  So the infrastructure is now getting there.
+[1018.12 --> 1020.86]  And I really, really appreciate that.
+[1020.98 --> 1029.12]  And that's what I'm saying that Go wasn't ready until, so I guess in the last year or so where we've, we're just right there right now.
+[1029.18 --> 1034.22]  We're at the point where I'm like, yeah, I can actually recommend this to you and not laugh when you turn your back.
+[1035.30 --> 1040.42]  And do you think that the fact that Go was open source was something that held it back?
+[1040.76 --> 1047.62]  And I'm speaking like in the past, I've worked in some big companies and this may not be as true.
+[1047.62 --> 1048.54]  And hopefully it isn't.
+[1048.64 --> 1054.32]  But definitely there used to be this attitude of, you know, open source wasn't proper software.
+[1054.74 --> 1056.96]  And if you want proper software, you have to pay for it.
+[1057.18 --> 1060.62]  Do you think that held Go back at all when it came to the enterprise?
+[1061.38 --> 1061.72]  It did.
+[1062.10 --> 1062.86]  And guess what?
+[1063.24 --> 1065.70]  50 years ago, black people had to sit at the back of the bus.
+[1066.08 --> 1067.10]  Or no, it's more than that now.
+[1067.20 --> 1069.34]  Now I would say 60, 70 years ago.
+[1069.96 --> 1070.76]  Was it right then?
+[1070.86 --> 1071.08]  No.
+[1071.14 --> 1071.82]  Is it right now?
+[1071.92 --> 1072.22]  No.
+[1072.36 --> 1073.20]  And it's the same thing.
+[1073.20 --> 1081.32]  The reason why a lot of these things held true for so long is because someone influential said these things.
+[1081.66 --> 1082.28]  And was it right?
+[1082.46 --> 1082.86]  No.
+[1083.16 --> 1090.74]  And actually, it's great now because as we go into this new decade, we're seeing that there's a lot more diversity and development.
+[1090.74 --> 1095.48]  Because of David Fowler that I mentioned from Microsoft earlier, I actually follow .NET development.
+[1096.22 --> 1100.12]  And just think a few years ago, .NET, wow, is that a great idea?
+[1100.38 --> 1101.60]  I don't even want to run Windows.
+[1101.76 --> 1105.44]  You know, now people are successfully deploying it off of Windows.
+[1105.44 --> 1110.32]  And there's some really cool tech in there made by some really good people.
+[1110.72 --> 1113.26]  And we need to start looking in other directions.
+[1113.72 --> 1116.66]  And also because, think about it, Java, what are they ready to do?
+[1116.72 --> 1117.30]  Java 14?
+[1117.88 --> 1119.40]  Look at what Java 14 is doing.
+[1119.68 --> 1123.62]  They're getting to the point now where they're like, uh-oh, the world is catching up.
+[1123.74 --> 1125.66]  Maybe we need to do some great things.
+[1125.66 --> 1134.38]  And some of the new language features coming in the new version of Java are definitely, you can definitely see that languages like Go had some impact on them.
+[1134.38 --> 1138.68]  But, you know, Java has had generics for forever, so I rest my case.
+[1139.02 --> 1139.30]  Or Douglas.
+[1139.84 --> 1145.06]  Yeah, that's definitely true of the different kind of languages, language features and things.
+[1145.36 --> 1148.22]  I've seen Swift also has like a defer statement.
+[1148.28 --> 1149.46]  It doesn't quite work the same.
+[1149.62 --> 1155.74]  But the trouble is they tend to add on to what they already have, don't they, with these other languages.
+[1156.24 --> 1160.22]  One of the nice things about Go is that it is quite a tiny little language.
+[1160.22 --> 1171.38]  And there aren't too many ways of doing one thing, which turns out to be quite important because taking choice away means, oh, then it's kind of obvious what you're going to use to solve a particular problem.
+[1172.64 --> 1174.58]  What is the Java features?
+[1174.70 --> 1176.48]  Do you know what sort of things are they adding to it?
+[1176.78 --> 1177.34]  You know what?
+[1177.92 --> 1178.28]  No.
+[1178.46 --> 1180.16]  I didn't come prepared to talk about that.
+[1180.28 --> 1181.04]  Yeah, that's all right, mate.
+[1181.04 --> 1187.08]  But I do remember, and someone's going to check me on this and tell me on Twitter, and I'm going to tell you right now, I don't care.
+[1187.84 --> 1194.52]  What we're thinking about is one of the biggest complaints with Java is the dash x mx flag.
+[1195.36 --> 1201.78]  And, you know, whenever you're actually saying how big this thing is going to be and that Java needs a runtime.
+[1201.78 --> 1207.42]  And I think, actually, as we move on in years, I think Java is going to move past that.
+[1207.52 --> 1213.08]  We're just going to assume that the JRE is part of the linking, and we're just going to give you a binary that runs on your platform.
+[1213.68 --> 1217.66]  And I see them moving there, and I think it's actually a pretty good idea.
+[1217.74 --> 1220.34]  But years ago, I can definitely see why they had a JRE.
+[1221.16 --> 1222.20]  But you know what?
+[1222.32 --> 1226.20]  It's Oracle, and I don't understand how they are doing their development.
+[1226.20 --> 1230.16]  But I do know that there's definitely some super smart people over there.
+[1230.28 --> 1232.68]  So hopefully they're trending toward the right direction.
+[1233.46 --> 1241.70]  Speaking of enterprise, I mean, you know, Oracle, you're almost at the pinnacle there of what you can think of as enterprise.
+[1242.40 --> 1249.74]  So with all these projects, right, and I'm talking about the projects that you don't have to go and pay for, right, to bring back what you were talking about earlier.
+[1249.74 --> 1254.92]  For open source projects to succeed, like Go, for example, is an open source project.
+[1255.18 --> 1257.58]  It still has a corporate benefactor.
+[1257.70 --> 1259.38]  It still has a corporate entity behind it.
+[1259.66 --> 1274.60]  From what I can see, be it Go, be it, you know, other languages with backing from corporations, it seems like those are the projects that are faring quite well nowadays than the non-funded, the non-backed project.
+[1274.78 --> 1277.12]  Is that me, or is that something that you're noticing as well?
+[1277.12 --> 1277.64]  Yeah.
+[1278.06 --> 1279.52]  You know, I'm going to tell you all a secret.
+[1279.74 --> 1282.14]  It takes money to develop stuff.
+[1282.50 --> 1284.92]  And you're like, no, no, no, I developed this on my weekend.
+[1285.04 --> 1285.98]  All right, that's cool.
+[1286.34 --> 1290.40]  But to have a large, sustainable project, it takes lots of money.
+[1290.78 --> 1300.24]  So I don't know how big the Go team is, but I know that from the first time I went to GopherCon to right now, it's a huge difference.
+[1300.24 --> 1307.40]  And you think to have all these hugely popular open source software projects.
+[1307.40 --> 1311.12]  The reason why they all come from companies is because they need a lot of money.
+[1311.12 --> 1320.78]  And this is also why you have groups like the Linux Foundation, CNCF under them, or to a lesser extent, the Apache Foundation.
+[1320.78 --> 1328.64]  Because it takes money to not only write the code, because if you think writing the code is the easy part, ha, ha, ha.
+[1328.96 --> 1332.40]  But to take care of governance, all the legal aspects.
+[1332.76 --> 1334.50]  And who's going to pay for your security review?
+[1334.50 --> 1337.24]  And people don't think about that.
+[1337.34 --> 1338.76]  They think about, oh, no, this is easy.
+[1338.86 --> 1340.24]  I can just write this on the weekend.
+[1340.72 --> 1342.58]  And you can to a certain level.
+[1343.06 --> 1347.18]  But when a project like Go, think about all the pieces of Go.
+[1347.34 --> 1349.08]  There's the build infrastructure.
+[1349.80 --> 1352.04]  There is low-level language things.
+[1352.12 --> 1357.44]  There's people who are getting their PhDs to think about how to make a better defer.
+[1357.44 --> 1358.46]  I'm just guessing.
+[1358.72 --> 1359.46]  I don't really know.
+[1359.66 --> 1361.90]  Or how do we actually do generics?
+[1361.94 --> 1364.52]  You can't just code that up on a whim.
+[1365.50 --> 1373.14]  And then there's all the other support things, like the module infrastructure or the module depth infrastructure that we have for Go.
+[1373.28 --> 1374.82]  Someone has to take care of that.
+[1374.90 --> 1376.24]  And then all the advocacy.
+[1376.96 --> 1381.16]  Someone has to go out and talk about how we are using these things.
+[1381.42 --> 1387.14]  And work with their internal partners at Google to make sure that they're using it in a certain way.
+[1387.14 --> 1391.48]  When projects get to a certain size, we realize that it's not easy anymore.
+[1392.08 --> 1393.26]  Like Kubernetes itself?
+[1393.66 --> 1398.20]  Just go look at their repository, like the Kubernetes slash Kubernetes repository on GitHub.
+[1398.58 --> 1401.18]  And then look at all the work that goes into that.
+[1401.80 --> 1404.18]  You can't do that without financial backing.
+[1404.60 --> 1409.50]  And then I know that at one time Google was putting the bill for Kubernetes backend.
+[1409.78 --> 1410.68]  It was not cheap.
+[1410.94 --> 1414.96]  I mean, like, Johnny, you and I weren't going to chuck in a couple bucks and pay for that.
+[1415.16 --> 1416.06]  No, no, no, no.
+[1417.14 --> 1424.58]  How much time does your team spend building and maintaining internal tooling?
+[1424.84 --> 1426.84]  I'm talking about those behind-the-scenes apps.
+[1427.10 --> 1428.86]  The ones no one else sees.
+[1429.12 --> 1431.62]  The S3 uploader you built last year for the marketing team.
+[1431.86 --> 1435.34]  That quick Firebase admin panel that lets you monitor key KPIs.
+[1435.58 --> 1440.62]  Maybe even the tool your data science team hacked together so they could provide custom ad spend analytics.
+[1441.22 --> 1443.16]  Now, these are tools you need so you build them.
+[1443.16 --> 1444.30]  And that makes sense.
+[1444.84 --> 1451.36]  But the question is, could you have built them in less time, with less effort, and less overhead and maintenance required?
+[1451.68 --> 1453.86]  And the answer to that question is, yes.
+[1454.34 --> 1455.60]  That's where Retool comes in.
+[1455.98 --> 1459.46]  Rohan Chopra, engineering director at DoorDash, has this to say about Retool.
+[1459.46 --> 1468.30]  Quote, the tools we've been able to quickly build with Retool have allowed us to empower and scale our local operators, all while reducing the dependency on engineering.
+[1468.74 --> 1469.10]  End quote.
+[1469.10 --> 1475.84]  Now, the internal tooling process at DoorDash was bogged down with manual data entry, missed handoffs, and long turnaround times.
+[1476.16 --> 1485.36]  And after integrating Retool, DoorDash was able to cut the engineering time required to build tools by a factor of 10x and eliminate the error-prone manual processes that plagued their workflows.
+[1485.78 --> 1489.88]  They were able to empower back-end engineers who wouldn't otherwise be able to build front-ends from scratch.
+[1490.26 --> 1495.24]  And these engineers were able to build fully functional apps in Retool in hours, not days or weeks.
+[1495.24 --> 1499.42]  Your next step is to try it free at retool.com slash changelog.
+[1499.56 --> 1502.02]  Again, retool.com slash changelog.
+[1521.16 --> 1522.28]  Maybe I'm jaded.
+[1522.52 --> 1524.34]  I've been on this earth for a few years.
+[1524.34 --> 1525.26]  So I've seen a few things.
+[1525.92 --> 1528.36]  So once you accept money, right?
+[1528.58 --> 1532.70]  That means they, or the proverbial they, can tell you what to do, right?
+[1532.70 --> 1544.64]  Do you think that this model, based on what we're seeing out there right now, the examples of open source projects that are backed by some sort of corporate benefactor,
+[1544.64 --> 1555.42]  do you think that model is going to retain what works best for everybody instead of just what works for the corporate benefactor, right?
+[1555.56 --> 1563.36]  So, for example, I'm sure there are some who do not like the way the Go team, right, does certain things with the Go language, right?
+[1563.36 --> 1565.38]  They're all on Hacker News right now.
+[1565.38 --> 1566.06]  Yep.
+[1566.70 --> 1567.22]  Yep.
+[1567.64 --> 1573.38]  I mean, yeah, there's no shortage of opinions on things like Go modules, for example.
+[1573.72 --> 1581.94]  And, you know, like cementing import versioning is a hot topic these days, you know, which kind of, you know, kind of makes a few people, you know, kind of not happy with that.
+[1581.94 --> 1593.00]  Right. But to them, it may seem like, okay, well, the Go team decided to do something that works against the broader community's interests, something that may work well at Google, but not for everybody else.
+[1593.12 --> 1594.88]  But you're always going to have these things.
+[1595.02 --> 1596.68]  So is it fair, right?
+[1596.82 --> 1605.90]  Is it naive of me to think that the corporate benefactor is going to always be looking after the best interests of the entire community and not just for things that work well for it?
+[1606.70 --> 1610.04]  Corporate benefactor is never looking out for the community.
+[1610.04 --> 1611.06]  Oh, no, no, no.
+[1611.06 --> 1620.56]  So companies, CEOs have fiduciary responsibilities to make money for their shareholders, or you can go to jail in the United States.
+[1621.00 --> 1622.46]  Well, if you don't make enough money?
+[1622.96 --> 1624.84]  Well, the board will get rid of you.
+[1625.70 --> 1629.62]  But you mean there's like a legal duty, though, to look after shareholders?
+[1629.82 --> 1632.78]  Yeah, your job is to do things that make more money.
+[1633.04 --> 1635.08]  This is why companies should not be considered people.
+[1635.40 --> 1637.48]  They do not work on the same set of ethics.
+[1637.48 --> 1639.46]  It's just how the law works in the United States.
+[1639.88 --> 1640.48]  And that's fine.
+[1641.06 --> 1642.50]  But think about this way.
+[1642.98 --> 1657.54]  If your company, companies aren't looking at big open source projects, whether it be, and this is just speculation, but Google looking at Go, Google and whoever else looking at Kubernetes, they're not doing this to be nice to the community.
+[1657.54 --> 1665.70]  They're doing it because if you support this project properly, ultimately, you have one of two things.
+[1665.70 --> 1675.20]  You either have a better funnel of people who are coming to use things that you can charge money for, or you have better integration with other parts of the industry.
+[1675.46 --> 1678.30]  So, for example, with Kubernetes, it's actually like with Linux.
+[1678.30 --> 1694.68]  Linux made it easier for us to have a good operating system that was free, that allowed us to build and work on all sorts of different types of hardware.
+[1694.68 --> 1695.52]  Think about that.
+[1695.90 --> 1698.44]  We didn't have to pay Microsoft to do these things.
+[1698.84 --> 1703.90]  Kubernetes is now taking that a level up where now I can actually install Kubernetes.
+[1704.40 --> 1707.94]  And I'm going to really gloss over this because we're not selling Kubernetes right now.
+[1707.94 --> 1709.70]  But I can install Kubernetes.
+[1710.32 --> 1715.50]  And with that, I can run workloads that span over multiple machines.
+[1715.70 --> 1716.80]  It's making it easier.
+[1717.20 --> 1721.92]  It's actually the industry is looking at that as actually a boon.
+[1722.02 --> 1723.88]  It helps us sell this type of thing.
+[1724.00 --> 1725.52]  It helps us do these things better.
+[1725.82 --> 1726.68]  And I think that's great.
+[1726.80 --> 1730.76]  And actually, last year at KubeCon, I alluded to this in a talk.
+[1731.14 --> 1733.84]  Industry is a bunch of verticals.
+[1734.56 --> 1735.22]  Think about this.
+[1735.22 --> 1742.68]  Your company, whatever they're doing, they're looking up and down at how do we take from zero to 100 to make as much money as we can.
+[1743.14 --> 1745.28]  But they realize that they're not by themselves.
+[1745.40 --> 1746.06]  They're not islands.
+[1746.60 --> 1748.48]  And what the ecosystem is, the horizontal.
+[1749.12 --> 1758.46]  Now, there's certain things that we need to worry about in a horizontal, whether it be our operating systems or our platform or platforms like Kubernetes or certain standards.
+[1758.92 --> 1759.74]  Like, think about this.
+[1759.74 --> 1764.86]  If we didn't have an SMTP standard, how in the world did we send mail?
+[1765.94 --> 1774.58]  Or, you know, or even better, if we didn't have a TCP or any of the OSI stack, how would we actually communicate?
+[1774.58 --> 1780.34]  So there are certain things that we realize as companies in the industry that we need to standardize on.
+[1780.68 --> 1784.14]  And then we compete around the edges or we bring another angle to it.
+[1784.44 --> 1785.22]  And you know what?
+[1785.30 --> 1787.38]  I don't think that's a problem.
+[1787.76 --> 1793.88]  I think that if we try to judge companies and what they do as people, yeah, it looks like companies are jerks.
+[1793.88 --> 1801.30]  But if you look at companies as doing what they're supposed to do to actually make money, yeah, I guess they're doing a good job.
+[1801.44 --> 1802.44]  And here's the best part.
+[1802.92 --> 1804.44]  We don't have to like any of it.
+[1804.52 --> 1806.30]  That's the biggest thing that I think.
+[1806.30 --> 1811.30]  And here's a soapbox that I will stand on that us as adults, we need to understand.
+[1811.52 --> 1813.44]  And especially as a black dude, I can just tell you this.
+[1813.66 --> 1815.30]  A large part of the world sucks.
+[1815.36 --> 1816.32]  It wasn't made for me.
+[1816.68 --> 1817.88]  And you know what I learned to do?
+[1818.02 --> 1819.74]  I learned to maneuver within it.
+[1820.26 --> 1823.96]  You know, there's things that I cannot change without killing myself literally.
+[1824.94 --> 1825.88]  So guess what?
+[1826.04 --> 1827.76]  Knowing that, now what do I do?
+[1827.88 --> 1828.82]  And actually, you know what?
+[1828.84 --> 1829.26]  It's crazy.
+[1829.40 --> 1831.56]  All these developers out here should embrace that.
+[1831.78 --> 1832.78]  We love constraints.
+[1833.74 --> 1836.58]  And we actually work better in constraints.
+[1836.58 --> 1842.18]  And I've actually used these constraints society has given me to do better things in certain places.
+[1842.68 --> 1848.56]  And what we should do, you know, in our computing with our languages is we should, in our technologies, we should be doing the same thing.
+[1848.96 --> 1850.48]  And yeah, it sucks.
+[1850.82 --> 1851.46]  But guess what?
+[1852.12 --> 1852.80]  That's okay.
+[1852.80 --> 1854.64]  Not everything is made for you.
+[1855.28 --> 1856.28]  It's made for society.
+[1856.44 --> 1859.24]  And we should look at how it's benefiting society as a whole.
+[1859.60 --> 1863.62]  And I know there's another whole huge capitalist conversation that needs to be had here.
+[1863.74 --> 1864.34]  But guess what?
+[1864.58 --> 1866.36]  That's not what go time is for.
+[1868.76 --> 1870.40]  And we appreciate that.
+[1870.98 --> 1875.18]  We do have a segment on the show called Unpopular Opinions.
+[1875.26 --> 1878.20]  But I think you might have been dropping maybe a couple already.
+[1878.58 --> 1881.62]  Maybe we should change it to Popular Opinions for Brian.
+[1881.62 --> 1884.62]  I actually had another one too.
+[1884.78 --> 1885.84]  So we can do that later.
+[1886.78 --> 1887.04]  Yeah.
+[1887.08 --> 1887.48]  Okay, good.
+[1887.52 --> 1887.90]  We'll do that.
+[1888.06 --> 1888.40]  Good.
+[1888.44 --> 1889.26]  We'll reserve it.
+[1889.26 --> 1894.28]  On that about the enterprise kind of playing nicely with communities.
+[1894.28 --> 1909.32]  One of the biggest resistances I've seen in the past to people adopting open source or contributing even to open source is this kind of, it's almost a blanket default attitude that all of our software is our IP.
+[1909.32 --> 1910.32]  And it's valuable.
+[1910.32 --> 1910.78]  And it's valuable.
+[1911.02 --> 1912.70]  And we need to protect it.
+[1912.86 --> 1920.80]  You know, it's a kind of default position that companies take, often by people that don't necessarily understand the nuances in that.
+[1921.02 --> 1924.02]  So that's part of that, I think, is, and I've seen this happen.
+[1924.02 --> 1925.22]  And it's definitely getting better.
+[1925.34 --> 1929.90]  And Corey on the Slack channel mentioned Rails and Ruby.
+[1930.26 --> 1933.98]  And maybe Ruby did pave the way a little bit for Go.
+[1934.14 --> 1935.92]  Ruby was this open source project.
+[1936.06 --> 1941.34]  Ruby on Rails, of course, the framework that everyone was, became popular kind of at a grassroots level.
+[1941.34 --> 1945.10]  And it took a long time to get into enterprises.
+[1946.14 --> 1947.12]  Compare that to Go.
+[1947.30 --> 1952.68]  Go's kind of gone almost, well, it hasn't taken as long as it probably took Ruby.
+[1952.90 --> 1955.28]  So I feel like maybe Rails did help us there.
+[1955.40 --> 1964.94]  But I think the attitude shift that I've seen is suddenly people realize not everything's going to be that valuable to the company.
+[1964.94 --> 1971.88]  Some things, like you say, Brian, you're better off collaborating with because then you all get the benefit of that particular problem being solved.
+[1972.14 --> 1973.30]  So I think it's a great one.
+[1973.36 --> 1976.16]  But that's the attitude, I think, that we need to bear in mind.
+[1976.46 --> 1979.30]  So I have another metaphor for you.
+[1980.50 --> 1984.66]  The other day, I went and bought a sampler, an MPC1.
+[1985.44 --> 1986.38]  And it was not pricey.
+[1986.44 --> 1987.06]  Go look it up.
+[1987.16 --> 1988.88]  And you're going to be like, Brian, you have too much money.
+[1989.36 --> 1992.34]  And no, I just like having hobbies.
+[1993.10 --> 1994.22]  I bought the sampler.
+[1994.22 --> 1995.70]  Is it an audio sampler?
+[1995.88 --> 1996.74]  An audio sampler.
+[1997.22 --> 2001.14]  So I'm going to, in my spare time now, I'm going to make hot beats.
+[2001.76 --> 2002.68]  That's what I'm going to do.
+[2002.78 --> 2004.18]  You can call me Kanye Junior Junior.
+[2004.78 --> 2008.96]  You know, the less self-hating, more righteous beat maker.
+[2008.96 --> 2009.48]  More righteous.
+[2010.04 --> 2010.54]  More righteous.
+[2011.08 --> 2011.62]  More righteous.
+[2013.18 --> 2015.96]  So I bought this thing, this box.
+[2016.22 --> 2019.14]  It's about, it's the size of like 11 by 11.
+[2019.60 --> 2021.20]  It has lots of buttons on it.
+[2021.20 --> 2028.74]  And I went to go make a beat the other day and realized I can't even figure out how to sample.
+[2029.24 --> 2029.62]  Okay.
+[2030.24 --> 2035.84]  So I go to YouTube and I learn and I'm watching this guy and I'm like, holy crap, this is hard.
+[2035.84 --> 2040.94]  But here's the thing is that our software is this beat, is a sampler.
+[2041.96 --> 2044.44]  You make a software and it's a sampler.
+[2044.72 --> 2045.40]  What does it do?
+[2045.74 --> 2051.46]  Well, until you can actually use it in a proper way, in a novel way, it's not useful.
+[2051.46 --> 2056.06]  So I can make a song that you could hear on the radio, literally can make songs that you
+[2056.06 --> 2058.30]  hear on the radio off of this machine.
+[2058.58 --> 2063.00]  But because I don't have the knowledge and the expertise and all the practice, you know,
+[2063.04 --> 2065.28]  I'm just over here entertaining myself with loud noises.
+[2065.64 --> 2067.44]  And I think it's the same thing with software.
+[2067.62 --> 2070.62]  We approach software like my software is my IP.
+[2070.62 --> 2076.08]  I said, unless it's encryption codes or things that we can't export or something that was
+[2076.08 --> 2079.98]  really novel, like PageRank when it first came out back in the 90s.
+[2081.14 --> 2083.40]  Most of your software is not that serious, dude.
+[2083.46 --> 2086.48]  And probably to tell you the truth, and I did say dude, I meant that.
+[2086.92 --> 2089.46]  And to tell you the truth, that's probably not very good.
+[2090.16 --> 2091.76]  And that's something that we need to realize.
+[2091.84 --> 2095.18]  It's how we employ our software, how we are helping our users.
+[2095.90 --> 2097.06]  That is just a tool.
+[2097.06 --> 2102.14]  We just happen to be using software to help our users out because they're using computers.
+[2102.46 --> 2103.46]  It works that way.
+[2104.02 --> 2107.88]  But just the ability to have source is not super helpful.
+[2108.58 --> 2111.54]  Or it's not a thing that should be a differentiator.
+[2111.84 --> 2114.26]  And companies want to actually optimize for that.
+[2114.56 --> 2118.00]  And the reason they do is because it goes back to this whole thing.
+[2118.06 --> 2120.24]  I'm reading a book called Good Strategy, Bad Strategy.
+[2120.64 --> 2122.34]  And I've read a couple of strategy books.
+[2122.68 --> 2124.62]  It's because we're not good at strategy.
+[2124.62 --> 2128.48]  We don't understand how to actually get from here to there.
+[2128.62 --> 2129.86]  We paint goals.
+[2130.28 --> 2131.44]  I should be stronger.
+[2132.46 --> 2134.06]  And I'm going to work out.
+[2134.70 --> 2136.48]  And then we don't really follow through.
+[2136.70 --> 2139.68]  And we don't have actionable plans to actually get stronger.
+[2140.20 --> 2140.72]  And the same thing.
+[2140.78 --> 2141.94]  Companies do this all the time.
+[2142.12 --> 2143.18]  Developers do this all the time.
+[2143.28 --> 2143.66]  We're lazy.
+[2143.84 --> 2144.44]  We're people.
+[2145.60 --> 2150.68]  And I'm not going to try to argue any company's lawyer down because I would lose.
+[2150.68 --> 2158.96]  But I think that we can see by having something like Kubernetes or something like Go or something like what Microsoft has done with .NET.
+[2159.20 --> 2160.56]  Isn't it all open source now?
+[2160.70 --> 2161.52]  Think about that.
+[2161.94 --> 2162.70]  It doesn't matter.
+[2163.30 --> 2164.16]  Because guess what?
+[2164.34 --> 2167.22]  We are architecting now at a level that is so high.
+[2167.50 --> 2168.20]  Who cares?
+[2168.62 --> 2170.64]  You know, who cares what letters people are using?
+[2170.64 --> 2173.20]  I want novel applications of this thing.
+[2173.86 --> 2179.82]  And then soon, you know, maybe in 20 years, we might not even think about software in the same way we think about software now.
+[2180.08 --> 2181.06]  And this is all silly.
+[2181.32 --> 2182.06]  But you know what?
+[2182.18 --> 2183.32]  It keeps people entertained.
+[2183.52 --> 2185.48]  And it definitely makes lots of money.
+[2185.82 --> 2188.04]  So that's why people are sticking to it.
+[2188.82 --> 2189.80]  That's controversial.
+[2190.26 --> 2190.78]  A little bit.
+[2190.92 --> 2191.58]  And you know what?
+[2191.68 --> 2192.34]  It's fine.
+[2192.94 --> 2193.64]  I like it.
+[2194.02 --> 2196.78]  If you're a language geek, right?
+[2196.90 --> 2198.92]  Admittedly, this is niche, right?
+[2198.92 --> 2212.62]  If you're a language geek and you sort of, you know, you revel in the esoteric knowledge of how language does, you know, things and its constructs and all these things, then yeah, maybe writing code, maybe just the source alone is valuable to you.
+[2212.68 --> 2214.52]  But again, very niche.
+[2214.90 --> 2219.94]  And that's not going to really be of value to anybody other than yourself, right?
+[2219.98 --> 2221.54]  So it's okay to play.
+[2221.64 --> 2222.52]  It's okay to experiment.
+[2222.78 --> 2227.34]  But, you know, not everything we write as software developers is of value, right?
+[2227.34 --> 2231.66]  And I think that perhaps that is something that we always think that by virtue of writing code, right?
+[2231.70 --> 2233.38]  We're giving this gift to humanity.
+[2233.68 --> 2234.88]  Well, not exactly.
+[2235.72 --> 2237.98]  It actually has to be useful beyond you.
+[2238.16 --> 2240.66]  And just writing the code, that's just a start.
+[2240.72 --> 2241.92]  That's where you begin, right?
+[2242.02 --> 2245.84]  Like, can it be used and produce value for others, right?
+[2245.96 --> 2251.62]  Every time I hear somebody say, oh, Facebook and Twitter, that's just a stream you just scroll down and auto refresh.
+[2251.62 --> 2253.08]  I can build that in a weekend, right?
+[2255.38 --> 2258.88]  You know, every time you hear that, you look at them, you know, with a side eye.
+[2258.98 --> 2260.60]  You're like, uh-huh, sure.
+[2260.86 --> 2261.00]  Right?
+[2261.12 --> 2265.50]  So it's, for some reason, I think maybe it's programmer culture, coder culture.
+[2265.56 --> 2266.46]  I don't know what it is.
+[2266.94 --> 2269.94]  And usually from dudes, I think you were alluding to that earlier.
+[2270.32 --> 2274.10]  We have this sort of this ego about us, about, you know, the ability to create.
+[2274.10 --> 2280.14]  We call ourselves makers and builders because we think just the act of creating something makes it valuable.
+[2280.72 --> 2281.80]  That's only the beginning.
+[2282.58 --> 2284.94]  Yeah, and that's my unpopular opinion, by the way.
+[2285.06 --> 2286.58]  I'll share that when it comes time.
+[2287.42 --> 2287.76]  Okay.
+[2287.92 --> 2290.28]  That's true also even down at the code level.
+[2290.52 --> 2295.20]  I had a friend who was looking at Go and they started to learn it.
+[2295.28 --> 2298.82]  And they sort of dropped it because they were confused about arrays and slices.
+[2298.82 --> 2310.40]  And I was kind of heartbroken because actually you don't need to really know everything about how arrays and slices work in Go to be able to use them, to be useful.
+[2310.78 --> 2314.96]  And so this sometimes sounds a little bit anti-intellectualist or something, and it's not that.
+[2315.08 --> 2317.52]  I think the more you learn, the better, of course.
+[2317.84 --> 2323.84]  But there is something about getting useful and solving real problems for people.
+[2323.84 --> 2329.18]  If that's your focus, not nothing else matters, but almost nothing else matters.
+[2329.76 --> 2339.72]  I feel like that's a way to give yourself a best chance of doing something that's going to stand the test of time or be useful or be used or be successful, whatever it is you're doing.
+[2340.56 --> 2341.36]  Oh, yeah, definitely.
+[2341.72 --> 2343.04]  No, you're exactly right.
+[2343.26 --> 2347.78]  I think I have definitely indoctrinated you into my church of haters.
+[2347.78 --> 2351.54]  I was called a killjoy.
+[2352.04 --> 2354.72]  No, I'm a professional joy stealer or killer.
+[2355.44 --> 2356.52]  And it's not that.
+[2356.60 --> 2357.96]  It's just how I get through today.
+[2358.16 --> 2361.32]  I realize that the world is not friendly.
+[2361.94 --> 2366.92]  And our goal is to make it better for people that come behind us.
+[2367.22 --> 2369.26]  And that's really all we can do.
+[2369.44 --> 2372.00]  But we realize that you can't change the whole world.
+[2372.88 --> 2375.90]  I tried boiling an ocean once, and guess what happened?
+[2376.28 --> 2376.64]  Nothing.
+[2376.64 --> 2379.86]  Was it a digital ocean or analog?
+[2381.90 --> 2382.80]  That was good.
+[2383.26 --> 2383.56]  I know.
+[2384.22 --> 2385.56]  Was it a digital ocean?
+[2385.74 --> 2387.08]  Were you making a pun or not?
+[2387.48 --> 2388.44]  No, I wasn't.
+[2389.24 --> 2391.30]  I mean, I definitely boiled things while I was there.
+[2392.38 --> 2396.08]  I am definitely not one of those seen and not heard people.
+[2397.32 --> 2398.64]  You're going to feel me if I'm here.
+[2400.46 --> 2401.10]  Well, good.
+[2406.64 --> 2410.10]  What up, nerds?
+[2410.10 --> 2411.82]  I've got some pretty awesome news to share with you.
+[2412.18 --> 2415.96]  Pluralsight is totally free for the entire month of April.
+[2416.34 --> 2416.84]  I'm not kidding.
+[2417.10 --> 2417.42]  Seriously.
+[2417.64 --> 2420.82]  Head to Pluralsight.com slash changelog and skill up while you stay at home.
+[2421.14 --> 2428.50]  For the entire month of April, you'll get access to over 7,000 courses from experts in software development, security, cloud, and data.
+[2428.50 --> 2430.44]  There's never been a better time to skill up.
+[2430.66 --> 2432.36]  Head to Pluralsight.com slash changelog.
+[2432.50 --> 2435.14]  Again, Pluralsight.com slash changelog.
+[2435.14 --> 2456.06]  I'm curious if you've witnessed situations where Go didn't take, right, in an enterprise organization.
+[2456.78 --> 2458.82]  You know, I have not seen that yet.
+[2458.82 --> 2460.86]  I've seen places where Go didn't take over.
+[2460.86 --> 2471.50]  You know, VMware, I will guarantee that our ESXi, so that's our hypervisor that runs your virtual machines on bare metal.
+[2471.68 --> 2473.04]  That's never going to be written in Go.
+[2473.62 --> 2483.06]  But I think in places where it has been applied at a higher level where we are building APIs and doing distributed computing, I've never seen it loose.
+[2483.54 --> 2488.10]  But, I mean, I've been hearing more rumbles about Rust lately, and I've been learning it.
+[2488.10 --> 2489.50]  And I look at it like this.
+[2490.18 --> 2493.04]  A programming language is a programming language is a programming language.
+[2493.74 --> 2495.40]  And it's not who you say.
+[2495.48 --> 2496.52]  It's how many people are listening.
+[2497.30 --> 2498.84]  Or what you say is how many people are listening.
+[2499.64 --> 2504.92]  And, I mean, I like, in my professional career, I've touched over 20 languages.
+[2505.50 --> 2506.98]  And we're doing Go now.
+[2507.36 --> 2508.60]  Might not be doing it in the future.
+[2508.78 --> 2509.26]  And guess what?
+[2509.32 --> 2509.80]  That's fine.
+[2510.56 --> 2510.82]  Mm-hmm.
+[2511.18 --> 2513.26]  Yeah, that's like, this is a Go podcast, though.
+[2513.30 --> 2518.28]  That's like going on the Great British Baking Show and going, yeah, and ovens and ovens and ovens.
+[2519.36 --> 2520.30]  Food's all food.
+[2520.86 --> 2524.16]  I'm just going to make a casserole today.
+[2525.02 --> 2530.20]  But the best part about this being a podcast is that the majority of people listening to this are not live.
+[2530.20 --> 2535.28]  So, if they have ill feelings about this, I'll just be like, I don't know what you're talking about.
+[2536.70 --> 2539.16]  Or, even better yet, that wasn't me.
+[2539.94 --> 2540.08]  Oh.
+[2540.78 --> 2541.32]  Yes.
+[2541.50 --> 2545.86]  There's a different Brian Liles floating about, you know, who's behind things like Octant.
+[2546.32 --> 2548.80]  Speaking of which, let's talk about Octant.
+[2549.02 --> 2552.68]  So, that is arguably an enterprise project, right?
+[2552.72 --> 2555.44]  Because it's solving problems that the enterprise have.
+[2555.72 --> 2556.66]  No, I disagree.
+[2557.20 --> 2557.72]  You disagree?
+[2558.12 --> 2558.20]  100%.
+[2558.72 --> 2559.00]  Yes.
+[2559.00 --> 2559.88]  Tell me about that.
+[2559.88 --> 2561.08]  It doesn't solve problems?
+[2561.40 --> 2561.66]  No.
+[2562.00 --> 2563.32]  First of all, shout out to Johnny.
+[2564.08 --> 2569.16]  I launched Octant to the world at Johnny's meetup in Baltimore last August.
+[2569.52 --> 2573.28]  He was the first person who had seen that outside of folks at VMware.
+[2573.50 --> 2578.14]  But, to go back even further, we started Octant at Heptio.
+[2578.32 --> 2582.52]  We were way less than 100 people at that time, or maybe we were around 100 people.
+[2583.54 --> 2585.62]  And it's not an enterprise.
+[2585.78 --> 2586.86]  It wasn't enterprise software.
+[2586.86 --> 2591.96]  The problem I have with software, just in general, or just tech in general, is all too hard.
+[2592.52 --> 2597.26]  Yeah, I can code in over 20 languages, and I have code in the Linux source.
+[2597.72 --> 2598.58]  Well, probably not anymore.
+[2598.70 --> 2599.82]  That's probably been dropped off.
+[2599.82 --> 2602.04]  And I've done all these crazy things.
+[2602.24 --> 2603.18]  But guess what?
+[2603.28 --> 2604.50]  My kid hasn't.
+[2605.00 --> 2610.04]  Or, and I hate to say my mom, but I do say my mom because she's particularly tech adverse.
+[2610.04 --> 2620.46]  And we need to realize that if we are trying to bring up technologies, and in my case, Kubernetes, we can't just say, oh, go use the command line.
+[2621.58 --> 2622.66]  That doesn't work.
+[2623.32 --> 2624.92]  But you're not saying that to your mom.
+[2625.20 --> 2626.64]  You're not saying that to your mother, are you?
+[2627.04 --> 2628.22]  Oh, no.
+[2628.28 --> 2630.92]  My mom doesn't know what Kubernetes is, nor does she care.
+[2631.16 --> 2633.04]  And she shouldn't care because it's not in her sphere.
+[2633.72 --> 2634.76]  I don't even care.
+[2635.16 --> 2636.54]  I don't even care about it.
+[2636.92 --> 2637.60]  And you shouldn't.
+[2637.80 --> 2639.76]  If it's not in your sphere, you shouldn't care.
+[2639.86 --> 2642.04]  And actually, we should get to the point where it doesn't really matter.
+[2642.64 --> 2648.50]  But it came down to, it was like, how do I know what's going on with my workloads in Kubernetes?
+[2649.18 --> 2650.44]  And that was the first premise.
+[2650.64 --> 2652.62]  And then it kind of evolved into a dashboard.
+[2653.04 --> 2654.12]  And people are like, it's a dashboard.
+[2654.32 --> 2656.82]  I'm like, well, no, it's not a dashboard.
+[2657.14 --> 2658.28]  It has a dashboard in it.
+[2658.56 --> 2662.76]  And now we're evolving it in a couple of different directions, where for a long time,
+[2662.76 --> 2663.66]  I was really resistant.
+[2664.12 --> 2668.60]  So Octant runs as a Go app and Angular and TypeScript.
+[2669.02 --> 2673.46]  But it's a Go app that runs on Windows, Mac, and Linux just fine.
+[2674.20 --> 2676.50]  We're moving it to two different directions.
+[2676.68 --> 2682.32]  We're moving it to the cluster, running in the cluster as a website, because come to find
+[2682.32 --> 2686.54]  out, and here's a lesson for everyone out there, you have to build software people need
+[2686.54 --> 2688.62]  to want to use, and you have to meet them where they are.
+[2689.00 --> 2692.64]  So we find with our enterprise customer, and this is the enterprise feature, we want to
+[2692.64 --> 2693.36]  run it in cluster.
+[2693.60 --> 2694.06]  That's cool.
+[2694.18 --> 2695.02]  We'll make that happen.
+[2695.58 --> 2701.46]  But for our other users and for our small office, home office people, or people like
+[2701.46 --> 2706.48]  me who just want a tool that can talk to different clusters, we're also moving it to an Electron
+[2706.48 --> 2707.08]  app as well.
+[2707.28 --> 2711.42]  I know people hate Electron, but go look at what else is out there, and then come back
+[2711.42 --> 2712.84]  to me and say that you hate Electron.
+[2713.20 --> 2717.00]  Electron is the best thing if you look at everything else out there.
+[2717.00 --> 2719.90]  And go look at what VS Code has done with them.
+[2720.04 --> 2721.54]  It is possible to make a good app.
+[2721.84 --> 2723.66]  If it's not great, guess what?
+[2723.94 --> 2724.94]  That's someone else's fault.
+[2725.08 --> 2727.62]  It's not Electron's fault every piece of the time.
+[2728.06 --> 2729.22]  So we're moving this app.
+[2729.46 --> 2730.42]  But we use Go.
+[2730.60 --> 2734.84]  And the reason we use Go is, first of all, because the best driver, the first client we
+[2734.84 --> 2736.54]  could find for Kubernetes wasn't Go.
+[2736.90 --> 2740.50]  But when you write an app that's small, writing whatever you want.
+[2740.58 --> 2743.22]  But when the app gets bigger, so Okten is super complex.
+[2743.22 --> 2747.66]  It probably has about 12 or 13 different domains of different things that it does.
+[2748.32 --> 2752.54]  And we find that having a strongly typed language in this case makes it easier for people to
+[2752.54 --> 2753.18]  come step in.
+[2753.64 --> 2754.94]  We have no weird duct typing.
+[2755.52 --> 2758.72]  There's only three lines of reflect in the whole entire app.
+[2758.80 --> 2760.24]  And I figured out a way to get rid of those.
+[2760.82 --> 2767.02]  And it's easy for people to be able to hook up to something like IntelliJ or no, or like
+[2767.02 --> 2767.44]  Goland.
+[2767.76 --> 2770.68]  And now they can actually view this thing and they can navigate through it.
+[2771.02 --> 2772.52]  There's lots of benefits there.
+[2772.52 --> 2774.40]  But here's the crazy part.
+[2774.62 --> 2776.94]  Ever try writing a web app in Go?
+[2777.36 --> 2778.82]  God, no, don't do that.
+[2779.00 --> 2779.30]  It's painful.
+[2779.64 --> 2780.32]  Don't do that.
+[2780.68 --> 2785.68]  So Go just serves a website that actually is an Angular type of app.
+[2785.92 --> 2786.84]  And we do it that way.
+[2786.94 --> 2789.82]  So really, it's go find the best tool for your problem.
+[2790.30 --> 2795.34]  If you run around with a Go hammer for everything, you're just going to make Go-sized holes
+[2795.34 --> 2795.82]  everywhere.
+[2795.94 --> 2796.82]  And that's not going to work.
+[2798.68 --> 2799.36]  That's brilliant.
+[2799.46 --> 2800.72]  Now I want to buy a Go hammer.
+[2800.72 --> 2805.62]  But I actually remember the launch of Octon.
+[2805.82 --> 2811.78]  I was following the Twitter storm on when you announced it in the Baltimore meetup, right?
+[2812.28 --> 2812.58]  Mm-hmm.
+[2812.90 --> 2814.10]  I mean, I say storm.
+[2814.38 --> 2816.08]  It had likes in the high ones.
+[2816.70 --> 2818.14]  But that was still...
+[2818.14 --> 2818.78]  Hey, you know what?
+[2818.78 --> 2820.34]  But no, I mean...
+[2820.34 --> 2822.04]  Well, it was...
+[2822.04 --> 2822.66]  Go what?
+[2822.92 --> 2823.68]  Here's the thing.
+[2824.30 --> 2826.20]  You say that, but I look at it like this.
+[2826.42 --> 2827.68]  It's success to me.
+[2828.18 --> 2834.42]  At VMware, I have a team of people writing, working full-time on software that I made.
+[2834.88 --> 2835.30]  I won.
+[2836.58 --> 2837.02]  Yeah.
+[2837.02 --> 2840.90]  Well, actually, what I love about it is it just solves a real problem.
+[2841.14 --> 2846.56]  And the thing is, that is a fundamental thing that a lot of projects miss, I think.
+[2846.90 --> 2850.54]  You know, especially, like, we like to write packages.
+[2850.76 --> 2853.44]  I love making packages and open sourcing them.
+[2853.46 --> 2854.46]  I really love doing that.
+[2854.46 --> 2856.46]  And I used to do it all the time.
+[2857.10 --> 2861.80]  And the only reason I don't do it as often now is because that process can involve a lot
+[2861.80 --> 2867.72]  of sort of imagining things and building kind of hypothetical software, solving hypothetical
+[2867.72 --> 2868.38]  problems.
+[2869.00 --> 2875.00]  When you solve an actual real problem that you have, it's a whole different ballgame.
+[2875.00 --> 2881.06]  And it's almost, I think we should, everyone who's working on software should kind of understand
+[2881.06 --> 2885.38]  the why, really make sure that you are solving a real problem for somebody.
+[2885.50 --> 2888.96]  And it isn't just, hasn't just been imagined, I think.
+[2889.38 --> 2890.58]  Yeah, that's a real thing.
+[2890.82 --> 2891.60]  Now, don't get me wrong.
+[2891.90 --> 2894.40]  If you go look at my GitHub right now, I think I looked at it this morning.
+[2894.60 --> 2898.30]  I have 255 projects that I've created over the life of GitHub.
+[2898.72 --> 2901.76]  And that's not even everything I've created because I think I have some GitLab stuff too.
+[2901.96 --> 2903.44]  I enjoy writing software.
+[2904.38 --> 2910.48]  But you need to write every once in a while, you have to write software that is usable, either
+[2910.48 --> 2912.56]  solves your problem or solves someone else's problem.
+[2912.92 --> 2918.64]  You can't just go out there and be like, I'm writing software all the time, unless you're
+[2918.64 --> 2920.72]  like independently wealthy or you just don't care.
+[2921.60 --> 2925.80]  But if you want to like progress in this world, you need to actually write software that helps
+[2925.80 --> 2926.14]  people.
+[2927.00 --> 2927.14]  Yeah.
+[2927.44 --> 2930.28]  I mean, that's what we all want to do really anyway.
+[2930.28 --> 2935.24]  I think it's just easy to kind of forget that or often it gets deferred as well.
+[2935.30 --> 2938.80]  It's like, well, that's the product, that's that team's responsibility.
+[2938.80 --> 2941.50]  So I just take instructions or something.
+[2941.66 --> 2947.14]  And I just think it's never as good as when you understand the problem you're really solving
+[2947.14 --> 2947.62]  yourself.
+[2947.88 --> 2950.00]  Well, that's another problem that we have.
+[2950.26 --> 2952.60]  You know, we're way over the, we're all over the map on this conversation.
+[2952.60 --> 2956.16]  But that attitude, well, that's someone else's problem.
+[2956.52 --> 2959.36]  No, no, no, no.
+[2959.48 --> 2963.38]  I mean, it might be someone else's job to solve it.
+[2963.38 --> 2965.94]  But if it's blocking you, it's your problem too.
+[2967.04 --> 2972.30]  And we, we're really easy, I guess today, especially, you know, people I work with that
+[2972.30 --> 2974.08]  I see, and this is like no complaints.
+[2974.08 --> 2975.40]  It's actually great where I work.
+[2975.76 --> 2976.82]  But I see this now.
+[2976.82 --> 2982.64]  And I see it on social media, where people are like, oh, it's not perfect for me.
+[2983.08 --> 2986.68]  Throw the hands up, you know, throw the whole, throw the whole baby out.
+[2987.04 --> 2988.72]  And you really, we can't do that.
+[2989.34 --> 2994.08]  We have to be, you know, villages are full of people.
+[2994.52 --> 2997.50]  And if someone doesn't do their job, you know, the village could fail.
+[2997.62 --> 3000.86]  Or we have to kill that person and find someone else.
+[3001.48 --> 3002.06]  But we can't kill.
+[3002.46 --> 3004.82]  That's quite a strong policy.
+[3004.82 --> 3007.04]  It is, but we can't kill people anymore.
+[3007.18 --> 3009.20]  And we don't want to kill people at all because that's wrong.
+[3009.76 --> 3012.88]  So we need to realize that no one's perfect.
+[3013.42 --> 3016.24]  And we have to work with everyone around us to make that world.
+[3016.40 --> 3016.72]  We want it to be.
+[3016.72 --> 3018.42]  Why did that sound so forced, Brian?
+[3018.60 --> 3023.04]  I feel like you've had a lawyer tell you before that you have to say, no, no, no.
+[3023.20 --> 3024.88]  Remember, killing people is wrong.
+[3025.28 --> 3029.46]  No, I mean, if you sat down with me, I am anti a whole bunch of bad things.
+[3029.54 --> 3030.82]  Like, I don't like any of that stuff.
+[3030.98 --> 3033.54]  So this is all metaphorical.
+[3033.54 --> 3034.86]  The village doesn't even exist.
+[3035.00 --> 3036.54]  Oh, you don't actually do any murders or anything?
+[3037.10 --> 3037.76]  Oh, gosh, no.
+[3038.16 --> 3038.72]  Okay, right.
+[3038.78 --> 3038.92]  Good.
+[3038.98 --> 3039.10]  Yeah.
+[3039.18 --> 3039.66]  Okay, good.
+[3039.78 --> 3040.30]  No, that's great.
+[3040.34 --> 3041.62]  Yeah, this would be a different podcast.
+[3042.02 --> 3042.18]  Yeah.
+[3042.38 --> 3044.84]  Or at least save it for the unpopular opinion bit.
+[3045.80 --> 3048.48]  Speaking of which, we're at time for unpopular opinions.
+[3048.60 --> 3051.44]  This episode has absolutely flown, I think.
+[3051.82 --> 3052.44]  I know, right?
+[3052.60 --> 3052.82]  Yeah.
+[3052.90 --> 3053.08]  Yeah.
+[3053.60 --> 3056.10]  Welcome to my world of random Brian thoughts.
+[3056.98 --> 3058.54]  And now you see what my Twitter's like.
+[3060.02 --> 3061.22]  All right, drop it like it's hot.
+[3061.22 --> 3061.72]  Oh.
+[3061.72 --> 3062.16]  Oh.
+[3062.16 --> 3062.30]  Oh.
+[3062.30 --> 3062.34]  Oh.
+[3062.34 --> 3062.36]  Oh.
+[3062.36 --> 3062.42]  Oh.
+[3062.42 --> 3062.94]  Oh.
+[3062.94 --> 3063.26]  Oh.
+[3063.26 --> 3063.30]  Oh.
+[3063.30 --> 3063.52]  Oh.
+[3063.52 --> 3063.72]  Oh.
+[3064.98 --> 3066.34]  Unpopular opinion.
+[3067.06 --> 3069.04]  I actually think you should probably leave.
+[3069.64 --> 3070.02]  Oh.
+[3072.22 --> 3072.56]  Oh.
+[3072.56 --> 3074.12]  Unpopular opinion.
+[3080.06 --> 3080.62]  All right.
+[3080.68 --> 3081.68]  My unpopular opinion.
+[3082.12 --> 3082.36]  All right.
+[3082.40 --> 3083.06]  Here it comes.
+[3083.28 --> 3085.22]  In my mind, the world owes you nothing.
+[3085.94 --> 3092.24]  So whenever you, if you go to a job and you're a beginning developer and you say, well, I've
+[3092.24 --> 3092.92]  worked four years.
+[3092.92 --> 3093.52]  I should be a senior.
+[3093.52 --> 3093.76]  I should be a senior.
+[3093.76 --> 3093.78]  I should be a senior.
+[3093.78 --> 3094.26]  Nope.
+[3094.26 --> 3094.30]  Nope.
+[3094.30 --> 3094.86]  Nope.
+[3094.86 --> 3095.94]  You should not be a senior.
+[3095.94 --> 3100.50]  What the world owes you though, is not blocking you from moving forward.
+[3100.50 --> 3101.70]  And that's the difference.
+[3101.70 --> 3105.16]  So my unpopular opinion is that you don't deserve anything.
+[3105.16 --> 3109.52]  You should have to go get and work and earn everything you have.
+[3109.52 --> 3110.74]  And it should be fair.
+[3110.74 --> 3114.10]  So whenever I see people out there saying, well, we deserve this.
+[3114.10 --> 3114.62]  No, you don't.
+[3114.62 --> 3116.60]  And people hate when I say that to them.
+[3116.60 --> 3117.80]  But guess what?
+[3118.76 --> 3119.94]  Your life is great.
+[3120.14 --> 3120.86]  You know why it is?
+[3120.86 --> 3122.16]  Because you can see my tweets.
+[3123.98 --> 3125.38]  And you're like, what does that mean?
+[3125.38 --> 3130.70]  Well, there's a whole part of this world that either can't because they're sick, can't because
+[3130.70 --> 3134.34]  they can't afford the tech to get here, or can't because they're just looking in other
+[3134.34 --> 3138.78]  directions because they're so busy trying to actually get through life and think you're
+[3138.78 --> 3139.68]  not one of those people.
+[3139.68 --> 3144.04]  And I'm not saying, this is not basically saying that there's someone worse than us.
+[3144.36 --> 3145.12]  But here's the thing.
+[3145.26 --> 3145.96]  Think about this.
+[3146.60 --> 3148.52]  You don't deserve anything.
+[3148.86 --> 3150.94]  Everything you have is you earned it.
+[3151.12 --> 3153.10]  And this is how you get rid of imposter syndrome.
+[3153.46 --> 3154.64]  You know how you're here right now?
+[3154.68 --> 3155.54]  Because you earned it.
+[3155.70 --> 3156.94]  There is no imposter syndrome.
+[3157.54 --> 3160.30]  What it is, you're here because you're supposed to be here.
+[3160.92 --> 3161.82]  And that's it.
+[3162.08 --> 3162.88]  Don't question it.
+[3163.80 --> 3164.82]  I can live with that.
+[3164.94 --> 3166.22]  Oh, hold on one second though.
+[3166.58 --> 3167.28]  Oh, another one?
+[3167.58 --> 3167.88]  No.
+[3167.88 --> 3171.54]  So in the peanut gallery, somebody said that's not unpopular.
+[3172.34 --> 3172.76]  All right.
+[3173.04 --> 3174.24]  So one more quick one.
+[3174.76 --> 3175.78]  Here's what it is.
+[3176.22 --> 3180.28]  Twitter fame, work fame, it's all crap.
+[3180.54 --> 3181.94]  Maybe this is not unpopular either.
+[3182.36 --> 3183.06]  But here's the thing.
+[3183.14 --> 3187.72]  That famous person you know is only famous because they impressed another famous person.
+[3188.42 --> 3192.30]  And people who are seeking that, trying to be famous and try to do this, basically,
+[3192.30 --> 3195.16]  you're not even measuring up to your own levels if that's what you're seeking.
+[3195.16 --> 3199.62]  What you're actually seeking is someone who probably doesn't really care about you, acceptance.
+[3200.24 --> 3201.66]  And you really have to stop that.
+[3202.14 --> 3205.12]  So that is my unpopular opinions.
+[3205.30 --> 3207.16]  And actually, here's the most unpopular opinion.
+[3207.34 --> 3208.34]  And this will leave it.
+[3208.34 --> 3213.44]  I hate to say this, but most likely, I am smarter than you.
+[3213.56 --> 3214.94]  And you might want to debate it.
+[3215.00 --> 3216.20]  And you might think, oh, no, Brian's wrong.
+[3216.30 --> 3218.34]  No, literally, I am probably smarter than you.
+[3218.70 --> 3220.68]  So we can debate that.
+[3220.74 --> 3221.88]  But you will be wrong.
+[3222.44 --> 3224.28]  And I will leave it at that, definitely.
+[3224.28 --> 3229.04]  That is a great ending.
+[3232.04 --> 3233.00]  There's no rebuttals.
+[3233.16 --> 3234.66]  They can't come debate you on the show.
+[3235.82 --> 3238.24]  It's like a mic drop right there.
+[3238.40 --> 3239.56]  Because I'm smarter than them.
+[3243.12 --> 3244.10]  Oh, my goodness.
+[3244.30 --> 3244.66]  Awesome.
+[3244.84 --> 3245.12]  Awesome.
+[3245.38 --> 3250.76]  Well, Brian, this has been both an educational and entertaining show.
+[3250.96 --> 3252.48]  Thank you so much for coming on the show.
+[3252.70 --> 3253.84]  It's been a pleasure having you.
+[3254.82 --> 3255.26]  Yeah.
+[3255.42 --> 3257.68]  Any parting gifts, parting words?
+[3258.36 --> 3260.00]  Yes, I do have a parting word.
+[3261.50 --> 3263.42]  Leave the world better than you found it.
+[3264.22 --> 3266.06]  There's always someone doing worse than you.
+[3266.48 --> 3268.00]  You don't have to help everyone doing.
+[3268.16 --> 3269.72]  But turn around and help that next person.
+[3270.12 --> 3274.20]  And tell them the only way that you can continue doing this is if they turn around and help that next person.
+[3275.06 --> 3279.68]  So realize that the world gets better when we all work together to make it better.
+[3280.08 --> 3281.10]  It ain't about politics.
+[3281.20 --> 3282.22]  It ain't about anything else.
+[3282.22 --> 3284.98]  It's about people helping people be better people.
+[3285.42 --> 3285.88]  That's it.
+[3285.88 --> 3291.30]  I love that you don't have to help everybody, but just help the next person in the line.
+[3291.42 --> 3293.96]  It's like some kind of kindness blockchain.
+[3294.78 --> 3297.20]  And I think, yeah, I like it.
+[3297.48 --> 3298.06]  I like that one.
+[3298.06 --> 3298.54]  Awesome.
+[3298.54 --> 3298.62]  Awesome.
+[3300.86 --> 3303.70]  Well, listener, we hope you've enjoyed the show.
+[3304.30 --> 3307.14]  And please tune in next time on the next Go Time.
+[3307.14 --> 3313.26]  Thank you for listening to this episode of Go Time.
+[3313.60 --> 3316.72]  More like this at changelog.com slash go time.
+[3316.82 --> 3321.92]  There you'll find our latest as well as the lists of our most popular episodes and the ones we recommend.
+[3321.92 --> 3325.08]  I personally recommend episode 110, the fireside chat.
+[3325.40 --> 3326.12]  It's a great listen.
+[3326.12 --> 3330.78]  Thanks again to Brian Lyles for inviting us into his world of random thoughts.
+[3331.28 --> 3332.20]  Follow him on Twitter.
+[3332.36 --> 3333.60]  He's at Brian L.
+[3333.76 --> 3335.92]  That's B-R-Y-A-N-L.
+[3336.58 --> 3339.58]  This episode was hosted by Johnny Borsico and Matt Reier.
+[3340.06 --> 3349.22]  If you got a chuckle out of Matt's kindness blockchain joke, hop in the Go Time FM channel of Gopher Slack and ask about the follow-on joke that we had to cut to keep the show family friendly.
+[3349.98 --> 3354.48]  Our music is produced by the mysterious Breakmaster Cylinder and we're brought to you by awesome sponsors.
+[3354.92 --> 3357.26]  Special thanks to Fastly, Linode, and Rollbar.
+[3357.78 --> 3358.70]  That's all for now.
+[3359.02 --> 3360.04]  We'll talk to you next week.
+[3360.04 --> 3361.04]  Bye.
+[3369.58 --> 3370.58]  Bye.
+[3399.58 --> 3420.94]  Okay, yeah.
+[3421.40 --> 3429.10]  I was thinking actually earlier, do you prefer to edit, to like change code when you're refactoring something?
+[3429.10 --> 3430.16]  To just change it?
+[3430.22 --> 3433.52]  Or do you prefer to kind of rewrite it wholesale, that piece?
+[3434.56 --> 3435.62]  Does anyone have any preferences?
+[3436.62 --> 3437.58]  It depends.
+[3437.78 --> 3442.06]  That's a really weird question, but it really depends.
+[3442.20 --> 3447.10]  It depends on really how bad Past Brian was.
+[3447.44 --> 3449.16]  I was looking at this code earlier.
+[3449.78 --> 3451.14]  It was yesterday or the day before.
+[3451.92 --> 3457.74]  And I was trying to figure out how to refactor it because I was like, we need to change this because I need to add new features.
+[3457.74 --> 3461.82]  And I was going through the call tree and I usually just write it down.
+[3461.98 --> 3463.88]  I don't even use the debug print stack trace.
+[3463.96 --> 3466.68]  I just went through and went through all the method calls to see what it was doing.
+[3467.10 --> 3470.06]  And it was pretty much eight more calls than it needed to be.
+[3470.50 --> 3473.44]  In that case, I am actually going to rewrite it.
+[3473.44 --> 3476.78]  And just wholesale, I actually said, nope, this is done.
+[3476.92 --> 3478.36]  I'm going to rewrite the whole thing.
+[3478.68 --> 3485.18]  But generally, if it's just a small piece and I'm either one, confident I have test coverage or two, don't really care.
+[3485.44 --> 3488.68]  I might just go in and say, we're going to live once.
+[3488.68 --> 3491.20]  Just tweak it, hack it.
+[3491.66 --> 3494.06]  Testing, test coverage does enable that, doesn't it?
+[3494.12 --> 3497.88]  You can, with quite a lot of confidence, you can sort of be quite bold.
+[3498.40 --> 3506.18]  But I always find still just like if I have to change something that already exists, it depends if you're right.
+[3506.24 --> 3509.46]  If the design's changing a lot, it's different.
+[3509.56 --> 3512.48]  But I like to just, I think, always just sort of rewrite it.
+[3512.48 --> 3516.14]  It's just like now I know what it's meant to be, I can do it again.
+[3516.26 --> 3523.82]  But I wondered if that was just a preference thing or if others do have different thoughts.
+[3523.82 --> 3528.96]  But I think it's a testament to, one, how well your tests are written.
+[3529.22 --> 3533.66]  And two, how well your abstraction is around what you're trying to do.
+[3534.18 --> 3540.94]  And I guess I'm not going to sound like old guy, but really when it comes down to it, if it's hard to change, you messed up.
+[3541.46 --> 3542.34]  And that's fine.
+[3542.48 --> 3543.36]  I mean, we're developers.
+[3544.50 --> 3549.76]  They wouldn't give us backspaces and get history if we were going to get it right the first time.
+[3550.10 --> 3551.14]  Wait, there's a backspace?
+[3552.32 --> 3554.94]  I've just been, oh, I could have really used that.
+[3555.28 --> 3558.28]  I've been just starting again when I make a mistake.
+[3559.62 --> 3560.86]  Yeah, backspace.
+[3562.16 --> 3564.28]  I like that it's called backspace as well.
+[3564.28 --> 3567.76]  It's like, properly comes from old typewriters probably, doesn't it?
+[3567.82 --> 3569.80]  Like underscore, right?
+[3570.16 --> 3572.30]  Backspace literally is just moving it back a space.
+[3572.30 --> 3572.70]  I suppose.
+[3573.98 --> 3575.02]  Yeah, we could have.
+[3575.14 --> 3579.66]  We definitely dropped a good chance to give it a much better name.
+[3580.14 --> 3580.46]  Yeah.
+[3580.62 --> 3582.56]  And then delete, of course.
+[3582.94 --> 3583.30]  Does it?
+[3583.60 --> 3585.72]  It's this strange deleting the other way.
+[3585.72 --> 3588.14]  On Windows it was that anyway, wasn't it?
+[3589.20 --> 3590.62]  Yeah, I don't know how it works.
+[3590.80 --> 3593.08]  I have a delete key on my keyboard.
+[3593.64 --> 3595.02]  I don't think I ever use it.
+[3595.48 --> 3596.26]  No, I do.
+[3596.60 --> 3599.24]  I use it because my keyboard can control Spotify.
+[3599.24 --> 3603.86]  And I think that's the stop command if I use it with the function key.
+[3604.20 --> 3604.86]  So that's what I do.
+[3604.86 --> 3608.44]  I thought you were going to say you could delete songs or something on Spotify.
+[3608.44 --> 3608.90]  No, I never.
+[3610.30 --> 3612.18]  Anything that you just don't like, gone.
+[3612.86 --> 3613.74]  That'd be good, wouldn't it?
+[3613.74 --> 3615.88]  All right.
+[3615.92 --> 3617.28]  I'm ready if you are.
+[3618.36 --> 3619.66]  Yes, I believe so.
+[3619.72 --> 3620.80]  Yeah, this is just the pre-show.
+[3621.20 --> 3626.36]  So if the pre-show is that good, how are we in the rest of it?
+[3626.76 --> 3627.98]  It's going to be, you know.
+[3628.30 --> 3629.10]  It's going to be great.
+[3629.62 --> 3630.64]  Unless we've peaked.
+[3630.64 --> 3632.86]  No, not yet.
+[3633.10 --> 3634.68]  We haven't peaked yet.
+[3635.00 --> 3635.58]  I hope not.
+[3635.82 --> 3637.24]  We haven't jumped a shark, as they say.
+[3637.98 --> 3641.48]  I mean, this is my second time on go time.
+[3642.86 --> 3645.64]  So, yeah, maybe it has jumped the shark.
+[3645.84 --> 3649.66]  If you're back to me, my gosh, there's so many interesting people out there.
+[3650.28 --> 3651.34]  Yep, maybe we're done.
+[3651.76 --> 3652.28]  This is it, people.
+[3652.28 --> 3654.20]  This is the last episode we didn't know about.
+[3654.64 --> 3655.58]  Yeah, last one.
+[3658.34 --> 3658.90]  All right.
+[3658.90 --> 3659.92]  Let me get in character here.
+[3659.92 --> 3662.10]  Let me get my voice.
+[3663.02 --> 3664.86]  I should have drank some tea before this.
+[3665.02 --> 3666.08]  You're not wearing your wig.
+[3668.14 --> 3670.00]  Normally wears a podcasting wig.
+[3671.06 --> 3674.16]  Yeah, my podcast wig and socks.
+[3674.46 --> 3674.56]  Yeah.
+[3674.64 --> 3674.86]  Yeah.
+[3674.86 --> 3675.48]  That's it.

Focusing in on PostgreSQL_transcript.txt

@@ -0,0 +1,533 @@
+**Mat Ryer:** Hello, and welcome to Go Time. I'm Mat Ryer. Today we're talking about Postgres. I'm joined by regulars, Jon Calhoun and Johnny Boursiquot. Hello, gentlemen.
+
+**Johnny Boursiquot:** Hello.
+
+**Jon Calhoun:** Hey, Mat.
+
+**Mat Ryer:** And we're also joined by a special guest, it's Johan Brandhorst. Hello, Johan.
+
+**Johan Brandhorst:** Hello, Mat.
+
+**Mat Ryer:** Welcome to the show. I should say welcome back to the show... I noticed your Twitter bio says that you're actually a maintainer of quite a few gRPC projects.
+
+**Johan Brandhorst:** Yeah, that's right.
+
+**Mat Ryer:** So you'll have to come back and do a different show about gRPC sometime.
+
+**Johan Brandhorst:** Oh, I'd love that.
+
+**Mat Ryer:** Yeah, we should do that. Aren't you working on gRPC, or something? Or you've just started somewhere?
+
+**Johan Brandhorst:** Yeah, so I've just started a new job at a company called Buf, where we're working on stuff related to gRPC, but more specifically, around API management and making protobuf easier to use, basically.
+
+**Mat Ryer:** Very cool. A lot of the gRPC repos you have are about making gRPC easy to use, like on the web, and things, aren't they?
+
+**Johan Brandhorst:** Yeah, exactly. Especially at the start of gRPC, when it first came out, it was kind of dropped out of Google, and then obviously, you're expected to know as much about it as anyone at Google... So there was a lot of space for blogging and creating exemplary posts for \[unintelligible 00:03:18.09\] so familiar with the packages.
+
+**Mat Ryer:** Yes. Great. And you also recently spoke at GopherCon EU about... What was the subject?
+
+**Johan Brandhorst:** Postgres, yeah.
+
+**Mat Ryer:** Postgres. What a great link. Not like an href, I mean like a proper one. Radio link.
+
+**Johan Brandhorst:** Like a foreign key? \[laughter\]
+
+**Johnny Boursiquot:** Oh, nice...!
+
+**Mat Ryer:** Here we go...!
+
+**Johnny Boursiquot:** This is gonna be a fun show.
+
+**Mat Ryer:** It's gonna be either fun or chaotic. Let's find out. We should maybe start just at the very beginning... For anyone really new to programming, what is Postgres, and why would you ever use it?
+
+**Johnny Boursiquot:** Oh, stop them. Stop them. \[laughter\]
+
+**Mat Ryer:** If we've stomped in there, we are gonna have a troublesome show... \[laughter\]
+
+**Johan Brandhorst:** \[04:09\] I kind of gave a whole introduction about this at the presentation, so I should know... Basically, what I said then was whenever you want to store some data that needs to live beyond the lifetime of your application, you'll probably want to use a data store. You can write to disk, for example, you can create a file, or whatever... But as soon as you want to do things like filter on the result sets, or read concurrently, or things like that - everything becomes much easier with a purpose-built application. And Postgres is one such application.
+
+**Mat Ryer:** Yeah, that's great. So yeah, saving data... Not just dumb saving, like you say; sticking JSON into a text file, or something, but slightly more sophisticated things than that... And of course, Postgres has been around a while as well, hasn't it?
+
+**Johan Brandhorst:** Yeah, it's over 20 years old now, actually. It came out of Berkeley, I think. It was a project that they were working on internally, and then they created an open source project out of it. That's why it's called Postgres; it's a post to something called -- I can't remember now. That was a poor lead on the--
+
+**Mat Ryer:** That's alright. It's the post to something called Gres.
+
+**Johan Brandhorst:** Exactly. That's kind of where the name comes from, actually.
+
+**Mat Ryer:** Yeah, right, it makes sense. So Gres maybe was like a database before it.
+
+**Johan Brandhorst:** Yeah. \[laughter\]
+
+**Mat Ryer:** I mean, we can guess... Next week we're gonna be guessing the origins of the name Rust. \[laughter\] Okay, so Postgres then - it's been an option for a while; as you say, it's been around a while. What makes it a particular favorite of yours, compared to alternatives? Because there are lots of alternatives... And MySQLite for example is quite an interesting--
+
+**Johan Brandhorst:** SQLite is also another popular database, like Postgres. The cool thing about SQLite is that it all works in a single file. Postgres is a little bit more complicated than that. But the big thing that Postgres has over SQLite in Go specifically is really good library support... Unfortunately, the SQLite driver that everyone uses is a cgo driver, and as most of you probably know, cgo means building with C, it means longer build times, it means less portable binaries, and stuff like that. In Go with Postgres we actually have several different pure Go libraries to speak with Postgres, which is really great. So that's just one reason to use Postgres.
+
+But other things, such as the stability of the software, for example - as you say, it's over 20 years old. It's been used by thousands of companies worldwide. It's not gonna just corrupt your files, because those bugs have been ironed out by now. So within databases I think you often say you don't wanna use something that's less than ten years old, because this data needs to live for a long time. You wanna make sure it doesn't corrupt on the disk, or whatever... And Postgres is one such stable, mature solution.
+
+It's also very fast, because it's written in C, and C is fast... And it's easy to run; you can provide containers for running in the cloud, or running it locally. You can also install it locally; you'll be able to install it on basically any operating system that you're running... And many cloud providers have especially built databases like RDS and GCP Cloud's SQL, for running Postgres-compatible servers... So you can just talk to Postgres anywhere, basically.
+
+**Mat Ryer:** Yes, because it has a SQL interface, doesn't it, at its root?
+
+**Johan Brandhorst:** Yeah. Well, the wire interface is a little bit different from other things. When I say "on the wire", it means the interface that you're really talking over the network. Obviously, to the user it provides a SQL, the structured query language interface, so you can do things like select and update and insert. So as a user it provides a similar interface to other SQL databases that you may have used... But on the wire, so to speak, it's a little bit different from, for example, MySQL.
+
+But actually, Postgres has kind of spawned its own little sub-genre of databases. Because Postgres is so popular, a lot of other databases implement their wire format in the format of Postgres. For example, CockroachDB, which is a large, popular, well-funded startup that's basically taken Postgres and made it more easy to deploy at a very large scale, with the clustering and geo-distribution. For example, that implements the Postgres-wire interface, so that you can basically pretend that your CockroachDB database is actually a Postgres server, and you can use the normal Postgres Go drivers to work with CockroachDB.
+
+**Mat Ryer:** \[08:40\] Hm. So as a programmer, what's it like to use -- presumably, you create tables at some point, and a table has columns, and there's data types, and things... Is it that familiar, the sort of spreadsheet kind of mindset? If you imagine a spreadsheet, it's that sort of thing, isn't it? Is it the same for Postgres?
+
+**Johan Brandhorst:** Yeah, so you can say that it has rows and columns, of course... So you can imagine a matrix where each row is a new entry of data, and each column is a different data type or field in your schema... So it's built on SQL. That means that you have to define your tables in the SQL language before you start working with the data. So in contrast to something like a document database, where you can just take a blob of JSON and insert it, and again read it out when in Postgres and other SQL databases like it, you have to predefine the structure of your data.
+
+So you will say "I want a table of users, and I want the users to have exactly these fields", and you can only insert data with that structure and you can know that when you're reading data out of it, it will have that structure. So it's much -- I kind of prefer saying that it's like going from using Python to using Go, in that you get that typing included in your data structures.
+
+**Mat Ryer:** Yeah, absolutely. There is a nice freedom you get with document stores, because it feels like you can just throw anything in and query it and it works... And of course, if there's different shapes of data, that's no problem.
+
+In practice, it turns out to be a little bit of a poisoned chalice, because often you've just moved the problem. You still need a schema, you still need data structures. They exist. You may not have discovered them upfront. But with a technology like Postgres, you do. You have to sort of plan a bit ahead.
+
+So for example a table then, if we were modeling a blog and we had blogs, and we had posts, and then comments - we might have a table for blogs, a table for posts, and a table for comments. And that's because each of those data types looks different. A comment might have an author, and a timestamp, as actually they all might. But those other fields might be different. I can't think of another example.
+
+So that gives you this sort of idea. And then they link together by referring to the primary keys in those other tables. I just wanna give a very basic kind of overview for anyone that really hasn't had a chance yet to play with databases properly.
+
+**Jon Calhoun:** I think your spreadsheet analogy is a pretty good one for anybody who's never used Postgres, or any SQL database... Where essentially, you just have to predefine the columns up top, and then you can't have multiple sheets that are connected. So like when Mat's talking about having IDs that link to other tables, it's almost like you have your user spreadsheet, and then there it might say "Okay, if you wanna find this user's purchases or something, you have to go to the purchases spreadsheet and find all the ones where the user ID column has that user ID." I think that's probably one of the biggest differences between a lot of other types of databases, is that in a SQL database you're generally running queries to join that data, to connect them all... Whereas in a lot of NoSQL databases you do a lot of that work upfront, to get whatever format you need it in... And that can be good or bad, depending on the situation.
+
+**Johan Brandhorst:** Yeah.
+
+**Johnny Boursiquot:** \[12:11\] I've heard from people who deal with very large scale that at some point the relational model - be it with MySQL, Postgres, Microsoft Access, whatever you wanna use... \[laughter\] Microsoft Access - I have nightmares about that still... Yeah, so at some point the relational model - they say that it starts to break down. I'm curious, has any of us on the show here ever reached that sort of theoretical boundary where having a relational model really starts to break down in terms of speed of access... Or really, once you start having such complicated business logic that you have the number of tables you're joining start to create a performance barrier, and basically how you'd go around that.
+
+**Jon Calhoun:** I can say for me -- I've had cases where very complex queries joining a bunch of different tables caused us to have really slow queries... But almost every time it wasn't that the database couldn't handle it, it's that we were doing something else wrong. Or we had to optimize something else.
+
+The simplest example of this is if you set up a database and you don't have any indexes, so there's no sorting of your fields there, and you're looking up users by email address - well, that'll work pretty fine; it'll work well until you have a certain number of users. And then at that point, eventually, it's just gonna take too long to scan the entire database for all those users. But the way you fix that is you add an index and all of a sudden that query becomes quick again. And basically, my experience for -- now, granted, I think there is some upper threshold, but all of my experiences have been that we've basically just gotten into one of those situations where we need to understand the problem better, and define indexes or something to simplify it in some way.
+
+**Mat Ryer:** Yeah, denormalization is something that comes up when I think about this kind of thing, because often that's a decision that you make really for performance reasons... And with relational databases - they're nice, because you can kind of query them in different ways later. You can join in the query itself, make different joins on the data, to mix data in different and new ways. So that's very nice, if you're exploring possibilities with an existing application. If you contrast that to document stores, where you're just storing basically JSON blobs, you have to pre-prepare everything in that JSON blob really, or have some tech around it to simulate that.
+
+Maybe there's a way when a user changes their profile picture, maybe there's a task that goes back in a document store and updates all the previous times that that was copied... And you wouldn't need to do that in a relational database, because of course you're joining it at runtime. And that's another time when - at least when I think about this at massive scale, that's why document stores exist, isn't it? ...it's because at massive scale that data is probably distributed geographically, and just joining it is not as simple as just reading from one place. That's one of the problems we have with it.
+
+**Johan Brandhorst:** I think that's one of the problems that Cockroach is trying to solve, as well. So they used a Raft algorithm to have a leader/follower relationship between nodes, so that the data is replicated across different geographical nodes. That's how they try to solve that sort of thing. But yeah, that's definitely one of the problems with Postgres specifically as well, like "How do you do that sort of data replication otherwise?"
+
+**Johnny Boursiquot:** \[15:45\] I wanna pull on that thread a little bit, because depending on how much data you're dealing with, it's usually unclear... If you're reading a blog post, reading off content out there, you'll eventually run across \[unintelligible 00:15:57.16\] as well once you reach a certain scale; the number of, say, read replicas you're gonna have stops being good enough... And there's various metrics you can use for that. I think it's gonna vary from team to team, and probably domain to domain.
+
+Then you're gonna see some advice around "Well, yes, it's good to store your 'transactional' data in a relational store", but if you wanna do things like reporting, for example, these kinds of things should live in a separate database, whereby you have a lot more denormalization going on, maybe you're using a data warehouse, maybe you're using a different approach altogether; maybe you're using star schemas, hybrids, or whatever the case may be, but you're really moving away from the use cases. You're splitting the use case, whereby you might have an application which is - this is where your users are using your application, and pumping a lot of data in there... And maybe you might have a read replica for their read-heavy actions... But for your internal business users, or even for a separate product, that's realized quite heavily on computation and reads, and sort of reporting, creating dashboards, and things like that... Maybe you wanna move that data to a different database system altogether.
+
+I've seen Postgres using both of those cases. You can create your schema, to use a common terminology from my data warehousing days... You can create a star schema also in your relational database. It could be a separate user, it could be a separate database, but you can use the same traditional relational databases to actually create those.
+
+So I'm curious what our collective experiences have been with regards to how soon do you make that split in your applications? Or do I go for as long as you can, where I say "Hey, you know what - using Postgres is just fine. I don't need to bring in new tech to do reporting. I can just create another schema inside of my existing server and I'm good." So what is the advice, first of all, with regards to transactional versus pure reporting OLAP data, versus - you know, do you keep everything in one system? What are the things you're considering before you make that decision?
+
+**Johan Brandhorst:** Well, that's a big question...
+
+**Johnny Boursiquot:** We ask big questions on this show.
+
+**Jon Calhoun:** I'm very much a "If it isn't broke, don't fix it" type person, but that's only because it's bitten me so many times to do that... So I very much will stick with Postgres until I'm absolutely certain I cannot solve this problem without it.
+
+**Johnny Boursiquot:** And when is that for you? What is that?
+
+**Jon Calhoun:** It just depends, is the problem. I've talked to people -- in multiple cases, I've heard hundreds of millions of rows works in SQL databases, fine. But I also know there are some tables where that probably doesn't work because of how much data you're storing, or whatever else it is. There could be something weird there. I think it just really depends on this specific problem what you're working on... But for me, it basically just came down to like, if I'm actually seeing bad performance, then I start looking at "Is there a way to fix this within Postgres?" Or "Is there gonna be a better long-term solution?" is sometimes also the thing you look at, to switch to something else. So it's really hard to tell somebody when they should do it, because it just really depends.
+
+**Johnny Boursiquot:** It sounds like you've narrowed it down to just performance though, right?
+
+**Jon Calhoun:** For me that's what it's mostly been... But it's kind of hard to say, because if I could put a little bit of effort into this and get the Postgres to work, that's one option. But if I know it's gonna continuously cause more issues down the road for us developers to maintain, then at some point, even if we can get the performance there in Postgres, it might not make sense to stay there, if it's gonna cause a lot bigger developer headache over the lifetime of it.
+
+**Mat Ryer:** You could always just stick a cache in front of it, and then it doesn't really matter how fast it is at all... You can relax a bit then.
+
+**Johnny Boursiquot:** Mat, is your cache also on your database server?
+
+**Mat Ryer:** \[19:45\] Actually, that's a good point, because sometimes you want to have data ready in a different format... So sometimes these queries might be slow-running, so you wouldn't do them while a user is waiting in the browser for the answer, but maybe you do it on a schedule (I don't know) and save the results each time, and then they're easier to read. And then I potentially would do that in the same database... They're the sorts of things you do to address performance problems. I think we get a little bit obsessed with the performance of individual components, and sometimes when you step back and look at the whole system, it might end up being insignificant compared to other things.
+
+We should do an entire episode on caching, because that is its own world of pain... You know, very powerful, but not something you have to worry about too much, unless you get to big scale. I think that's the other thing - most projects stay quite small, and that's why you don't have to worry so much about performance.
+
+**Johnny Boursiquot:** Along those lines, when we do have that show - and I'm gonna hold you to that - on caching (and I think we do need to have one), I do want us to talk about views, specifically materialized views, in database technology... Because what I've noticed - it should be no surprise on folks listening to this show - I work for Heroku and we probably have the largest fleet of Postgres databases for our cloud platform - that's our bread and butter. And we see all kinds of different patterns of usage for Postgres. And my single biggest takeaway from my experience there has been that folks underestimate how much power Postgres has. They're quick to bring in other technologies to solve certain kinds of problems that they don't necessarily need to. Maybe that's being driven by hype... "Oh, let me go try Cockroach." Well, why? "Well, I wanna use NoSQL." Well, have you tried to model your domain? Have you tried to model your business problem? What kind of data are you dealing with? Is it graph-like, is it relational in nature? There's a lot of work that is missing, and some pre-work that is missing, before you make that database technology choice. And if you don't know enough now to make the "appropriate choice" for your domain, start with a relational model, start with Postgres... You don't have to use Postgres; you can use MySQL if you want, but start with a relational model, because the tooling that exists for relational databases... I mean, relational databases have been around for many, many, many years; a lot of hard problems that have already been solved.
+
+So the tooling, and the domain knowledge, the developer mindshare - all that stuff is there. The NoSQL stuff is still -- I mean, there's wide adoption for it, but there's not nearly as much content on how to solve different kinds of problems - performance problems, operational problems - as it is in the relational world.
+
+So you can start with -- just start out with Postgres, it's fine. And then just like Johan's saying, you address your performance or scale problem once you reached that point where you can actually pinpoint, you can actually see based on metrics, based on your utilization, hopefully you have monitoring going on and you can see what your performance looks like from day to day, week to week, month to month... Then you're making decisions based on data. Not on a hunch, not on the latest hype, not on what big, famous tech company has just published a blog post about using something else instead of Postgres, or something like that... Base your decision on what your actual needs are. And I'm curious to hear what Johan has to say about that, because he's been quiet... And I don't want you to be quiet.
+
+**Johan Brandhorst:** \[laughs\] No, I was sitting here, nodding along to what you were saying - start with Postgres... Because I think a lot of users come into this space knowing what to use, and there's was kind of a hope with my talk just giving you introduction to something you can use, some opinionated tips on what libraries to use, and so on... Giving you a head start, just getting started with actually moving away from the problem of choosing a technology, and actually using a technology to solve your problems.
+
+I really agree with that sentiment, that a lot of people will try and google "What should I use to store my data?" or even read some popular programmer's blog where they have tested out the latest technology and found it to be really good for their very specialized use case, and then applied that to all of their problems, because that's all they know. You know, starting with something well tested, well established... Like you said, it's good to develop a mindshare like Postgres; it will take you very far before you need to change.
+
+**Break:** \[24:17\]
+
+**Jon Calhoun:** I think there's also probably the issue that MongoDB is so easy just to throw data at, like Johnny said, without even thinking about what your schema is going to be or what data you're going to have... But I think when people are first learning, sometimes it's easier just to have something you can throw stuff at and not care about. And I think that just sort of sticks, because -- at some point, if you want to use Postgres, or really any SQL database, you have to learn SQL to some degree... And then as you get to more and more complex queries, and you get more and more data, you have to learn it even further... And I think some people just don't like that learning curve; they just don't like to have to invest into that. But I do think that's often a mistake, because -- we've talked about all these different use cases for SQL, and I've seen so many great cases where you get so much by using a SQL database.
+
+One company I worked for, our entire metrics dashboard was just a couple of SQL queries that you fed into the graphing software, and it just spit the whole thing out. So we didn't have to do any work, and we could get pretty much any metric we wanted graphed, and anybody who was on the sales team or whatever could track these things really easily. That was really powerful, because it wasn't a massive developer effort to get them these things they needed. It was "Okay, just tell me what you need and I'll write the query for it and you're good."
+
+And I think people get worried that -- like Johnny said, they think that it's gonna cap them somewhere, and I think that they don't realize that more than likely if it actually caps you, you're at a stage where you can invest in experts, people who specifically this is what they focus on for their entire career, and they can help you figure out good solutions... And you're gonna need somebody like that at that point. But if you're not at that point, you can almost just bring somebody in to consult for like one day, who could probably come in there and get your database running 100 times faster than it was, just by looking at obvious issues and teaching you how to fix them. That sounds like a lot, but it's really not, in comparison to investing in a technology that might not solve your problems and might lead to way worse issues.
+
+**Mat Ryer:** Yeah, I think that's great advice. It's the same thing like in JavaScript and Go. Go has that type-safety, so if it asks for a string, you can only give it a string, and it's enforced by the compiler... A bit like how these table schemas - kind of enforcing that schema for the data, and any exceptions to that will be an error, probably. And of course, in JavaScript if something is asking for a string, you can say "No, here's a photograph of a cat, and that's it." JavaScript will say "Okay, cool. Thanks", and it just carries on.
+
+**Johan Brandhorst:** I hate it when that happens.
+
+**Johnny Boursiquot:** \[laughs\]
+
+**Mat Ryer:** \[27:49\] Yeah, it's a good point; you don't solve any problems, you're just moving them. That's the thing. So it feels good just throwing stuff in a data store, but at some point you have to read it and use it, and then if it's not in the same schema, you've got even more work to do to figure that mess out.
+
+But one of the approaches that I remember when I used to do relational database work in Ruby days were migrations... And I wonder if you can do migrations with Postgres. Migrations are, for anyone that doesn't know - essentially, in order to describe the schema, you run SQL commands. You might say "Create this table, and add these columns. Add this column that says String", and things like this. And of course, you can use the same kinds of commands to remove things as well. So migrations are essentially a little script that makes a little change to the database, accompanied by another little script to undo that change... And that allows you to move forwards and backwards through your database schema quite easily.
+
+Do migrations exist in Postgres? Is it just something you would build yourself, or are there other tools already that help you do it?
+
+**Johan Brandhorst:** In Postgres - obviously, as you were saying already, Postgres uses SQL, so any migrations that you want to do will have to be done with the @sql. I have a library that I quite like to use, which is called golang-migrate. Funnily enough, it's probably the first thing that shows up if you try to google search for it...
+
+**Johnny Boursiquot:** That's good SEO right there. That is excellent SEO right there. \[laughs\]
+
+**Johan Brandhorst:** Basically, it just has this structure of files where you order your migrations by lexicographically-sorted filenames... So basically, \[unintelligible 00:29:34.25\] the first kind of snippet of SQL commands that you want to run when you first start working with a database, which will then gonna be the thing that creates the tables and sets up any relationships between the tables, and then nothing else.
+
+So the more interesting use case for migrations is later on when you decide that "Hey, we actually need to add some more information to this table here, but we already have a lot of customers in the table", we'd have to manually go through and add the email to all of them, or whatever. So a migration script looks at the existing data - you write some SQL to do this, of course... You write some SQL to look at the existing data and then adapt the existing data for the new table data that you've added with your changes. So that's really the key reason why you need migrations, really.
+
+If you're a novice user and you find yourself never making any changes to the database, you could probably get away with not using migrations... But I quite like doing it from the start, because it means you can very easily pick it up again when you actually need to make those changes... And I often find that you do need to make those changes eventually. It's also a really nice operation between querying, which is something that you can -- there's a few different libraries in Go for querying data that don't actually do anything about creating data, or creating tables... And the migrations script, which is responsible for setting up and also tearing down the schema itself. So it's kind of a nice approach in between the two.
+
+**Mat Ryer:** Yeah, and so when you instantiate a new database, it essentially just runs through those migrations in order, applying them one by one, and you know that you'll end up in the same state as your latest release, I guess.
+
+**Johan Brandhorst:** Yeah. So you have to be a little bit careful, because often when you have a database, you'll want it to be used by several clients. And if you do something like an upgrade of a system where you shut down one server and you start up a new server that has a new feature, so you have both an old server and a new server speaking to the database at the same time, you'll need to make sure that the migrations are being applied in a backwards-compatible fashion, so that both of those clients can continue to operate at the same time.
+
+But if you have a single server and a single database, then it's very easy, because you can just apply all the migrations when you start up, because you know no one else is talking to the database. That's really the simplest case.
+
+**Mat Ryer:** \[31:55\] Yeah. It's nice also because you used to be able to try things out, and undo them, essentially. Often it's difficult to do -- sometimes you might be making destructive changes. If one of your migrations deletes a column, then you're gonna lose data. In the down script you can't put that data back... But assuming you've thought about it properly - yeah, you could be a bit more experimental with different data schemas, and then you either just don't commit the migrations, or you can roll back to a previous version, delete the future ones and try something else. Yeah, it is really cool as a tech.
+
+Well, you mentioned the clients in Go... I'm interested, from a Go perspective, how do you actually use Postgres? And I'm even thinking in the context of, say, a website that is hosting a blog, at what point would you make the connection to Postgres? Do you tend to make one connection per instance of your code running, and then create sessions off that? Or would each handler make its own connection? How does it actually work from a Go developer's point of view?
+
+**Johan Brandhorst:** Yeah, so if we take a step back and look at the standard library database SQL package, that actually has a connection pool built in. So if you come from another language like Python or Ruby, you might be familiar with putting something -- I think there's a PgBouncer or something like that which does connection pooling between your database and your client... In Go you don't need to do that sort of thing, because it's already built into the standard library. So what you would normally do when you connect to the database is just create a single SQL.db handle, and then that's safe for concurrent use. So you can use that in all of your handlers, even though they're being called from different goroutines, from different clients.
+
+You can also configure things such as max connections on the SQL connection, but normally that's all handled by the Go standard library, so you don't really have to worry about it, which is really nice.
+
+**Mat Ryer:** So you do use the Go standard library then. Or do you use other third-party libraries?
+
+**Johan Brandhorst:** This is not very common, actually, but my Postgres driver of choice, jackc/pgx - it actually has its own little interface as well for interacting with Postgres, which is a little bit faster, because it doesn't have to go through that compatibility layer, so it can structure things a little bit differently. It has binary encoding that it uses when it talks to Postgres, whereas the database SQL uses a slightly different encoding, as far as I understand.
+
+However, I still try to use the standard library, just because it's compatible with the rest of the Go ecosystem, really. For example, if you have a query builder or some sort of generator to make calls to the database on the SQL.db handle, that's what they expect. And if you're using pgx, then you have the choice of choosing between the two. For example, I would use the SQL.db interface as much as possible, because most of the time, that performance is gonna be absolutely fine for you. But if you do have a specific use case, or you find that actually you're being bottlenecked by the standard library here, you can switch to the more efficient binary interface. I would only really consider it if you have to insert thousands of rows, or something like that, and you want to do it very quickly... Then that's something that you can do with jackc/pgx. You can switch to the binary encoding interface and do that within a single connection, and then switch back to the standard library one for normal interface handling.
+
+**Jon Calhoun:** So one of our listeners had actually asked "How hard is the switch from lib/pq to jackc/pgx?" Because I think historically lib/pq was recommended as a library, but I think that one uses cgo, doesn't it?
+
+**Johan Brandhorst:** \[35:44\] I don't think so. I think it's pure as well. But I used to use lib/pq as well back when I started using Go. But it hasn't really kept up with the advances that jackc/pgx has. I think jackc was later to the stage, but it's really picked up because it has this really rich type package, which implements a lot of the Postgres types, so you can work in a Go type-safe way with the Postgres types, like money, and date, and things like that. Postgres has really a lot of interesting types that you can take advantage of with jackc. And then it also has this second interface, which is optimized for speed. It has a really cool way of copying from an interface, so you can give it an interface and then it will kind of call a function on that until it's done. It's like, you give it an iterator. So I think jackc/pgx has innovated a lot in this space in the last 3-4 years, which is why it's taken over the crown from lib/pq.
+
+**Johnny Boursiquot:** So I have to confess... I like using the standard library. And some of the things that bother folks, like having to map fields to different fields, from the database and over to your Go types - I don't find these things annoying as much. But when I do, I do like to use sqlx, which is probably one of the early nice little abstractions, nice little wrapper around the standard library database SQL package. That made it easy to ingest all your fields into a type all at once, as opposed to do all the scanning for different things... So it was for the vast majority of the projects where I need a relational data store, that I do rely on a combination of the standard library and just the sqlx wrapper.
+
+I have no doubt that lib/pq, which probably still is by far the most widely-used Postgres library within the Go ecosystem - but I have no doubt that it works quite well, and I've used it as well... The new kid on the block. I'm not sure if it's new at this point; like, 3-4 years old. I haven't tried it though myself, but from what I've read, from the documentation, it has some very nice efficiency and performance gains that it has added. It doesn't use the standard database SQL approach.
+
+I don't wanna be a wet blanket. It always comes as "Well, don't jump to the new fangle thing right away" kind of thing. Understand your needs first. And if you're gonna pick a driver, that's fine. I'm not saying use all the different drivers, and have different drivers do different things in your application... But again, perhaps that's the right approach. Perhaps using the standard library and sqlx for the vast majority of your application is good enough, and then the area where you actually need performance - maybe you're doing heavy batch processing, maybe you need to do lots and lots of inserts and you don't want the index getting in your way for your reads... Maybe you do use something specialized like that, right?
+
+So for me, all of this comes back to pretty much optimization. If you pick something just because you hear it's fast, and you're making that decision "Well, I'm gonna go with that one", not knowing why you're going with a particular library, then that's where I'm usually "Okay, let's slow down a little bit. Let's understand why using the standard library or a thin wrapper around the standard library is not good enough for your use case." I have no problem with using the latest and greatest, if you can justify why you wanna use that.
+
+At this point in my career I'm kind of boring. I don't like the exciting stuff, especially as an SRE. I don't want exciting with the things that I'm responsible for. It's really like "Beyond the standard library, what else do you need?" and "Can you get away with using the standard library, or some very thin wrapper around that?"
+
+**Jon Calhoun:** A question for you, Johnny... You like using the standard library for it, but I feel like in my experience Postgres is the one that interacts the most poorly with the standard library. And more specifically, I'm referring to -- like, `lastInsertId` doesn't work... And then I think -- I don't remember which ones use which, but the characters you use for variables you're putting in...
+
+**Johan Brandhorst:** Dollars, yeah.
+
+**Jon Calhoun:** ...I know Postgres uses the dollar sign, and then the numbers...
+
+**Johnny Boursiquot:** Dollar sign, yup.
+
+**Jon Calhoun:** \[39:54\] ...and I think all the others use question marks... So one of the benefits in my mind of using the standard library is I can run my tests with SQLite, and have it in memory, and that's real nice, because I can just throw the whole thing out afterwards... But that's really hard to actually achieve with the standard library, unless you put in extra work that I generally don't wanna put in, because it's just not fun to do... So I guess what are the benefits you see from the standard library from that perspective, and for somebody who is just getting started with it, do you have any advice for those gotchas?
+
+**Johnny Boursiquot:** Let me throw that back you... When was the last time you switched databases? When was the last time you said "You know what - I'm gonna start with Postgres, but I might end up with MySQL"?
+
+**Jon Calhoun:** So I've never switched them in that sense. The only reason that it matters to me is that I like running SQLite for a lot of tests... Because that means that actually setting things up is a lot simpler. It's not "Oh, you need to install Postgres, and you need to make sure your database is up and running, and you need to do all this stuff..." It's literally just "Okay, I'm just gonna throw this thing in memory", and I can run all the migrations from that point and have the test go. So it's more of a simplicity type thing there... But I often find that it's hard to actually make work, because all the differences that exist.
+
+**Johnny Boursiquot:** Well, I would say that while I see your point, if I knew that I needed to do some integration testing, I need to actually hit -- if I don't have enough abstractions around my testing, that I actually need to use a data store and communicate over the wire to a data store, I would just run a Docker version of Postgres. Heck, I have Postgres installed locally on my machine. No abstraction layer needed.
+
+So I think it's just a choice, and I think Johan is gonna have some flavor to add to that... But personally, I wouldn't wanna use two different databases for my testing, however similar they might be... And I know you have a slightly different view on that, but I'm interested in seeing what Johan has to say about that, too.
+
+**Johan Brandhorst:** It's great that you brought up testing, because that's one of the things I want to talk about today... And I agree with Johnny that if you have to make your tests so that you have to use one flavor of SQL to test, and another flavor to run, then - well, to start with, your tests might not be giving you the confidence you want, but also, there's a better solution today, which is that you can use automated Docker testing to create a Docker container during the lifetime of your test, and talk to that as if it's a real Postgres -- it's a real Postgres container, so you can talk to it like a real Postgres server, and then just shut it down after the test finishes.
+
+That was something I demoed during my talk, and I think maybe the most revelatory thing that I really brought onto the whole talk - because people afterwards were like "Oh, this is so cool!" and yeah, I agree, it is so cool; it completely revolutionizes the way that you do database testing... You don't really have to have a Docker Compose script or whatever where you spin up a database, then you run your integration tests with a build tag on them or something like that, because you don't want to run them if you run go test, because someone needs to have a database running... And you also don't need to annotate your CI at testing with extra containers, or whatever; everything just works. You just spin up a database... This is during the runtime of your tests; it uses a Docker socket API to just spin up a database, take the IP and port of that database, talk to it during a test, and then shut it down immediately after it... And it runs in like three seconds total. It's mind-blowing when you see it for the first time, and I highly recommend you try and check it out.
+
+**Jon Calhoun:** I'll have to check it out, because one of the reasons I like SQLite stuff is that if I'm running a bunch of tests concurrently, I can have a bunch of them loading up in separate instances of memory, and do stuff like that... But I do fully agree that you still need to test against Postgres, which is why generally speaking I'd more consider that something I'd run in my CI tool... So it happens there before it actually goes anywhere, but I don't have to do it locally as much.
+
+And it depends from company to company. I think some of this stems from -- you know, coming from a Rails background, I think it was pretty common Rails to run one flavor of database locally and another one depending... There were definitely times where that bit you, but it handled a lot of stuff for you sometimes, so you could do it.
+
+**Johan Brandhorst:** \[44:06\] Actually, another thing that's kind of related to this is one of my hopes for generics in Go is that the database interface in the standard library is going to get a bit of an overhaul... Because I think it's probably one of the weakest-typed interfaces in the entire standard library... Because you have to query, and then you have to give it a variadic slice of interface, which is as bad as it gets... I can't wait to see what we can come up with to make that easier to use... And I think that would also help maybe in this case, where we have a different sort of placeholders. I guess the dream of a database/sql standard library package is that you should be able to use it with any SQL database, right? Maybe even at initialization times say "I want placeholders to be question marks", and then everything else should just work... As long as you're using the SQL standard. That's the dream. And I think practically that probably doesn't work at all.
+
+I think if you were trying to write a package that is supposed to be agnostic between different flavors of SQL and different databases, you're gonna end up making a lot of compromises that will compromise your type safety, compromise performance... And it's not really there, but I hope that generics could help in that area.
+
+**Break:** \[45:27\]
+
+**Jon Calhoun:** When you're talking about the performance stuff - Johnny, you mentioned actually just using the standard library, and writing your queries and actually inserting it into whatever field you need to... I've used ORMs and I find them useful in certain scenarios, especially when people don't know SQL that well, as a way to get them a taste for it... But I've also found that even if you're using sqlx, which is really not that much, you tend to write queries that allow you to use their tools to basically take all the data that you're pulling and putting it into a Go struct... Rather than writing the correct query, that might require you to do more work, but you're like "I don't wanna do that, so I'll run two queries instead" or something.
+
+I've noticed that just weird queries like that tend to happen, where you almost write bad code because you're trying to leverage a tool a little bit more than just putting in the work to get it done the right way the first time.
+
+**Johnny Boursiquot:** That's in my world -- at least in my mind, ORM is a dirty word. \[laughs\] I don't know, I'm generalizing here; obviously, it has its uses... But I've been a part of so many teams where we rely so heavily on ORMs, and they are great for the 80% use case; and then there's that extra 20% where you have -- you try and twist the ORM so much... And it's not the fault of the ORM, it's more like basically we're abusing it, or maybe we're nesting some things, or maybe we're using it in a way it wasn't intended, or it's too easy to make certain common mistakes and we end up creating way too much thrashing with the database, and you're doing N+1 type queries, and it becomes a performance issue, and then often you don't find out about these kinds of problems during testing. You will find out about them in production, because locally, typically, you have a smaller-sized dataset; you're not querying as much data, or maybe things tend to be hyper-optimized from a performance perspective when you're testing locally, because you don't want data or your querying to impact the speed of your tests, and things like that.
+
+So you want that quick feedback. But at staging or in production, where the data volume is much higher, you go find out about these kinds of things in that environment, \[unintelligible 00:48:29.23\] And always somehow the ORM ends up being the culprit when I start to dive deep into "Okay, let's troubleshoot this problem. Let's figure out what's causing this problem." It's because the ORM allowed the developer to make those kinds of mistakes.
+
+\[48:50\] Basically, when I talk about using a standard database SQL package, it's not because I believe it's superior to all other approaches, it's because my pendulum has swung from really liking the abstractions of the ORM to basically seeing being bitten by it so many times to basically say "You know what - I wanna write the actual SQL queries." I wanna know everything I'm gonna be writing in there; what the interaction is gonna be with the database, and I can do an explain plan on it, and I can see exactly what the cost is gonna be in the database server. I can actually see it; the same sort of values we hold dear in the Go community, that explicitness... It might be a little verbose defined, but I'm willing to pay that cognitive cost in order to get that clarity and know exactly what the query is gonna look like on the other side.
+
+So yes, ORMs are great, and I think they're great for prototyping, but for certain parts of your codebase where you really need to keep an eye on performance... Maybe you have a really complex query, lots of things you're joining, and if you can't prove exactly how that's gonna perform in production, maybe you write the SQL yourself.
+
+**Mat Ryer:** For anyone that doesn't know what you're actually talking about, ORMs - can we just illuminate that a little bit? What does it literally stand for? Is it like Object Relationship--
+
+**Johan Brandhorst:** Relational mapping model, yeah...
+
+**Jon Calhoun:** Object Relational Model, or something...? I don't know. Mapping? I don't know.
+
+**Mat Ryer:** Okay, something like that. But what is it?
+
+**Jon Calhoun:** So there's a couple different ways you can think about interacting with your database. ORMs are -- you essentially write code in whatever programming language you're in; that's Go for us. And your database, generally speaking, gets mapped to whatever that code is, and then you write queries in Go code, and that Go code gets mapped to some sort of query in the database. I think at a high-level that's the easiest way to put it.
+
+**Mat Ryer:** Right.
+
+**Jon Calhoun:** So you don't really have to learn SQL, you just write Go code using this library, and it handles all of that for you. Now, the next step that you occasionally see is generators, which -- I think SQLBoiler is one that's popular in Go... Basically, they'll look at your database and they'll try to generate Go code from the database. So they look very similar to ORMs, but they still use the database as the source of truth, and they don't try to translate the other way.
+
+And then there's the kind of pure "Use SQL and translate back and forth yourself" approach, which is - you have a lot more control that way, so you can fine-tune everything. And I should say, Johnny, when you were talking about performance, and stuff - I generally view ORMs as a stopgap. You can use them, but you should have the mindset of you might have to replace them at some point, and you should be aware of that potentially being a risk.
+
+**Johnny Boursiquot:** Actually, Mat, you may be able to shed some light on this, because there are ways to actually write your data access layer - in Go otherwise, but we're not specifically talking about Go here - that would allow you to do what Jon's talking about, right? To be able to swap out that data access layer with maybe another data access layer that doesn't use an ORM, but uses raw SQL instead. Like, what is the approach? What is the recommended approach here? What is the best practice, at least within the Go community, for writing -- for not spreading, sprinkling your SQL all over your codebase?
+
+**Mat Ryer:** \[52:02\] That's a good question.
+
+**Jon Calhoun:** That's probably my biggest issue with ORMs - anybody who's come from Rails has seen a view that has a SQL query in it... And somebody doesn't realize it's a SQL query because they're just writing Ruby code... But you see that and you're like "Why are you running queries inside of a view? That makes no sense." And I think that that's one of the biggest issues you run into with ORM - you get what should be database access layer code scattered throughout the rest of your application. So I don't know what your approach is, Mat, but mine is generally to define the few queries I need, or however many there are, and to basically make little services that are like "Okay, if I need to --" Let's say for users, like authenticating, creating accounts, that sort of stuff; I'll sort of define a little struct that has all the methods for that, and then the rest of my code just expects an interface that has those methods... And I pass in -- depending on the application, what I pass in will kind of change, but I'll often try to pass in something that's unspecific to the database as I can make it.
+
+Now, there are exceptions... IDs in databases often tend to be integers of some sort, so that ends up getting into your code, in some sense... But there are other things you can -- or maybe not integers; maybe you're using UUIDs or something, but still, you get some of that in there... But you can still pull a lot of it out.
+
+**Johnny Boursiquot:** Don't use auto-increment. Just a side note... Just use UUIDs. Trust me.
+
+**Jon Calhoun:** Johnny. Let me use auto-increment.
+
+**Johnny Boursiquot:** \[laughs\]
+
+**Johan Brandhorst:** I don't!
+
+**Mat Ryer:** So Johan, do you use an ORM?
+
+**Johan Brandhorst:** Actually, I have never had the pleasure of using an ORM, so what am I doing on this show talking about databases...? Very early on in my career I was kind of persuaded against using an ORM, because naturally, as a beginner programmer, I was like "Oh, this looks cool..." But I never really had to use it, because I was told by someone who knew better than me that that was a bad idea... So here I am, saying to other beginners like myself once was that you should try not to use the ORM, even though it looks really appealing at first. You should probably just learn to use SQL. It's really not that bad, and you'll learn to love it, like I do.
+
+**Jon Calhoun:** I would agree with that, having come from the other side... Because everything specific to Rails that I learned is useless to me now. And everything I learned about SQL along the way when I couldn't get that to work is much more useful to me, and it'll carry over to any language.
+
+**Johnny Boursiquot:** That's right.
+
+**Mat Ryer:** Well, that would have been an Unpopular Opinion section, but everyone agrees...
+
+**Johnny Boursiquot:** \[laughs\] It's not that unpopular, I guess...
+
+**Mat Ryer:** It's not happened before, where we all agree...
+
+**Jon Calhoun:** Alright, so I have a question for the rest of you. Johnny, earlier you had mentioned that when you go to production, you sometimes don't notice performance issues, and that sort of thing... And earlier we talked about migrations, and that sort of stuff, so my question is "How do you guys handle testing for things like migrations, or things like performance, when generally speaking you don't have a dataset the same as production?" Do you have a nightly type environment or a test environment that clones the database, or do you take other approaches?
+
+**Johnny Boursiquot:** We've taken several approaches... The one I've liked the most, and I like it because it was sufficiently complex that we had an elegant way of solving it... And because we were dealing with data, some PI data (personally identifiable information) in the system, we had to find a way to effectively not expose, not copy that data into multiple environments. We kind of created a compliance risk...
+
+So what we did is basically watch the data -- when you have primary and other replicas, the primary typically needs to communicate changes to those read-only replicas... So if you tap into that stream, going back and forth - or rather typically going one way, from the primary (actually where you're writing your data) to the read-onlies, if you tap into that stream, you will see exactly what the changes are.
+
+\[56:08\] So now we can sort of inject ourselves - basically, you can think of it as a listener - whereby we can actually see the data and actually keep our separate test environment up to date with that stream of information... And then during that copying over, we actually apply transformation to that data to remove (or to anonymize rather) the data before it actually gets stored on disk with the test environment.
+
+That's an elaborate and more complicated way of doing it, but that's one where it was the most fun for me, because one, that's when I figured out basically how the replication was working between the primary and secondary nodes, and basically being able to transform the data in flight, before you write it to disk - that was a really fun technical challenge... But that's definitely one of the more complicated ways to do it.
+
+**Jon Calhoun:** To make sure I understand -- so you have the lead database... I'm so used to bad terms, sorry. And then you have the follower. Did you spin up a new follower that you intended to use for test, one that was popping all the data over to that new follower, and that's when you did all the transformations?
+
+**Johnny Boursiquot:** Well, it wasn't in the follower pool. Basically, the primary wasn't aware of that separate -- it's not really a follower, it's just a separate instance somewhere else...
+
+**Jon Calhoun:** Okay.
+
+**Johnny Boursiquot:** ...and basically we would be writing that data to it. Obviously, it was a little slower, but it was a test environment. It didn't need to be to-the-millisecond up to date. We could apply our transforms before the data got written. It was a way of actually interjecting ourselves in that process.
+
+**Jon Calhoun:** Oh, so you kept this going all the time then?
+
+**Johnny Boursiquot:** Well, yeah. Yeah.
+
+**Jon Calhoun:** Okay. So it wasn't a one-time thing; you just kept it up to date, and then you could use it in the future.
+
+**Johnny Boursiquot:** Yeah, we kept it up to date as the system changed, and we just kept our staging environment up to date.
+
+**Mat Ryer:** It's really clever, because if you'd just read it normally, that of course would have been activity on the database, potentially...
+
+**Johnny Boursiquot:** Exactly. It would have been another client, yeah.
+
+**Mat Ryer:** Yeah... It's funny, because we're kind of looking at this exact thing, and it's basically -- as Johnny described, what we're gonna do is read from one place, I do the anonymizing thing, and what we really want is the shape of the data. We're not gonna read it; we're not gonna sit and read through things... So we'll probably scramble everything; we'll probably mix all the words up and everything, just so that you don't get -- because customer data, if you really care about that, then it's not acceptable to just copy the production database to test, to work on... It doesn't matter really what the data is, but... That's really clever though. I love the way that it was just -- it must have been some listener API, so it could receive the changes; that's a really smart solution.
+
+**Johan Brandhorst:** Another way you can also test migrations specifically is actually write the test where -- if you have access to the migration files, you can kind of migrate to the first step, you can insert some data... Now, obviously, it won't be entirely representative of your production database, where you may have hundreds of thousands of customers, or whatever... But you insert some data, and then you do the migrations in the next step, and you check that that data was updated correctly. And you can do that for all of your migrations as well, and then run that as a normal test... Especially if you spin up a database with the dockertest container. So that's what I've been doing for some of my projects...
+
+**Mat Ryer:** In only three minutes...
+
+**Johnny Boursiquot:** Yeah. \[laughter\]
+
+**Mat Ryer:** To be fair, I saw that demo that you did, and it didn't look real. It was so quick... Yeah.
+
+**Johan Brandhorst:** I wanna say as well, the first time I ran that -- I had just formatted my PC, so the first time I ran it, I had to pull down the Docker image as well, and it still took 12 seconds in total... But the second time I ran it, it took three seconds.
+
+**Mat Ryer:** \[59:48\] \[laughs\] We'll post a link to dockertest in the show notes, dear listener...
+
+**Jon Calhoun:** I need to look at that myself, so...
+
+**Mat Ryer:** Yeah. It's good. It's a good point, because you wanna test as close to that production environment as possible.
+
+**Jon Calhoun:** Especially with like migrations and all that sort of stuff - if you're doing anything complex, it can be really hard because you might have data that you just don't expect to be in the format it is, or something. I've definitely had one or two cases where we were running Rails migrations during a deploy, and all of a sudden something breaks and everybody just loses it, because you're like "I don't know what's happening right now." And it's just hard if you somehow missed it in a test, or something; you have to find good solutions for that.
+
+**Johan Brandhorst:** Yeah. Even the solution that Johnny was proposing may suffer from that issue, because you anonymize the data, and all of a sudden you took out all of the Unicode names, or something... You can never really be sure, unless you're doing it straight up against your Postgres test database, right? \[laughs\]
+
+**Johnny Boursiquot:** And production. \[laughter\]
+
+**Jon Calhoun:** That's terrifying though.
+
+**Mat Ryer:** That can't be our advice at the end of the show... "Test in production."
+
+**Johnny Boursiquot:** "Test your data changes in production." \[laughs\]
+
+**Jon Calhoun:** One last question then... Somebody on Twitter had asked "Why do we import the underscore for SQL packages?
+
+**Mat Ryer:** I think this is left over from not a great design decision in the early days. I don't think anyone would design packages today that use this underscore import thing. Isn't it where there's an init function, and obviously when it's imported, the inits in a package will run, and they can sometimes then interfere. They can import other packages, they can set variables, they can do these magical things in global space.
+
+**Johnny Boursiquot:** It's a huge side effect.
+
+**Mat Ryer:** Yeah, yeah.
+
+**Jon Calhoun:** The worst part is I think all they do is they call `sql.register`, and they pass in a name, and then they pass in the driver. I think that's literally the line.
+
+**Johnny Boursiquot:** Well, you hope, yeah... You hope... \[laughter\]
+
+**Jon Calhoun:** You hope that's all they're doing... But the worst part is I look at it and I'm like "If I had to import this package, I could have just written that line myself and been done with it."
+
+**Mat Ryer:** That's right, and I think that was it. It was before we had a lot of experience with Go, I think. It was early days, where that just wasn't obvious. It seems really quite obvious now in retrospect. It's the same with the image packages. If you wanna support gif and JPEG and things, you just have to make sure the packages are imported; you don't use them. It's so weird, and I really don't like it. It's no harm to just say `sql.register` and pass the thing in. It's a a bit more verbose... So if you're designing packages, then please just make it verbose. It's so tempting to want to be very nice and make users' lives very easy, and do everything for them... But yeah, I think that's one example where you shouldn't try. Don't try and be too clever, just let them import a package and then use it. And if they don't use it, don't import it. Don't have any side effects to importing a package.
+
+**Johnny Boursiquot:** This is one of the cases where ignorance is not bliss.
+
+**Mat Ryer:** \[laughs\] Yeah.
+
+**Jon Calhoun:** Once you understand how it works, it's still not ideal, but it's not as terrifying. But I completely understand the first time anybody sees it, they're like "Look, there's magic in Go", and I'm like "Not really..." It just requires you -- it forces you to learn about init, which in some ways I'm like "We don't really wanna teach people about this."
+
+**Johnny Boursiquot:** \[laughs\] Stay away from globals, generally speaking, and stay away from init...
+
+**Mat Ryer:** Yeah.
+
+**Johnny Boursiquot:** Is it time for Unpopular Opinions? I want an unpopular opinion from Johan.
+
+**Johan Brandhorst:** Oh, I can do that, actually.
+
+**Mat Ryer:** Okay, well then let's do it!
+
+**Johan Brandhorst:** I could do an unpopular opinion...
+
+**Jingle:** \[01:03:59.22\]
+
+**Johan Brandhorst:** I have this library that I like to use, which is called Squirrel, and it's a query builder. It uses the builder pattern. Everyone hates the builder pattern in Go, right? And for good reason, because the builder pattern doesn't work well with the static typing that Go provides, because we don't have generics... And Squirrel suffers from this problem as well, but it also provides a lot of power at the same time.
+
+So it's like, the one exception to the rule of "Don't use the builder pattern anywhere" is for query building use the Squirrel package, because it's really easy to use at constructing queries with.
+
+**Mat Ryer:** So the builder package then, for anyone not familiar - this is where you get these fluent APIs where every method returns the main object itself, and--
+
+**Johnny Boursiquot:** Right, you chain your calls.
+
+**Mat Ryer:** ...then lets you chain them. Yeah, right
+
+**Jon Calhoun:** And I should say, I hate these... Because you can't define interfaces that work with them, at all. It's just a nightmare. So I do have one question, Johan... Could you reconstruct the Squirrel package using functional options?
+
+**Johan Brandhorst:** Potentially...
+
+**Jon Calhoun:** Because I've looked at GORM, which is a big ORM for Go, and I'm pretty sure you could rewrite a vast majority of it using that; I just don't think that was as popular when those libraries were written.
+
+**Johan Brandhorst:** You're right that it's annoying to use, because you can't construct an interface that works with it... In the cases where I have been using it, it's usually just like in a single API layer; you're just using it through that package, and you don't really have to pass around in any kind of generic way.
+
+And also, Squirrel does provide a few interfaces for working with the standard library. It has a BaseRunner, or whatever... An interface that describes how sql rows behaves, so that you can operate on that and pass that through functions if you want to.
+
+I find it the one exception to this rule, but generally, the builder pattern and (like Mat described) fluent interfaces like that just don't work very well in Go. That's my unpopular opinion.
+
+**Mat Ryer:** It is a bit unpopular, because I don't know that I would agree. I'm just looking at the syntax... You're sort of writing strongly-typed code, so there's that; because obviously, the alternative is just a SQL string, I assume... But you lose that sqlness, don't you? It doesn't say "select star from table." You lose that a little bit. But you do get type safety.
+
+**Jon Calhoun:** The reason I ask about the functional options one is that if everything returns the same object, that means that every option function is essentially just going to be a function that accepts that one argument, and then you can just list a bunch of them as things you're passing into like squirrel.query, and then you pass in all your options for it. I think that's the type of library that probably could be rewritten like that, and it might work a little bit easier in some ways... I'm not positive...
+
+**Johan Brandhorst:** Me too. \[laughs\]
+
+**Mat Ryer:** It'd be a good exercise. I'd like to see you try that, Jon... Live, on Twitch.
+
+**Johnny Boursiquot:** \[laughs\]
+
+**Jon Calhoun:** Well, I've done it with GORM a little bit. Not the whole thing, because that would just take forever, but I've definitely toyed with doing some of it, so I know there's -- the hard part is there might be some weird edge case I don't know about, and you almost have to do the entire thing exhaustively to see if it works.
+
+**Mat Ryer:** Yeah. But talking about these abstractions, and things - we have in Pace a very light abstraction over the data store. It's a schemaless document store. So the get and the put operations are abstracted in this kind of really light way, but what it allows us to do is check -- because obviously, it's a remote service, and sometimes that can fail with temporary errors... So what we can do in that abstraction is check if there was like just a network error, and just try again. And then you get a kind of robustness for free.
+
+And again, a bit like the example earlier, of using SQL most of the time, and then you can break that rule and use different ones, I think that is quite a nice way to do it. Use the abstraction -- it's not gonna be 100%, but use it for the 80% of the cases, and then you can do perhaps more in-depth things after, in those cases where you really need it. So yeah, I like that... I've definitely got a lot to think about for this.
+
+\[01:08:16.15\] Just one last question then, Johan... One of the nice things about these kinds of builder tools and packages is that you protect against some security considerations like SQL injection. This is where you're talking about using question marks and dollar symbols for parameters - it's instead of just building strings up yourself, isn't it? Why is that important?
+
+**Johan Brandhorst:** The danger of an SQL injection is that you provide some sort of user input that maybe prematurely interrupts your SQL statement and then construct its own SQL statement, and in such a way could make changes in the database, or extract data from the database in a very dangerous way. And the way that you protect against that sort of thing is by using these placeholders or extrapolated variables. They have many different names...
+
+And the way that Squirrel makes this easier is by -- like, when you normally use the Go standard library database/SQL interface, you may be tempted to just use `fmt.sprintf` to construct your queries... This is kind of the major danger, where you might end up causing an SQL injection without thinking about it.
+
+If you've ever found yourself "Oh, I don't really want to write out this very long SQL query. I don't want to have to map the names myself. I'll just use a `fmt.sprintf` in a helper package somewhere", and then accidentally you may have gotten the argument wrong or something like that, and you ended up with something that the user can manipulate inside of your query - that's super-dangerous, and a sure way to just avoid that thing altogether is to use Squirrel, because all of the variables that you put into the builder automatically become interpolated variables.
+
+**Johnny Boursiquot:** I need to ask, have you met Bobby Tables?
+
+**Johan Brandhorst:** I have met Bobby Tables. \[laughter\] This is a great one, actually... So the reference that Johnny is making there is to an XKCD comic, which has a very illustrative way of showing just exactly what SQL injection means. There's a school principal who's making a call to a concerned parent (I suppose), asking about their son, Bobby Tables -- DROP TABLE students, or something like that... The parent says "Yes, we called him Bobby Tables" and then the principal says "Well, I hope you're happy; we've lost this year's student records."
+
+It's really funny, because obviously, the implication there is that they had to enter their kid's name somewhere, in some sort of form, and they thought "Hey, it'd be fun to see if this is vulnerable to a SQL injection." And then they put in a command that if it was vulnerable to a SQL injection, dropped the table called STUDENTS... And of course, the joke then is that it actually did, and the principal was furious about it. And the lesson, I guess, is that you shouldn't have been vulnerable to SQL injection.
+
+**Mat Ryer:** Then he's got no job though, so it's no good telling him that, is there? His life's devastated by that, but... \[laughter\] Yeah, I did start telling a story once, and we had Felippo, a security engineer on the Go team actually, just stopped me halfway and said "Just don't tell that story..." It was a similar kind of story about SQL injection. A good lesson for all... Yeah, he stopped me from telling it, which is a good -- but if you see me in the real world, I'll tell you my story about SQL injection. Yeah, but don't worry, kids, you won't be SQL injected. I don't know what that means...
+
+**Johnny Boursiquot:** \[laughs\]
+
+**Mat Ryer:** Don't worry, that bit will be cut out...
+
+**Johnny Boursiquot:** That didn't come out right. \[laughs\]
+
+**Mat Ryer:** No, no. I was gonna say - don't worry; if you use it the right way, it's not something to be scared of... If you use Squirrel, and things.
+
+**Jon Calhoun:** You're gonna have to change the rating for the podcast now, Mat... \[laughter\]
+
+**Mat Ryer:** I'm just gonna make a quick note of the timestamp that I said that at, for editing purposes...
+
+**Jon Calhoun:** It's like "PG up until the 58-minute mark, and then - nope."
+
+**Mat Ryer:** Yeah. That's actually how we should do the podcast. It should be an hour, and then an alarm goes off, and then it's like the purge. We can say whatever we want. \[laughter\]
+
+**Johan Brandhorst:** It's the watershed in the U.K, right?
+
+**Jon Calhoun:** If your kids are in the car, you need to close their ears...
+
+**Mat Ryer:** Yeah.
+
+**Johnny Boursiquot:** \[laughs\] Awesome.
+
+**Mat Ryer:** Well, this was a great show. Thank you so much. It was great to learn so much about Postgres. I think it's nice for junior people to know as well that it's a perfectly reasonable choice. Pick it up, get going with it, learn it, see what you can do with it, see what you can build.
+
+Thank you so much to our special guest, Johan Brandhorst, and Johnny Boursiquot and Jon Calhoun who were here also. We will see you next time on Go Time. Goodbye!
+
+**Outro:** \[01:13:20.27\]
+
+**Johnny Boursiquot:** And it's Johaen, not Johan.
+
+**Johan Brandhorst:** Yeah, it depends where you are, to be honest... Johaen works...
+
+**Jon Calhoun:** Do you prefer Johan?
+
+**Mat Ryer:** How do you say it?
+
+**Johan Brandhorst:** Johaen, I think...
+
+**Johnny Boursiquot:** You think? \[laughs\]
+
+**Mat Ryer:** You don't really say your own name...
+
+**Johan Brandhorst:** I don't. Exactly.
+
+**Mat Ryer:** When do you need to say your own name?
+
+**Jon Calhoun:** When your mom is mad at you, what does she yell?
+
+**Johan Brandhorst:** That's a different question, because my mom is Swedish, and in Sweden we would say Juwan... But in English -- no one with an English background will get that right, unfortunately... And it sounds butchered, no matter how much they try.
+
+**Johnny Boursiquot:** It's actually "You won" something.
+
+**Johan Brandhorst:** Yeah, but it's like the Irish Ewan, like Ewan McGregor.
+
+**Johnny Boursiquot:** Ewan, right, right.
+
+**Jon Calhoun:** So if we wanna properly introduce you, we have to call your mom up, record her saying it...
+
+**Johnny Boursiquot:** Ewan.
+
+**Jon Calhoun:** ...and just slip that in there.
+
+**Mat Ryer:** Pardon, why is your name different when you're in trouble?
+
+**Johnny Boursiquot:** \[laughs\]
+
+**Johan Brandhorst:** Yeah...
+
+**Mat Ryer:** Your names change in Sweden if you get in trouble. It's like good cop/bad cop, or something. I don't know, I just wanna learn...
+
+**Johan Brandhorst:** We should have another episode about that.
+
+**Mat Ryer:** \[laughs\]
+
+**Johnny Boursiquot:** Why? Why not both? We can do both right now.
+
+**Mat Ryer:** Yeah. Well, okay... Okay, hint taken. \[laughter\] Drop it, don't mention it. Okay, fair play. I'll take a hint, and then I won't mention it again. That's me; you can rely on me for that.
+
+**Jon Calhoun:** I feel like Mat is making me lie today. I'm like "We're talking about Postgres in Go", and it's like "No, we're not. We're not talking about that at all right now."
+
+**Mat Ryer:** We should though. That would be good, if we could.
+
+**Jon Calhoun:** Alright, is everybody recording locally?

Füźžįñg_transcript.txt

@@ -0,0 +1,1503 @@
+[0.00 --> 1.10]  Fuzzy Wuzzy was a bear.
+[1.20 --> 2.12]  Fuzzy Wuzzy had no hair.
+[2.60 --> 4.10]  Fuzzy Wuzzy wasn't very fuzzy, was he?
+[4.36 --> 4.80]  That's impressive.
+[4.82 --> 6.56]  Oh, mate, that's got to be the opening, isn't it?
+[7.06 --> 7.92]  Yeah, it really does.
+[9.06 --> 9.58]  Yeah, deal.
+[9.88 --> 10.24]  Done.
+[13.06 --> 15.64]  Bandwidth for Changelog is provided by Fastly.
+[16.00 --> 17.88]  Learn more at Fastly.com.
+[18.14 --> 21.22]  We move fast and fix things here at Changelog because of Rollbar.
+[21.22 --> 23.02]  Check them out at Rollbar.com.
+[23.28 --> 25.46]  And we're hosted on Linode cloud servers.
+[25.80 --> 27.80]  Head to Linode.com slash Changelog.
+[30.00 --> 33.06]  This episode is brought to you by Digital Ocean.
+[33.50 --> 34.10]  Droplets.
+[34.44 --> 35.22]  Managed Kubernetes.
+[35.58 --> 36.44]  Managed databases.
+[36.96 --> 37.48]  Spaces.
+[37.80 --> 38.68]  Object storage.
+[38.96 --> 40.20]  Volume block storage.
+[40.46 --> 43.94]  Advanced networking like virtual private clouds and cloud firewalls.
+[44.14 --> 50.38]  Developer tooling like the robust API and CLI to make sure you can interact with your infrastructure the way you want to.
+[50.38 --> 54.32]  Digital Ocean is designed for developers and built for businesses.
+[54.32 --> 61.36]  Join over 150,000 businesses that develop, manage, and scale their applications with Digital Ocean.
+[61.72 --> 65.16]  Head to do.co slash Changelog to get started with a $100 credit.
+[65.58 --> 67.64]  Again, do.co slash Changelog.
+[80.18 --> 81.14]  Let's do it.
+[81.72 --> 82.78]  It's go time.
+[82.78 --> 84.94]  Welcome to go time.
+[85.12 --> 88.24]  Your source for diverse discussions from around the go community.
+[88.82 --> 95.08]  We would like to thank each and every one of you who joined Changelog++ during our soft launch in August.
+[95.54 --> 97.74]  We truly appreciate you supporting the show.
+[97.74 --> 104.16]  If you have no idea what I'm talking about, check it out and learn more at changelog.com slash plus plus.
+[104.74 --> 106.48]  Okay, it's time to talk fuzzing.
+[107.04 --> 108.28]  Here we go.
+[112.78 --> 117.50]  Hello and welcome to go time.
+[117.72 --> 118.68]  I'm Matt Ryer.
+[118.92 --> 120.96]  Today we're talking about fuzzing.
+[121.24 --> 126.28]  We're going to find out what it is and how we can use it to make our code better.
+[126.82 --> 134.64]  And we're going to take a close look at a new draft design that discusses bringing fuzzing as a first class concern to go.
+[135.06 --> 135.70]  It's very exciting.
+[135.82 --> 140.32]  And we're lucky to be joined by the author of that draft design, Katie Hockman.
+[140.42 --> 140.92]  Hello, Katie.
+[141.32 --> 141.82]  Hi, Matt.
+[141.82 --> 142.32]  How's it going?
+[142.86 --> 143.18]  Good.
+[143.28 --> 144.04]  Welcome to the show.
+[144.14 --> 144.68]  Thanks for coming.
+[145.12 --> 146.04]  Thank you for having me.
+[146.64 --> 149.72]  We're also joined by Filippo Valsorda.
+[149.84 --> 150.80]  Hello, Filippo.
+[151.12 --> 151.60]  Hey, Matt.
+[151.66 --> 152.26]  Good to be back.
+[152.60 --> 154.96]  Always a pleasure to have you here, sir.
+[155.38 --> 156.20]  Same, same, same.
+[156.50 --> 157.40]  Looking forward to it.
+[157.68 --> 158.20]  Very formal.
+[158.78 --> 159.00]  Yeah.
+[159.14 --> 159.56]  Thank you.
+[160.24 --> 163.52]  And we're also joined by Roberto Clapis.
+[163.76 --> 164.62]  Hello, Roberto.
+[165.98 --> 166.30]  748.
+[167.80 --> 168.24]  Okay.
+[168.24 --> 170.72]  Is that a fuzzed response?
+[170.72 --> 170.96]  Yeah.
+[171.38 --> 173.26]  I wanted to see if you crashed with Integer.
+[174.30 --> 176.34]  I've not crashed, though, nor am I panicking.
+[176.68 --> 177.46]  I've continued.
+[178.46 --> 180.90]  In fact, that was in my unit test earlier.
+[181.22 --> 183.18]  So I was ready for it.
+[183.66 --> 184.50]  But thank you very much.
+[184.56 --> 185.30]  Welcome to the show.
+[185.52 --> 185.82]  Thanks.
+[185.82 --> 190.88]  Can we take a second to acknowledge how Matt rolled the R's for both the Italian names?
+[191.60 --> 191.72]  Yeah.
+[191.78 --> 192.52]  Oh, it's my pleasure.
+[192.62 --> 192.96]  That was good.
+[195.34 --> 196.46]  It's a lovely accent.
+[196.70 --> 200.36]  So I always like to listen to it and have you on for that purpose, really.
+[200.74 --> 204.88]  So if that's all you contribute to this show, then that's fine by me.
+[205.36 --> 206.26]  That's our intention.
+[206.26 --> 209.86]  Well, so maybe we should start at the beginning then.
+[209.94 --> 214.08]  For anybody not familiar, what is fuzzing and what's it for?
+[214.42 --> 216.62]  Yeah, I can give a quick summary of that.
+[216.80 --> 223.44]  So basically, fuzzing is a form of automated testing that can manipulate inputs in a way
+[223.44 --> 228.04]  that can find bugs that maybe you wouldn't otherwise be able to find on your own.
+[228.04 --> 233.24]  So in my mind, it's kind of a supplement to some of the existing testing that people already do.
+[233.52 --> 237.12]  That's pretty common, like unit testing or integration testing.
+[237.34 --> 244.04]  But what sets it apart is it actually does things on its own and runs and can run continuously.
+[244.34 --> 246.02]  So it's kind of smart in a way.
+[246.16 --> 251.26]  So if it has some interesting inputs, it can actually use some intelligence to go in and
+[251.26 --> 256.62]  mutate those inputs in interesting and meaningful ways to find crashes and panics
+[256.62 --> 262.44]  that wouldn't easily be otherwise found if the developer had to try to identify them themselves.
+[262.88 --> 265.02]  So that's interesting then you talk about this intelligence.
+[265.32 --> 267.64]  It isn't just random then.
+[267.80 --> 269.62]  There's something else going on.
+[270.02 --> 276.26]  Yeah, and I think it's really tricky because there's no industry standard on how these kinds
+[276.26 --> 277.16]  of things work.
+[277.16 --> 281.26]  I mean, there are definitely tons of different ways that you can mutate things randomly.
+[281.76 --> 285.96]  And there's also a lot of interesting discussion around how do you prioritize
+[285.96 --> 288.12]  which corpus entries?
+[288.76 --> 292.60]  And I'll talk a little bit more about what corpuses are later, but basically which inputs
+[292.60 --> 297.56]  to modify and how to modify them and how smart it should really be.
+[297.94 --> 303.02]  And all those things are kind of up in the air and a lot of different fuzzers work differently,
+[303.60 --> 305.22]  which is actually kind of cool in my mind.
+[305.88 --> 306.66]  Yeah, that's interesting.
+[306.66 --> 310.32]  So what situations is it good for helping out in then?
+[310.32 --> 315.92]  So let's say the string.split example from the standard library, you pass in a string
+[315.92 --> 321.12]  and you pass in a separator and it basically just splits that string wherever it finds that
+[321.12 --> 325.46]  separator and returns a slice of the components, the segments that it found.
+[325.72 --> 327.62]  Would that be a good candidate for fuzzing?
+[328.18 --> 329.32]  Yeah, I think it could be.
+[329.52 --> 334.10]  And I think it'd also be, I think Filippo and Rob will have a lot of really good things
+[334.10 --> 339.02]  add to in terms of who's used fuzzers in the past and how they've usually had a security
+[339.02 --> 340.24]  context around them.
+[340.80 --> 345.46]  And what this proposal has been trying to do is actually get fuzzing into the hands of
+[345.46 --> 349.38]  non-security experts and non-security developers and have other people use them.
+[349.52 --> 354.02]  So in the string split example, you know, if there's an off by one error somewhere or maybe
+[354.02 --> 359.84]  some issue that can cause a panic or some input that doesn't meet some specific property,
+[359.84 --> 361.62]  it might be easy to find with fuzzing.
+[361.62 --> 366.14]  And I think it would be a really good package to test or a good function to test in that
+[366.14 --> 366.50]  package.
+[366.70 --> 366.72]  Yeah.
+[367.10 --> 367.30]  Yeah.
+[367.32 --> 372.38]  Because you hear a common use cases are things like parsers and things that are doing decoding
+[372.38 --> 378.68]  because, you know, they are dealing with usually unknown in advance kind of structures that
+[378.68 --> 380.78]  maybe they have to sort of infer along the way.
+[381.02 --> 386.68]  So there is a lot of room in that kind of operation for things to go wrong or unexpected input,
+[386.78 --> 390.28]  just things that you would never imagine anyone would pass in.
+[390.28 --> 395.20]  And so, yeah, that's what separates it out from unit tests, I guess, really, because
+[395.20 --> 397.12]  unit tests are very deliberate, aren't they?
+[397.42 --> 397.58]  Yeah.
+[397.90 --> 398.40]  Yeah, they are.
+[398.50 --> 404.64]  I mean, you give a set of inputs and you run something and then you look at the output and
+[404.64 --> 405.34]  it's very clear.
+[405.48 --> 409.90]  And you have to say, these are the inputs that I think are important that should test
+[409.90 --> 410.52]  it well enough.
+[410.52 --> 412.20]  And then it should have this output.
+[412.90 --> 417.40]  And fuzzing, I think, can apply to a lot of context beyond parsers and things like that,
+[417.40 --> 420.58]  because there's a reason that we have unit tests everywhere.
+[420.58 --> 426.16]  And there's a reason we don't just test parsers or difficult cryptography or things like that.
+[426.22 --> 427.42]  There's a reason we test everything.
+[427.50 --> 430.92]  And it's because we don't always know where the bugs in our code are.
+[430.92 --> 433.54]  You know, we have default assumptions that our code works.
+[433.66 --> 439.44]  And so we just kind of test it kind of just in good faith sometimes just, you know, to
+[439.44 --> 440.32]  prove that it works.
+[440.80 --> 445.24]  And I think fuzzing, a fuzzing engine should be pretty agnostic in the sense that it doesn't
+[445.24 --> 449.36]  assume that it's going to work and it's going to go and maybe find things that you didn't
+[449.36 --> 453.54]  really realize could actually break or something that you had overlooked that you didn't realize
+[453.54 --> 455.44]  is a dependency somewhere else that might break.
+[455.44 --> 456.36]  Right.
+[456.52 --> 463.20]  I would also add that since when you write the fuz test target, kind of, you want to
+[463.20 --> 466.44]  expect on properties of the stuff that you work on.
+[466.44 --> 470.32]  Instead, when you work on unit tests, you expect some output.
+[470.74 --> 474.76]  For example, in the string split case, you can say, I'm going to call a string split with
+[474.76 --> 475.32]  two parameters.
+[475.84 --> 481.78]  And I'm going to check that the second one never appears in the return slices because the
+[481.78 --> 482.96]  separator should never appear.
+[482.96 --> 483.44]  Right.
+[483.68 --> 487.38]  And that is something you would generally not test in a unit test.
+[487.46 --> 492.94]  Or like you're going to check that the returned slices are less than the characters of the
+[492.94 --> 493.20]  string.
+[494.12 --> 498.50]  So like if you return more than characters there are, there must be a problem.
+[498.98 --> 501.78]  And this is stuff that normally doesn't get tested.
+[502.32 --> 503.50]  I'm pretty bad at writing tests.
+[503.66 --> 506.94]  But when I write unit tests, I don't test for this kind of condition.
+[506.94 --> 507.54]  Yeah.
+[507.70 --> 512.76]  In another example of something that would be good to check in a fuzz test of the split
+[512.76 --> 518.32]  function is that if you put it back together, putting the separators between the things you
+[518.32 --> 520.78]  split, do you get back the regional string?
+[521.26 --> 523.52]  If you do, it probably did its job right.
+[523.94 --> 528.10]  And that's the kind of stuff that fuzzers are pretty good at finding because they can just
+[528.10 --> 533.42]  go and find some input where, I don't know, the separator is at the end and it's missing
+[533.42 --> 538.28]  one character or I don't know, where the thing doesn't round trip.
+[538.72 --> 543.00]  That gives you even more because then you now are testing for an additional property,
+[543.18 --> 547.46]  which is if you string split and then you string join, you must get the same thing out,
+[547.62 --> 548.86]  which is a normal expectation.
+[549.06 --> 551.94]  I mean, when I use the strings package, I expect that to be true.
+[552.32 --> 555.76]  But I don't know if there is anyone that has been fuzzing that to make sure that that
+[555.76 --> 561.76]  is actually true, especially on edge cases like nil slices or slices of empty strings.
+[561.76 --> 564.40]  What happens would be interesting to see.
+[565.30 --> 565.44]  Yeah.
+[565.58 --> 568.08]  So there is an element then of design here.
+[568.22 --> 573.82]  You have to think of that kind of thing, that property to then model it in a fuzz test,
+[573.90 --> 574.12]  right?
+[574.66 --> 579.86]  It's not just you just point it to a method and it just fills the method up with nonsense.
+[580.42 --> 581.32]  I think yes and no.
+[581.38 --> 582.04]  I think it can.
+[582.12 --> 584.02]  I think it depends on what you're using it for.
+[584.42 --> 588.84]  I mean, you could just throw random input at a function and just see if it panics.
+[588.84 --> 592.96]  Like that is a property that can be tested and you don't have to know anything about
+[592.96 --> 593.14]  it.
+[593.54 --> 597.94]  I think it can also be used for things like differential testing or property testing or
+[597.94 --> 601.18]  a lot of different things that can be a supplement to your unit test, but it can also just go
+[601.18 --> 601.80]  find a crash.
+[602.18 --> 604.30]  And you could probably do that in a couple of lines with a little thought.
+[604.90 --> 609.44]  Differential testing is something that honestly works a little too well.
+[610.00 --> 613.70]  The idea is that there are multiple implementations of the same thing, right?
+[613.70 --> 618.00]  For example, big number implementations.
+[618.36 --> 619.78]  It doesn't matter what library you use.
+[619.80 --> 625.94]  If you multiply two arbitrary precision decimals, you should get the same arbitrary precision
+[625.94 --> 626.88]  decimal out.
+[627.46 --> 628.40]  Sounds right, right?
+[629.24 --> 633.82]  Oh, my friend, how many bugs fuzzers have found just by telling them.
+[634.22 --> 634.40]  Yep.
+[634.54 --> 635.46]  So here's two functions.
+[635.58 --> 636.62]  They need to return the same thing.
+[636.84 --> 637.02]  Cool.
+[637.18 --> 637.34]  Go.
+[637.34 --> 642.74]  I get emails because one of them, the one that are tested is the Go one.
+[643.20 --> 648.66]  And I get emails when there's a mismatch between the Go one and some other implementation.
+[649.12 --> 650.02]  And oh, boy.
+[651.86 --> 652.34]  Yep.
+[652.62 --> 653.56]  Multi-precision is hard.
+[654.08 --> 655.58]  So yeah, that's an excellent example.
+[656.18 --> 662.54]  One thing that I did with differential testing was at one point in Go, a bug was fixed with a
+[662.54 --> 664.00]  problem with header parsing.
+[664.00 --> 668.70]  And I thought, this looks easy to find with a fuzzer.
+[668.94 --> 675.50]  So I just imported fast HTTP and the standard HTTP libraries, both in Go, run GoFuzz for
+[675.50 --> 676.20]  25 minutes.
+[676.34 --> 677.00]  And I found the bug.
+[677.62 --> 680.70]  The bug that was just fixed and has been there for 12 years.
+[681.46 --> 687.00]  So yeah, if you want to assert for a property, and the case was, I want the header set to
+[687.00 --> 687.54]  be identical.
+[687.96 --> 690.14]  It's quite easy to find problems.
+[690.14 --> 696.88]  And if I recall correctly, at one point, the JSON package was optimized, heavily optimized,
+[697.34 --> 702.00]  and there was a differential fuzzer in place that checked that the old version and a new
+[702.00 --> 704.28]  version would parse the JSON the same way.
+[704.66 --> 710.94]  And it found a bug before it hit a stable release, which would have been kind of bad.
+[710.94 --> 718.06]  So that was another kind of success story of fuzzing, not for security reasons.
+[718.42 --> 720.06]  That would just get another test.
+[720.66 --> 723.06]  So how can you do fuzzing in Go today then?
+[723.32 --> 725.12]  What are the choices that we have?
+[725.34 --> 725.98]  There are a few.
+[726.20 --> 728.94]  I mean, I can speak to at least one or two.
+[729.06 --> 731.20]  I mean, I think the common one is GoFuzz.
+[731.40 --> 733.34]  That's the one that everyone knows about.
+[733.44 --> 734.94]  That's Go-Fuzz.
+[734.94 --> 735.46]  Fuzz.
+[735.92 --> 738.66]  And that was written primarily by Dmitry Vykov.
+[739.36 --> 742.58]  And yeah, I mean, it's really, really amazing.
+[742.84 --> 744.86]  And I've spoken to him about it.
+[744.90 --> 750.28]  And he's actually given a lot of really, really good feedback into the proposal that's out
+[750.28 --> 750.66]  there now.
+[751.08 --> 755.50]  So it's been nice to partner with him a little bit on that, too, and have him get some feedback
+[755.50 --> 756.02]  on that.
+[756.60 --> 758.24]  And yeah, I mean, it's really neat.
+[758.32 --> 760.34]  And if you haven't used it, you should definitely check it out.
+[760.34 --> 764.20]  And then another tool that somebody wrote was FuzzGo.
+[764.44 --> 765.72]  It was F-Z-G-O.
+[766.24 --> 770.84]  And I think that was kind of a proof of concept written mostly by the pods, it sounded like,
+[771.20 --> 775.34]  to try to integrate it a little bit more kind of with the Go command and making it look
+[775.34 --> 780.36]  more like kind of an end-to-end tool that wouldn't have to have so many build steps like
+[780.36 --> 781.20]  GoFuzz has.
+[781.58 --> 785.04]  And add a little bit of support for modules, I think, was part of that.
+[785.10 --> 786.56]  Or maybe that was a part of GoFuzz.
+[786.56 --> 790.70]  But there's been different features that both of them have tried to basically model
+[790.70 --> 791.72]  and see how they would work.
+[792.10 --> 796.68]  And I think FuzzGo was meant to be kind of a prototype or an experiment of what it might
+[796.68 --> 797.90]  look like as a final approach.
+[799.12 --> 801.24]  And you mentioned build steps there then.
+[801.42 --> 804.56]  So it isn't just something at all that runs at runtime.
+[804.86 --> 806.26]  There's other things that happen.
+[806.36 --> 810.90]  Is there some kind of introspection that happens or reflection on the types and things?
+[810.96 --> 812.58]  Is it kind of generic in some way?
+[812.58 --> 817.72]  Well, when I say build steps, I don't remember all of the exact details of how GoFuzz works.
+[817.74 --> 820.78]  But I do know that it has kind of like a GoFuzz build.
+[820.88 --> 823.98]  And you have to kind of build the binary that will be fuzzed.
+[824.10 --> 827.34]  And then you have to run it separately and kind of manage your own corpus.
+[827.48 --> 828.82]  And so there's a lot of different steps.
+[828.90 --> 832.86]  You can't just run one command with the Go toolchain as it is today.
+[832.94 --> 836.78]  You kind of have to learn a different workflow, which was a bit of a...
+[836.78 --> 841.08]  Just like an impediment for some people to try to start it.
+[841.84 --> 843.38]  Because they didn't want to learn a new tool.
+[843.84 --> 847.98]  I would say that that is one of the main reasons why people are not using it.
+[848.06 --> 850.82]  It's because it's external kind and feels different.
+[851.06 --> 854.12]  Also, one thing that it does, it does a source to source transformation.
+[854.32 --> 857.58]  So it takes your source code and implements some sort of checkpoints.
+[857.58 --> 862.48]  So basically, when your code runs, it can check at which point it got.
+[863.06 --> 867.08]  So basically, while your code executes, it can check how much of the code was covered.
+[867.18 --> 868.56]  More or less like the cover tool.
+[869.24 --> 873.20]  But it needs to do it more heavily than the cover tool and in a more efficient way.
+[873.82 --> 877.72]  And this is one of the reasons why it was quite hard to make it support modules.
+[878.10 --> 880.04]  Because it actually rewrites the sources.
+[880.04 --> 887.04]  Yes, some context here is that part of what makes Fuzzer's magic is that...
+[887.04 --> 891.48]  Well, the recent generation of Fuzzer's magic since, I think, AFL.
+[891.86 --> 898.80]  Is that they use coverage to figure out what mutations are the ones that are interesting to look at.
+[899.12 --> 902.64]  Katie was talking about how there's different strategies for these.
+[903.00 --> 909.02]  But in general, the common denominator is that they all look at the cover of your code.
+[909.02 --> 913.18]  If you ever run go test-cover profile.
+[913.54 --> 915.30]  Oh boy, I don't remember the flag.
+[915.40 --> 919.52]  But anyway, if you ever generated the coverage report, you know, with the green and the red.
+[919.92 --> 920.98]  That's what Fuzzer's do.
+[921.08 --> 924.72]  They run the input and check which parts light up.
+[925.14 --> 929.64]  And if they change the input and some new code lights up, the Fuzzer goes like,
+[929.70 --> 929.98]  Aha!
+[930.30 --> 931.34]  Okay, this is useful.
+[931.44 --> 936.08]  I can keep changing this and maybe I'll hit another path that takes from there.
+[936.08 --> 941.18]  Or maybe I'll be able to combine two paths in a way that we're not tested together.
+[941.18 --> 946.62]  And that's what makes them honestly kind of freakishly effective.
+[947.26 --> 954.32]  There's this demo of AFL slowly building a valid JPG out of nothing.
+[954.32 --> 962.36]  And it slowly makes a picture and it figures out the letters to put in the tags and everything.
+[962.66 --> 963.36]  It's very good.
+[963.68 --> 969.42]  One thing that really scared me was when I run gofuz against the HTTP library.
+[969.72 --> 975.02]  And after a while, I saw that in the corpus, something that looked like random started appearing.
+[975.28 --> 976.30]  And I was like, oh, cool.
+[976.30 --> 983.24]  So the Go package, the GoH standard package started accepting something that is not HTTP because it was HTTP2.
+[983.66 --> 988.38]  Basically, I started constructing valid HTTP2 requests from nothing.
+[989.14 --> 990.56]  And that was scary.
+[990.94 --> 993.36]  And also, I was ashamed because I didn't recognize it.
+[993.42 --> 996.92]  And I had to manually write to decompress it and see what was going on.
+[997.28 --> 1002.66]  Rob, if you can ever read HTTP2 to the naked eye, you need to tell me.
+[1002.66 --> 1006.62]  Yeah, because that is a strange superpower.
+[1007.06 --> 1011.46]  I don't know what has had to bite you for that to be the power that then manifests.
+[1011.94 --> 1013.00]  There are support groups.
+[1013.48 --> 1014.44]  We've all been there.
+[1014.72 --> 1015.74]  Mine is TLS.
+[1016.60 --> 1017.84]  Used to be DNS.
+[1018.28 --> 1019.20]  It's okay.
+[1020.36 --> 1020.92]  There's help.
+[1022.18 --> 1022.78]  Thanks.
+[1023.06 --> 1024.22]  Thanks for keeping them in mind.
+[1024.84 --> 1032.28]  So that is really interesting then that it's not just shifting the inputs like by some external means.
+[1032.28 --> 1039.34]  It actually has an insight into the code that's running inside in your own code, in your own binary.
+[1039.56 --> 1043.42]  And it uses that information to also influence what it's doing.
+[1043.60 --> 1045.92]  So that is kind of like spooky.
+[1046.16 --> 1047.26]  I could definitely imagine.
+[1047.50 --> 1053.28]  It's a little bit like adversarial training in machine learning where you have a model and you have another model.
+[1053.42 --> 1055.06]  And they sort of compete with each other.
+[1055.06 --> 1058.52]  And then they both just keep getting better, you know, together.
+[1058.70 --> 1060.06]  And that's kind of a great way to...
+[1060.58 --> 1062.38]  It's almost like feels like cheating in some way.
+[1062.38 --> 1067.94]  But you can end up with a mirror of something else, you know, by this technique.
+[1068.02 --> 1069.48]  So it is kind of amazing.
+[1069.60 --> 1072.38]  And yeah, to see it, it really will start to feel intelligent.
+[1073.14 --> 1075.48]  And a few of you have said it's kind of spooky, this thing.
+[1076.44 --> 1083.38]  Another, just a note of another spooky thing is you can also kind of like reverse engineer your code such that it can figure out...
+[1083.38 --> 1090.40]  There are certain tools that can figure out what the input is actually supposed to be and then kind of do that for you.
+[1090.44 --> 1095.54]  So it can actually basically tell the fuzzing engine, like this is what input will make this if statement pass.
+[1095.80 --> 1102.42]  And then it'll just do that to kind of get unstuck from wherever you're at with the fuzzing engine.
+[1103.00 --> 1104.56]  And that's something maybe you do...
+[1104.56 --> 1110.86]  Like I think GoFuzz does this once every thousand mutations just to try to unstick it, but not every time because it's too expensive.
+[1110.86 --> 1114.22]  And so it's a lot of trade-offs of like how random do you want this to be?
+[1114.62 --> 1118.90]  How much do you want to use prioritization of certain inputs?
+[1119.04 --> 1123.92]  How much coverage is coverage a metric in terms of what is that in terms of feedback loop?
+[1124.02 --> 1126.78]  Like how much do you care about it in terms of other things?
+[1126.82 --> 1130.82]  And so it's kind of creepy and it's a judgment call from the developer on how they want to design that too.
+[1131.36 --> 1134.40]  Yeah, it does sound like a kind of hacker's tool, doesn't it?
+[1134.44 --> 1137.92]  And in fact, did it have its origins in the security world?
+[1137.92 --> 1143.70]  Yeah, but I like what Katie just said that it's a trade-off that the developer has to make.
+[1144.18 --> 1148.38]  And I think she meant the developer of the fuzzing tool, correct me if I'm wrong?
+[1148.38 --> 1149.04]  Yes, yeah.
+[1149.38 --> 1150.78]  Because that's the thing.
+[1151.10 --> 1163.82]  The thing I like about the proposal is that it does not leave all these decisions and the necessity to learn about all this stuff to the end users, to the Go developers that are just trying to test their code.
+[1163.82 --> 1171.44]  Yeah, and also if you look at the proposal, it tries to make fuzz test targets as close as possible to what a test looks like nowadays.
+[1172.42 --> 1179.46]  So basically the friction to adopt fuzzing if you're used to write unit tests, and if you're not, you should, is going to be very low.
+[1180.30 --> 1187.02]  Because it's going to basically slightly change the pattern, but it's going to be as close as possible.
+[1187.02 --> 1190.82]  Yeah, we should talk more about that proposal.
+[1191.04 --> 1194.18]  But before we do, I just want to get a few other concepts kind of clear.
+[1194.70 --> 1196.96]  There's this concept of seeding the corpus.
+[1197.10 --> 1201.38]  There's this concept of kind of giving the fuzzing tool some kind of head start.
+[1201.88 --> 1207.40]  A bit like with unit tests where you say, you know, we know these are the inputs and these are the expected outputs.
+[1207.88 --> 1211.96]  You also kind of seed the fuzzing tool in a similar way, don't you?
+[1211.96 --> 1227.24]  Yes, and I think it's also kind of a goal of the proposal to try to make it such that the unit tests that people have now and the use cases that they've already come up with can basically just be directly used as seed corpus.
+[1227.24 --> 1234.42]  And so the seed corpus is kind of filling two needs, at least in terms of this goal proposal.
+[1235.52 --> 1238.08]  It's first of all seeding the mutation engine.
+[1238.24 --> 1242.26]  It's seeding that the corpus trying to tell it this is a good starting point for you.
+[1242.84 --> 1251.00]  Build off of this and then it can manage its own corpus on its own as it wants to and build it up as it finds new coverage and new interesting things.
+[1251.20 --> 1255.24]  But it's also can serve as a regression test of sorts.
+[1255.24 --> 1259.74]  The seed corpus is either checked into basically your test data directory.
+[1259.88 --> 1265.40]  It's basically checked in directly into your module or into your package or it's in there programmatically.
+[1265.58 --> 1266.90]  It's in your test in code.
+[1267.86 --> 1270.98]  And so that's run every single time go test is run.
+[1271.80 --> 1273.70]  And so it's also meant to act as a regression test.
+[1273.84 --> 1275.80]  So you can use existing things.
+[1275.88 --> 1282.82]  You can use new crashes and you can build out that seed corpus as you find new regressions that you want to make sure you're testing.
+[1282.82 --> 1293.50]  Yeah, so that's a really cool feature that if something fails, that automatically gets contributed to the testing so that next time that will explicitly get tested.
+[1293.72 --> 1295.22]  Is that how it works?
+[1295.22 --> 1306.80]  Yeah, so that is very cool because, of course, the value of unit testing, if in the case where you find a bug and then you write a test to prove that bug, which you do if you follow TDD tightly.
+[1307.34 --> 1315.12]  And in some cases, I find that to be a kind of great way to work because you get a kind of to-do list for free from the tool chain.
+[1315.12 --> 1322.80]  You know, as you write your test, if things aren't working, they fail, you get kind of errors that you then have to unblock.
+[1322.98 --> 1327.88]  And it's a kind of nice way to decide what you have to do to get something to pass.
+[1328.46 --> 1336.32]  And yes, it has that same kind of idea is if you find a bug and you've written a test to prove it, you then save that test.
+[1336.44 --> 1339.94]  And next time you run all your test suite, it'll check for that bug again.
+[1339.94 --> 1342.36]  So this is what we mean by protecting from regression.
+[1342.64 --> 1348.00]  You can never have that same bug again if you've fixed it and you keep the unit test.
+[1348.66 --> 1350.66]  What do we do with that corpus, though?
+[1351.02 --> 1355.80]  Dominic Roos on Twitter asked, what are the best practices for the corpus?
+[1356.14 --> 1357.56]  Should you put it into Git?
+[1357.74 --> 1359.12]  Should it go into some other repo?
+[1359.26 --> 1360.66]  Do you share it amongst the team?
+[1361.14 --> 1363.76]  Is it just something you run on your own dev machine?
+[1364.30 --> 1365.54]  Where does this go in practice?
+[1366.04 --> 1367.16]  I think it's going to depend.
+[1367.16 --> 1372.28]  I also think this is kind of a bit of an open question in terms of what kind of practices do we want?
+[1372.36 --> 1374.54]  Like best practices do we want to lay out for this?
+[1374.62 --> 1378.58]  But also, that part is also kind of up to the developer, too.
+[1378.94 --> 1380.16]  It could be programmatic.
+[1380.36 --> 1387.96]  Like, let's say, like I mentioned before, you have existing unit tests and you just want to move it into basically change your t.run into an f.fuzz.
+[1388.36 --> 1390.44]  Something like that should be basically possible.
+[1390.64 --> 1393.06]  So if it's already programmatic, keep it programmatic.
+[1393.46 --> 1394.74]  And if it fails, it fails.
+[1394.80 --> 1395.40]  And that's great.
+[1395.40 --> 1405.62]  If you have a bunch of test data, like let's say you have a bunch of big HTTP requests or binary files or something like that that you already have somewhere, you can just use those, too.
+[1406.02 --> 1408.44]  And the fuzzing engine will look at that.
+[1408.50 --> 1413.44]  Or not the fuzzing engine, but GoTest will look at test data as part of the seed corpus, too.
+[1413.44 --> 1417.44]  And so I think it also depends on what the seed corpus is.
+[1417.50 --> 1418.94]  Is it a huge binary?
+[1419.62 --> 1421.36]  Is it a small thing?
+[1421.48 --> 1423.36]  Is it something that's best built programmatically?
+[1423.36 --> 1427.32]  And what the best practices for that will be, I think, are still kind of an open question.
+[1427.44 --> 1428.18]  At least it is to me.
+[1428.58 --> 1433.50]  I think there's also an angle of maturity of the ecosystem in there, of maturity of the technique.
+[1433.50 --> 1444.90]  Because when fuzzing is just this tool that some security researchers use to smash against a program once, try to get something out of it, and then move on.
+[1444.90 --> 1449.48]  Of course, they just run the corpus wherever they're keeping it.
+[1449.90 --> 1458.08]  But I feel like just like with testing, we set up continuous integration, and we trust machines to do the heavy lifting for us.
+[1458.46 --> 1463.60]  I expect that fuzzing also take that path once it's built into developer workflows.
+[1464.16 --> 1470.32]  So you would have a small corpus locally on your machine, and Katie's proposal puts it automatically in a cache folder.
+[1470.32 --> 1476.86]  And that will, you know, do a very quick pass, but you're not going to run the fuzzer mostly on your laptop.
+[1477.20 --> 1483.54]  Part of what makes fuzzers work is that computers are fast, but also you can keep throwing more cores at it.
+[1484.02 --> 1491.94]  And then you upload it, and some CI or OSS fuzz or some continuous integration system can just run the fuzzer.
+[1491.94 --> 1499.46]  And it should persist the corpus, so it will keep running the same corpus against it so that you make changes.
+[1499.46 --> 1505.08]  And the corpus is already hot and large, but it's not checked into your repository.
+[1505.54 --> 1509.94]  Because most people don't want megabytes and megabytes of corpus checked in.
+[1510.36 --> 1514.84]  Right. One thing that I also like about fuzzers is that there is usually a way to tell them,
+[1515.22 --> 1519.28]  don't feed me input that is bigger than this amount, either directly or indirectly.
+[1519.40 --> 1525.52]  The indirect way is you take whatever the fuzzer passes you, and if it is bigger than a certain size, you just return,
+[1525.82 --> 1526.94]  no, I don't want this.
+[1526.94 --> 1531.70]  And after a while, the fuzzer will stop seeding the corpus with anything bigger than the size you want.
+[1531.82 --> 1539.24]  So if you're testing string split, yes, you can get up to a megabyte, but it doesn't make sense to split a gigabyte of string.
+[1539.36 --> 1547.42]  Because, I mean, you know the code that you're fuzzing, and you shouldn't be too exaggerating on how liberal you are in the input you feed it to.
+[1547.68 --> 1550.88]  It's like, yes, you're fuzzing, but you know what you're fuzzing.
+[1550.88 --> 1554.32]  It's like, if you're fuzzing a JPEG parser, yes, feed it big stuff.
+[1554.50 --> 1560.90]  If you're fuzzing a string splitter, it's very hard that there is a bug at the three gigabytes mark.
+[1561.54 --> 1566.68]  Yeah, that's a good point, though, because you do get the sense that this is just, you'd switch it on,
+[1566.82 --> 1569.70]  and it just points to your methods, and it's just going to go and do it.
+[1570.06 --> 1573.94]  That is interesting, though, that this is a continuous thing.
+[1573.94 --> 1577.92]  It's not something that you would do like a benchmark, where you just run that on your laptop.
+[1579.14 --> 1583.26]  But there is, in the proposal, there is like a new flag to run the fuzz,
+[1583.34 --> 1590.06]  but is the expectation that that would run in some kind of continuous integration or some other place?
+[1590.72 --> 1591.86]  I think it probably depends.
+[1592.02 --> 1596.16]  I think, yeah, I think it kind of depends on how long someone wants to run a fuzzer.
+[1596.46 --> 1600.86]  If they are willing to just let it run on their machine for a while, maybe that's okay.
+[1600.86 --> 1604.04]  If they want to just run it for the weekend, that's totally fine.
+[1604.48 --> 1606.96]  If it's a company, and they have a ton, or just an individual,
+[1607.10 --> 1609.32]  and they have a ton of different things they want to try to fuzz at once,
+[1609.82 --> 1611.72]  I'm not really sure if that's even going to be supported,
+[1612.08 --> 1615.12]  to be able to run multiple fuzzers at once.
+[1615.20 --> 1617.40]  I don't know what would happen, like if there's a race condition.
+[1618.38 --> 1620.68]  There's a lot of different things that I'm not totally sure would be supported.
+[1620.80 --> 1623.26]  If it crashes something somewhere, it's hard to know where it's coming from.
+[1623.42 --> 1627.96]  And so it may make more sense in situations like that to have it on some kind of continuous integration.
+[1627.96 --> 1632.22]  I wonder if we're going to end up in a situation where, like with Bitcoin miners,
+[1632.38 --> 1637.54]  we've just got all these machines that are just spending all their time crunching through fuzzing stuff.
+[1638.04 --> 1639.16]  When we've got fuzz coin.
+[1639.66 --> 1641.08]  OSS Fuzz already exists.
+[1641.20 --> 1647.74]  It's this project by Google that basically provides what internally we call cluster fuzz,
+[1648.48 --> 1651.72]  which I don't know if I was allowed to say, but yeah, we're rolling.
+[1651.72 --> 1658.54]  For open source projects where any open source project can submit.
+[1658.86 --> 1659.32]  I don't know.
+[1659.50 --> 1660.76]  There are criterias, of course.
+[1660.88 --> 1664.72]  I don't know what they are exactly, but they will just run your fuzzers for you.
+[1664.84 --> 1667.78]  And if we make it standard how to do that with Go,
+[1667.88 --> 1670.24]  it would be extremely easy to submit Go projects.
+[1671.08 --> 1673.34]  Yeah, that gets very exciting, actually.
+[1673.48 --> 1674.46]  That's really cool.
+[1674.72 --> 1676.22]  I think cluster fuzz is open source.
+[1676.60 --> 1676.90]  Cool.
+[1677.38 --> 1677.60]  Okay.
+[1677.86 --> 1679.22]  I'm not getting fired today.
+[1679.22 --> 1681.88]  Yeah, don't get fired, please.
+[1682.14 --> 1685.02]  But if you do want to get fired, please do it this way.
+[1685.14 --> 1687.70]  Come on the show and reveal something that you shouldn't reveal.
+[1688.20 --> 1689.62]  So cool for us.
+[1689.82 --> 1690.50]  Such a scoop.
+[1690.72 --> 1693.86]  I've got a history with that and let's leave it at that and move on.
+[1694.38 --> 1695.52]  Yeah, don't encourage him.
+[1696.10 --> 1696.26]  Yeah.
+[1696.72 --> 1701.80]  Last time Filippo was on the show, he stopped me from admitting to a crime before I said it,
+[1701.80 --> 1702.96]  which was brilliant.
+[1703.66 --> 1705.08]  Really useful service.
+[1705.08 --> 1711.02]  We can take a short break if anyone needs to.
+[1711.90 --> 1714.58]  And people at home can take a break anytime they want to, really.
+[1714.96 --> 1720.42]  Probably just carrying us around on their portable devices so they can just do what they like.
+[1720.94 --> 1721.94]  I don't know why I'm explaining that.
+[1722.88 --> 1724.96]  I was just going to say some bits will get cut out.
+[1725.04 --> 1726.98]  If you need anything cut out, let us know.
+[1727.70 --> 1728.36]  We'll do that.
+[1728.36 --> 1734.30]  Oh, Matt, I've listened to so many episodes of this in which you say this will be cut out and that never happens.
+[1734.94 --> 1735.32]  I know.
+[1735.52 --> 1738.30]  They don't do it for me, but they will do it for you three.
+[1738.94 --> 1739.26]  Okay.
+[1741.80 --> 1742.32]  Thank you.
+[1742.36 --> 1745.38]  They add bits for me from other times I've embarrassed myself.
+[1745.82 --> 1748.44]  I find them in extra shows.
+[1748.68 --> 1750.56]  I didn't embarrass myself then.
+[1750.56 --> 1752.66]  That was a different time when I embarrassed myself.
+[1752.66 --> 1753.98]  It's directly on the soundboard.
+[1753.98 --> 1755.98]  Yeah, exactly, yeah.
+[1756.68 --> 1756.92]  Yeah.
+[1757.22 --> 1758.74]  It's just got me embarrassing myself.
+[1758.84 --> 1759.48]  This is one of them.
+[1759.92 --> 1760.80]  This is one of the clips.
+[1763.80 --> 1768.82]  How much time does your team spend building and maintaining internal tooling?
+[1769.08 --> 1771.08]  I'm talking about those behind-the-scenes apps.
+[1771.32 --> 1773.10]  The ones no one else sees.
+[1773.36 --> 1775.86]  The S3 uploader you built last year for the marketing team.
+[1775.86 --> 1779.60]  That quick Firebase admin panel that lets you monitor key KPIs.
+[1779.94 --> 1782.70]  Maybe even the tool your data science team hacked together
+[1782.70 --> 1784.88]  so they can provide custom ad spend analytics.
+[1785.46 --> 1787.38]  Now, these are tools you need so you build them.
+[1787.58 --> 1788.54]  And that makes sense.
+[1789.08 --> 1792.40]  But the question is, could you have built them in less time,
+[1792.54 --> 1795.58]  with less effort, and less overhead and maintenance required?
+[1795.92 --> 1798.10]  And the answer to that question is, yes.
+[1798.58 --> 1799.84]  That's where Retool comes in.
+[1800.22 --> 1803.70]  Rohan Chopra, engineering director at DoorDash, has this to say about Retool.
+[1803.70 --> 1808.02]  Quote, the tools we've been able to quickly build with Retool have allowed us to empower
+[1808.02 --> 1812.54]  and scale our local operators, all while reducing the dependency on engineering.
+[1813.00 --> 1813.34]  End quote.
+[1813.82 --> 1817.60]  Now, the internal tooling process at DoorDash was bogged down with manual data entry,
+[1818.02 --> 1820.08]  missed handoffs, and long turnaround times.
+[1820.30 --> 1824.08]  And after integrating Retool, DoorDash was able to cut the engineering time required
+[1824.08 --> 1828.68]  to build tools by a factor of 10x and eliminate the error-prone manual processes
+[1828.68 --> 1829.60]  that plagued their workflows.
+[1829.60 --> 1833.42]  They were able to empower backend engineers who wouldn't otherwise be able to build front
+[1833.42 --> 1834.12]  ends from scratch.
+[1834.52 --> 1838.32]  And these engineers were able to build fully functional apps in Retool in hours,
+[1838.50 --> 1839.50]  not days or weeks.
+[1839.92 --> 1843.66]  Your next step is to try it free at retool.com slash changelog.
+[1843.82 --> 1846.24]  Again, retool.com slash changelog.
+[1859.60 --> 1872.82]  So the new proposal, which we'll post a link to in the show notes,
+[1873.54 --> 1878.74]  it kind of has a very nice Go feel to it, like the design of it.
+[1878.74 --> 1883.74]  So in the same way that we're used to test functions, being how we describe unit tests,
+[1883.74 --> 1889.20]  there are fuzz functions now which take a different argument, the testing.f.
+[1889.86 --> 1892.26]  And is that like an interface then?
+[1892.48 --> 1893.98]  What is that testing.f type?
+[1894.82 --> 1899.02]  That testing.f type is very similar to a testing.t or testing.b.
+[1899.58 --> 1903.12]  So it'll implement the testing.tb interface.
+[1903.52 --> 1905.36]  Will there be a testing.f interface then?
+[1905.50 --> 1907.30]  Or is that like a strong type?
+[1907.54 --> 1908.38]  It's a strong type.
+[1908.38 --> 1908.98]  Right.
+[1909.30 --> 1916.80]  And that has methods on it that lets you then interact with the fuzzing stuff.
+[1916.98 --> 1920.18]  But it's a relatively simple API, isn't it?
+[1920.48 --> 1922.08]  Just two methods.
+[1922.18 --> 1922.58]  Is that right?
+[1923.24 --> 1928.86]  Well, I didn't include in that proposal all the other methods that are in the testing.tb interface,
+[1928.94 --> 1930.54]  which it will support.
+[1930.74 --> 1933.24]  Like, for example, if you have some pre-work that you need to do
+[1933.24 --> 1935.34]  and you want to fatal the test or something like that,
+[1935.34 --> 1937.44]  because something failed, you can do that.
+[1937.54 --> 1938.08]  Things like that.
+[1938.58 --> 1944.76]  Originally, some earlier designs had the testing.f function accepted testing.f
+[1944.76 --> 1948.48]  or the f.fuzz function accepted testing.f.
+[1948.64 --> 1952.40]  And then it ended up kind of being not as clear, I think.
+[1952.56 --> 1954.54]  And it was going to complicate things quite a bit.
+[1954.96 --> 1957.14]  And that was some discussions that Filippo and I had.
+[1957.16 --> 1961.68]  And we ended up basically keeping it as a testing.t within that function.
+[1961.68 --> 1965.12]  So it basically should look almost exactly like a t.run.
+[1965.12 --> 1968.20]  And if you have a t.run, you can kind of copy it over directly.
+[1968.38 --> 1974.14]  So it really, it should look and feel exactly like a unit test within that f.fuzz function,
+[1974.32 --> 1975.78]  which just runs kind of as a unit test.
+[1975.90 --> 1980.28]  And then anything you need to do before that, like set things up, add to the corpus,
+[1980.76 --> 1983.96]  whatever you need to do, you can use the testing.f for that part.
+[1985.04 --> 1991.24]  Unlike the run function, where the only argument you can pass into that function is a testing.t,
+[1991.24 --> 1994.58]  you can have additional arguments in these functions.
+[1994.58 --> 1997.80]  And they seem somewhat dynamic.
+[1998.40 --> 1999.90]  Can you explain how they work?
+[2000.30 --> 2000.46]  Yeah.
+[2000.58 --> 2004.90]  So inside this f.fuzz function, those first parameters,
+[2005.48 --> 2009.74]  what you're basically telling it is it's going to take a testing.t.
+[2009.88 --> 2011.84]  Basically, it's scoped to this t.
+[2011.84 --> 2018.42]  And then you're just telling it what things you want the fuzzing engine to be generating for you.
+[2018.64 --> 2022.14]  What is the structure, basically, of each input in your corpus?
+[2022.64 --> 2028.44]  So in the proposal, the example is it takes a testing.t, an a, which is a string,
+[2028.68 --> 2030.28]  and then a big int, which is num.
+[2030.78 --> 2034.40]  And what that's telling it is, okay, we have an f.fuzz function.
+[2034.60 --> 2036.76]  That's what's going to be run with the fuzzing engine.
+[2037.18 --> 2040.74]  That function is going to be run for every input.
+[2041.10 --> 2042.32]  It's bound by that t.
+[2042.74 --> 2046.54]  And then the corpus is an a with a string and a big int.
+[2046.68 --> 2050.04]  And those are those, that's basically the structure of the corpus.
+[2050.04 --> 2054.68]  And so every time it runs, it should be running with a new string and big int.
+[2054.68 --> 2062.32]  Does it dynamically look at the arguments that you've passed there and change the code?
+[2062.42 --> 2064.70]  Does it like respond to the arguments?
+[2064.90 --> 2066.58]  Or do you have to define them somewhere?
+[2066.84 --> 2068.32]  Or are there patterns you have to follow?
+[2068.76 --> 2071.54]  I'm not sure exactly that I want to make sure I'm explaining it right.
+[2071.64 --> 2074.52]  But basically, that string and that big int,
+[2074.52 --> 2076.80]  if you look up a little bit higher in the proposal,
+[2077.04 --> 2079.78]  and you're looking at this f.add function,
+[2080.00 --> 2082.68]  what that's doing is it's adding to the corpus.
+[2082.68 --> 2084.86]  And it's adding a string and a big int,
+[2085.50 --> 2092.60]  which must look exactly the same as the a and the string and the big int in that order,
+[2093.06 --> 2094.80]  in that f.add fuzz function.
+[2094.92 --> 2100.12]  So what's that basically defining is this is the definition of the corpus entries
+[2100.12 --> 2103.74]  that will be added manually and it will be generated by the fuzzing engine.
+[2104.52 --> 2107.00]  And it works with a slice of empty interface.
+[2107.24 --> 2109.22]  So it's kind of generic code in a way.
+[2109.66 --> 2112.30]  If Go got generics, would that change?
+[2112.30 --> 2114.40]  Would that affect this design in any way?
+[2114.52 --> 2117.94]  Or do you think you'd still probably use it in a similar way?
+[2118.44 --> 2121.22]  I'm not actually sure that it would impact the design.
+[2121.38 --> 2123.70]  It might impact the implementation a little bit,
+[2123.98 --> 2125.54]  but I haven't really thought too much about it.
+[2125.60 --> 2127.52]  But I also like just thinking about it now,
+[2127.68 --> 2129.70]  I'm not actually sure that it would change much.
+[2129.80 --> 2131.64]  I think what this function is supposed to do,
+[2131.78 --> 2134.54]  this f.fuzz is just kind of like a,
+[2135.14 --> 2136.28]  it's a little bit magic-y,
+[2136.28 --> 2139.34]  but it's basically just trying to tell the fuzzing engine,
+[2139.46 --> 2143.12]  the structure that it should be aware of and be using.
+[2143.32 --> 2146.94]  It's a nice API to be able to just define the function
+[2146.94 --> 2150.94]  and have it kind of notice that or work at least.
+[2151.32 --> 2154.62]  But what happens if you've added different kind of data
+[2154.62 --> 2155.58]  or you change the structure?
+[2155.76 --> 2156.92]  What happens in that case?
+[2156.92 --> 2158.98]  Like if, for example, you added,
+[2159.22 --> 2162.32]  you did an f.add with two ints or something like that?
+[2162.48 --> 2163.14]  Yeah, exactly, yeah.
+[2163.60 --> 2164.92]  I expect it would probably panic.
+[2166.52 --> 2168.74]  Because what you're doing is you're basically telling it,
+[2168.82 --> 2172.58]  here's two ints and it's expecting a string and a big int.
+[2172.66 --> 2174.12]  And maybe that can work with static check
+[2174.12 --> 2176.60]  and things like that to find those things at build time.
+[2176.60 --> 2179.06]  For anybody who hasn't read the proposal,
+[2179.56 --> 2182.78]  f.add is the function you use to seed the corpus.
+[2183.26 --> 2185.36]  So it's the function that you use to say,
+[2185.58 --> 2187.38]  here's the starting points,
+[2187.76 --> 2190.60]  which by the way is one of my favorite things of the proposal
+[2190.60 --> 2193.96]  because usually you have to just create a bunch of files,
+[2194.22 --> 2196.68]  one for each input, then put them in a full.
+[2196.98 --> 2198.32]  Actually, I'm going to do something else.
+[2198.86 --> 2200.80]  And instead here, you just write f.add
+[2200.80 --> 2202.74]  and here's my ECTSA certificate,
+[2202.92 --> 2204.48]  here's my RSA certificate.
+[2204.66 --> 2206.12]  These are examples, go for it.
+[2206.46 --> 2209.72]  Well, so f.add is the function that adds to the corpus,
+[2209.94 --> 2211.92]  while f.fuzz is the function
+[2211.92 --> 2214.44]  that actually runs the fuzzer
+[2214.44 --> 2218.16]  and it runs a function that takes the same types of arguments.
+[2218.34 --> 2219.02]  Yeah, thanks for the problem.
+[2219.04 --> 2220.34]  Just mentioning it because
+[2220.34 --> 2222.92]  in case people haven't read the proposal yet.
+[2223.44 --> 2224.14]  Thank you, brilliant.
+[2224.38 --> 2226.74]  And I love the fact that it kind of still,
+[2226.90 --> 2229.96]  I mean, it's designed to fit into what we already have.
+[2229.96 --> 2231.66]  So it knows about GoTest
+[2231.66 --> 2234.58]  and it kind of cooperates with GoTest as well, doesn't it?
+[2235.06 --> 2238.10]  Yeah, and actually that was my main goal with all of this.
+[2238.10 --> 2240.48]  I wasn't going to be okay with the design
+[2240.48 --> 2244.48]  that didn't feel like testing that we have now.
+[2244.80 --> 2246.00]  Someone should be able to look at this
+[2246.00 --> 2247.68]  and hopefully understand it pretty quickly.
+[2248.04 --> 2250.32]  And the goal is that if you know how to write a unit test,
+[2250.46 --> 2251.72]  you know how to write a fuzz target.
+[2252.04 --> 2254.06]  And it should be approximately as easy.
+[2254.50 --> 2258.28]  I wanted it to be able to work with the Go command as it is now.
+[2258.28 --> 2263.06]  And if people run GoTest, it should just run the same way.
+[2263.28 --> 2265.68]  And it shouldn't have to use anything terribly special.
+[2265.82 --> 2266.88]  It shouldn't have to do anything.
+[2267.18 --> 2268.48]  Learn that much new.
+[2268.86 --> 2270.36]  I wanted kind of the barrier to entry
+[2270.36 --> 2272.64]  to be as low as humanly possible.
+[2272.94 --> 2274.98]  And so if it looks like Go code, that's the goal.
+[2275.82 --> 2276.62]  And I'm glad to hear it.
+[2277.08 --> 2278.14]  I love that about the design
+[2278.14 --> 2279.94]  because I've done some research
+[2279.94 --> 2281.78]  and I've seen people using,
+[2282.02 --> 2285.06]  creating fuzz targets in the wild for parsers.
+[2285.06 --> 2291.08]  And what usually happens is that they take whatever the fuzzer engine passes them
+[2291.08 --> 2293.98]  and put it in the parser and that's it.
+[2294.42 --> 2295.76]  So they just basically check.
+[2295.96 --> 2298.96]  The only property they check for is if it panics.
+[2299.76 --> 2301.26]  So that is kind of sad
+[2301.26 --> 2304.38]  because it's so much easier to feed something into your parser
+[2304.38 --> 2306.48]  and then maybe serialize it where I can parse it again
+[2306.48 --> 2307.46]  and check if it is the same.
+[2307.90 --> 2311.76]  So it's kind of easier to write fuzz targets than people assume.
+[2311.76 --> 2314.96]  But since fuzz seems to be such an alien concept,
+[2315.58 --> 2318.12]  I've seen most fuzz targets to assert nothing.
+[2318.32 --> 2320.66]  They just feed the input to the function they want to test.
+[2320.74 --> 2325.64]  It's like if testing strings.join would just join.
+[2326.04 --> 2326.46]  That's it.
+[2326.58 --> 2327.48]  And then you don't check.
+[2327.90 --> 2329.46]  You even get the string back.
+[2330.00 --> 2332.68]  So there is a type system for that, but that's what you get.
+[2332.86 --> 2336.26]  So I'm really looking forward for this to be first class
+[2336.26 --> 2341.12]  and to be so close to the original test target
+[2341.12 --> 2344.44]  to see what people actually start asserting as a property.
+[2344.44 --> 2347.30]  Because doesn't panic seems to be a little bit too weak.
+[2347.50 --> 2347.72]  Yeah.
+[2348.60 --> 2351.86]  If people get one thing away from this conversation,
+[2352.22 --> 2356.00]  it should really be that fuzzing is going to be built in Go.
+[2356.24 --> 2358.44]  It's not just about finding panics.
+[2358.88 --> 2362.12]  It's not just feed some input and wait for it to crash.
+[2362.68 --> 2365.96]  It's about writing as many invariants as you can think of
+[2365.96 --> 2367.72]  and as many checks as you can think of
+[2367.72 --> 2371.32]  and then letting the fuzzer find the inputs
+[2371.32 --> 2374.10]  for which the thing doesn't do what you want it to do.
+[2374.46 --> 2376.80]  So would you say that fuzzing makes a lot of sense
+[2376.80 --> 2379.82]  if you're working with multiple methods?
+[2379.94 --> 2381.98]  I mean, in that example that Roberto gave,
+[2382.14 --> 2383.90]  where you're encoding and decoding,
+[2384.24 --> 2387.02]  because you can say something about the way
+[2387.02 --> 2388.78]  that those two things should interoperate.
+[2388.96 --> 2391.14]  But how can you make assertions on something
+[2391.90 --> 2393.74]  if the input is completely random?
+[2393.74 --> 2395.68]  What kind of assertion are you going to make?
+[2395.88 --> 2396.68]  One thing that I did,
+[2396.88 --> 2398.90]  I was fuzz testing a cache that I implemented.
+[2399.30 --> 2401.24]  Caches are harder than people would normally assume.
+[2401.68 --> 2402.86]  So I wanted to make sure that,
+[2403.10 --> 2405.14]  for example, what I put in, I got back.
+[2405.84 --> 2407.38]  So to test my cache,
+[2407.54 --> 2409.64]  I did differential fuzzing with a hash map.
+[2410.86 --> 2413.14]  So a hash map is a perfect cache, right?
+[2413.22 --> 2414.32]  I mean, it grows indefinitely,
+[2414.54 --> 2415.98]  but that was not, I didn't care.
+[2416.06 --> 2416.84]  It was just fuzz testing.
+[2417.18 --> 2419.58]  So I just fed stuff to my cache
+[2419.58 --> 2420.60]  and when I retrieved it,
+[2420.66 --> 2422.56]  if it wasn't there, meh, it was evicted.
+[2422.56 --> 2423.50]  But if it was there,
+[2423.62 --> 2426.04]  it should be identical to whatever was in that map.
+[2426.92 --> 2429.56]  So you can have simpler number implementation
+[2430.24 --> 2432.38]  of the algorithm you want to implement,
+[2432.48 --> 2433.54]  or maybe a slower one.
+[2434.24 --> 2435.66]  Like if you optimize your code,
+[2435.80 --> 2436.78]  you can keep the old code,
+[2436.84 --> 2438.42]  the slow one to test against.
+[2438.92 --> 2440.66]  And usually slow code is easier to debug
+[2440.66 --> 2441.96]  and it's more reliable
+[2441.96 --> 2443.22]  and it's easier to write.
+[2443.22 --> 2444.84]  It's slower and you can see what's happening.
+[2445.14 --> 2445.54]  Yeah, exactly.
+[2445.98 --> 2447.04]  Not that slower.
+[2448.78 --> 2451.70]  But yeah, that's kind of the point.
+[2451.70 --> 2454.48]  Another example that I had written up
+[2454.48 --> 2458.32]  for the Cloud4Blog is that I had this parser.
+[2458.80 --> 2460.14]  No, sorry, not this parser, actually,
+[2460.22 --> 2461.00]  this serializer.
+[2461.64 --> 2464.80]  And you're like, how do you test the serializer?
+[2465.00 --> 2466.80]  Like, how do you know if the thing it generated is good?
+[2467.16 --> 2468.54]  Well, the thing I wanted to know
+[2468.54 --> 2471.76]  was whether it would work reusing buffers
+[2471.76 --> 2473.20]  for performance reasons.
+[2473.20 --> 2475.08]  I didn't want to allocate a new buffer
+[2475.08 --> 2476.54]  or zero the buffer every time.
+[2476.64 --> 2478.70]  I just wanted to give it the old packet
+[2478.70 --> 2481.10]  packet and say, just serialize over this one.
+[2481.84 --> 2483.96]  So what I did was write a fuzzer
+[2483.96 --> 2485.96]  that would parse a packet.
+[2486.32 --> 2488.10]  And then, but in this case,
+[2488.26 --> 2489.52]  with the Go proposal,
+[2489.76 --> 2492.04]  I would not even maybe do the parse step.
+[2492.10 --> 2492.90]  I would just tell it,
+[2493.28 --> 2496.08]  give me a random packet structure
+[2496.08 --> 2499.64]  and then serialize it on both empty buffer
+[2499.64 --> 2500.70]  of all zeros
+[2500.70 --> 2504.50]  and on a full buffers of all one bits.
+[2504.50 --> 2506.42]  And if they come out different,
+[2506.94 --> 2509.76]  it means that it's not setting the zeros
+[2509.76 --> 2511.26]  in some of the fields.
+[2511.40 --> 2511.88]  And it did.
+[2512.22 --> 2514.44]  And that might or might not have been
+[2514.44 --> 2517.30]  why some stuff in the cloud for DNS server
+[2517.30 --> 2517.84]  wasn't working.
+[2518.44 --> 2519.38]  And that's the kind of stuff
+[2519.38 --> 2520.68]  you can find with fuzzers.
+[2521.08 --> 2523.88]  In general, testing should really be about
+[2523.88 --> 2526.76]  defining expected behaviors.
+[2527.28 --> 2529.10]  And that's true of all kinds of testing.
+[2529.38 --> 2532.02]  It's not just about defining expected inputs
+[2532.02 --> 2532.52]  and outputs.
+[2532.72 --> 2535.24]  It's about locking in expectations.
+[2536.02 --> 2537.62]  Any expectation that you can define
+[2537.62 --> 2540.58]  not strictly in terms of this input
+[2540.58 --> 2542.30]  needs to have this output,
+[2542.78 --> 2544.80]  but just the output needs to be longer
+[2544.80 --> 2545.48]  than the input.
+[2545.62 --> 2546.78]  The output needs to be shorter
+[2546.78 --> 2547.52]  than the input.
+[2548.06 --> 2548.86]  Anything like that,
+[2548.96 --> 2550.60]  you can put in a fuzzer in a fast target.
+[2552.30 --> 2554.42]  It's kind of like meta testing
+[2554.42 --> 2556.76]  or some kind of abstract testing
+[2556.76 --> 2557.64]  in a sense.
+[2557.82 --> 2559.52]  You're not dealing with the specific values,
+[2559.52 --> 2562.00]  but you still deal with the ideas,
+[2562.16 --> 2562.66]  the variables.
+[2562.92 --> 2564.70]  Yes, which is kind of takes away
+[2564.70 --> 2566.70]  one big risk that there is
+[2566.70 --> 2567.76]  when you write unit tests.
+[2567.98 --> 2568.88]  When you write unit tests,
+[2568.94 --> 2570.56]  you have those assumptions in mind.
+[2570.80 --> 2572.60]  Like what you're trying to test is like,
+[2572.96 --> 2575.10]  I want string split to actually split the string.
+[2575.88 --> 2577.94]  And then you go and test your stuff
+[2577.94 --> 2578.90]  and you put the input
+[2578.90 --> 2579.62]  and you put the output,
+[2579.92 --> 2581.38]  but you're just giving examples.
+[2581.96 --> 2583.74]  You're not testing the actual property
+[2583.74 --> 2584.32]  that you want.
+[2584.68 --> 2587.28]  So I think that writing a property assertion
+[2587.28 --> 2589.46]  for a fuzz target is actually closer
+[2589.46 --> 2591.10]  to what you want to do usually in tests.
+[2591.86 --> 2593.92]  Now, unit tests are always going to be needed.
+[2594.56 --> 2596.14]  But if you put on top something
+[2596.14 --> 2597.46]  that asserts the actual property
+[2597.46 --> 2598.10]  that you meant,
+[2598.44 --> 2599.98]  I think you're adding a lot of value.
+[2600.50 --> 2601.46]  One opinion I heard
+[2601.46 --> 2603.26]  that I'm not supporting,
+[2603.40 --> 2605.50]  you know, retweets are not endorsements,
+[2605.96 --> 2606.94]  but was that,
+[2607.30 --> 2608.70]  why would you write unit tests
+[2608.70 --> 2610.02]  if you already know
+[2610.02 --> 2612.18]  what your program is going to break on?
+[2612.58 --> 2614.08]  Just don't write the bug.
+[2614.08 --> 2616.02]  And I mean,
+[2616.12 --> 2618.22]  yes, yes, yes.
+[2619.02 --> 2619.74]  I know.
+[2620.14 --> 2624.18]  But there is a degree of truth to that.
+[2624.80 --> 2627.04]  The things you can write unit tests,
+[2627.28 --> 2629.16]  unit tests are actually kind of more useful
+[2629.16 --> 2630.96]  for refactoring later
+[2630.96 --> 2633.08]  and for regressions.
+[2633.62 --> 2634.40]  But that's the thing.
+[2635.06 --> 2637.52]  It's unlikely you will think of inputs
+[2637.52 --> 2640.62]  that break on the program you just wrote
+[2640.62 --> 2642.70]  because you thought about those edge cases.
+[2642.70 --> 2643.14]  Yeah.
+[2643.60 --> 2647.36]  And fuzzing will just not care
+[2647.36 --> 2648.48]  about what you thought about.
+[2648.78 --> 2650.80]  Fuzzing will find where it hurts.
+[2651.58 --> 2652.00]  Right.
+[2652.44 --> 2653.98]  And one thing that I like to say
+[2653.98 --> 2656.10]  is that I write test targets
+[2656.10 --> 2658.72]  for my future interaction with the code
+[2658.72 --> 2662.86]  because I also used to do TDD most of the time.
+[2662.96 --> 2664.12]  So I write the tests
+[2664.12 --> 2665.54]  and then I write the code that implements
+[2665.54 --> 2667.22]  whatever I'm testing for.
+[2667.62 --> 2668.16]  And in the future,
+[2668.22 --> 2668.78]  when I refactor,
+[2668.90 --> 2669.84]  I want the tests to pass.
+[2669.84 --> 2672.40]  When I said that I write fuzzes
+[2672.40 --> 2674.30]  for the tests I wrote in the past,
+[2674.38 --> 2675.60]  for the code I wrote in the past.
+[2676.18 --> 2676.74]  So basically,
+[2676.90 --> 2677.66]  the fuzzer makes sure
+[2677.66 --> 2678.42]  that whatever is there
+[2678.42 --> 2680.02]  is actually what it's meant to do
+[2680.02 --> 2681.98]  and the tests are there
+[2681.98 --> 2683.98]  so that the future code will keep doing it.
+[2684.42 --> 2687.66]  I really like what Filippo said about
+[2687.66 --> 2689.88]  kind of like the fuzzing engine
+[2689.88 --> 2692.12]  doesn't care what the developer thought about.
+[2692.28 --> 2692.40]  I mean,
+[2692.44 --> 2695.60]  I think that's kind of the benefit of having,
+[2696.24 --> 2697.30]  that's why, for example,
+[2697.30 --> 2698.26]  well, code reviews exist
+[2698.26 --> 2699.80]  because you need another person
+[2699.80 --> 2702.52]  who's kind of more objective to look at it.
+[2702.58 --> 2704.38]  And I think that a fuzzing engine
+[2704.38 --> 2707.18]  can kind of be this third-party objective,
+[2707.70 --> 2708.58]  you know,
+[2708.64 --> 2710.10]  being that just goes in
+[2710.10 --> 2711.44]  and does everything it can
+[2711.44 --> 2712.28]  to try to break it.
+[2712.58 --> 2713.46]  And it has no idea
+[2713.46 --> 2714.38]  what you thought about it.
+[2714.62 --> 2715.58]  It doesn't care about that.
+[2715.64 --> 2716.44]  It just cares about
+[2716.44 --> 2719.30]  trying to find as much coverage as it can
+[2719.30 --> 2720.20]  and try to find bugs.
+[2720.68 --> 2722.30]  And that kind of like third-party entity
+[2722.30 --> 2723.80]  is kind of a cool concept to me.
+[2724.42 --> 2724.60]  Katie,
+[2724.68 --> 2726.20]  aren't you worried about the fuzzing thing
+[2726.20 --> 2727.24]  becoming self-aware
+[2727.24 --> 2728.36]  and then just going around
+[2728.36 --> 2729.66]  doing loads of random crime?
+[2730.12 --> 2732.04]  That's like actually my goal with this.
+[2732.28 --> 2734.74]  I'm actually trying to build a,
+[2734.74 --> 2735.22]  yeah,
+[2735.68 --> 2736.54]  self-learning robot
+[2736.54 --> 2738.12]  that'll just take over the language.
+[2738.20 --> 2738.98]  How do you know?
+[2739.38 --> 2740.16]  Based on fuzzing.
+[2740.36 --> 2740.70]  Exactly.
+[2741.02 --> 2741.90]  How do you know
+[2741.90 --> 2743.54]  that's not already what happened
+[2743.54 --> 2745.46]  and we're here pitching fuzzing
+[2745.46 --> 2749.26]  to just make our fuzzer overlords happy?
+[2750.18 --> 2751.52]  I'm actually a fuzzing engine.
+[2751.84 --> 2752.42]  All this time,
+[2752.48 --> 2753.22]  it's been a simulation.
+[2753.96 --> 2754.84]  Well, it is a good one.
+[2754.84 --> 2755.58]  Yeah, you're right.
+[2755.64 --> 2756.44]  It does a good job.
+[2758.04 --> 2758.40]  Yeah.
+[2759.20 --> 2759.82]  But the thing is,
+[2759.84 --> 2760.06]  yeah,
+[2760.26 --> 2761.60]  I'd love that though.
+[2761.88 --> 2762.36]  Not really.
+[2764.50 --> 2766.08]  Doesn't know how to interact with you
+[2766.08 --> 2767.32]  now that he knows you're a robot.
+[2767.32 --> 2771.86]  I love it when the machines
+[2771.86 --> 2774.38]  do kind of get this emergent intelligence.
+[2774.60 --> 2776.50]  I find that to be really quite amazing,
+[2776.64 --> 2778.04]  especially when there's so much chaos
+[2778.04 --> 2780.10]  in what's actually going on.
+[2780.10 --> 2780.90]  So yeah,
+[2781.00 --> 2782.64]  the fact that I think the thing
+[2782.64 --> 2783.34]  that I've learned
+[2783.34 --> 2785.48]  and I'll take away is
+[2785.48 --> 2787.78]  it's less about random input
+[2787.78 --> 2790.02]  and it's more about kind of variations
+[2790.02 --> 2792.82]  of the realistic kind of input
+[2792.82 --> 2794.76]  that you're going to pass in, right?
+[2794.94 --> 2795.10]  Right.
+[2795.18 --> 2796.72]  Or that didn't resonate
+[2796.72 --> 2797.92]  because I can tell on my screen
+[2797.92 --> 2799.10]  that there's no...
+[2799.90 --> 2800.24]  Go on.
+[2800.46 --> 2801.56]  Correct me if that's...
+[2801.56 --> 2802.28]  No, it's just...
+[2802.28 --> 2803.18]  I wanted to say that
+[2803.18 --> 2805.74]  I was putting stuff on top of this,
+[2805.88 --> 2806.36]  which is
+[2806.36 --> 2808.00]  the fuzzer doesn't care
+[2808.00 --> 2809.66]  about what the code does.
+[2809.70 --> 2810.32]  And that's important
+[2810.32 --> 2811.64]  because if we had like
+[2811.64 --> 2812.76]  machine learning algorithm
+[2812.76 --> 2813.82]  fuzzing our code,
+[2814.34 --> 2815.14]  just, you know,
+[2815.18 --> 2816.62]  trying to learn how the code behaves,
+[2816.86 --> 2817.80]  at one point they would do it
+[2817.80 --> 2818.48]  as humans would.
+[2818.74 --> 2819.84]  They would understand
+[2819.84 --> 2821.14]  what the code is supposed to do
+[2821.14 --> 2822.14]  and kind of, you know,
+[2822.86 --> 2824.24]  accept the code works.
+[2824.78 --> 2826.26]  And instead,
+[2826.42 --> 2827.60]  if you just use an algorithm
+[2827.60 --> 2829.22]  that just tries to bash
+[2829.22 --> 2830.10]  with random stuff,
+[2830.16 --> 2831.14]  at one point you find
+[2831.14 --> 2832.98]  like after two years
+[2832.98 --> 2834.42]  you have been fuzzing a target,
+[2834.96 --> 2836.48]  a new edge case that crashes.
+[2837.28 --> 2838.36]  And this is something that I love
+[2838.36 --> 2839.30]  because a human
+[2839.30 --> 2840.16]  or an intelligent
+[2840.16 --> 2841.56]  like kind of design
+[2841.56 --> 2844.16]  in our way of defining intelligence
+[2844.16 --> 2845.34]  would not find it.
+[2845.82 --> 2847.08]  Because why would you keep doing
+[2847.08 --> 2847.68]  for two years
+[2847.68 --> 2848.20]  the same thing
+[2848.20 --> 2849.34]  expecting a different result?
+[2849.58 --> 2850.66]  Isn't that the definition of madness?
+[2851.74 --> 2853.00]  Yeah, but we are going to end up
+[2853.00 --> 2854.18]  with fuzzing Terminators
+[2854.18 --> 2855.68]  literally just running around
+[2855.68 --> 2857.16]  trying all kinds of different things
+[2857.16 --> 2857.70]  to get you
+[2857.70 --> 2859.54]  and just like goes and hacks something,
+[2859.70 --> 2860.20]  smashes it,
+[2860.28 --> 2860.82]  kicks a puppy,
+[2860.82 --> 2861.94]  throws a baby in the sea.
+[2862.28 --> 2862.64]  Do you know what I mean?
+[2862.70 --> 2864.10]  Just doing all kinds of...
+[2864.10 --> 2865.40]  Just to see what works.
+[2865.90 --> 2866.22]  Do you know what I mean?
+[2866.30 --> 2866.54]  It's not...
+[2866.54 --> 2867.72]  It's a risk we're willing to accept.
+[2867.84 --> 2868.32]  It's a risk.
+[2868.96 --> 2869.24]  Okay.
+[2869.50 --> 2870.02]  You are, are you?
+[2870.58 --> 2871.14]  You really are
+[2871.14 --> 2872.62]  chocolate factory boffins over there.
+[2872.68 --> 2873.38]  It's a sacrifice
+[2873.38 --> 2874.42]  we're willing to make.
+[2874.60 --> 2874.78]  Really.
+[2874.78 --> 2875.78]  Okay.
+[2875.78 --> 2891.74]  What's up, gophers?
+[2891.74 --> 2892.76]  Are you looking for a way
+[2892.76 --> 2894.08]  to instantly debug
+[2894.08 --> 2895.50]  and troubleshoot your applications
+[2895.50 --> 2896.76]  and services running
+[2896.76 --> 2897.30]  in production
+[2897.30 --> 2898.72]  on Kubernetes?
+[2898.92 --> 2899.60]  That's a mouthful.
+[2899.90 --> 2901.06]  Well, Pixie gives you
+[2901.06 --> 2902.20]  a magical API
+[2902.20 --> 2903.76]  to get instant debug data.
+[2903.76 --> 2904.98]  And the best part
+[2904.98 --> 2906.30]  is this doesn't involve
+[2906.30 --> 2907.14]  changing code.
+[2907.48 --> 2908.72]  There are no manual UIs
+[2908.72 --> 2909.98]  and all this lives
+[2909.98 --> 2911.14]  inside Kubernetes.
+[2911.82 --> 2913.18]  Pixie is an API
+[2913.18 --> 2915.00]  which lives inside your platform,
+[2915.38 --> 2916.40]  harvests all of your data
+[2916.40 --> 2917.04]  that you need
+[2917.04 --> 2919.18]  and exposes a bunch of interfaces
+[2919.18 --> 2920.06]  that you can ping
+[2920.06 --> 2921.34]  to get data you need.
+[2921.70 --> 2922.94]  Pixie is essentially
+[2922.94 --> 2924.58]  like a decentralized Splunk.
+[2924.78 --> 2925.80]  It's a programmable
+[2925.80 --> 2927.16]  edge intelligence platform
+[2927.16 --> 2928.50]  which captures metrics,
+[2928.70 --> 2929.04]  traces,
+[2929.26 --> 2929.58]  logs,
+[2929.58 --> 2930.30]  and events
+[2930.30 --> 2931.62]  without any code changes.
+[2932.22 --> 2933.32]  And the team behind Pixie
+[2933.32 --> 2934.10]  is working hard
+[2934.10 --> 2934.88]  to bring it to market
+[2934.88 --> 2935.72]  for broad use
+[2935.72 --> 2936.94]  by the end of 2020.
+[2937.46 --> 2938.62]  But I'm here to tell you
+[2938.62 --> 2939.24]  how you can get your hands
+[2939.24 --> 2939.98]  on the beta today.
+[2940.46 --> 2941.58]  Links are in the show notes
+[2941.58 --> 2942.38]  so check them out
+[2942.38 --> 2943.30]  so you can click through
+[2943.30 --> 2943.82]  to the beta
+[2943.82 --> 2944.98]  and their Slack community.
+[2945.30 --> 2945.72]  Once again,
+[2945.82 --> 2946.54]  links are in the show notes.
+[2946.60 --> 2947.24]  Check them out
+[2947.24 --> 2948.48]  and look forward to Pixie Day
+[2948.48 --> 2949.06]  coming soon.
+[2949.06 --> 2969.20]  I actually think
+[2969.20 --> 2970.20]  you should probably leave.
+[2970.20 --> 2975.06]  I'm talking to the region.
+[2979.06 --> 2979.98]  So,
+[2980.56 --> 2981.78]  does anybody have
+[2981.78 --> 2983.00]  an unpopular opinion
+[2983.00 --> 2983.96]  for us today?
+[2984.72 --> 2985.90]  It can be fuzzing related
+[2985.90 --> 2987.04]  but it doesn't have to be.
+[2987.30 --> 2988.00]  It can be anything.
+[2988.38 --> 2988.82]  Yeah,
+[2988.88 --> 2989.40]  I've got one.
+[2990.24 --> 2991.34]  I can throw in the ring.
+[2991.62 --> 2992.16]  Throw it in.
+[2992.48 --> 2992.82]  So,
+[2993.14 --> 2994.16]  I think that
+[2994.16 --> 2995.64]  it's kind of more like a,
+[2996.12 --> 2996.92]  I don't know if it's an opinion
+[2996.92 --> 2998.50]  so much as a personal experience
+[2998.50 --> 2999.82]  but I actually got into
+[2999.82 --> 3000.42]  computer science
+[3000.42 --> 3001.48]  because math wasn't
+[3001.48 --> 3002.48]  social enough for me.
+[3003.42 --> 3003.86]  So,
+[3003.96 --> 3004.62]  I think that
+[3004.62 --> 3006.52]  the thing that I like the most
+[3006.52 --> 3007.54]  I think is the best part
+[3007.54 --> 3008.46]  about computer science
+[3008.46 --> 3009.46]  is actually building
+[3009.46 --> 3011.08]  things with other people
+[3011.08 --> 3011.98]  and I think like
+[3011.98 --> 3012.86]  having social skills
+[3012.86 --> 3014.10]  can take you a really long way
+[3014.10 --> 3015.60]  and is kind of
+[3015.60 --> 3016.36]  undervalued
+[3016.36 --> 3017.24]  in tech.
+[3017.40 --> 3017.54]  So,
+[3017.66 --> 3018.36]  you're telling me
+[3018.36 --> 3020.26]  that you got into CS
+[3020.26 --> 3022.50]  because of the social aspect.
+[3023.34 --> 3023.74]  Exactly.
+[3023.88 --> 3024.70]  I didn't want to sit alone
+[3024.70 --> 3025.56]  in a corner all day
+[3025.56 --> 3026.98]  and just solve math problems
+[3026.98 --> 3027.50]  but I was like,
+[3027.56 --> 3027.68]  oh,
+[3027.70 --> 3029.14]  I can build stuff with people.
+[3029.60 --> 3030.56]  That sounds more fun
+[3030.56 --> 3031.36]  so I'm going to do that
+[3031.36 --> 3032.60]  which I realize
+[3032.60 --> 3033.18]  is the opposite
+[3033.18 --> 3033.78]  of a lot of people.
+[3034.36 --> 3035.02]  And then you ended up
+[3035.02 --> 3035.74]  in security
+[3035.74 --> 3037.54]  because the InfoSec community
+[3037.54 --> 3038.36]  is,
+[3038.36 --> 3039.50]  you know,
+[3039.78 --> 3040.86]  shining example
+[3040.86 --> 3041.92]  of excellent
+[3041.92 --> 3043.06]  community support.
+[3043.58 --> 3044.28]  That's a highly
+[3044.28 --> 3045.26]  socialist field though
+[3045.26 --> 3046.38]  because you need to be able
+[3046.38 --> 3048.08]  to talk to people
+[3048.08 --> 3049.26]  and understand
+[3049.26 --> 3050.80]  like if they disclose
+[3050.80 --> 3051.30]  a report
+[3051.30 --> 3051.80]  you need to be able
+[3051.80 --> 3052.86]  to communicate with them
+[3052.86 --> 3053.84]  and understand them
+[3053.84 --> 3054.56]  and be able to
+[3054.56 --> 3055.40]  communicate back
+[3055.40 --> 3056.14]  and you need to be able
+[3056.14 --> 3056.84]  to communicate
+[3056.84 --> 3058.52]  really complicated things
+[3058.52 --> 3059.62]  in a really simple way
+[3059.62 --> 3060.26]  that other people
+[3060.26 --> 3060.86]  can understand
+[3060.86 --> 3061.68]  which is really hard.
+[3062.28 --> 3062.86]  And I think this is,
+[3063.04 --> 3063.48]  that's a field
+[3063.48 --> 3064.60]  where it's even more important
+[3064.60 --> 3065.12]  that you have
+[3065.12 --> 3065.88]  good social skills
+[3065.88 --> 3066.58]  because it's so,
+[3067.22 --> 3068.36]  the stakes are so high.
+[3068.92 --> 3069.20]  Yeah,
+[3069.30 --> 3069.94]  to be fair,
+[3070.24 --> 3071.26]  I should point out
+[3071.26 --> 3072.72]  that the Go security
+[3072.72 --> 3073.24]  community
+[3073.24 --> 3074.40]  is extremely nice.
+[3074.54 --> 3075.20]  The kind of people
+[3075.20 --> 3076.68]  that email us reports
+[3076.68 --> 3077.46]  are usually
+[3077.46 --> 3078.38]  a delight
+[3078.38 --> 3079.42]  to work with.
+[3079.76 --> 3080.30]  I was just
+[3080.30 --> 3081.40]  making a cheap shot
+[3081.40 --> 3082.22]  at the,
+[3083.44 --> 3084.20]  right,
+[3084.46 --> 3085.16]  let's say
+[3085.16 --> 3086.64]  traditional
+[3086.64 --> 3088.82]  security community.
+[3089.20 --> 3089.46]  Traditional.
+[3089.56 --> 3090.64]  What can they do
+[3090.64 --> 3090.98]  to get you?
+[3090.98 --> 3091.42]  That's one way
+[3091.42 --> 3091.86]  to put it.
+[3091.98 --> 3092.42]  You're safe,
+[3092.48 --> 3092.72]  aren't you,
+[3092.78 --> 3093.28]  from that lot?
+[3093.28 --> 3094.26]  What can they ever do?
+[3094.66 --> 3095.00]  Right.
+[3095.22 --> 3095.54]  Yeah!
+[3097.76 --> 3098.80]  Now that you say that,
+[3098.88 --> 3099.02]  Katie,
+[3099.36 --> 3099.90]  I think that one
+[3099.90 --> 3101.16]  of the important things
+[3101.16 --> 3102.18]  about the human aspect
+[3102.18 --> 3102.84]  of software
+[3102.84 --> 3103.48]  is like,
+[3103.70 --> 3104.58]  when you design an API,
+[3105.14 --> 3105.94]  you have to design it
+[3105.94 --> 3106.52]  in a way that people
+[3106.52 --> 3107.06]  will understand.
+[3107.56 --> 3107.72]  Like,
+[3108.10 --> 3109.22]  I hate when people say,
+[3109.56 --> 3109.76]  like,
+[3109.96 --> 3110.86]  users of this API
+[3110.86 --> 3111.44]  are stupid
+[3111.44 --> 3112.24]  because they can't
+[3112.24 --> 3112.96]  use it right.
+[3114.06 --> 3114.18]  No,
+[3114.36 --> 3115.84]  when you're designing
+[3115.84 --> 3116.28]  something,
+[3116.38 --> 3117.08]  you're communicating
+[3117.08 --> 3117.82]  to the user.
+[3118.52 --> 3119.64]  People keep forgetting
+[3119.64 --> 3120.78]  that issue.
+[3121.70 --> 3122.10]  Yeah,
+[3122.18 --> 3122.70]  that is true,
+[3122.76 --> 3123.06]  actually,
+[3123.06 --> 3124.06]  because you do think,
+[3124.36 --> 3124.94]  in the beginning,
+[3125.20 --> 3126.30]  I thought APIs
+[3126.30 --> 3127.14]  were for machines
+[3127.14 --> 3128.12]  to talk to each other,
+[3128.28 --> 3128.92]  but they aren't.
+[3129.00 --> 3129.72]  They're for humans
+[3129.72 --> 3131.04]  to build the thing
+[3131.04 --> 3131.92]  that allows the machines
+[3131.92 --> 3132.76]  to talk to each other.
+[3133.30 --> 3133.46]  Yeah,
+[3133.52 --> 3134.56]  so that is true.
+[3134.92 --> 3135.50]  But I don't know,
+[3135.60 --> 3136.42]  Pythagoras could have been
+[3136.42 --> 3137.64]  a laugh at a party,
+[3137.76 --> 3139.42]  might have had a great time
+[3139.42 --> 3139.64]  with him.
+[3140.60 --> 3141.66]  He's probably measuring
+[3141.66 --> 3142.36]  all the stuff
+[3142.36 --> 3142.66]  and you're like,
+[3142.74 --> 3143.20]  Pythagoras,
+[3143.32 --> 3144.42]  just put your ruler down
+[3144.42 --> 3145.24]  for five minutes,
+[3145.38 --> 3145.60]  mate.
+[3145.92 --> 3146.80]  Have a sandwich,
+[3146.98 --> 3147.82]  I've cut them into triangles
+[3147.82 --> 3148.58]  while you like them.
+[3148.92 --> 3149.08]  You know,
+[3149.10 --> 3149.68]  that kind of thing.
+[3151.22 --> 3151.50]  Okay,
+[3151.58 --> 3152.88]  any other unpopulists?
+[3152.88 --> 3154.28]  I have a whole list
+[3154.28 --> 3155.56]  of photography
+[3155.56 --> 3156.42]  and popular opinions,
+[3156.42 --> 3158.04]  but the thing is,
+[3158.70 --> 3159.80]  I don't think anybody
+[3159.80 --> 3161.68]  actually has an opinion
+[3161.68 --> 3162.38]  on these things
+[3162.38 --> 3163.96]  and it's just these 10 people
+[3163.96 --> 3164.66]  and we're all
+[3164.66 --> 3165.60]  on the same slack
+[3165.60 --> 3166.34]  and we just,
+[3166.50 --> 3166.94]  you know,
+[3166.98 --> 3167.88]  discuss these things
+[3167.88 --> 3168.42]  between us.
+[3168.54 --> 3169.48]  So I'm not going to go there.
+[3169.48 --> 3170.62]  Instead,
+[3170.94 --> 3172.18]  my unpopular opinion
+[3172.18 --> 3173.28]  is that,
+[3173.74 --> 3175.20]  and Katie will,
+[3176.06 --> 3177.00]  I know she understands,
+[3177.20 --> 3179.04]  but dogs in the office
+[3179.04 --> 3179.92]  are bad.
+[3181.22 --> 3181.96]  Just bad.
+[3182.46 --> 3183.44]  Dogs in the office.
+[3183.50 --> 3184.90]  There should be no dogs
+[3184.90 --> 3185.62]  in the office.
+[3186.12 --> 3186.34]  Yeah,
+[3186.44 --> 3186.68]  go on,
+[3186.72 --> 3187.02]  elaborate.
+[3187.22 --> 3188.10]  Are you allergic to them,
+[3188.24 --> 3188.56]  Filippo?
+[3188.70 --> 3189.68]  I'm allergic to them.
+[3189.76 --> 3190.54]  I know a bunch of people
+[3190.54 --> 3191.42]  who are allergic to them.
+[3191.50 --> 3192.16]  I know a bunch of people
+[3192.16 --> 3193.30]  who are scared of them
+[3193.30 --> 3194.90]  and don't feel like
+[3194.90 --> 3195.56]  they can say,
+[3195.92 --> 3196.16]  hey,
+[3196.38 --> 3197.96]  so beautiful dog.
+[3198.24 --> 3198.98]  I'm scared of it
+[3198.98 --> 3200.16]  so you don't get to bring it
+[3200.16 --> 3201.34]  to the office anymore
+[3201.34 --> 3202.50]  because I'm scared of dogs.
+[3202.62 --> 3204.48]  No one wants to be that guy.
+[3205.00 --> 3205.12]  Yeah,
+[3205.16 --> 3206.00]  I know you love it
+[3206.00 --> 3206.50]  but to me,
+[3206.58 --> 3208.32]  that's basically a little monster
+[3208.32 --> 3209.82]  from a nightmare.
+[3209.82 --> 3211.04]  somebody might have gotten bitten,
+[3211.36 --> 3211.98]  you know,
+[3212.12 --> 3213.32]  and they're just like,
+[3213.40 --> 3213.60]  yep,
+[3213.64 --> 3215.06]  that makes me extremely uncomfortable
+[3215.06 --> 3216.32]  but I just joined
+[3216.32 --> 3217.98]  and I don't want to be that guy.
+[3218.38 --> 3219.78]  So they're not going to tell you
+[3219.78 --> 3221.84]  and they're just going to walk around
+[3221.84 --> 3222.22]  and be like,
+[3222.48 --> 3222.74]  yep,
+[3222.94 --> 3223.12]  yep,
+[3223.30 --> 3223.54]  cute,
+[3223.66 --> 3223.84]  cute,
+[3223.94 --> 3224.16]  cute,
+[3224.16 --> 3225.86]  walks along the border.
+[3226.16 --> 3227.92]  And to be clear,
+[3228.02 --> 3229.56]  I think Filippo said that about me
+[3229.56 --> 3230.66]  because I love dogs
+[3230.66 --> 3231.48]  like more than,
+[3231.76 --> 3232.86]  anybody who's ever talked to me
+[3232.86 --> 3233.78]  for more than five minutes
+[3233.78 --> 3235.18]  knows that I love dogs
+[3235.18 --> 3237.46]  more than pretty much anything,
+[3237.54 --> 3238.04]  I would say.
+[3238.30 --> 3238.56]  Whoa.
+[3238.86 --> 3239.04]  Yeah.
+[3239.60 --> 3239.90]  I mean,
+[3239.92 --> 3242.36]  I do actually agree with you
+[3242.36 --> 3244.64]  that it makes things complicated.
+[3244.78 --> 3244.98]  I mean,
+[3245.02 --> 3245.18]  like,
+[3245.24 --> 3245.38]  yeah,
+[3245.42 --> 3246.72]  it can bring a source of joy
+[3246.72 --> 3247.96]  for people like me
+[3247.96 --> 3249.14]  who aren't allergic
+[3249.14 --> 3249.86]  and that love them
+[3249.86 --> 3250.44]  but also
+[3250.44 --> 3252.74]  if it's a source of conflict
+[3252.74 --> 3253.82]  and discomfort
+[3253.82 --> 3254.70]  or worse
+[3254.70 --> 3256.06]  for people that I work with
+[3256.06 --> 3257.32]  or people around me
+[3257.32 --> 3259.70]  then that isn't ideal either
+[3259.70 --> 3260.66]  and,
+[3260.66 --> 3261.76]  you know,
+[3261.76 --> 3262.42]  aside from the case
+[3262.42 --> 3263.20]  of like a service dog
+[3263.20 --> 3264.42]  which I know that Filippo
+[3264.42 --> 3266.60]  agrees that's totally fine.
+[3267.00 --> 3267.32]  Honestly,
+[3267.44 --> 3269.24]  I think it's a really reasonable opinion.
+[3269.68 --> 3271.32]  Service dogs are well trained
+[3271.32 --> 3272.70]  and in general,
+[3273.26 --> 3273.52]  you know,
+[3273.62 --> 3275.90]  if accommodations have to be made,
+[3276.12 --> 3276.70]  you know,
+[3276.86 --> 3279.56]  one can work case by case
+[3279.56 --> 3280.44]  but honestly,
+[3280.52 --> 3282.24]  I never had the problem of,
+[3282.24 --> 3282.66]  oh no,
+[3282.74 --> 3283.40]  I'm really allergic
+[3283.40 --> 3284.46]  to the service dog
+[3284.46 --> 3286.84]  that I can't not be around
+[3286.84 --> 3288.54]  but I did have the problem
+[3288.54 --> 3289.52]  with pets a bunch
+[3289.52 --> 3290.80]  because there's just
+[3290.80 --> 3291.74]  many more pets.
+[3292.00 --> 3293.20]  It's just a numbers problem.
+[3293.76 --> 3294.68]  But how will the management
+[3294.68 --> 3295.74]  show how cool they are
+[3295.74 --> 3296.80]  if they don't allow dogs
+[3296.80 --> 3297.54]  in the office?
+[3298.16 --> 3298.56]  Right.
+[3298.76 --> 3299.72]  What are you going to ban next,
+[3299.84 --> 3300.12]  Filippo?
+[3300.18 --> 3300.40]  Right.
+[3300.58 --> 3301.48]  Foosball tables.
+[3301.66 --> 3302.40]  Pim-pom tables
+[3302.40 --> 3303.44]  have gotten old.
+[3304.50 --> 3304.76]  Yeah.
+[3305.02 --> 3305.38]  There you go.
+[3306.78 --> 3306.98]  Yeah.
+[3307.96 --> 3308.36]  Roberto,
+[3308.80 --> 3309.52]  what do you think?
+[3309.58 --> 3310.44]  How do you feel about dogs
+[3310.44 --> 3310.98]  in the office?
+[3312.54 --> 3314.72]  I'm kind of scared of big dogs
+[3314.72 --> 3316.50]  so I'm on Filippo's side
+[3316.50 --> 3317.56]  but also I have friends
+[3317.56 --> 3319.32]  that are allergic to dogs
+[3319.32 --> 3320.66]  so yeah,
+[3320.90 --> 3321.68]  I agree.
+[3322.28 --> 3323.28]  Unless they are needed
+[3323.28 --> 3324.62]  like they are service dogs,
+[3325.12 --> 3326.52]  I'm not in favour of that.
+[3327.66 --> 3327.86]  Well,
+[3328.12 --> 3328.58]  folks,
+[3328.78 --> 3329.68]  your unpopular opinions
+[3329.68 --> 3330.92]  have so much to discuss
+[3330.92 --> 3331.06]  on.
+[3331.10 --> 3332.00]  My unpopular opinion
+[3332.00 --> 3332.60]  was going to be
+[3332.60 --> 3333.70]  I like yellow
+[3333.70 --> 3335.00]  so wow,
+[3335.18 --> 3335.68]  this is just
+[3335.68 --> 3336.90]  important topics
+[3336.90 --> 3337.36]  that you brought
+[3337.36 --> 3338.50]  to the conversation there.
+[3338.80 --> 3340.04]  Mine was completely useless.
+[3341.72 --> 3342.96]  That's a terrible opinion,
+[3343.12 --> 3343.28]  Rob.
+[3343.36 --> 3343.94]  Take it back.
+[3344.50 --> 3345.68]  Is it the colour you like
+[3345.68 --> 3346.52]  or you just love
+[3346.52 --> 3347.44]  that Coldplay song?
+[3347.68 --> 3348.04]  No,
+[3348.10 --> 3348.50]  the colour.
+[3348.76 --> 3349.44]  Just the colour
+[3349.44 --> 3350.46]  and the effect
+[3350.46 --> 3351.50]  it has on people.
+[3352.04 --> 3352.88]  There are so many
+[3352.88 --> 3353.60]  better colours.
+[3354.82 --> 3355.26]  Right.
+[3355.58 --> 3356.02]  Yeah,
+[3356.02 --> 3356.96]  beautiful colours.
+[3357.46 --> 3358.24]  Yellow is one of them.
+[3358.54 --> 3358.80]  I mean,
+[3358.82 --> 3359.68]  your headphones are yellow.
+[3360.26 --> 3360.64]  That's true.
+[3360.70 --> 3361.18]  Now I was looking
+[3361.18 --> 3362.04]  for clues of yellow
+[3362.04 --> 3362.92]  to verify.
+[3363.24 --> 3363.72]  For some reason
+[3363.72 --> 3364.38]  I'm sceptical
+[3364.38 --> 3365.62]  when Roberto says
+[3365.62 --> 3366.26]  he likes yellow.
+[3366.26 --> 3366.56]  I think,
+[3366.64 --> 3366.78]  well,
+[3366.84 --> 3367.66]  this is a trick.
+[3367.78 --> 3367.98]  Well,
+[3368.12 --> 3369.34]  I have something else
+[3369.34 --> 3370.04]  to prove it.
+[3370.62 --> 3370.82]  Oh,
+[3370.88 --> 3371.74]  a yellow gopher.
+[3372.16 --> 3372.60]  Oh,
+[3373.10 --> 3374.22]  I don't have one.
+[3374.22 --> 3374.98]  It's a podcast.
+[3375.44 --> 3376.42]  This is a podcast.
+[3376.82 --> 3376.98]  Yeah.
+[3377.38 --> 3377.78]  So,
+[3377.86 --> 3378.92]  I already tweeted
+[3378.92 --> 3379.88]  this picture today
+[3379.88 --> 3381.16]  so people just need
+[3381.16 --> 3381.88]  to go back.
+[3382.02 --> 3382.82]  I'll tweet it again.
+[3382.84 --> 3383.98]  I'm scared of those gophers
+[3383.98 --> 3384.42]  by the way.
+[3385.04 --> 3386.12]  That's my unpopular opinion.
+[3386.32 --> 3387.10]  Those little things.
+[3387.32 --> 3388.36]  I have nightmares about them.
+[3388.70 --> 3389.92]  They do look a little weird,
+[3390.06 --> 3390.20]  huh?
+[3390.60 --> 3391.42]  That one in particular,
+[3391.54 --> 3392.52]  that yellow one.
+[3392.52 --> 3393.28]  Well,
+[3393.34 --> 3393.82]  it's a podcast.
+[3394.06 --> 3394.80]  So this really is.
+[3394.80 --> 3395.52]  It's still a podcast.
+[3396.00 --> 3396.30]  Basically,
+[3396.56 --> 3397.72]  I'm getting the gopher
+[3397.72 --> 3398.42]  closer and closer
+[3398.42 --> 3398.96]  to the webcam
+[3398.96 --> 3400.10]  until Matt screams.
+[3400.46 --> 3401.04]  He didn't scream.
+[3401.28 --> 3402.06]  You passed the test.
+[3402.18 --> 3402.34]  Sorry,
+[3402.40 --> 3403.62]  I wasn't suggesting
+[3403.62 --> 3404.78]  do an audio commentary
+[3404.78 --> 3405.16]  of it.
+[3405.20 --> 3405.78]  I was suggesting
+[3405.78 --> 3406.86]  let's not do that
+[3406.86 --> 3407.74]  in the first place
+[3407.74 --> 3408.90]  and focus on the audio.
+[3409.22 --> 3409.66]  Right.
+[3410.80 --> 3411.12]  Okay,
+[3411.20 --> 3411.40]  well,
+[3411.58 --> 3412.10]  unfortunately,
+[3412.30 --> 3413.42]  that's all the time
+[3413.42 --> 3414.54]  we have today.
+[3414.90 --> 3416.16]  Thank you so much
+[3416.16 --> 3417.12]  for joining us,
+[3417.26 --> 3417.62]  Katie,
+[3418.12 --> 3418.62]  Filippo,
+[3418.70 --> 3419.22]  and Roberto.
+[3419.22 --> 3421.16]  And we'll see you next time.
+[3424.86 --> 3426.14]  If you're not following
+[3426.14 --> 3427.00]  GoTime on Twitter,
+[3427.26 --> 3428.32]  let's fix that bug.
+[3428.76 --> 3430.30]  We tweet live show notifications,
+[3430.82 --> 3431.24]  clips,
+[3431.52 --> 3432.16]  and highlights
+[3432.16 --> 3433.28]  from past episodes.
+[3433.50 --> 3434.32]  We take polls
+[3434.32 --> 3435.54]  about unpopular opinions
+[3435.54 --> 3436.66]  and have a lot of fun.
+[3437.08 --> 3437.90]  Join the conversation.
+[3438.16 --> 3439.64]  We're at GoTime FM.
+[3440.40 --> 3441.82]  This episode was hosted
+[3441.82 --> 3442.78]  by Matt Reier
+[3442.78 --> 3443.72]  with special guests
+[3443.72 --> 3444.46]  Katie Hockman,
+[3444.64 --> 3445.48]  Roberto Clapis,
+[3445.70 --> 3446.86]  and Filippo Valsorda.
+[3447.02 --> 3447.64]  It was produced
+[3447.64 --> 3448.50]  by Jared Santo.
+[3448.50 --> 3449.32]  That's me.
+[3449.54 --> 3450.68]  And we get our music
+[3450.68 --> 3451.64]  from the Beat Freak
+[3451.64 --> 3452.84]  Breakmaster Cylinder.
+[3453.34 --> 3454.50]  Thanks to this episode's
+[3454.50 --> 3454.92]  sponsors,
+[3455.06 --> 3455.66]  Digital Ocean,
+[3455.86 --> 3456.28]  Retool,
+[3456.46 --> 3457.00]  and Pixie.
+[3457.16 --> 3457.68]  And of course,
+[3457.78 --> 3458.80]  our long-time partners
+[3458.80 --> 3459.34]  Fastly,
+[3459.58 --> 3459.94]  Linode,
+[3460.14 --> 3460.80]  and Rollbar.
+[3461.32 --> 3462.18]  That's all for now.
+[3462.48 --> 3463.30]  We'll talk to you again
+[3463.30 --> 3463.80]  next week.
+[3463.80 --> 3475.98]  We'll see you again next week.
+[3476.14 --> 3477.34]  Yeah.
+[3477.34 --> 3507.32]  Thank you.
+[3507.34 --> 3537.32]  Thank you.

Füźžįñg_transcript.txt

@@ -0,0 +1,407 @@
+**Mat Ryer:** Hello, and welcome to Go Time. I'm Mat Ryer. Today we're talking about fuzzing. We're gonna find out what it is and how we can use it to make our code better... And we're gonna take a close look at a new draft design that discusses bringing fuzzing as a first-class concern to Go. It's very exciting, and we're lucky to be joined by the author of that draft design, Katie Hockman. Hello, Katie.
+
+**Katie Hockman:** Hi, Mat. How's it going?
+
+**Mat Ryer:** Good! Welcome to the show. Thanks for coming.
+
+**Katie Hockman:** Thank you for having me.
+
+**Mat Ryer:** We're also joined by Filippo Valsorda. Hello, Filippo.
+
+**Filippo Valsorda:** Hey, Mat. Good to be back.
+
+**Mat Ryer:** Always a pleasure to have you here, sir.
+
+**Filippo Valsorda:** Same, same, same. Looking forward to it.
+
+**Mat Ryer:** \[unintelligible 00:02:35.25\] Thank you. And we're also joined by Roberto Clapis. Hello, Roberto.
+
+**Roberto Clapis:** Seven hundred and forty eight.
+
+**Mat Ryer:** \[laughs\] Okay... Is that a fuzzed response?
+
+**Roberto Clapis:** Yeah. I wanted to see if you crash to an integer. \[laughter\]
+
+**Mat Ryer:** I've not crushed though, nor am I panicking. I've continued... In fact, that was in my unit test earlier, so I was ready for it. But thank you very much, welcome to the show.
+
+**Roberto Clapis:** Thanks.
+
+**Filippo Valsorda:** Can we take a second to acknowledge how Mat rolled the r's for both the Italian names?
+
+**Roberto Clapis:** Yeah, that was good. \[applause\]
+
+**Mat Ryer:** Oh, it's my pleasure. It's a lovely accent, so I always like to listen to it and have you on for that purpose, really. So if that's all you contribute to this show, then that's fine by me.
+
+**Roberto Clapis:** That's our intention.
+
+**Mat Ryer:** So maybe we should start at the beginning then, for anybody not familiar... What is fuzzing, and what's it for?
+
+**Katie Hockman:** Yeah, I can give you a quick summary to that. Basically, fuzzing is a form of automated testing that can manipulate inputs in a way they can find bugs that maybe you wouldn't otherwise be able to find on your own. In my mind, it's kind of a supplement to some of the existing testing that people already do, and it's pretty common, like unit testing or integration testing... But what sets it apart is it actually does things on its own, and kind of runs continuously; so it's kind of smart, in a way. If it has some interesting inputs, it can actually use some intelligence to go in and mutate those inputs in interesting and meaningful ways, to find crashes and panics that wouldn't easily be otherwise found if the developer had to try to identify them themself.
+
+**Mat Ryer:** \[04:22\] That's interesting then. You talk about this intelligence. It isn't just random then. There's something else going on.
+
+**Katie Hockman:** Yeah, and I think it's really tricky, because there's no industry standard on how these kinds of things work. I mean, there are definitely tons of different ways that you can mutate things randomly, and then there's also a lot of interesting discussion around how do you prioritize which corpus entries - and I'll talk a little bit more about what corpuses are later, but basically which inputs to modify and how to modify them, and how smart it should really be. All those things are kind of up in the air, and a lot of different fuzzers work differently, which is actually kind of cool in my mind.
+
+**Mat Ryer:** Yeah, that's interesting. So what situations is it good for helping out in then? Let's say the strings.split example from the standard library. You pass in a string and you pass in a separator, and it basically just splits that string wherever it finds that separator, and returns a slice of the components, the segments that it found. Would that be a good candidate for fuzzing?
+
+**Katie Hockman:** Yeah, I think it could be, and I think it would also be -- and Filippo and Rob will have a lot of really good things to add too, in terms of who has used fuzzers in the past, and how they've usually had a security context around them... And what this proposal has been trying to do is actually get fuzzing into the hands of non-security experts and non-security developers, and have other people use them.
+
+In the string split example, if there's an off-by-one error somewhere, or maybe some issue that can cause a panic, or some input that doesn't meet some specific property, it might be easier to find with fuzzing... And I think it would be a really good package to test, or a good function to test in that package, yeah.
+
+**Mat Ryer:** Yeah, because you hear that common use cases are things like parsers, and things that are doing decoding, because they are dealing with usually unknown in advance structures that maybe they have to infer along the way... So there is a lot of room in that kind of operation for things to go wrong, or unexpected input. Just things that you would never imagine anyone would pass in. So yeah, that's what separates it out from unit tests really, I guess, because unit tests are very deliberate, aren't they?
+
+**Katie Hockman:** Yeah, they are. You give a set of inputs, and you run something, and then you look at the output; it's very clear, and you have to say "These are the inputs that I think are important, that should test it well enough, and then it should have this output." Fuzzing, I think, can apply to a lot of context beyond parsers and things like that, because there's a reason that we have unit tests everywhere, and there's a reason we don't just test parsers, or difficult cryptography, or things like that.
+
+There's a reason we test everything, and it's because we don't always know where the bugs in our code are. We have default assumptions that our code works, and so we just kind of test it, kind of just in good faith sometimes, just to prove that it works, and I think a fuzzing engine should be pretty agnostic, in the sense that it doesn't assume that it's gonna work... And it's gonna go and maybe find things that you didn't really realize could actually break, or something that you had overlooked, that you didn't realize is a dependency somewhere else that might break.
+
+**Roberto Clapis:** Right. I would also add that when you write the fuzz test target, you want to expect on properties of the stuff that you work on. Instead, when you work on unit tests, you expect some output. For example, in the strings.split case you can say "I'm going to call a strings.split with two parameters, and I'm going to check that the second one never appears in the returned slices", because the separator should never appear.
+
+\[08:02\] And that is something you would generally not test in a unit test. Or you're going to check that the returned slices are less than the characters of the string. If you return more characters than there are there must be a problem, and this is stuff that normally doesn't get tested... I'm pretty bad at writing tests, but whenever I unit-test, I don't test for this kind of condition.
+
+**Filippo Valsorda:** Yeah, and another example of something that would be good to check in a fuzz test of the split function is that if you put it back together, putting the separators between the things you split, do you get back the original string? If you do, it probably did its job right. And that's the kind of stuff that fuzzers are pretty good at finding, because they can just go and find some input where -- I don't know, the separator is at the end, and is missing one character, or I don't know, where the thing doesn't roundtrip.
+
+**Roberto Clapis:** That gives you even more, because then you now are testing for an additional property, which is if you string split and then you string join, you must get the same thing out... Which is a normal expectation. When I use the strings package, I expect that to be true... But I don't know if there is anyone that has been fuzzing that to make sure that that is actually true... Especially in edge cases like nil slices, or slices of empty strings, what happens would be interesting to see.
+
+**Mat Ryer:** Yeah. So there is an element then of design here. You have to think of that kind of thing, that property, to then model it in a fuzz test, right? It's not just you just point it to a method and it just fills the method up with nonsense.
+
+**Katie Hockman:** I think yes and no. I think it can. I think it depends on what you're using it for. You could just throw random input at a function to see if it panics. That is a property that can be tested, and you don't have to know anything about it. I think it can also be used for things like differential testing, or property testing, or a lot of different things. It can be a supplement to unit tests, but it can also just go find a crash, and you could probably do that in a couple lines with a little thought.
+
+**Filippo Valsorda:** Differential testing is something that honestly works a little too well. The idea is that there are multiple implementations of the same thing. For example, big number implementations - it doesn't matter what library you use; if you multiply two arbitrary precision decimals, you should get the same arbitrary precision decimal out. Sounds right, right? Oh, my friend, how many bugs fuzzers have found just by telling them "Yup, so here's two functions that need to return the same thing. Cool. Go!" I get emails because one of the ones that are tested is the Go one, and I get emails when there's a mismatch between the Go one and some other implementation, and... Oh, boy. Yup, multi-precision is hard. So yeah, that's an excellent example.
+
+**Roberto Clapis:** One thing that I did with differential testing was at one point in Go a bug was fixed - a problem with header parsing - and I thought "This looks easy to find with a fuzzer", so I just imported fasthttp and the standard HTTP libraries, both in Go; I ran go-fuzz for 25 minutes and I found a bug. The bug there was just fixed, and had been there for 12 years. So yeah, if you want to assert for a property and the case was "I want the header set to be identical", it's quite easy to find problems. And if I recall correctly, when at one point the JSON package was heavily optimized, and there was a differential fuzzer in place that checked the old version and the new version would parse the JSON the same way... And it found a bug before it hit a stable release, which would have been kind of bad. So that was another success story of fuzzing not for security reasons, that was just yet another test.
+
+**Mat Ryer:** So how can you do fuzzing in Go today then? What are the choices that we have?
+
+**Katie Hockman:** \[12:03\] There are a few. I can speak to at least one or two. I think the common one is go-fuzz. That's the one that everyone knows about. That was written primarily by Dmitry Vyukov. It's really amazing; I've spoken to him about it, and he's actually given a lot of really good feedback into the proposal that's out there now. So it's been nice to partner with him a little bit on that too, and have him give some feedback on that. It's really neat, and if you haven't used it, you should definitely check it out.
+
+And then another tool that somebody wrote was fzgo. I think that was kind of a proof of concept written mostly by the \[unintelligible 00:12:45.20\] it sounded like, to try to integrate it a little bit more kind of with the go command, and making it look more like an end-to-end tool that wouldn't have to have so many build steps like go-fuzz has... And add a little bit of support for modules, I think, was part of that... Or maybe that was a part of go-fuzz. But there's been different features that both of them have tried to basically model and see how they would work... And I think fzgo was meant to be kind of a prototype or an experiment of what it might look like as a final approach.
+
+**Mat Ryer:** And you mentioned build steps there then... So it isn't just a tool that runs at runtime; there's other things that happen. Is there some kind of introspection that happens, or reflection on the types, and things? Is it kind of generic, in some way?
+
+**Katie Hockman:** Well, when I say build steps, I don't remember all of the exact details of how go-fuzz works, but I do know that it has a go-fuzz build, and you have to build the binary that will be fuzzed, and then you have to run it separately, and kind of manage your own corpus... So there's a lot of different steps. You can't just run one command with the Go toolchain as it is today. You kind of have to learn a different workflow, which was a bit of a -- just like an impediment for some people to try to start it, because they didn't wanna learn a new tool.
+
+**Roberto Clapis:** I would say that that is one of the main reasons why people are not using it, is because it's external, and it feels different. Also, one thing that it does - it does a source-to-source transformation. So it takes your source code and implements some sort of checkpoints. Basically, when your code runs, it can check at which point it got. So basically, why your code executes, it can check how much of the code was covered... More or less like the cover tool, but it needs to do it more heavily than the cover tool, and in a more efficient way... And this is one of the reasons why it was quite hard to make it support modules, because it actually rewrites the sources.
+
+**Filippo Valsorda:** Some context here is that part of what makes fuzzers magic is that - well, the recent generation of fuzzers, since I think AFL - they use coverage to figure out what mutations are the ones that are interesting to look at. Katie was talking about how there's different strategies for this, but in general, the common denominator is that they all look at the cover of your code. If you ever run gotest-cover for a file - I don't remember the flag, but anyway, if you ever generate a coverage report, with the green and the red, that's what fuzzers do; they run the input and check which parts light up. And if they change the input and some new code lights up, the fuzzer goes like "A-ha! Okay, this is useful. I can keep changing this and maybe I'll hit another path that takes from there, or maybe I'll be able to combine two paths in a way that were not tested together." And that's what makes them, honestly, kind of freakishly effective.
+
+There's a demo of AFL slowly building a valid JPG out of nothing, and it slowly makes a picture and it figures out the letters, it puts in the tags, and everything. It's scary good.
+
+**Roberto Clapis:** \[16:01\] One thing that really scared me was when I ran go-fuzz against the HTTP library, and after a while I saw that in the corpus something that looked like random started appearing. I was like "Oh, cool." The Go standard package started accepting something that is not HTTP, because it was HTTP/2. Basically, it started constructing valid HTTP/2 requests from nothing. That was scary, and also, I was ashamed because I did not recognize it, and I had to manually write to decompress it and see what was going on.
+
+**Filippo Valsorda:** Rob, if you can ever read HTTP/2 with the naked eye, you need to tell me.
+
+**Roberto Clapis:** \[laughs\]
+
+**Mat Ryer:** Yeah, because that is a strange super-power. I don't know what has had to bite you for that to be the power that then manifests.
+
+**Filippo Valsorda:** There are support groups. We've all been there. Mine is TLS; it used to be DNS. It's okay... There's help.
+
+**Roberto Clapis:** \[laughs\] Thanks. I'm keeping them in mind.
+
+**Mat Ryer:** So that is really interesting then... It's not just shifting the inputs by some external means. It actually has an insight into the code that's running inside in your own code, in your own binary, and it uses that information to also influence what it's doing. So that is kind of like spooky, and I could definitely imagine -- it's a little bit like adversarial training in machine learning, where you have a model and you have another model, and they sort of compete with each other, and then they both just keep getting better together. It almost feels like cheating in some way, but you can end up with a mirror of something else by this technique. So it is kind of amazing to see -- it really will start to feel intelligent, and a few of you have said it's kind of spooky, this thing...
+
+**Katie Hockman:** Mm-hm. Just a note of another spooky thing is it can also reverse-engineer your code such that it can figure out -- there are certain tools that can figure out what the input is actually supposed to be, and then do that for you. So it can actually basically tell the fuzzing engine "This is what input will make this if statement pass", and then it'll just do that to kind of get unstuck from wherever you're at with the fuzzing engine. That's something maybe you do -- I think go-fuzz does this once every thousand mutations, just to try to unstick it, but not every time, because it's too expensive... So it's a lot of trade-offs of like "How random do you want this to be? How much do you want to use prioritization of certain inputs? Is coverage a metric? What is that in terms of feedback loop? How much do you care about it in terms of other things?" So it's kind of creepy, and it's a judgment call on the developer and how they wanna design that, too.
+
+**Mat Ryer:** Yeah, it does sound like a kind of hacker's tool, doesn't it? And in fact, didn't it have its origins in the security world?
+
+**Filippo Valsorda:** Yeah, but I like what Katie just said, that it's a trade-off that the developer has to make... And I think she meant the developer of the fuzzing tool; correct me if I'm wrong...
+
+**Katie Hockman:** Yes, yes.
+
+**Filippo Valsorda:** ...because that's the thing - the thing I like about the proposal is that it does not leave all these decisions and the necessity to learn about all of this stuff to the end users, to the Go developers that are just trying to test their code.
+
+**Roberto Clapis:** And also, if you look at the proposal, it tries to make fuzz test targets as close as possible to what a test looks like nowadays... So basically, the friction to adopt fuzzing if you're used to write unit tests - and if you're not, you should - is going to be very low, because it's going to basically slightly change the pattern, but it's going to be as close as possible.
+
+**Mat Ryer:** Yeah, we should talk more about that proposal, but before we do, I just wanna get a few other concepts clear. There's this concept of seeding the corpus; there's this concept of kind of giving the fuzzing tool some kind of head start... A bit like with unit tests, where you say "We know these are the inputs and these are the expected outputs." You also kind of seed the fuzzing tool in a similar way, don't you?
+
+**Katie Hockman:** \[20:09\] Yes, and I think it's also kind of a goal of the proposal to try to make it such that the unit tests that people have now and the use cases that they've already come up with can basically just be directly used as seed corpus. The seed corpus is filling two needs, at least in terms of this Go proposal. First of all, seeding the mutation engine, seeding the corpus is trying to tell it "This is a good starting point for you. Build off of this", and then it can manage its own corpus on its own, as it wants to, and build it up as it finds new coverage and new interesting things. But it also can serve as a regression test of sorts. The seed corpus is either checked into your test data directory, it's basically checked in directly into your module or into your package, or it's in there programmatically. It's in your tests in code. That's run every single time go test is run, and it's also meant to act as a regression test. So you can use existing things, you can use new crashes, and you can build up that seed corpus as you find new regressions that you wanna make sure you're testing.
+
+**Mat Ryer:** Yeah, so that's a really cool feature that if something fails, that automatically gets contributed to the testing. So the next time, that will explicitly get tested is how it works. That is very cool, because of course, the value of unit testing, in the cases where you find a bug and then you write a test to prove that bug, which you do if you follow TDD tightly... And in some cases I find that to be a great way to work, because you get a kind of to-do list for free from the toolchain. And as you write your test, if things aren't working, they fail, you get errors that you then have to unblock, and it's a kind of nice way to decide what you have to do to get something to pass.
+
+Yeah, so it has that same kind of idea - if you find a bug and you've written a test to prove it, you then save that test, and the next time you run all your test suite, it'll check for that bug again. So this is what we mean by protecting from regression - you can never have that same bug again if you've fixed it and you keep the unit test.
+
+What do we do with that corpus though? Dominic Rouse on Twitter asked "What are the best practices for the corpus? Should you put it into Git, should it go into some other repo? Do you share it amongst the team? Is it just something you run on your own dev machine? Where does this go in practice?"
+
+**Katie Hockman:** I think it's gonna depend, and I also think this is a bit of an open question in terms of what kind of best practices do we wanna lay out for this... But that part is also kind of up to the developer, too. It could be programmatic - like I mentioned, before you have existing unit tests, and you just wanna change your t.run into an f.fuzz; something like that should be basically possible. So if it's already programmatic, keep it programmatic. And if it fails, it fails, and that's great.
+
+If you have a bunch of test data, like let's say you have a bunch of big HTTP requests, or binary files or something like that that you already have somewhere, you can just use those too, and go test will look at test data as part of the seed corpus, too. And so I think it also depends on what the seed corpus is - is it a huge binary? Is it a small thing? Is it something that's best built programmatically, and what the best practices for that will be I think are still kind of an open question... At least it is to me.
+
+**Filippo Valsorda:** \[23:46\] I think there's also an angle of maturity of the ecosystem in there, of maturity of the technique... Because when fuzzing is just this tool that some security researchers use to smash against a program once, try to get something out of it and then move on - of course, they just run the corpus wherever they're keeping it. But I feel like just like with testing we set up continuous integration and we trust machines to do the heavy lifting for us, I expect that fuzzing also takes that path once it's built into developer workflows.
+
+So you would have a small corpus locally on your machine, and Katie's proposal puts it automatically in a cache folder... That will do a very quick pass, but you're not gonna run the fuzzer mostly on your laptop. Part of what makes fuzzers work is that computers are fast, but also you can keep throwing more cores at it. And then you upload it, and some CI or OSS-Fuzz or some continuous integration system can just run the fuzzer, and it should persist the corpus, so it will keep running the same corpus against it, so that you make changes and the corpus is already hot and large, but is not checked into your repository, because most people don't want megabytes and megabytes of corpus checked in.
+
+**Roberto Clapis:** Right. One thing that I also like about fuzzers is that there's usually a way to tell them "Don't feed me input that is bigger than this amount", either directly or indirectly. The indirect way is you take whatever the fuzzer passes you, and if it is bigger than a certain size, you just return "No, I don't want this." After a while, the fuzzer will stop seeding the corpus with anything bigger than the size you want; so if you're testing strings.split - yes, you can get up to a megabyte, but it doesn't make sense to split a gigabyte of string... Because you know the code that you're fuzzing, and you shouldn't be too exaggerated on how liberal you are in the input you fit it to. It's like, yes, you're fuzzing, but you know what you're fuzzing. If you're fuzzing a JPEG parser - yes, feed it big stuff. If you're fuzzing a string splitter, it's very hard that there is a bug at the 3 gigabytes mark.
+
+**Mat Ryer:** Yeah, that's a good point, because you do get the sense that you'd switch it on and it just points to your methods, and it's just gonna go and do it. That is interesting though, that this is a continuous thing; it's not something that you would do like a benchmark, where you just run that on your laptop. But in the proposal there is a new flag to run the fuzz, but is the expectation that that would run in some kind of continuous integration, or some other place?
+
+**Katie Hockman:** I think it probably depends on how long someone wants to run a fuzzer. If they are willing to just let it run on their machine for a while, maybe that's okay. If they wanna just run it for the weekend, that's totally fine. If it's a company - or just an individual - and they have a ton of different things they wanna try to fuzz at once, I'm not really sure if that's even gonna be supported to be able to run multiple fuzzers at once. I don't know what would happen, like if there's a race condition.
+
+There's a lot of different things I'm not totally sure would be supported. If it crashes something, somewhere, it's hard to know where it's coming from, so it may make more sense in situations like that to have it on some kind of continuous integration.
+
+**Mat Ryer:** I wonder if we're gonna end up in a situation like with Bitcoin miners, where we've got all these machines that are just spending all the time crunching through, fuzzing stuff... \[laughter\] When we've got Fuzzcoin.
+
+**Filippo Valsorda:** OSS Fuzz already exist. There's a project by Google that basically provides what internally we call ClusterFuzz, which I don't know if I was allowed to say; but yup, we're rolling! For open source projects, where any open source project can submit... And there are criteria, of course; I don't know what they are exactly, but they will just run your fuzzers for you. And if we make it standard how to do that with Go, it would be extremely easy to submit Go projects.
+
+**Mat Ryer:** Yeah, that gets very exciting, actually. That's really cool.
+
+**Roberto Clapis:** I think ClusterFuzz is open source.
+
+**Filippo Valsorda:** Cool! I'm not getting fired today.
+
+**Katie Hockman:** \[laughs\] Yeah, don't get fired, please.
+
+**Mat Ryer:** But if you do want to get fired, please do it this way - come on the show and reveal something that you shouldn't reveal. \[laughter\] It's so cool for us! It's such a scoop!
+
+**Filippo Valsorda:** I've got a history with that... And let's leave it at that and move on.
+
+**Katie Hockman:** Yeah, don't encourage him. \[laughs\]
+
+**Mat Ryer:** \[28:12\] Yeah. Last time Filippo was on the show he stopped me from admitting to a crime before I said it, which was brilliant. Really useful service. \[laughter\]
+
+**Break:** \[28:20\]
+
+**Mat Ryer:** We can take a short break if anyone needs to, and people at home can take a break anytime they want to, really. They're probably just carrying us around on their portable devices, so they can just do what they like... I don't know why I'm explaining that. \[laughter\]
+
+I was just gonna say, some bits will get cut out. If you need anything cut out, let us know and we'll do that.
+
+**Roberto Clapis:** Oh, Mat, I've listened to so many episodes of these in which you say "This will be cut out", and that never happens...
+
+**Mat Ryer:** I know. They don't do it for me, but they will do it for you three.
+
+**Roberto Clapis:** Okay. \[laughter\]
+
+**Katie Hockman:** Thank you.
+
+**Mat Ryer:** They add bits for me from other times I've embarrassed myself... \[laughter\] \[unintelligible 00:29:00.18\] extra show, and I'm like "I didn't embarrass myself then... That was a different time when I embarrassed myself."
+
+**Filippo Valsorda:** It's directly on the soundboard.
+
+**Mat Ryer:** Yeah, exactly. \[laughs\] It's just got me embarrassing myself. This one of them. This is one of the clips.
+
+**Break:** \[29:21\]
+
+**Mat Ryer:** So the new proposal - which we'll post a link to in the show notes - kind of has a very nice Go feel to it, like the design of it. So in the same way that we used to test functions being how we described unit tests, there are fuzz functions now which take a different argument, the testing.F. Is that like an interface then? What is that testing.F type?
+
+**Katie Hockman:** That testing.F type is very similar to a testing.T or testing.B. So it'll implement the testing.TB interface.
+
+**Mat Ryer:** Will there be a testing.F interface then, or is that like a strong type?
+
+**Katie Hockman:** It's a strong type.
+
+**Mat Ryer:** And that has methods on it that lets you then interact with the fuzzing stuff... But it's a relatively simple API, isn't it? It's just two methods... Is that right?
+
+**Katie Hockman:** Well, I didn't include in that proposal all the other methods that are in the testing.TB interface which it will support. For example, if you have some pre-work that you need to do and you wanna fail the test or something like that, because something failed, you can do that... Things like that. Originally, some earlier designs had the testing.F function; the f.fuzz function accepted testing.F, and then it ended up being not as clear, I think, and it was gonna complicate things quite a bit...
+
+\[32:30\] And there were some discussions that Filippo and I had, and we ended up basically keeping it as a testing.T within that function. So it basically should look almost exactly like a t.run. If you have a t.run, you can copy it over directly.
+
+So it should look and feel exactly like a unit test within that f.fuzz function, which runs kind of as a unit test... And then anything you need to do before that, like set things up, add to the corpus, whatever you need to do, you can use the testing.f for that part.
+
+**Mat Ryer:** Hm. Unlike the run function, where the only argument you can pass into that function is a testing.T, you can have additional arguments in these functions... And they seem somewhat dynamic. Can you explain how they work?
+
+**Katie Hockman:** Yeah, so inside this f.fuzz function, those first parameters, what you're basically telling it is it's gonna take a testing.T, basically it's scoped to this T, and then you're just telling it what things you want the fuzzing engine to be generating for you. What is the structure basically of each input in your corpus?
+
+In the proposal, the example is "It takes a testing.T, and A, which is a string, and then a big int, which is num." And what that's telling it is "Okay, we have an f.fuzz function. That's what's gonna be run with the fuzzing engine." That function is gonna be run for every input, it's bound by that T, and then the corpus is an A with a string and a big int. That's basically the structure of the corpus... So every time it runs, it should be running with a new string and big int.
+
+**Mat Ryer:** Does it dynamically look at the arguments that you've passed there and change the code? Does it respond to the arguments, or do you have to define them somewhere? Or are there patterns you have to follow?
+
+**Katie Hockman:** I'm not sure exactly, and I wanna make sure I'm explaining it right, but basically that string and that big int -- if you look up a little bit higher in the proposal and you're looking at this f.add function, what that's doing is it's adding to the corpus, and it's adding a string and a big int, which must look exactly the same as the string and the big int, in that order in that f.fuzz function. So what's that basically defining is "This is the definition of the corpus entries." That will be added manually and it will be generated by the fuzzing engine.
+
+**Mat Ryer:** And it works with a slice of empty interface, so it's kind of generic code, in a way.
+
+**Katie Hockman:** Yeah.
+
+**Mat Ryer:** If Go got generics, would that change/affect this design in any way, or do you think you'd still probably use it in a similar way?
+
+**Katie Hockman:** I'm not actually sure that it would impact the design... It might impact the implementation a little bit, but I haven't really thought too much about it. But also, just thinking about it now, I'm not actually sure that it would change much. I think what this function is supposed to do, this f.fuzz is just kind of like a -- it's a little bit magicky, but it's basically just trying to tell the fuzzing engine the structure that it should be aware of and be using.
+
+**Mat Ryer:** It's a nice API to be able to just define the function and have it notice that, or work at least... But what happens if you've added different kind of data, or you changed the structure? What happens in that case?
+
+**Katie Hockman:** Like if for example you didn't f.add with two ints, or something like that?
+
+**Mat Ryer:** \[35:57\] Exactly, yeah.
+
+**Katie Hockman:** I expect it would probably panic... Because what you're doing is you're basically telling it "Here's two ints", and it's expecting a string and a big int. And maybe that can work with static check and things like that, to find those things at build time...
+
+**Filippo Valsorda:** For anybody who hasn't read the proposal, f.add is the function used to seed the corpus; it's the function that you use to say "Here's the starting points." Which by the way, is one of my favorite things of the proposal, because usually you have to just create a bunch of files, one for each input, and put them in a foo.. Actually, I'm gonna do something else... And instead, here you just write f.add, and "Here's my ECDSA certificate, here is my RSA certificate. These are examples. Go for it." So f.add is the function that adds to the corpus, while f.fuzz is the function that actually runs the fuzzer, and it runs a function that takes the same types of arguments... I'm just mentioning it in case people haven't read the proposal yet.
+
+**Mat Ryer:** Thank you, brilliant. And I love the fact that it kind of still -- I mean, it's designed to fit into what we already have. So it knows about go test, and it kind of cooperates with go test as well, doesn't it?
+
+**Katie Hockman:** Yeah, and actually that was my main goal with all of this. I wasn't going to be okay with a design that didn't feel like testing that we have now. Someone should be able to look at this and hopefully understand it pretty quickly... And the goal is that if you know how to write a unit test, you know how to write a fuzz target. It should be approximately as easy. I wanted it to be able to work with the go command as it is now, and if people run go test, it should just run the same way, and it shouldn't have to use anything terribly special, and it shouldn't have to do anything, learn that much new... I wanted the barrier to entry to be as low as humanly possible. So if it looks like Go code, that's the goal, and I'm glad to hear it.
+
+**Roberto Clapis:** I love that about the design, because I've done some research and I've seen people creating fuzz targets in the wild for parsers, and what usually happens is that they take whatever the fuzzer engine passes them and put it in the parser, and that's it. So they just basically check. The only property they check for is if it panics. That is kind of sad, because it's so much easier to feed something into your parser and then maybe serialize it, where I can process it again and check if it is the same.
+
+So it's kind of easier to write fuzz targets than people assume, but since fuzz seems to be such an alien concept, I've seen most fuzz targets to assert nothing; they just feed the input to the function they want to test. It's like, if testing strings.join, we just join, that's it; and then you don't check if you even get a string back. So there is a type system for that, but that's what you get. So I'm really looking forward for these to be first-class, and to be so close to the original test target, to see what people actually start asserting as a property... Because "Doesn't panic" seems to be a little bit too weak as a property.
+
+**Filippo Valsorda:** Yeah. If people get one thing away from this conversation, it should really be that fuzzing that is gonna be built in Go is not just about finding panics. It's not just "Feed some input and wait for it to crash." It's about writing as many invariants as you can think of and as many checks as you can think of, and then letting the fuzzer find the inputs for which the thing doesn't do what you want it to do.
+
+**Mat Ryer:** So would you say that fuzzing makes a lot of sense if you're working with multiple methods? I mean, in that example that Roberto gave, where you're encoding and decoding, because you can say something about the way that those two things should interoperate... But how can you make assertions on something if the input is completely random? What kind of assertion are you gonna make?
+
+**Roberto Clapis:** \[39:50\] One thing that I did - I was fuzz-testing a cache that I implemented... Caches are harder than people would normally assume, so I wanted to make sure that, for example, what I put in, I got back. So to test my cache, I did differential fuzzing with a HashMap. A HashMap is a perfect cache. I mean, it grows indefinitely, but I didn't care; it was just fuzz testing. So I just fed stuff to my cache, and when I retrieved it, if it wasn't there - meh; it was evicted. But if it was there, it should be identical to whatever was in that map. So you can have a simpler, dumber implementation of the algorithm you want to implement, or maybe a slower one. If you optimize your code, you can keep the old code, the slow one, to test against... And usually, slow code is easier to debug, and it's more reliable, and it's easier to write.
+
+**Mat Ryer:** Because it's slower and you can see what's happening. \[laughter\]
+
+**Roberto Clapis:** Not that slower, but yeah... That's kind of the point.
+
+**Filippo Valsorda:** Another example that I had written up for the Gopher blog is that I had this parser... No, sorry, not this parser, actually... This serializer. And you're like "How do you test a serializer? How do you know if the thing it generated is good?" Well, the thing I wanted to know was whether if it would work reusing buffers, for performance reasons; I didn't want to allocate a new buffer or zero the buffer every time. I just wanted to give it the old packet and say "Just serialize over this one."
+
+So what I did was write a fuzzer that would parse a packet, but in this case with the Go proposal I would not even maybe do the parse step. I would just tell it "Give me a random packet structure, and then serialize it on both empty an empty buffer of old zeroes, and on full buffers of all one bit." If they come out different, it means that it's not set in zeroes in some of the fields... And it did. That might or might not have been why some stuff in the cloud for DNS server wasn't working... And that's the kind of stuff you can find with fuzzers.
+
+In general, testing should really be about defining expected behaviors, and that's true of all kinds of testing. It's not just about defining expected inputs and outputs, it's about locking in expectations. Any expectation that you can define not strictly in terms of "This input needs to have this output", but just "The output needs to be longer than the input. The output needs to be shorter than the input." Anything like that you can put in a fuzzer, in a fuzz target.
+
+**Mat Ryer:** Hm... Kind of like meta-testing, or some kind of abstract testing, in a sense. You're not dealing with the specific values, but you still deal with the ideas, the variables...
+
+**Roberto Clapis:** Yes, which kind of takes away one big risk, the reason when you write unit tests. When you write unit tests, you have those assumptions in mind. What you're trying to test is like "I want strings.split to actually split the string." Then you go and test your stuff, and you put the input and you put the output, but you're just giving examples, you're not testing the actual property that you want. So I think that writing a property assertion for a fuzz target is actually closer to what you want to do usually in tests.
+
+Now, unit tests are always going to be needed, but if you put on top something that asserts the actual property that you meant, you're adding a lot of value.
+
+**Filippo Valsorda:** One opinion I heard that I'm not supporting - retweets are not endorsements - was that "Why would you write unit tests if you already know that your program is going to break on? Just don't write the bug." I mean... \[laughter\] Yes, yes, yes. I know. But there is a degree of truth to that. The things you can write unit tests -- unit tests are actually more useful for refactoring later and for regressions. But that's the thing - it's unlikely you will think of inputs that break on the program you just wrote, because you thought about those edge cases. And fuzzing will just not care about what you thought about. Fuzzing will find where it hurts.
+
+**Roberto Clapis:** \[44:04\] Right... And one thing that I like to say is that I write test targets for my future interaction with the code, because I also used to do TDD most of the time. So I write the tests, and then I write the code that implements whatever I am testing for... And in the future, when I refactor, one of the tests will pass, when I said that I write fuzzers for the code that I wrote in the past. So basically, the fuzzer makes sure that whatever is there is actually what it meant to do, and the tests are there so that the future code will keep doing it.
+
+**Katie Hockman:** I really like what Filippo said about "The fuzzing engine doesn't care what the developer thought about." I think that's the benefit of having -- that's why, for example, code reviews exist, because you need another person who's more objective to look at it, and I think a fuzzing engine can be this third-party objective being that just goes in and does everything it can to try to break it, and has no idea what you thought about it. It doesn't care about that, it just cares about trying to find as much coverage as it can, and trying to find bugs. That kind of third-party entity is a cool concept to me.
+
+**Mat Ryer:** But Katie, aren't you worried about the fuzzing thing becoming self-aware and then just going around doing lots of random crime?
+
+**Katie Hockman:** That's actually my goal with this... I'm actually trying to build a self-learning robot that'll just take over the language.
+
+**Filippo Valsorda:** How do you know--
+
+**Mat Ryer:** Based on fuzzing.
+
+**Katie Hockman:** Definitely. Exactly.
+
+**Filippo Valsorda:** How do you know that's not already what happened, and we're here pitching fuzzing to just make our fuzzer overlords happy? \[laughter\]
+
+**Katie Hockman:** I'm actually a fuzzing engine. All this time it's been a simulation. I hope that's okay.
+
+**Mat Ryer:** Well, it is a good one. You're right, it does a good job, yeah. But the thing is -- yeah, I'd love that though... Not really. \[laughter\]
+
+**Roberto Clapis:** He doesn't know how to interact with you now that he knows you're a robot... \[laughter\]
+
+**Katie Hockman:** \[unintelligible 00:46:00.08\]
+
+**Mat Ryer:** I love it when the machines do kind of get this emergent intelligence. I find that to be really quite amazing, especially when there's so much chaos in what's actually going on. I think the thing that I've learned and I'll take away is it's less about random input and it's more about variations of the realistic kind of input that you're gonna pass in, right? ...that didn't resonate, because I can tell on my screen that there's no... Go on, correct me if that's wrong.
+
+**Roberto Clapis:** No, I just wanted to say -- I was putting stuff on top of this, which is... The fuzzer doesn't care about what the code does, and that's important, because if we had a machine learning algorithm fuzzing our code, just trying to learn how the code behaves, at one point it would do as humans would. It would understand what the code is supposed to do, and kind of accept the code works. Instead, if you're just using an algorithm that just tries to bash with random stuff, at one point you find -- like, after two years you've been fuzzing a target, a new edge case that crashes... And this is something that I love, because a human, or an intelligent design - in our way of defining intelligence - would not find it... Because why would you keep doing for two years the same thing, expecting a different result? Isn't that the definition of madness?
+
+**Mat Ryer:** Yeah, but we are gonna end up with fuzzing terminators literally just running around, trying all kinds of different things to get you... And it just like goes and hacks some things, smashes it, kicks a puppy, throws a baby in the sea... Do you know what I mean? Doing all kinds of -- just to see what works.
+
+**Roberto Clapis:** It's a risk we're willing to accept.
+
+**Mat Ryer:** It's a risk... Okay \[unintelligible 00:47:40.12\]
+
+**Filippo Valsorda:** It's a sacrifice we're willing to make, really. \[laughter\]
+
+**Break:** \[47:48\]
+
+**Mat Ryer:** So does anybody have an unpopular opinion for us today? It can be fuzzing-related, but it doesn't have to be... It can be anything.
+
+**Katie Hockman:** Yeah, I've got one I can throw in the ring.
+
+**Mat Ryer:** Throw it in.
+
+**Katie Hockman:** So I think that -- it's kind of more like a... I don't know if it's an opinion so much as a personal experience, but I actually got into computer science because math wasn't social enough for me... So I think that the thing that I like the most, the best part about computer science is actually building things with other people, and I think having social skills can take you a really long way, and is kind of undervalued in tech.
+
+**Roberto Clapis:** Wait, you're telling me that you got into CS because of the social aspect?
+
+**Katie Hockman:** Exactly. I didn't wanna sit alone in the corner all day and just solve math problems, but I was like "Oh, I can build stuff with people. That sounds more fun, so I'm gonna do that." Which I realized is the opposite of a lot of people.
+
+**Filippo Valsorda:** And then you ended up in security, because the InfoSec community is a shining example of excellent community \[unintelligible 00:50:33.09\] \[laughs out loud\]
+
+**Katie Hockman:** That's a highly social skilled one, because you need to be able to talk to people and understand -- like, if they disclose a report, you need to be able to communicate with them, and understand them, and be able to communicate back, and you need to be able to communicate really complicated things in a really simple way, that other people can understand, which is really hard... And I think that's a field where it's even more important that you have good social skills, because the stakes are so high.
+
+**Filippo Valsorda:** Yeah... To be fair, I should point out that the Go community is extremely nice. The kind of people that I used to work for are usually a delight to work with. I was just making a cheap shot at the...
+
+**Roberto Clapis:** Right...
+
+**Filippo Valsorda:** ...let's say traditional security community.
+
+**Mat Ryer:** Well, what can they do to get you?
+
+**Roberto Clapis:** That's one way to put it.
+
+**Mat Ryer:** You're safe, aren't you, from that? What can they ever do...?
+
+**Roberto Clapis:** Right... \[laughter\] Now that you say that, Katie, I think one of the important things about the human aspect of software is like when you design an API, you have to design it in a way that people will understand. I hate when people say "Users of this API are stupid, because they can't use it right." When you're designing something, you're communicating to the user... People keep forgetting that issue.
+
+**Mat Ryer:** Yeah, that is true, actually... Because you do think -- in the beginning I thought APIs were for machines to talk to each other, but they aren't. They're for humans to build the thing that allows the machines to talk to each other. Yeah, so that is true.
+
+\[52:07\] But I don't know, Pythagoras could have been a laugh at a party. He might have had a great time with it, I don't know... \[laughter\] You're probably measuring all the stuff and you're like "Pythagoras, just put your ruler down for five minutes, mate! Have a sandwich. I've cut them into triangles, how you like them." You know, that kind of thing. Okay, any other unpopulars?
+
+**Filippo Valsorda:** I have a whole list of cryptography unpopular opinions, but the thing is I don't think anybody actually has an opinion on these things, and it's just these ten people, and we're all on the same Slack, and we just discuss these things between us... So I'm not gonna go there. Instead, my unpopular opinion is that - and Katie, I know she understands, but... Dogs in the office are bad. Just bad.
+
+**Mat Ryer:** Dogs in the office.
+
+**Filippo Valsorda:** There should be no dogs in the office.
+
+**Mat Ryer:** Yeah. Go on, elaborate. Are you allergic to them, Filippo?
+
+**Filippo Valsorda:** I'm allergic to them, I know a bunch of people who are allergic to them... I know a bunch of people who are scared of them, and don't feel like they can say "Hey... Beautiful dog. I'm scared of it, so you don't get to bring it to the office anymore, because I'm scared of dogs." No one wants to be that guy.
+
+**Mat Ryer:** "Yeah, I know you love it, but to me, that's basically a little monster from a nightmare..."
+
+**Filippo Valsorda:** Yeah. Like, somebody might have gotten bitten, and you're just like "Yeah, that makes me extremely uncomfortable, but I just joined, and I don't want to be that guy", so they're not gonna tell you... They're just gonna walk around being like "Yup, yup... Cute, cute...", and walks along the border of the office...
+
+**Katie Hockman:** And to be clear, I think Filippo said that about me, because I love dogs more than -- anybody who's ever talked to me for more than five minutes knows that I love dogs more than pretty much anything, I would say.
+
+**Roberto Clapis:** Whoa...
+
+**Katie Hockman:** Yeah. I mean, I do actually agree with you that it makes things complicated. I mean, yeah, it can be a source of joy for people like me, who aren't allergic and I love them, but also, if it's a source of conflict and discomfort, or worse, for people that I work with or people around me, then that isn't ideal either, and... Yeah, aside from the case of like a service dog, which I know that Filippo agrees that that's totally fine...
+
+**Filippo Valsorda:** Oh yeah, of course.
+
+**Katie Hockman:** ...honestly, I think it's a really reasonable opinion.
+
+**Filippo Valsorda:** Service dogs are well-trained, and in general, if accommodations have to be made, one can work case-by-case. But honestly, I never had the problem of "Oh no, I'm really allergic to a service dog that I can't not be around." But I did have the problem with pets a bunch, because there's just many more pets. It's just a numbers problem.
+
+**Mat Ryer:** But how will the management show how cool they are if they don't allow dogs in the office?
+
+**Filippo Valsorda:** \[laughs\] Right?
+
+**Mat Ryer:** \[54:51\] What are you gonna ban next, Filippo? Foosball tables?
+
+**Filippo Valsorda:** Ping-pong tables have gotten old... \[laughter\]
+
+**Mat Ryer:** Roberto, how do you feel about dogs in the office?
+
+**Roberto Clapis:** I'm kind of scared of big dogs, so I'm on Filippo's side... But also, I have friends that are allergic to dogs, so yeah, I agree... Unless they are needed, like they're service dogs, I'm not in favor of that. Folks, your unpopular opinions - I have so much discuss on. My unpopular opinion was gonna be I like yellow, so... \[laughter\] It's important topics you brought to the conversation there. Mine was completely useless.
+
+**Katie Hockman:** \[laughs\] That's a terrible opinion, Rob. Take it back.
+
+**Mat Ryer:** Is it the color you like, or you just love that Coldplay song?
+
+**Roberto Clapis:** No, the color. Just the color. And the effect it has on people.
+
+**Filippo Valsorda:** There are so many better colors... \[laughter\]
+
+**Roberto Clapis:** Right. Yeah, beautiful colors... Yellow is one of them.
+
+**Katie Hockman:** I mean, your headphones are yellow...
+
+**Mat Ryer:** That's true. I was looking for clues of yellow to verify. For some reason I'm skeptical when Roberto says he likes yellow; I think "Well, is it a trick?"
+
+**Roberto Clapis:** Well, I have something else to prove it...
+
+**Mat Ryer:** Oh, a yellow gopher...
+
+**Katie Hockman:** Aww...
+
+**Mat Ryer:** You'll have to send a picture of that. This is a podcast.
+
+**Roberto Clapis:** Yeah... So I already tweeted this picture today, so people just need to go back. I'll tweet it again.
+
+**Mat Ryer:** I'm scared of those gophers, by the way. That's my unpopular opinion. Those little things... I have nightmares about them.
+
+**Filippo Valsorda:** They do look a little weird, huh?
+
+**Mat Ryer:** That one in particular, that yellow one. Well, it's a podcast, so this really is--
+
+**Filippo Valsorda:** This is still a podcast... \[laughs\]
+
+**Roberto Clapis:** Basically, I'm getting the gopher closer and closer to the webcam, until Mat screams... He didn't scream. You passed the test.
+
+**Mat Ryer:** I wasn't suggesting you do an audio commentary of it, I was suggesting "Let's not do that in the first place, and focus on the audio." \[laughter\]
+
+**Roberto Clapis:** Right...
+
+**Mat Ryer:** Well, unfortunately that's all the time we have today... Thank you so much for joining us, Katie, Filippo and Roberto. We'll see you next time!

GitHub's Go-powered CLI_transcript.txt

@@ -0,0 +1,319 @@
+**Johnny Boursiquot:** Hello, and welcome to this episode of Go Time. Welcome back, for those of you who are joining us once more, and for those of you who are new to the show - yeah, welcome for the first time. Hopefully, this is not your last; hopefully, you enjoy today's panel... Which actually includes Mr. Jon Calhoun. How are you doing, Jon?
+
+**Jon Calhoun:** Good, Johnny. How are you?
+
+**Johnny Boursiquot:** You know, generally speaking I just answer that "Yeah, I'm fine", whatever it is, but I think I'm gonna give a different answer today. I think I am -- not everything in my life is going quite right, but I'm choosing to focus on the things in my life that are going quite right. That way, I can be a bit more... How do you say this...? I can take stock of everything that's going on, and be thankful for the things that are going right. Because this is 2020. It could be a lot worse. So yeah, that's how I'm gonna answer that today.
+
+And joining us today - a special guest - is Mr. Mislav... And I know I'm gonna mispronounce your name... Marohnić. Yeah, I knew I was gonna mess it up. \[laughs\]
+
+**Mislav Marohnić:** It's pretty good.
+
+**Johnny Boursiquot:** Thanks for joining us, Mislav.
+
+**Mislav Marohnić:** Thank you.
+
+**Johnny Boursiquot:** Yes, yes, awesome to have you. So Mislav, for those of you who do not know, is the maintainer of a project you probably use, or have used in the past, called hub. So Mislav is gonna give us hopefully a little bit of a history around hub, how it came to be, what he's been doing with it for the last few years...
+
+\[04:00\] And also, if you haven't heard, there' s a new GitHub CLI that's been sort of a release recently, that Mislav also had the opportunity to work on at GitHub. So we're looking forward to unpacking that, and getting to know how you got so lucky... And again, to learn from his learnings. Also, basically following up on our Intro to Go, or rather Introducing Go to Your Team episode from last week - hopefully, this will add some flavor to the stories that Mislav is gonna tell us about Go at GitHub in general.
+
+So yeah, let's get into this. Mislav, do you wanna give us -- I mean, I've already introduced you as the person who maintains hub, and now the GitHub CLI officially, but can you give us a little bit of an intro to yourself? Who are you, my friend?
+
+**Mislav Marohnić:** Sure. I worked for GitHub for seven years now, and that's the place I feel like I have some impact, because I get to -- like with these command line tools, they let me experiment a lot. And I get to open source a lot what I do, and I get to ultimately help this platform, which is a large part of what it is for - it's about sustaining the open source world. That's what makes me happy, and this is what I like to contribute my free time into. Nowadays I'm lucky enough to also be doing open source as my full-time job.
+
+**Johnny Boursiquot:** Awesome. You are the envy of many out there, my friend; you get to work on things you love, and open source, and contribute back, and somebody pays your bills for that. That is pretty awesome. So do give us a little bit of background on hub. How did you come to inherit that project as a maintainer on it, what was its original goal as a project, and how did that evolve under your stewardship?
+
+**Mislav Marohnić:** The story of hub starts about ten years ago (or more; eleven), when it was just a short script made by Chris Wanstrath (know as @defunkt on his online handle), then CEO and co-founder at GitHub... And it was basically a little gimmick; it was meant to extend the interface of Git in a way that it just feels slightly more GitHubby, and defaults for certain shorthands to GitHub URLs, as opposed to somewhere else... And just makes Git have more sense working with GitHub.
+
+That was really well received also by me, who was really at the time nerding out about just CLIs in general, and Git, and had been already a very active GitHub user. So I started contributing a lot, and eventually, as I guess other people who have - especially Chris, who kicked off the whole project, as his participation kind of faded out (as often becomes of open source projects), somebody sometimes comes along and takes it over, and I just appeared to be that person.
+
+So since then, I continued writing it, but over time it had this organic adoption into a tool that eventually was considered GitHub's official CLI tool... Which it was definitely not, but I guess now in hindsight I can feel how that might have been moving the project under the GitHub org eventually, even though it was maintained ultimately by a then non-GitHub employee. It was something that signaled strongly that this is something that has an organization backing it, whereas it was mostly just like a pet project still, and it was an experiment. It just kind of grew beyond just a little gimmicky tool that people use in the command line.
+
+\[07:51\] Eventually, I started to feel a really large responsibility about it, because so many people have been using it, so I kind of stepped up my project maintenance to give it more and more time. Eventually, GitHub noticed that this is something that is really worthwhile having, to the extent of investing a whole team, resources of a whole team into it. They asked me would I want to participate, because I had advertised that I wanna switch teams at that point, so it worked out great for me. I got to change up my job after six years, which I feel I had a good run, and then do something completely different.
+
+**Jon Calhoun:** So you started working on hub before you were working at GitHub, correct?
+
+**Mislav Marohnić:** Yeah.
+
+**Jon Calhoun:** Okay. So did that help you when you were applying to GitHub? How did that work, I guess? Did that help you get through the interview process a lot easier, or was that something you talked about with them?
+
+**Mislav Marohnić:** I'm sure it did. The way that GitHub hired then, about 7-8 years ago, and the way that it hires right now, of course, is very different, because GitHub is a very different organization since then... But I do remember having this privilege of having known at that point most people who have either founded GitHub, or were otherwise there, in a really high clout capacity. And not in the way just like buddies or something, but in a way that I have actually spent tons of hours, and even with some of them up to several years collaborating on open source projects.
+
+So I think as like an interview, if you can see a person in front of you, and instead of them solving a blackboard problem for you, you know that with this person you have years of experience, collaborating, coding, reviewing PRs, and something like that. I think that all that experience before that was basically a very prolonged interview process... And I must have made a good impact, because they had the trust in me that I'll be really passionate about what I'm doing, even though I was just a remote employee who is always traveling and working from different timezones. And I think the trust paid off eventually. Since then, I was really excited to work here.
+
+**Jon Calhoun:** So you said the first hub started off as like a Ruby script. Or you said a script; I think I read that it was a Ruby script. Is that correct?
+
+**Mislav Marohnić:** Yeah, the idea was that hub was a single-file script, so it can be just easily copied over to any system.
+
+**Jon Calhoun:** So the initial version was kind of meant to replace Git, and I think over time it evolved; at least the GitHub CLI that we have now doesn't feel like it's meant to be a Git replacement. It's not supposed to be an alias for that. So around what time did you start to feel that wasn't the case with hub, as it evolved?
+
+**Mislav Marohnić:** Well, maybe we should first define what does replacing Git mean. The way I see it, if somebody wanted to replace Git, it would probably have two main potential ways of doing it. So they could either abstract it away in the sense of replacing its entire API, which is on a command line all these commands that we use: log, commit, rebase, and things like that... And replacing it with a smaller API, so something that makes sense for an abstraction. Of course, abstractions want to have a smaller surface area.
+
+And another way would be extending it. So on top of all those commands we had some more commands, and we had certain extra flags. And even though Git is by itself, by nature extendable as its core feature, that it can invoke other git/something executables if they're found in the path, even though Git is extendable in that way, it's not extendable in a way that really likes extra flags to be added to its existing commands. So that part is kind of really hacked on in a little bit of a Frankenstein manner that's hard to maintain.
+
+\[11:51\] We also considered eventually doing an abstraction of Git, in the sense that what if we could capture the essence of Git's API mostly as it matters to 80% or 90% of GitHub users, and only expose that? But that was such a scary concept to hold as a team, both then when we were considering that with hub, and now, when before we released the CLI we were considering that for the CLI.
+
+In both instances we decided not to do it, so I think somebody who decides to do it, that's a really bold undertaking, that I wouldn't necessarily wanna discourage people from... But I think that is much less feasible than doing other kinds of extensions, which allow still Git to be used in its full capacity.
+
+**Jon Calhoun:** So just to add a little context... When we talk about extending it, we're talking about doing things like -- normally, if you type "git clone", you have to give a URL. So the extensions that I remember, at least with hub, were things like you'd type "git clone" and then a username/repo, and it would just know "Okay, I'm gonna go to GitHub and pull it from that user and that repo", and it would sort of expand that stuff. Are there other ones you can think of that might have stuck out? But I think that's probably a good example of what type of extensions it was doing...
+
+**Mislav Marohnić:** That's a very good example, yeah. That one is basically taking an argument and transforming it, massaging it a little bit before it reaches Git proper. Other kinds could be adding a certain flag that only makes sense in conjunction with GitHub, that doesn't exist there with Git otherwise.
+
+Some others would be adding a completely separate command. For instance hub has a sync command, which if you wrap Git as hub, then you can type "git sync" and have all the local branches sync-ed up with the remote ones.
+
+So all these different types of additions to Git were shipped as the same tool, and that was powerful in the way that it adds a lot of features at once, and it can feel to somebody who studied it as a really good toolset to add to their arsenal... But I think overall, it just was too much of different layers of additions that people would mostly pick one and benefit from some of them, but then to some others they wouldn't even notice, nor appreciate, or sometimes they would even get in their way by subtle bugs that would be in the layer that's added on top of Git. So it was hard to maintain in that way.
+
+**Jon Calhoun:** Yeah, that makes sense. I imagine it being incredibly hard to get something that sits on top of Git and still doesn't alter something or somehow break some functionality of Git without realizing it.
+
+So you talked about things like sync and cloning, that would sort of add functionality that was GitHub-specific. When you were building hub, did you think about adding stuff that was not GitHub-specific, it was just sort of extensions onto Git that you wanted, or did you leave that to the -- you said that you could extend Git itself with just like git/ -- I think they're Bash scripts; is that what they normally are? Or scripts of some sort... Did you usually leave that to people, or were you doing stuff like that as well with hub?
+
+**Mislav Marohnić:** Well, hub does add some of its extra commands to Git while GitHub CLI as the next iteration of the tool decidedly doesn't wrap Git at all. But going back to hub, it doesn't prevent the user from also adding their own extensions... It just would -- well, the extensions of the same name would clash so somebody would add a custom command, which could be implemented in Bash, but the beauty of Git is it would execute anything executable...
+
+So it really doesn't matter in which language something is written in; it will be invoked with a certain set of parameters, and then it's up to that thing to do whatever it needed to do. And hub did add some extensions and sometimes those extensions, in rare cases, but those people were vocal; they would clash with their own. So it was also a little bit testing the limit of this extension framework of Git, which is very barebones, and it's not really meant to be taken to the extent that hub was taking it...
+
+\[16:12\] So I feel that maybe it was a little bit intruding on this extension system, that was really just meant as a very simple system for users, specifically for their direct environment to maintain. As third-party tools like hub and others would try to move in on the system and plug into the same mechanic, it just wouldn't scale past that point anymore... So I felt that was a little bit of a misuse of that mechanism in the first place.
+
+**Jon Calhoun:** Okay, that makes sense. So you said at one point when we talked about hub that it was written in Ruby, and now - I don't think it would be on the podcast if it wasn't written in Go; or if the new CLI wasn't written in Go.
+
+**Johnny Boursiquot:** I don't know, maybe... \[laughs\]
+
+**Jon Calhoun:** Maybe... But it might not make as much sense. So how did that evolution happen? How did you go from Ruby to Go? What led to you trying out Go, I guess?
+
+**Mislav Marohnić:** That was probably the most ridiculous thing that happened in any of my projects. Imagine your open source project, that already has a huge codebase, and thousands or tens of thousands of users - somebody comes along and rewrites it all from Go to Rust, or something like that, and says "Here you go." You're now supposed to merge this thing, deleting all of your code and replacing it with new code. This is basically what happened.
+
+There was this person who maintained for a while his fork of hub, that was a total rewrite from Ruby. And this person's - his name is Owen - primary motivation was that Ruby was really slow. The Ruby interpreter took about 60 seconds then on my machine, on a really high-end MacBook, to only start... And that was often not even including things like the standard library. So as soon as you would then require NetHttp of the same name as it is in Go from the standard library, that would add maybe 20 more, or something like that... So we were talking about almost we're now at 100 milliseconds and the program hasn't even started doing anything yet.
+
+The other thing is portability. People had to install Ruby on the system, and they had to be not at a very certain version of Ruby, but over time Ruby versions changed that are pre-installed system, or some things getting checked into the Ruby loader by default that is not compatible when it boots up for hub's purposes... There were all sorts of these problems, and it was just really hard to make it really seamlessly portable, unless somebody already had a Ruby development environment. That's all okay, because they understand their environment.
+
+But to somebody who would just like to use hub, that they now have to commit to maintaining their Ruby version over system upgrades over the years, it was a pain for a lot of users. So a precompiled binary that is just cross-compiled for different systems and just dropped in there - that sounded like a dream. And I actually didn't believe it, because I had no experience. Go was this new thing, and I'm not a very fast learner, or early adopter of things... And it took me a while to warm up to the idea.
+
+Slowly, I started working with Owen to really solidify the test suite around the transition, so we can have some confidence that we didn't break too much. We knew that every complete rewrite will introduce a lot of bugs; not the normal amount of bugs as a PR or something, but a lot of them. And we just at least tried to minimize this, not to lose the trust of the hub community. It took us six months of mostly addressing edge cases, and in the meantime every new features that was merged in the Ruby version was ported over by us to the Go version. So that was almost like a full-time job of itself. I did that then in my afternoons or weekends, and things like that, as did Owen.
+
+\[20:12\] In the end, Owen had the privilege of just hitting Delete on all Ruby code, so erasing the entire code in a commit, in the next commit after the thing got merged... And suddenly, the project was -- it was a pretty solid transition. We really did well on minimizing the bugs, because people largely could upgrade, and never realize that the program changed, up until the point where months later they got the idea to add a feature; they would open the project in its main branch and see "Well, nothing familiar like before." The organization structure of a Ruby project is really simple; it's just a few Ruby files, that's it. And they went into something like this and they were really confused when they opened an issue what happened... So yeah, you weren't here for it. You blinked, and...
+
+**Johnny Boursiquot:** Now it's in Go. \[laughs\]
+
+**Jon Calhoun:** In one of the past episodes - I think it was a couple weeks ago - we talked about introducing your team to Go. You know, like if you're working on a team that doesn't use Go for anything, some different ideas for getting them to try it with some sort of project... So after using go for a CLI, do you think that's a good fit for that type of thing, or if you're trying the new language out?
+
+**Mislav Marohnić:** Yeah, absolutely. I think that it's a great opportunity to introduce it this way. I think also CLIs could be considered as like internal tools as a way of showing in the organization how Go can be productive.
+
+The way it was introduced to GitHub, I remember, and also what helped the transition of hub to Go, what gave me confidence is that I started seeing my colleagues around me, that I didn't directly work, but they were really engineers that I looked up to - they started using Go for microservices. And the first microservice that was extracted from the monolith, that is a Rails app, at github.com, is the one that still today delivers avatars; it stores users' avatars. Organization ones and team ones, and things like those. And those developers had a really good time writing this in Go, even if that was unheard of before in the org. I think they just had enough of their hands untied that they could just ship a new service written in whatever they seemed fit.
+
+And seeing their success with that, and also their involvement and contribution to hub around that time also really helped develop this Go version. And we likely couldn't have shifted without that, and I also wouldn't have the confidence to merge in a rewrite to a language that I'm kind of still unfamiliar with had it not been for my colleagues that were at the same time introducing Go to the rest of GitHub... And now Go at GitHub is just this huge slice of the org. So many services are written in it, and I would say that it's almost as fundamental right now to engineering in GitHub as Ruby is.
+
+**Break:** \[23:13\]
+
+**Jon Calhoun:** Going in a slightly different direction... When you talked about the transition from Ruby to Go, you said that you had to sort of take the test suite and make it so it covered both of them... Can you talk a little bit about how you were testing the CLI in a way that you could actually use it for both languages? Because I think a lot of us when we think testing, we think about running go test and having unit tests run, or integration tests of some sort... But I don't think that's what you were doing if it's something you could run with both Ruby and Go. So what did that look like?
+
+**Mislav Marohnić:** One thing to keep in mind while I'm talking of anything about hub is that because of the history of hub and because of its nature of just starting up as a mere proof of concept and evolving far more rapidly in popularity than it actually evolved technically, hub was always this treasure trove of anti-patterns, I would say. So it was not definitely a project that I would advise anyone to look for either good Ruby practices in terms of testing, or later good Go practices. I made probably every Go mistake in the book with the hub project, because it was my first Go project, and I hadn't been before in teams of other people working on Go, so that I can see how people with experience were using it. I was mostly just inventing its use as I went along, and that was not really great. But of course, we all have to learn somewhere.
+
+So in the Ruby side, the testing approach was -- at first, there were some unit tests, but the test coverage wasn't really great. Just a very few isolated functions were unit-tested. And overall, hub had a very solid and good coverage, but it was done by end-to-end testing through a tool called Cucumber, and I would say story-driven development, because we would write it in this human-parsable Cucumber format, which would then execute those feature files, as they were called, and drive the usage of hub from the outside as if a user is typing into the terminal.
+
+Some of the tests took it to the extreme, where literally we would use tmux, a terminal multiplexer basically, to spawn an internal terminal to the test, a headless one, to literally send key strokes in an interactive shell, and type "hub pull request these and these flags", press enter, and then inspect what's happening after that.
+
+And so much of the test coverage was done that way, and I really spent a lot of time making sure that we have really good test coverage across the whole codebase... But it's not an approach that I would recommend in the long-run for the next person who's listening to this who might wanna use Go to create a command line app. But somehow, in a bizarre twist of things, because of the rewrite from one language to another, we could have kept the entire test suite, because the test suite never knew what hub was written in. It just ran the hub executable.
+
+So when we rewrote it in Go, mostly what we had to change is those parts where we stubbed out things... For instance, the GitHub API is completely stubbed out; we don't run the test and then have it talk to the GitHub API. That would just not be -- well, first, not great performance, but second of all, they would be really hard to maintain when it comes to write actions, as opposed to get actions.
+
+\[27:53\] But the way this test was set up, a separate server pretending to be GitHub API was spun up in Ruby. And after the rewrite, we had really no reason to rewrite this test server. So the codebase continued to have a test runner, which is really in the Cucumber language, which is executed in Ruby and also uses Sinatra that pretends to be a GitHub API server.
+
+And in the end, I think that's why I used this Frankenstein expression earlier, because it was this stitched abomination of different texts, which made no sense if you were to drop into a project and wanted to open up a pull request. You would think "I thought I was contributing to a Go thing, and now I'm editing a Sinatra API endpoint, or something like that?" But it worked beautifully as long as nobody touched it ever, and as long as there was precisely mostly over the time one person working on it who understood how it all worked.
+
+It's fine when you have one developer who understands all of this; it would have been a nightmare if this was any sort of project where there is actual business value to it, or some pressure, or a shipping cadence, or something like that. Or a team of people working on it. So that would not be something I recommend as a developing practice, but it somehow worked out and it brought us that far.
+
+**Johnny Boursiquot:** I was chuckling, partly because I know of some projects that are still in production today, that fit that criteria... \[laughs\] Nobody touch it, because nobody knows how this thing works, and the Frankenstein that we've got here, and whatnot... That's how software evolves over time, and as you bolt on pieces, and different developers, different perspectives, different hands touching that thing - yeah, it can certainly get that way, for sure.
+
+I did wanna mention that Owen, actually, who was that first transition, used to be a colleague of mine over at Heroku. He's moved on recently... He's a pretty smart fella, and I think you were lucky to have had somebody like him to help you in your journey. In the back of my mind, I'm thinking "Man, when I was learning Go - wouldn't it have been amazing to have a super-mentor who knows the ins and outs of the thing to help you along?" kind of thing... But yeah, I think you lucked out there, for sure.
+
+**Mislav Marohnić:** Yeah, it was great to jump into his codebase and then to learn Go by literally just kind of tweaking the variables and functions that already he laid out... And I think like many people I prefer learning not from a blank slate. So a new language, if I had to write out the new program from a blank directory, I'm not gonna do so well. But if I jump in an existing one, then I'm just having this kickstart and wind in my back, and I can start editing things, and seeing his Git log of changes on a certain command and how he got to the point there. It was almost as if I got to sit next to his shoulder while he's coding. And eventually, I did.
+
+We met up once when I was passing through Vancouver, and we got to hack together. But of course, most of our collaboration was asynchronous, and across continents. I could thank him primarily to get me through my first year of learning Go, definitely.
+
+**Jon Calhoun:** So you had mentioned that you eventually rewrote hub into Go, and now we have the new CLI, which I believe is written in Go as well, but I think it's a complete rewrite from the ground-up. Do you wanna talk a little bit about that rewrite, what caused you to decide to throw out what you had and write something from scratch? And since you were writing from scratch, again, what made you decide to use Go this time?
+
+**Mislav Marohnić:** Well, Go was a very short discussion with the rest of my team, all of whom other than me were unfamiliar with Go; they were familiar with it, but they were not using it for anything of substance up to that point in their engineering careers.
+
+\[31:56\] And I had pitched the Go idea mostly as a way to preserve what already we know worked well, and that was - well, I'm a big fan of its compiler; I think it's very robust and it gives me a lot of confidence in it while I write code, together with a good integration with Gopls for instance right now in the text editor, and having this confidence that everything is wired up properly through static typing... And for me, that was a big departure coming from very dynamic languages like Ruby, and into learning how to let go of that mentality and become very secure in the static typing mentality.
+
+Another of the qualities that we wanted to preserve is portability. I remember it being a very short conversation, because my colleagues were just evaluating those things and thinking about potentially other languages we can write it in... But all of that I just mentioned resonated with them, so they were absolutely on board with "Alright, we're learning Go now", and through the next few weeks and months they went from basically zero to also doing Go like me right now, on a daily basis. I would say that they surpassed already my abilities, because I sometimes feel I'm catching up to the rest of my team.
+
+**Jon Calhoun:** So I guess I'm asking you about the rewrite, because typically, when you hear people talk about projects -- or to go back to your Frankenstein. You said that hub was kind of a Frankenstein, with the test suite being in Ruby, with some Sinatra, and other things like that... I think when you're a new college graduate or you're newly coming into the field, you learn all these things about best practices.
+
+Then you go in an organization and you see a repo like that, and you think "What are these people thinking? This is a terrible idea", and you don't really think about how projects evolve over time... And if you actually saw the whole history, it would make complete sense. But when you just see it brand new, you're like "This doesn't make sense." So a lot of newcomers to the field will think "We need to rewrite this", but almost always that's a bad idea, because you spend so much time rewriting and trying to get feature parity that it's just really hard to do. But in your case, it seems like you successfully rewrote, and it sounds like you think that was the right decision. And I'm not saying it wasn't, I just -- so, I guess, can you share a little bit about what really motivated you to be like "This needs to be rewritten from the ground up"?
+
+**Mislav Marohnić:** Well, when it came from learning what really worked with the hub tool, and then choosing how much of that we wanna promote of that spirit, to eventually being GitHub CLI as an official tool, we mostly went over its feature set and decided that its fundamental design paradigm was not something that we wanted to port over. And then after that, considering inputting any of its code, it's actually really already -- it doesn't align with that first and foremost decision, which we eventually were pretty secure in... Because if you don't wanna preserve the spirit, a design of how a tool works, then it's just really hard to get anything from it, especially due to the fact that being my first Go project, I let the Go packages become basically huge, to the extent where -- I think there were largely just two Go packages where most of the hub implementation lied... So to cherry-pick the good parts out of that and leave out the bad parts would have been something that's not really feasible technically, and I think it's very bug-prone.
+
+\[35:48\] Another thing is that we didn't wanna go with the same testing approach, so now we're copying over parts of the implementation, but we're actually not gonna port over to Ruby Test because we wanna commit to the Go stack and the default Go tooling to make the project more approachable. With that line of thinking, it was a little bit obvious that starting from scratch would the be right decision.
+
+I appreciate that you brought it up - it's not an easy decision. It should never be made lightly. And I think rewrites should never be made lightly as well, because these are technologically really risky endeavors. But what made it a little bit less risky in our sense is that we were promoting kind of like a semi-official tool to another one. And even if we broke a lot of things, or didn't port a lot of functionality over - well, we didn't actually erode trust, because we were launching this new tool, which starts over from version 0.0.1, and who wants to follow us on our journey can, and who wants to stay safely embedded with the tool that already works for them also can.
+
+And I felt that we couldn't have kept the trust if we tried to make radical changes, and then mostly just disappoint people (I would say) that used hub to do a lot of automation; I also personally loved doing that. They would be the ones who were most affected then, trying to upgrade to a newer version and finding out that their scripts are broken, and that that tool that they've used as a reliable Swiss Army knife is not as reliable. That would be my nightmare scenario personally, and we avoided it.
+
+**Johnny Boursiquot:** Speaking of evolution of these tools... I imagine a some point you are going to start sun-setting hub, because I can't imagine you trying to keep up with basically development of both of these things at the same time... They have both their own sub-communities, and each one is gonna have their own needs, kind of thing... So what is your plan for ultimately retiring hub, and putting all your efforts towards the official CLI tool? And also guiding, helping people who rely on hub today sort of transition over to the new official tool.
+
+**Mislav Marohnić:** I tried to reassure people around the time that I was gonna be hired onto the new project that I won't just like ride away off hub, and archive the project, and nobody gets any updates anymore... So I did release a few - or at least one that I remember - bug fix release in this year, that also I have parallely developed CLI... And I feel that I fell a little bit short on my own promise how much I'll be invested into it, because as it turns out, my primary motivation with hub was that I got to nerd out on this CLI that talks to this platform where I host most of my projects.
+
+So GitHub is not just the place where I work, but it's literally the platform where I host all of my open source projects, and where I communicate with a lot of people on a daily basis as part of my hobby. So that was really important to me, and I got to develop these tools to help me accomplish more with it. Eventually, that is integrally a part of my job as well. So I don't have the same itch to scratch anymore after-hours. In fact, after-hours what I'm thinking is "Well, I don't need to now switch over the VS Code tab to another project that is also a command line client, for the same platform that I just worked for eight hours on."
+
+And I guess because my need for tinkering osn CLIs was satisfied, I had felt that I had not followed up as much -- I didn't make a strong promise, but as much as I imagined in my head. And I will have to publicly admit, in a sense, that I will have to signal that better to the community, about how much I'm actually gonna de-escalate my involvement. But I do wanna make a series of updates before that, handle things more in the long run, like authentication to GitHub, which is subtly changing in its API versions... And maybe potentially exposing things that people have been asking for a while.
+
+But ultimately, I was just imagining investing more into the features that are about extensibility, and people writing their own scripts, like the hub API tool that has a completely equivalent counterpart in the GitHub CLI, which is called the gh api...
+
+\[40:13\] And I feel investing in social tools is great for everyone in the long run, because I can make a minimum amount of changes to enable other people making a lot more changes on top of that without necessarily shipping updates. So I wanna leave it in a place where it's still gonna be useful for years to come, and extensible for years to come, but not necessarily have to receive new commands over the future.
+
+**Jon Calhoun:** I think that makes sense. So you've written two CLIs in Go, or you've worked on two at this point... Are there any libraries or tools that you've found especially useful or especially -- what are the ones you'd recommend, what are the ones you've used that you sort of didn't care for? For somebody who wants to build a CLI, what are you recommending to them?
+
+**Mislav Marohnić:** Well, when I started looking at that in Go, and wanted to apply that to hub, by that time Owen had already made his dispatcher command from scratch, so there was no third-party library that we imported for that purpose. And maybe tools like Cobra right now, which is really popular, or your urfave/cli -- sorry, some of these projects are only known by their GitHub owner/repo pairs, because that's how we refer to them from Go import statements, I guess.
+
+**Johnny Boursiquot:** Right.
+
+**Mislav Marohnić:** Those projects - I'm not even sure they existed; it was a bit while ago... And even if they did, we couldn't have used them, because as it turns out, the problem of writing a dispatcher that is an extension or a wrapper to something else, like its own commands, and having one that is just its own self-contained command, like a completely new CLI, like kubectl - that part would have made none of those tools really usable. So it was first from scratch. I would not generally recommend that, unless it's kind of like an exercise. If somebody is doing this for hobby, for instance learning Go, I would actually recommend it. It's a great exercise. If you love writing CLIs and exploring how you can structure them in Go from scratch, it's a great way of learning.
+
+I would not recommend it for a work project where maybe the CLI that you're trying to introduce to your team should immediately do something useful, and not be just a code exercise. I feel that's a great way to get other people's buy-in on a certain piece of technology that is not just a gimmick, but it's also very useful, and that you can iterate fast with.
+
+To Go, actually, with those libraries - I would heartily encourage any of those that I mentioned. But we have a closer relationship with the Cobra project. We chose it for the GitHub CLI project. And I think over time that Cobra really changed in neither direction which people were responsible for it... And I know that it's from first-hand maintainer perspective; I know how hard it is to maintain projects for many years, especially when they have a lot of eyes on them and a lot of dependents... Because what that means is extra pressure to the maintainers. I think our risk of burning out is actually rising with popularity of our projects...
+
+So popularity is not always a good thing. And I feel for Cobra that this necessity to maintain backwards-compatibility, which I absolutely agree with, eventually kind of coordinated into this stalemate in which it's hard to make any kind of significant change to the project, even though some of the initial decisions that they did about which stream to output to, how do they do error handling, how do they do help command, and the help flag, and things like that - eventually, a lot of that didn't work for us in GitHub CLI, and we started working around it or implementing parts of Cobra outside, orthogonally, in a separate package.
+
+So by now, I feel that for our purposes it would have been maybe a better decision to go with something simpler, that we don't have to fight against... Or to have just a better overview of what Cobra is and what Cobra isn't and trying not to delegate too much to the tool if you can't handle the load.
+
+\[44:18\] So I kind of feel that in hindsight, I wish that some of the corporate documentation was pushing you towards the better practices, rather than encouraging you like "Here's how to get started. Generate this file, and generate this file, and add this command here and here." But the way that the tutorial is set up eventually creates a Go command structure that ultimately I feel doesn't scale... As evident by all of the Cobra large projects that I've studied, for instance the kubectl, which is an incredible CLI; there's so much to study there. I had found that they're using Cobra in such an unusual way... And eventually that made sense, but it was not really apparent why they did so before we ran into all of those roadblocks... I feel that they must have run into them as well, because now the project is organized as they did.
+
+But I would recommend I guess not relying too much on the CLI library, and using it more as an accessory, as an underlying implementation detail of the library... But structuring the library in a manner that you could imagine that the specific CLI implementation like Cobra or urfave/cli could have been swapped out with minimal disruption. That's what I would recommend.
+
+**Jon Calhoun:** I think that makes sense, especially if you're building something large that needs to withstand the test of time. Were there any other tools that stood out to you? If I recall correctly, the CLI has some color-coded text, and some other text formatting, and things like that... Did you find specific tools or libraries were helpful with that sort of stuff?
+
+**Mislav Marohnić:** For me, off the top of my head it's hard to remember these --
+
+**Johnny Boursiquot:** We're cheating, because we're probably looking at the go.mod file. \[laughter\]
+
+**Mislav Marohnić:** Authors, and things like that... But I can also quickly just open it, to refresh my memory. There is definitely some things that I find myself reaching over and over. Not just myself, but I see common dependencies in the projects. For instance, the Testify library is not just used by our team for testing, but it's also used wider at GitHub in other Go teams as well. That's the tool that we reach for often, even though we try to stay as close (with the exception of using Testify) to the Go standard library for testing, and not deviate in that too much. And then some tools that we are using, authored by a GitHub user @MitchellH, and another GitHub user called @muesli. So a lot of the tools are by those two users. And of course, @MattN. MattN published tools like go-colorable, go-isatty, and it had seemed that almost like a lot of the itches that we had, a lot of problems that there were these prolific GitHub and Go contributors who have already encountered these problems and made this super-tiny, hyper-specialized libraries... And I was really a fan of those.
+
+I'm not necessarily always a fan of -- in the JavaScript world with the npm, of the super-tiny, micro-specialized JavaScript libraries, but I was very much a fan of that here, because it was something that we could then easily reuse across projects and rely on. And I think if somebody compared the hub codebase and the CLI codebase, they would have found plenty of the same library dependencies.
+
+\[47:54\] For markdown library we're actually really impressed by the renderer. It's a project from charmbracelet/glamour. Without looking it up, I would say that it uses Blackfriday for markdown parsing, which I've also found to be a very useful library. And I guess - yeah, a lot of tools that were already there really made it possible for us to just launch ourselves in this space.
+
+But also, all of these tools that I've enumerated - they don't necessarily have to do with writing a CLI. They didn't have to necessarily specifically apply with writing a CLI. And I feel maybe that specifically tools that interact with the capabilities of the terminal, and are able to output different colors, but also in a way that respects user settings, and the capabilities of the terminal, things like that - I think the fact that those tools are hyper-specialized and so scattered around makes it kind of hard to discover them and assemble them in a proper way.
+
+I had experience in the JavaScript world writing CLIs that CLI-related libraries were much more mature, and I had not experienced as much that with Go. I had more felt that a person really needs to spend a lot of time researching these tools, and sometimes under a time pressure to ship. That does not always work out. Maybe it works out for a hobby project.
+
+**Johnny Boursiquot:** It sounds like a good blog post.
+
+**Mislav Marohnić:** It could be. But I feel even as a person who always feels like I wanna contribute back to all this plethora of tools - like the Go project, that itself was made by people in an open source fashion. I'm sometimes thinking "Well, if there's a big hole that is there, why not invest some time in filling it?" So I'll try to divest some of my learnings from doing GitHub CLI to maybe start creating more of the Go libraries, because I feel it's one of the ways that I always give back to the Ruby community and the JavaScript community, and to the Bash community; even though the Bash community doesn't exist, I like to make the joke, because one of my favorite languages is Bash, and I feel that I'm a little bit too overqualified in it for a sort of limited deployment potential tool... But I see in my future, and hopefully maybe in my team's future as well, that we take some of those learnings and we learn a little bit more about what does it take to make these reusable libraries in a space that there is not so many alternatives that we feel should be doing a certain thing; that right now projects are commonly reimplementing and reinventing over and over. So maybe we'll see something like that.
+
+**Break:** \[50:43\]
+
+**Jon Calhoun:** I have one more question, and I think we have to jump to the Unpopular Opinion segment soon...
+
+**Johnny Boursiquot:** No, we don't have to; we want to. \[laughs\]
+
+**Jon Calhoun:** We want to. \[laughter\]
+
+**Johnny Boursiquot:** People look forward to that stuff, man...
+
+**Jon Calhoun:** I understand. When you're talking about tooling, I know for me when I'm building web servers, tools like Sentry are kind of a go-to; something that will allow you to track bugs or errors and log them somewhere. But with the CLI, I imagine that's not really -- you're not running on a web server, you're running on everybody's computer. So how did you handle that challenge of actually figuring out what these bugs were, and getting people to report them, and actually handling all of that, I suppose?
+
+**Mislav Marohnić:** At the same time, we feel like the non-addition of monitoring and error capturing and reporting and things like that to our tool saved us a lot of trouble about how to do this consciously and as transparently as possible, and as respectful to the user as possible. People aren't necessarily always comfortable with their clients reporting everything that they do, for reasons that I don't really have to elaborate. And of course, in our case it's a little bit different, because I guess that people hosting their projects on the GitHub platform already means that they have to some extent given GitHub trust, and that's while they're interacting with a GitHub set of features they wouldn't mind as much that we report what is the most used command, or what are the most used flags for certain commands, for instance. That would have been a very good insight for us. Unfortunately, we have none of that, because we haven't built any of that in a tool. It's not out of the question, but I think that I feel in hindsight that we could have structured the project better to lend itself to that case, because I think eventually we want to, even just for the purpose of debugging and gathering statistics about the execution of the tool, or how much time is spent in shelling out to Git, and how much time is spent in API requests, and things like that; I feel that if we designed the tool as a microservice instead, the capability of monitoring and error reporting, that all of this would be easier to find.
+
+So I see it in our potential future, but I also think that it was a big load off our chests that we didn't do it initially, and I think if we ever do it, of course, we'll have to do it in a way that's probably opt-in, because right now people are using the tool and we just can't slide in monitoring where people are not necessarily expecting it.
+
+One place where especially that hurt us is that we don't have any way of crash reporting. And a lot of my colleagues, especially initially in the CLI project, were people who also worked on GitHub Desktop, which is another GitHub client, but it's graphical; it has nothing to do with CLIs. GitHub Desktop, on the other hand, has an excellent crash reporter. GitHub Desktop's crash reporter was also always a very good smoke test if there was a bad deploy, or something.
+
+It also has a beta stream for updates, so people could opt in to getting beta updates, and then those users would then -- from their reports it would have been evident if there was something really crashy, a ship that is a potential blocker... And we don't have that kind of visibility with GitHub CLI at all. So we just have to do extra diligence so we didn't break it for everyone... And it's very easy to break the CLI tool for people, because unlike graphical apps, the CLI tools can execute in so many different environments, under so many different permutations of circumstances.
+
+I guess their versatility is part of the appeal of the CLI tools, that there's a very low-barrier to executing them or running them on maybe an embedded system somewhere. But we do have less visibility into it, and these are all trade-offs that we've considered. And I wish in the future that we can have some more visibility into that, because I feel that will empower us to then make better decisions about what really matters in the tool, rather than right now relying on self-reporting from users by asking them "Okay, what are your most-used commands?"
+
+**Jon Calhoun:** I think that makes sense.
+
+**Jingle:** \[56:21\]
+
+**Jon Calhoun:** Do you wanna lead into the Unpopular Opinions, Johnny?
+
+**Johnny Boursiquot:** Oh, no, no, no. I will cede the floor to Mislav. I wanna hear some unpopular opinions... I heard a few were brought in, so let us hear them. \[laughter\]
+
+**Mislav Marohnić:** I'll start with a Go-related one, because the other one was not specifically Go-related. A lot of what we were excited to do with the GitHub CLI - so the next iteration after hub - was we wanted to really try out how it feels using the GraphQL version of the GitHub API, which shipped in between. Of course, hub originally has used the REST version, and there was not enough added value into migrating completely to another version of the GraphQL API, so we only did that experiment with GitHub CLI when we eventually started working on it, thinking that that would be this massive win in this new API paradigm, which is supposedly really more powerful... And I've found that the exact features of the Go language, static typing and compiling, that it's not actually lent itself well to being a good GraphQL client.
+
+While I'm talking about this, just keep in mind that I'm mostly just talking about an experience of writing in Go a GraphQL client, so something that makes and parses GraphQL requests. I have zero experience of making a GraphQL server in Go, which some of my other colleagues at GitHub have experience with, but I don't have first-hand experience... So this is not about making a server, which I feel that there is more solid tooling. But when we look at the offering of the different GraphQL clients that are written in Go right now, and mostly used as a de-facto standard when we look at the largest, most prolific projects that are open source right now, if we look at how they make requests, not just to GitHub's GraphQL API, but to any other, I feel that all of those libraries right now are missing the mark on what makes GraphQL really stand out.
+
+GraphQL is not a query language that wanted to be used by having a pre-generated query, which is always the same per compiled version of an app, and then having different requests come in separately, because they were all statically-generated from maybe a schema, or something like that... GraphQL wanted to first of all allow people to bundle several queries at once, or even several mutations. I don't think it will allow bundling a query and a mutation acting on the results of those queries; I think that's decidedly against its design. But it definitely can execute an arbitrary number of queries at the same time, and also an arbitrary number of mutations. So if I wanted to change labels in a hundred GitHub issues in the same request, theoretically I can do that. And I was really excitedly searching for Go tools that allow you to kind of batch up a bunch of queries, and then they all execute transparently over GraphQL. It wasn't a thing that I was able to find by weeks of searching and studying the other libraries that were -- well, the ones that were open source, of course.
+
+\[59:47\] And another thing that GraphQL really lends itself well into is to stopping over-fetching. When you make a request to GitHub's REST API, you don't get to choose what you get back. You always get this enormous object back. We mostly always return absolutely everything about, let's say, a pull request that you're interested in. We return everything about the author of this pull request; all the fields of an author, all the fields of the repository that the pull request is embedded in.
+
+As you can imagine, in a lot of back-and-forth communication eventually a lot of redundant data is not just being exchanged and parsed, but it's also just being needlessly collected, and presents some overhead on both the client and the server. In GraphQL, the idea is to only request the fields that you are really interested in. But sometimes, between runs, that number of fields for a certain query changes based on user input parameters. So now we're back again on this square one, where I mentioned with the static compilation of the language we mostly embed a static struct, which may be used as a parse destination for a GraphQL JSON response, for deserialization that a lot of libraries do in a similar way - they deserialize into static structs; or at least they always generate a query from the static representation of the resource itself. There's no such consciousness about adding APIs that will let us choose the fields that are being queried, right?
+
+So I feel that a lot of Go projects right now are using -- well, a lot of projects in general right now are using GraphQL because it's trendy, but I feel that Go is a little bit lagging behind, because I feel that it wants to use GraphQL because it's trendy, but I feel that the features of the language are precisely what make it a little bit unsuitable... And I'm not saying unsuitable in an absolute sense, but I'm saying it's a little bit harder to achieve that theoretical idea of what GraphQL is best in.
+
+So I guess my unpopular opinion would be that I'm not really convinced that it's being used in a really good way now. And right now I'm also not the person who is offering a better way to do it, but I'm really interested in exploring a better way to do it, and I'm really interested in bouncing ideas with potentially better Go developers to figure out how to solve this problem, and potentially create another client, that could be used not just with the GitHub GraphQL API, but for any other. I would be the first project to migrate over to that, because I would really be keen on figuring out how we can batch and squash queries together, and also use more concurrency features that Go is so good in.
+
+**Johnny Boursiquot:** So I heard Mislav say that all current Go implementations of GraphQL clients suck. \[laughs\] And that a new one ought to exist, and PRs are welcome, or brand new projects are welcome.
+
+**Mislav Marohnić:** To say "suck" would be a hard word, but I'm very thankful for -- right now we're using `schurcool/graphql` for GitHub CLI, and it's an excellent library. I would recommend checking it out. But I also would like to explore what can we do on top of that approach. How can we do take that approach even further.
+
+**Jon Calhoun:** On one hand I'm not too shocked that that's the current state of everything... One, because Go, like you said, does not strike me as a language that is really meant -- it doesn't seem as flexible as some of the other languages. If you're writing JavaScript, it's a lot more flexible in what you can get away with. But then the other aspect of it is so many front-end UIs for websites are built in JavaScript - pretty much all of them - or something that compiles into JavaScript at some point.
+
+\[01:03:54.23\] So as a result, you expect the libraries to be there. But with CLIs, while there are a decent bit of CLIs, I think, being written in Go, I don't think that numbers -- like, the sheer number is not quite there, especially for... You're building a CLI and you happen to be interacting with an API that's GraphQL. That's gotta be a pretty small number right now. Now, that could grow (I don't know), but I could definitely see that being not a huge audience right now.
+
+**Mislav Marohnić:** That's a fair assessment. But I believe it will grow. For instance, I would definitely right now prefer to use GitHub's GraphQL API. And I'm not just saying this because of course I'm biased - I'm a GitHub employee - but I was first and foremost always a GitHub user, and I had used all of their APIs since the first day they shipped until now, through their different iterations. And those powers of GraphQL that I described are definitely where I see going for it in the future they are harder to set up, but they have a bigger pay-off, and I feel tech is inevitably going towards that.
+
+Look at Kubernetes, for instance. Harder to set up, but there's a huge pay-off at the end of that. So I think this space will evolve.
+
+**Jon Calhoun:** I'm curious if that's one of those fields or areas where if Go doesn't find a good solution, it possibly becomes less useful as a CLI language.
+
+**Johnny Boursiquot:** Whoa, whoa, whoa... Jon. Jon.
+
+**Jon Calhoun:** If a lot of APIs end up being GraphQL and Go doesn't do it well, I can see that being problematic.
+
+**Johnny Boursiquot:** Blasphemy. \[laughs\]
+
+**Jon Calhoun:** I mean, I haven't seen any other static language do it any better, to my knowledge... But I don't know. I'm sure somebody will tell me Rust does it better.
+
+**Johnny Boursiquot:** There's always somebody who's gonna tell you Rust does it better. \[laughs\] So Mislav, you have two unpopular -- or was that like a two-in-one?
+
+**Mislav Marohnić:** No, that wasn't two-in-one. I have another one, if you have time.
+
+**Johnny Boursiquot:** Oh, please.
+
+**Mislav Marohnić:** Okay, so the other thing that I'm really opinionated on - and it's directly related to a lot of things that we're talking about, especially early on with hub - is just generally Git. And when I say Git in this context of today's show, I mean the Git CLI, the Git command line interface... Which I would also dare to say is the primary interface to Git itself... Because Git is a concept, it's a storage mechanism, and it's also a protocol, and reference implementations in it can also be in C libraries, it can be imported in the projects... So Git is all of that. But I think most clients of Git still right now wrap Git on the command line.
+
+And a lot of users - I can't say most, because I have no data on the thing - like myself also still primarily use Git on a command line as a primary interface to Git. So my unpopular opinion is that Git is actually so hard to not just learn, but to use consistently... And I say that as a person who used it for probably over ten years, because I used it since GitHub was in beta, and I heard of this Git thing, and it was trending, and it was cool, and probably my only early adoption thing kicked in around that time, when I wanted to check it out... Since then I was using it probably every day. At least every day that I'm on my computer - which is not every day, but every day that I'm at a computer, I was using it. And I have interfaced with it so often, and read all of its man pages, and documentation, everything, and still to this day, ten years later, sometimes it's hard for me to explain.
+
+When people come to me with a very basic question of like "Oh, I just pushed a change. I really didn't wanna push that, so how do I undo it?" So from their perspective, that's like a really reasonable ask to make. And then I'm just like "I'm sorry... This is gonna sound like I am teasing you or that I'm mocking you, but actually I'm really being frank, and I'm gonna give you my advice. It's just not gonna be great."
+
+\[01:07:58.26\] Or for instance, when people ask "How do I delete a branch?" and then I have to ask them "Well, what do you want to do? Do you wanna delete a local branch, and then just get it recreated when you pull again from the same remote, or delete the remote tracking branch, or delete the remote branch?" And now for the first time I'm realizing that they have never even considered that there's such a thing, because why would they. A branch is just a single concept in our head that is made needlessly complicated by -- I wouldn't say "needlessly", but it is made complicated by the inherent distributed nature of Git.
+
+It's not to say that I've become disillusioned with the tech itself. I think it's amazing, and I think that the tech itself is at the quality of the tech that is Git is a testament to GitHub being able to make it this far in the tech space... And I think it evolved amazingly as well. But I think in its evolution it's only getting bigger. It's just getting more commands. And even though its documentation is getting more approachable by newcomers every year, and there's really good man page documentation by now, and error messaging is fantastic because it often suggests you what you should do next to get yourself out of the mess that you've accidentally made...
+
+I feel that with all that power that it's gaining, instead of being a more approachable tool, that it's actually being a tool that is continuously making people feel frustrated, to the point where I feel that whatever the next version control system is - and it does not have to be something separate than Git; it should maybe be just a really powerful abstraction built on top of Git. But I think whatever the next iteration of the people's version control is, it should be something that is just more reflective of how we think about what is version control for us.
+
+And how people think about things is generally always very simple. "I have some changes. I wanna share it with Jon and Johnny, so they can tell me what they think. And then maybe they can add their ideas. And then we can have a merging of our ideas, and eventually test out if it works, and have it out there and ship it." That wasn't hard to explain. I think it's very easy for all of our listeners to understand that mental model in their heads. But then when we come to physically typing out all those commands, suddenly we need months or sometimes even years of learning this set of tools to become proficient enough with them.
+
+I have really initially resisted the idea of graphical tools for Git, because I was this heavy terminal nerd; I was very much in my terminal bubble of being really proficient with a lot of these things, because I was for ten years, before even my Git learning, I was using just terminal tools in general, because I was a very Linux nerd. And I feel that even though it was easy for -- not easy, but it was possible for me to learn that, I feel that nobody should need to have spent so much time in a terminal to be able to understand those things. I especially see it when somebody not from my background is approaching this.
+
+So I would definitely say it's not such an unpopular opinion; I've heard a lot of people express their anguish, especially on Twitter, with their inability to use the Git command line even after a long while. I definitely feel, like the Go GraphQL world is something, this is something in version control. The user hands-on aspect of version control, how we interact with it, needs to be built in something that is much more closer to how humans think about it... Rather than being "I will get you to think about a directed graph or as operations on a directed graph" or something like that. No human thinks about that. Humans think about "I'm gonna save my work and I'm gonna share it with other people. Then I'm gonna step off this computer and just leave for the day."
+
+**Johnny Boursiquot:** So I heard Mislav say we should all go back to using Visual SourceSafe or something. \[laughs\]
+
+**Jon Calhoun:** Oh, boy...
+
+**Johnny Boursiquot:** \[01:12:19.02\] I kid, I kid.
+
+**Mislav Marohnić:** I guess this short TL;DR version would be I feel that version control systems - the next version of them should not be something that was specifically made for the Linux Kernel community. It should be something that was specifically designed to be used by the wider community. And it can still be implemented on top of the Git tech, or it doesn't have to be.
+
+For me, I love version control and I'm gonna love it in any iteration it appears in. I feel that the next one should be with more broader users in mind than a bunch of people who are already really comfortable with their terminals, and they're reading the email from mailing lists in their terminals already, and unpacking patches by typing out a tar command in a single go. The new generations of users of GitHub that I witness are not those people. And they're not me, and they're not those Linux maintainers. Sometimes they're even designers.
+
+We have a designer on our team, and if left to her own devices, she would use a graphical tool for version control like GitHub Desktop. Not just because it's easier, but I think it's just a saner solution. I've also felt it, that I sometimes feel just very safely coddled by a tool that just is like "Okay, save. Here's a Save button. Okay, here's a nice rendering of what just happened."
+
+I've grown to now be in-between the worlds, of not being so seduced into thinking that the terminals are answers to everything, but also to always consider that there's a graphical equivalent of things, and there's better abstractions that we can do, and that we should be more inclusive with our software in general.
+
+**Jon Calhoun:** I think Git has sort of fallen victim to the fact that you have a bunch of power users who want to be able to do anything and everything, and it enables that. But like you said now, the average user wants to do 1% of what Git can do.
+
+**Johnny Boursiquot:** Average things, right?
+
+**Jon Calhoun:** Average things, like maybe 1%. And I think because of that it's just hard to -- it's kind of like you talked about... I think you talked about libraries earlier. Hub - you didn't wanna break it for anybody who's using it... But realistically, there almost needs to be two versions of Git. Like, "This is the average user's version, and this is the "you wanna be able to do everything under the sun" version", but that's hard to maintain. It's hard to make that work, which is a challenge.
+
+I'm even thinking of -- Git has the tools to do a binary search to figure out where something broke, and where a bug was introduced, and the average user probably has no idea how to do that... Which makes sense, because most people probably aren't doing that. But those things all are there, and they exist, and they're cool, but it's just -- every time you wanna use them, you're like "Let me go find a tutorial that teaches me how to do it again, because I sure don't remember."
+
+**Johnny Boursiquot:** Indeed. Well, it is that time... Sadly, we have to go away. I know you will miss us in our absence, but it's been a pleasure having you on the show, Mislav. Thank you so much for the insight on hub, how it came to be, and of course, its successor, gh, or the GitHub official CLI. We're glad you made the decision to write that in Go, even though it has its challenges. From what we hear, we think you think it was a good decision. Are we in the right ballpark?
+
+**Mislav Marohnić:** Definitely. One thing that I'm happy about having chosen it is that I get to now learn it better as well as result of it. That also happens sometimes with people's contributions who say "Well, I switched to this standard library thing", and then as a result I'm like "Wow, this exists. This is great." Something like this happens every week.
+
+**Johnny Boursiquot:** Awesome, awesome. Well, thank you for coming. Jon, thank you for being an excellent co-host. We will catch you on the next Go Time!

Go at Cloudflare_transcript.txt

@@ -0,0 +1,345 @@
+**Mat Ryer:** Hello, and welcome to Go Time. I'm Mat Ryer. Today we're gonna be taking a look at Cloudflare, which is kind of a real-life case study and somewhat success story for Go, I think.
+
+Today I'm joined by a couple of regulars, and a special guest. Jon Calhoun is here. Hello, Jon.
+
+**Jon Calhoun:** Hey, Mat.
+
+**Mat Ryer:** How are you doing?
+
+**Jon Calhoun:** I am doing well.
+
+**Mat Ryer:** Good. We're also joined by Jaana B. Dogan. Hello, Jaana.
+
+**Jaana Dogan:** Hello!
+
+**Mat Ryer:** How's it going?
+
+**Jaana Dogan:** Good!
+
+**Mat Ryer:** And we are also joined by our special guest, CTO of Cloudflare - it's only John Graham-Cumming. Welcome, John!
+
+**John Graham-Cumming:** Hello! Hi, nice to be here.
+
+**Mat Ryer:** Thanks for joining us, we're very excited about this. I suppose at the top of the show maybe you could give us a quick intro of yourself and Cloudflare, just so people have the background, for anyone that hasn't encountered it.
+
+**John Graham-Cumming:** Yeah, absolutely. If you don't know what Cloudflare is, what I always say to people is you probably used it today without realizing it. We are an infrastructure company that protects and accelerates and makes more reliable about 20 million domains on the internet. That's everything from websites you've probably used, to the API back-end for an app on your phone, all those kinds of things. We check those things, and -- you know, if you visited a website or go and use an app which uses us, we've taken a look to see if you're a hacker, and decided you're not, hopefully, and let you through, or we've made it faster or more reliable.
+
+So we're a piece of infrastructure on the internet that's kind of hidden, but it's there. And I'm Cloudflare's CTO; I've worked for Cloudflare for eight years. Started out as a programmer, writing a lot of Go, as it happens; that was the first thing I wrote at Cloudflare, was Go code... And now here I am, to talk about that.
+
+**Mat Ryer:** Great. And actually, John, I remember you spoke at the first GopherCon, didn't you? On channels.
+
+**John Graham-Cumming:** I think that's right. Or was it the second? It was one of the very early ones. I gave a talk about channels and why I like them a lot, which goes way back into my history of my Ph.D, because I did a whole lot of Ph.D. stuff - long before Go, but I actually used a similar paradigm.
+
+**Mat Ryer:** That's very cool. So did Go appeal when you saw channels in the language then? Was that something that appealed?
+
+**John Graham-Cumming:** \[03:54\] Yeah, that was an enormous part of it, because I had done -- we have to go right back into the late '80s, early '90s; I'd done a Ph.D. using a language called CSP (Communicating Sequential Processes, and a language related to it called Occam. Both of these have a concept essentially of channels and communication with synchronization over channels. When Go came along, one of the things in the actual Go introduction was -- it says "Oh yeah, we were inspired by CSP and the way in which did things." That was a big light bulb moment for me, because I was like "Wait, I was doing that a very long time ago." And I'd done loads of other programming, and I was like "Finally."
+
+I really liked that way of thinking and reasoning about how a program works, and so here it was... And here it was in a C-like language, I guess we could call it that, because I'd done a lot of C and C++. And something with garbage collection... I was like "Oh, this is an interesting combination of things", and that's what got me interested in Go right from the get-go.
+
+**Mat Ryer:** That's interesting. Is the design the same in CSP and channels in Go, or are there differences?
+
+**John Graham-Cumming:** Well, it depends. If you go right back to the beginning of CSP, there weren't actually explicitly channels, although quite quickly afterwards the idea of channels came along. The really interesting idea in CSP is you have -- well, Communicating Sequential Processes, so you have a bunch of processes that are doing things sequentially, i.e. goroutines that are just doing their own thing... And they synchronize to communicate. That's really the key thing. And if you think about an unbuffered channel in Go, there's this explicit moment of synchronization. When this gets transmitted, it gets received by this other goroutine.
+
+So that idea of this explicit synchronization for communication makes it quite easy to reason about what your program is doing. It doesn't eliminate, of course, all problems, but it is much simpler to reason about than something that's vastly asynchronous... So that really appealed to me, and that was fundamentally what was in CSP.
+
+**Jaana Dogan:** John, did you see this old paper from Bell Labs? Maybe it was from the days of Plan 9... They were conceptually explaining some of the concepts around synchronization over channels. I think there was a paper like that, and I think that Go is a proper implementation of some of those concepts... If I can actually find the paper, I will send it to you. It's really interesting to see the resemblance.
+
+**John Graham-Cumming:** Yeah. Well, what's interesting about Plan 9 -- so Plan 9 arrives in the '80s, and the fundamentals of CSP are 1977 and 1978. So I think a lot of this stuff has an origin in some of this thinking about synchronization... And also, people think about "What are we gonna do if we have really big, multi-processor machines?", which they didn't have at the time, but "How are we gonna tame them?" and this was a way of taming them, with this kind of thinking.
+
+**Mat Ryer:** That's interesting... The theory came just before we practically had the use.
+
+**John Graham-Cumming:** Yeah, well -- I mean, I think that's actually true with a lot of theory. I think if you look at a lot of computer science, going backwards, you have (especially in the 1970's) a very rich vein of papers which are talking about the problems of distributed systems, clock synchronization, how are we gonna think about this, how are we gonna reason about it... Because I think people could see on the horizon that it was possible that there was gonna be many multiprocessor machines. And there were some early ones. The CDC 6600 I think had multiple function units. So people could already see "Oh, this is gonna happen."
+
+\[07:43\] What's funny is in a 1978 paper on Communicating Sequential Processes there's actually a line in there which is something like "...however, developments of processor technology suggest that a multi-processor machine constructed from a number of similar self-contained processes may become more powerful, capacious, reliable and economical than a machine which is disguised as a monoprocessor..." So you could kind of see, "Oh, this is gonna happen", in 1978... And then it did.
+
+**Mat Ryer:** That's awesome. So how much Go is there then at Cloudflare? And was it there from the beginning, or were there other languages that pre-dated Go's use?
+
+**John Graham-Cumming:** I didn't join Cloudflare right from the beginning. When I joined, there were 24 people in the company, and Cloudflare was already operating. They had a service that was up and running for a few thousand domains, compared to the 20 million today... But it was operating, and it was quite a mixture of things.
+
+The core processing actually in Cloudflare for the actual request processing, which today is doing something like 10 to 15 million requests per second. That was written in PHP. There was some C, because there were modifications to NGINX as C modules, to make NGINX faster for certain things... There was some C++ for doing log file processing... So there was a little bit of a collection of languages. And then the company was just really getting started with Lua.
+
+The plan at that time was to reimplement the PHP in Lua, because NGINX has a good Lua integration, and there's a thing called OpenResty, which allows you to write RESTful programs, with the Lua and NGINX combination. That was actually underway when I joined, was that move away from PHP... But there was no Go, and part of the reason for that is we're talking 2011, and we've just had the ten-year anniversary of Go, so you're talking two years into the life of Go, and pre-version 1.0.
+
+When I joined we were on 0.98, pre the first release. I had seen it and had been messing around with it a little bit, because of what I thought was nice about it... But very quickly - actually, literally two weeks and a day after I joined Cloudflare - I had a few thousand lines of Go implemented, to show how we might use it in a real product. It was a product that still exists today, actually.
+
+**Mat Ryer:** Alright, that's interesting... So you actually tackled a real problem with it to demonstrate and show it off internally.
+
+**John Graham-Cumming:** Yeah, because I had a pretty sure idea that it was gonna be good for what we had, which was lots of computers, with lots of CPUs around the world, and stuff that was quite I/O-bound... Because we fundamentally do network stuff, and I think Go has a fantastic net library, and it was fairly easy to get this thing up and running, and it was literally two weeks after I joined that I had a prototype running.
+
+**Mat Ryer:** Did your prototype contain use of channels then?
+
+**John Graham-Cumming:** Oh, absolutely. Just to talk about this, there's a thing called Railgun, which is -- one of the things that happens with a service like Cloudflare is we sit between the end user and the web server... So that could introduce latency, because instead of going directly to the web server, the end user comes to us. Now, as it happens, because of the size of our network now, it doesn't actually cause a problem, because we're so massive... But at the time, there was a real concern about what could we do to speed up the link between us and the web server.
+
+So this Railgun thing, the idea was if we took over the connection between the two and we didn't use HTTP, we could do all sorts of interesting things with compression; in particular, we could do what's called delta compression, which is to realize that websites don't actually change that fast, and if you request the same page multiple times, it's probably only changed by a few bytes. So you can get crazy amounts of compression, like 99.5% smaller than the original size.
+
+\[11:48\] But we're a shared environment, so what you're gonna have is you're gonna have these Railgun things talking to potentially thousands of web servers on the back-end, with thousands of connections. So if for every connection we had a goroutine, and some coordinating goroutines, and then channels all over the place...
+
+**Mat Ryer:** And your talk covers some practices and things that you can do with channels as well in Go, which I always still to this day refer back to when I wanna use some of those patterns as well...
+
+**John Graham-Cumming:** Yeah, that was kind of fun, because a lot of that was just, in a way, stuff I'd done at university, and I was like "Well, these are the sorts of things you can do. You can make a specific synchronization, or you can coordinate things in a different way", and that was just to show people what the possibilities of channels were... And they're not just a mechanism for getting output from your program, or something; they're fundamentals. That was the idea of that talk, it was just to give people some idea. I think it's called a Channel Compendium... Is that what it's called? I don't quite remember now. It's been a while...
+
+**Jaana Dogan:** Did you end up refactoring any of those bits? I'm seeing a lot of people going over some of the previous patterns, say concurrency patterns they've come up with over time... So did that happen at Cloudflare as well?
+
+**John Graham-Cumming:** Absolutely, in terms of Railgun... Because one of the things that was interesting about Railgun was I was both learning the language and writing what was gonna be one of our products at the same time... So there were definitely times when I did things that could have been done better and it got refactored extensively... And of course, eventually a team of people took it over and in classic style said "What idiot wrote this? We need to refactor everything..." \[laughter\]
+
+**Mat Ryer:** "Let's take his permissions away..."
+
+**John Graham-Cumming:** Yes, exactly. \[laughter\] Exactly. So yes, of course... But the other thing that's really important to realize is we were on 0.98, so there were bugs, and there were things that weren't implemented. Most notoriously -- so we were running on Linux, because Cloudflare uses Debian everywhere... But we were shipping software -- this Railgun thing had a component which was on the customer's website, and one of our customers were using FreeBSD; Go supported FreeBSD, but unfortunately it would run out of memory very quickly... And it was one of these situations where you think "Why is it not running out of memory on Linux, but it is on FreeBSD?" And so that was my chance to learn all about the garbage collector and how the garbage collector works. That's the beauty of open source - I was like "Well, I can debug this myself. I'll go and read all the garbage collector code..."
+
+Eventually, I realized that everything seemed fine. What was happening was when the memory is being actually released back to the operating system -- so you've collected garbage, and then you've decided "I literally don't need this bit of the heap anymore. I can give the operating system that bit back...", it caused this thing in the Go runtime called sysUnused(). And sysUnused() is meant to call something in the operating system called madvise to say "I don't need this block of memory." It does that on Linux, and on FreeBSD it didn't. And the reason it didn't is I then obviously got the source code, and I come down and I find:
+
+*// TO DO (rsc)*
+
+So that's why we were running out of memory, because every time we gave memory back to the operating system, we said "We don't need it", but the operating system didn't get it...
+
+**Jaana Dogan:** That's amazing...
+
+**John Graham-Cumming:** So that was a fairly easy fix. I went in and fixed it... And if you can go back in time, you can find the request by me to say "By the way, I'm implementing this on FreeBSD", because it was missing. So there were things definitely in the early days.
+
+The other thing that was -- I mean, the syslog package needed some work; it wasn't quite compatible with the RFC, so I fixed that, because syslogging was very important... And the other thing that really was difficult right at the beginning was sync.Pool didn't exist.
+
+\[15:57\] In fact, there are some things I talk about in the Channel Compendium, which is "How do you recycle memory in Go pre the existence of sync.Pool?" and you can do it really nicely with channels. There's a really nice pattern for it. Because fundamentally, if you think about what was happening inside of Railgun, there was a lot of "We're doing this HTTP request, we're sending it over here", and then you're getting rid of it again. But then you need another one of these HTTP requests. If you go back and forth the heap, you just end up with this big mess and a huge amount of garbage... So I had to figure that out, and that was a big debugging effort.
+
+**Mat Ryer:** Did it put you off then, encountering these problems? Were these little red flags that you were sort of -- was it chipping away any sort of...? I don't know how to ask that question in a good way...
+
+**John Graham-Cumming:** Well, I think I was lucky in that Cloudflare was developing at the time; it was very much a small company. I wasn't coming into a big company where everything was already set... So I got away with a lot, having these problems. We were growing very rapidly and implementing a lot of stuff. And I think in the early days lots of things are difficult.
+
+But to be honest with you, I've run into problems with every programming language I've ever used, ever, for the last 25 years or whatever it is (maybe even 30). I think that you just deal with the problems of the thing you're dealing with - the operating system, the libraries... I did a lot of C++ programming, and don't get me started on horrible things that happened with C++ with libraries...
+
+**Mat Ryer:** \[laughs\] Yeah. And of course, now everyone can use Go with a lot more confidence thanks to efforts like yours, where people did go in and fix things that didn't work for them.
+
+**Jaana Dogan:** I think Cloudflare has been extraordinarily instrumental figuring out some of the critical things... I remember the leap second bug, that you wrote a post-mortem about. There was no monotonic clock, and there were a lot of discussions going on, but nobody was particularly interested in just considering it as an additional API, or in a way to the standard library. That situation actually became very critical, and I think that they released something in the next release, or something, right?
+
+**John Graham-Cumming:** Yeah, they did, and it's great that they did that, although I definitely believe in "A bad workman blames his tools" as a maxim... So we should never have had that bug.
+
+**Jaana Dogan:** Yeah.
+
+**John Graham-Cumming:** We were assuming that the time API was monotonic, when it wasn't... And that was quite scary. But yes, obviously by us having these mistakes, things do progress.
+
+The biggest thing for us though, I would say, is that up until Go 1.5, garbage collection pauses were real, especially as the heap got large, and that really caused these issues... And in particular -- so we operate what I believe is the world's largest authoritative DNS service, and that's entirely written in Go. It's called \[unintelligible 00:18:57.09\] which is our internal DNS server. If you've gone to our website that Cloudflare uses, or used an API, then you've done a DNS request against one of those... And what's interesting about it is that we could see the performance of our DNS server go down when the garbage collector paused, externally; there were external measuring services... And Go 1.5 really fixed that. So that was a huge thing.
+
+Then the other thing is we spend a lot of effort on optimization, particularly around crypto stuff. We've committed back a lot of Assembly optimizations of stuff, so that Go net and crypto is really, really far.
+
+**Mat Ryer:** That's awesome. Were there any situations where you felt like Go didn't fit? Where you maybe tried to solve a problem with it, but you were fighting with it too much and you opted for a different language, or something?
+
+**John Graham-Cumming:** \[19:52\] I mean, we don't use Go for everything... I think that it's very good if you have I/O bound stuff, and you need lots of parallelism, lots of concurrency. It's great for that kind of stuff. If we're doing low-level bit twiddling type stuff... For example, Cloudflare does a lot of image manipulation - resizing, resampling, all sorts of stuff. I think there are languages which are better suited for really fiddling around at a low level.
+
+I know in the early days there were some times where it felt like we were fighting a bit with the language, like in this garbage collection thing. Now I don't feel like that, and now it's a question of "Well, what's the right language to use for this particular problem we're solving?" I think Go is great for many, many things, and we use it a lot.
+
+**Jaana Dogan:** Do you ever have any guidelines around picking a language? I'm seeing some very large companies do that... "If this is the type of problem pattern you're solving, or whatever, just don't use this language." At least they're documenting some of the anti-patterns.
+
+**John Graham-Cumming:** No, we don't have any formal guidelines about it. We have a lot of discussions internally, depending on the project, and then people are fairly free to choose the language they want to use at Cloudflare. We don't actually have a large number of languages people use, but mostly I think from the engineers we have, they'll make a choice about what they wanna use, and it's usually within one of a very small number of languages.
+
+**Break:** \[21:21\]
+
+**Jon Calhoun:** When you were adopting Go early on - I know one of the things we hear a lot from people is that picking up a new language like that is very hard to build a team around, because you can't say "Oh, I want a Go developer with two years of experience" when the language has only been around two years. That's just not gonna happen. Somebody actually asked Mat this on Twitter, I believe... Is that right, Mat?
+
+**Mat Ryer:** Yes. Dylan M. on Twitter asked "Do the devs pick Go up on the job, or do you only now hire gophers? Has that changed over the years?"
+
+**John Graham-Cumming:** We definitely don't only hire gophers, or rustaceans, or whatever. I think that's an enormous mistake, to be like "Yes, you have to have these specific things." You end up excluding a lot of people who are great. And I think that programmers in general are very passionate about learning new stuff and getting new skills. We're very happy for people to learn Go on the job.
+
+There are lots of resources for learning about Go, it's an easy language to pick up, and we have lots of other Go programmers... Equally, we have lots of other Rust programmers. So I think that it would be a mistake if we said "You have to know Go." Yeah, it's great if you do. Super. But fundamentally, technology changes very rapidly, and I think of programming - especially in a very rapidly-changing environment like the one Cloudflare is in - as a learning job. You're gonna have to learn new stuff all the time; language is a part of it, libraries are a part of it. You've just gotta go for it.
+
+\[24:00\] I think if somebody came to us and said "I'm only willing to program in Go. That's the love of my life and that's what I do", I'd be a little bit worried that they would be somebody who'd just be in a rut, and never do anything else.
+
+**Mat Ryer:** That's interesting... Because that sort of attitude flies in the face of a lot of the way that the recruitment stuff works. I saw somebody tweet the other day, they said "Go is ten years old, so I can finally apply for some of those jobs that need ten years of Go experience..." \[laughter\]
+
+**John Graham-Cumming:** Yeah, but that's like -- you know, the people who are putting they need ten years of Go experience are really unimaginative. That's probably a sign that you shouldn't go and work for them.
+
+**Mat Ryer:** Yeah.
+
+**Jon Calhoun:** I think it's also a sign that the person writing that must not use Go much, because -- like, there are some things that people make mistakes with early on, and there are some patterns that they eventually learn to adopt or learn to avoid... But in my mind, one of the things I like about Go is that it doesn't strike me as a language you need ten years of experience with to be productive.
+
+**Mat Ryer:** Right.
+
+**John Graham-Cumming:** If you do need ten years of experience to be productive with a language - I mean, I think we should probably deprecate that language. \[laughter\]
+
+**Mat Ryer:** Yeah.
+
+**John Graham-Cumming:** Now, having said that...
+
+**Jaana Dogan:** I can name too many languages actually in that category... \[laughter\]
+
+**John Graham-Cumming:** I was gonna say, I'm not sure that it's actually possible to master C++ completely. I think everybody who writes C++ writes their own variant, which is probably what people like about it... But no, I think that fundamentally programmers like to learn; they want to work on new stuff, and early on, once Go took off and once we were pushing it, people wanted to work for us to work on Go. So they discovered it, or they were keen on it, and they were coming to us, and that was -- you know, what you want from programmers is motivation. You want people to be motivated. And if you have an intrinsic motivation, which is "Hey, I'm learning, I'm growing", that's fantastic. As a company, why would you not hire people who want to do that?
+
+**Jon Calhoun:** So with that, did you run into issues where -- one of the things people do when they pick up Go is they'll think "Go has these awesome channels. I wanna use them", and they'll use them in a lot of places where realistically they shouldn't be used, or they're overkill. Did you have to push back on that some, or was that just something where -- how did you handle that?
+
+**John Graham-Cumming:** That does happen a little bit... Although I think somehow that self-corrects, because people generate these monstrous things, and I'm sure I did, in the early versions of Railgun; I think I had weird goroutines that were reading from one channel and writing on another one, like some sort of foreign key table in a database, or something... But I think that sort of self-corrects, because people build things that are very complicated. The other thing is we do do code reviews internally, so people will point out "This is a weird way of doing this."
+
+**Mat Ryer:** Yeah. I was definitely guilty of that, too. When I first saw channels, I just thought "Okay, this is brilliant. I mean, do you want me to concatenate some strings? I'll use channels, no problem." \[laughter\] Yeah, absolutely, and it'll be great. And with a bit more experience, sometimes you think "Do you know what - a mutex is just gonna be some simple here. I'm just gonna go for that."
+
+**John Graham-Cumming:** Yeah, I mean... That is the one instance where -- yes, sometimes it's great. If you're accessing a map concurrently - yeah, okay, wrap it with a mutex and use that, and probably not have a goroutine processing a map through channels... I mean, I guess you could do that, but... Ultimately, a lot of this stuff comes down to optimization. It's not the end of the world if you build something that's not optimal, because if it works for your environment, then that's fine... And then later on you go and measure it and figure out what's bad or not.
+
+**Mat Ryer:** I think coming from the CTO of Cloudflare, that's quite an important thing that you've just said, John... Because too often programmers are a little bit obsessed with that too early, of making it perfect, and "Don't worry about whether it's easy to read and maintain, we just care about squeezing out all that performance..." And obviously, if you do that too soon - this is the famous mistake that we all still are making, which is if you do it too soon, you make bad assumptions and things. And the kind of scale that Cloudflare runs at - that's really encouraging to hear that, I think.
+
+**John Graham-Cumming:** \[28:02\] Well, the thing is I started life as a C and assembly language programmer, writing network device drivers, and there every cycle matters. You don't LDA 0 you XOR AA because that's only one byte and it's all about less cycles. But that's appropriate for that environment. Now we have multiprocessor CPUs... It is very obvious that you should measure it, because one of the things about measuring performance is you frequently get surprised by what is the problem. You come to a point and you're like "Wait, why is that happening?" Whereas if you used your gut, it's often completely wrong, especially in large systems; you just don't necessarily know where things are.
+
+I know that programmers like to be really clever, and it's really tempting to optimize things. "I'm gonna write my own strcmp and make it even faster." It's like, "No, you're not."
+
+What we've done, for example with Go, is we went and optimized the crypto stuff, because we're doing a lot of cryptography because of all those HTTPS requests... So it was appropriate to go and do that work, and we have someone who loves doing that work, that used to work for Intel... So I think you've just gotta measure it and figure out where your problem is, and not right from the beginning be worried about some of this stuff, because you just -- you just optimized the wrong thing.
+
+**Jaana Dogan:** Talking about measurements... Where do you measure? Do you measure the production performance? Where is the data coming from?
+
+**John Graham-Cumming:** Yes, we can measure production performance. There are lots of great tools for doing this. You can get in and you can use things like strace to figure out what your programs are doing in production. We do do that in production when we want to fully understand something.
+
+Obviously, we also have test environments, but I have to say, when you're operating at the scale of Cloudflare, one of the things that's surprising is how heterogeneous the traffic is. The traffic that goes to 20 million websites is not what you expect in your test environment; it's very hard to replicate... So the real world will surprise you, and so it is useful.
+
+One of the tools we've used a lot is a thing called a flame graph, which can show us so we can introspect and understand which functions and which parts of the code are spending a lot of time... So when we get to have to optimize things, we have tools to do that work.
+
+**Mat Ryer:** And in your case, code optimizations and things that the rest of us often think of really just as purely technical exercises, in your case they must in some situations have quite a significant business impact. The cost of doing things if you just have a few users, of course - you almost don't even have to think about it. But at your scale it matters... So have there been any situations where there's been a kind of tussle between business demands on one hand, and the technical on the other?
+
+**John Graham-Cumming:** Well, I don't think they have, not from an optimization perspective... Because actually, what happens is we have a financial model where we can say how much saving CPU time saves us in terms of money... Because we're growing very rapidly, and if we can not buy new hardware quickly, then that saves us a significant amount of money... So the optimization - you can look at a dollar amount on it.
+
+Now, what's interesting is we're optimizing for different things. One of them is CPU utilization, because we'd like to buy less machines. Another one is latency, because we'd like you to have the fastest experience when you use our service, you go to a website. To a certain extent they're the same thing, but not always; it depends on what the actual workload is on the edge machine. Some things are not involved in the critical path/latency part, but might be using a lot of CPU.
+
+\[31:56\] So the only real trade-off tends to be "Is it better to build this new feature, or is it better to do an optimization?" That's probably the trade-off that comes into play... Because there might be some new product you wanna build, that you wanna get out, and there'll be some predicted revenue around that product, or you want to be first, or you want to beat a competitor, or something... Versus "Well, that same engineering team could save us X million dollars by shaving x% off of this CPU utilization globally."
+
+Those decisions have to be made, and we have a product management team that helps us make those decisions, about where to spend engineering resources.
+
+One interesting thing that involves Go is we have an internal product which is a sort of load balancer. If you think about the growth of Cloudflare over the last 8-9 years, we've got multiple generations of hardware; we have 194 cities where we have hardware, and there are multiple generations, with different performance characteristics... And we wanted to make sure that basically every machine is running at the same CPU utilization within any city. That was not the case if you do naive load balancing, because all the machines can't handle the load.
+
+A quite typical example would be some machines running at 50% load and some at 75% or 80%, at the same time. So we wrote this coordination layer which is actually measuring the performance of the machines, understanding what they're capable of, and then real-time directing traffic between the machines... And the coordination of that is Go. That has actually brought us into line. There were quite dramatic graphs in Cloudflare where you can see this mixture of CPU percentages in a column and suddenly you turn on this thing called Unimog and the entire graph flattens out. Every machine is running exactly the same CPU utilization. So there's many things that Go is being used for.
+
+**Mat Ryer:** That must be so satisfying, to see those graphs change...
+
+**John Graham-Cumming:** If you use Cloudflare, you probably know that when you make a configuration change in Cloudflare, be that you click a button, or you upload some code to run on our edge computing environment, it goes global very fast. The way this works is we have this internal thing called Quicksilver, which is a distributed key-value store. That is written in Go... And it typically will distribute a change completely globally in under half a second. So everywhere from New Zealand to Atlanta, to Santiago de Chile. And that's, again, this is a fundamental part of what Cloudflare does is to be able to make those changes really, really fast... And again, that's a Go program.
+
+**Mat Ryer:** That's really cool. Can you tell us a little bit about how that works? Did I read a blog post about this (Quicksilver)?
+
+**John Graham-Cumming:** We've mentioned it a couple of times, but we have not actually given a lot of details. It's actually being worked on right now. We're going to talk about all the details about how Quicksilver works, and we're trying to prep the code so we can just open source it, so other people can use it.
+
+Essentially, we're running a log file system globally; you push stuff in, and then you can at any point find where you are in the log and you can catch up. So if a machine needs to catch up with the latest changes, it can just communicate and say "Okay, I'm at this checkpoint. Give me the delta."
+
+**Mat Ryer:** That's why you keep finding time bugs for us..
+
+**John Graham-Cumming:** Maybe, yeah... \[laughs\] I mean, the team who works on that could probably tell you about all of the things they found. One of the things that's interesting is a lot of -- we looked at a lot of other distributed key-value stores and they tend to be oriented around a lot of machines in a single data center. And Cloudflare has a lot of machines in a single data center, that data center being the planet. And the problem with that is there's varying packet loss around the world, and there's very varying latency. Coping with that was really what this was designed to do.
+
+**Mat Ryer:** Yeah, that's really cool. I'd love to learn more about that. I don't have a use case for it, but just sort of geeky curiosity, if you will...
+
+**John Graham-Cumming:** \[36:13\] Yes, absolutely.
+
+**Jaana Dogan:** Yeah, talking about that - you have tons of open source projects that people keep looking at as reference... How did it all start? Did you just want to just push things because that's part of your culture, or did you just specifically did it to share, because Go was very small in the beginning and you were one of the major users?
+
+**John Graham-Cumming:** Well, I'd done open source stuff in the past, so I thought open source was important. I think it's very satisfying for engineers to see their work used by others, so it wasn't difficult to encourage people to open source stuff. And obviously, if you're modifying something, then you should upstream it. The upstream doesn't always want your modification, but you should at least do that...
+
+And what we tried to do is -- so we have some rules around open sourcing stuff. First of all, it has to be in production. The reason for that is we don't actually want to pay people to write open source products that are the thing they fancy writing... They'd better be writing something Cloudflare wants. So our rule is we'll open source something as long as it's used in production, i.e. we really wanted it.
+
+Then there's a real question around the maintenance burden. One of the problems about open sourcing things is people are gonna make pull requests, and you're now gonna have to dedicate time to it. So we tended to open source stuff we thought others could use easily. That tends to be small programs, libraries, technologies which are standalone...
+
+There are bits of Cloudflare... For example, our DNS server - a lot of people say "Can you open source your DNS server?" and the problem is it's deeply-integrated with our business logic, because DNS is fundamentally one of the things we do. So we would have to abstract out that business logic and create some abstract structure for it, and then provide that thing. And that just becomes a lot of work that doesn't make any sense to us.
+
+**Mat Ryer:** Yes. The other thing as well about that is that rule of only open source things that are in production is first of all you know it's useful, but also you know it works as well. I always think that's a great piece of advice for anybody - don't just imagine a package and build it... Well, you can do that; it's a great way to learn and explore... But the best open source packages are ones where people have just solved their problem.
+
+But of course, you gave that good example, John - they're not always appropriate, even if for the greatest will you would do it. Is there ever a fight...? Sometimes people can look at that decision about open sourcing something or not, and get a bit nervous around company IP, and those sorts of issues. Do you ever have that kind of discussion, too?
+
+**John Graham-Cumming:** Honestly, no. The way in which open sourcing stuff works at Cloudflare is there is an internal mailing list. You just email it and say "Hey, I want to open source this thing. This is the license I'm thinking of using." We have a small number of approved licenses... And on that list there's myself, a couple of other senior technical people, and some of our legal team. And to be honest with you, the responses to that -- I mean, if I'm asleep, I obviously don't reply quickly, but mostly it's a yes within a few hours to open sourcing things.
+
+Now, a part of the reason we can do that is Matthew, the CEO of Cloudflare, has said many times that he doesn't believe there's any piece of code that Cloudflare has written that gives us long-term advantage. So that means there's no real danger in open sourcing most stuff.
+
+We think we could just dump the whole of our internal Git and then say "Hey, here is everything", and it ultimately wouldn't hurt us long-term. Now, we're not gonna do it, because of the maintenance burden of doing that.
+
+**Mat Ryer:** \[40:03\] Does it run on AWS? \[laughs\]
+
+**John Graham-Cumming:** God, no. No, definitely not.
+
+**Mat Ryer:** There's no point then.
+
+**John Graham-Cumming:** We definitely do not use AWS.
+
+**Mat Ryer:** No, yeah. That's really interesting... And it does sound like other companies could learn a little bit, I think, from Cloudflare's example here. The generosity and the community spirit and all that is -- I just see it only really rewards companies, but when I speak to people, there's a lot of nervousness around that. The same goes for your blog as well, John, which - for anybody that doesn't know it, it's blog.cloudflare.com. It's a great resource for all kinds of technical and otherwise stuff, isn't it?
+
+**John Graham-Cumming:** Yeah, I mean - that was fairly early on. When I joined, Cloudflare had a blog, and Matthew and others were writing for it. But I really wanted to write really in-depth technical stuff, partly because I like that, and I'm just writing about this stuff, and I think that other nerds like reading in-depth, nerd stuff... So myself and others started doing it, and today I'm the editor-in-chief of the Cloudflare blog.
+
+What I tell people is - when they write something for that - our goal is to educate the reader. So I will go back and I will say "You need to explain to the reader what you're talking about - the background, what the subject is here." So you'll find that some of our blog posts are really long; they take a lot of work.
+
+We have a really fantastic illustrator, Carrie, who does illustrations for the blog... So we really make a huge effort on that, partly for hiring people, because people read and go "I'd like to work for Cloudflare", partly because our customers then will know the sorts of things we're doing, and the sort of technical depth we're involved with, and also partly because it's just good, it's just fun. Other nerds like to know about this stuff.
+
+**Mat Ryer:** Absolutely, yeah. Please do keep it up, we do like it. That's right.
+
+**John Graham-Cumming:** Yeah, we're definitely not slowing down. I know there are -- because I review everything that goes on the blog, and I know there are engineers who probably sigh when they see that I've commented, because I'm probably gonna say "This isn't good enough. You need to add another 500 words explaining what you're talking about."
+
+**Mat Ryer:** Right, yes.
+
+**Break:** \[42:29\]
+
+**Mat Ryer:** \[44:30\] Changing the subject a little, every year at [GopherCon UK](https://www.gophercon.co.uk/) (the U.K. Go conference) we have now a regular little tradition where we visit Bletchley Park. John, you have a connection to Alan Turing, don't you? I'd love for you to tell that story, if you could.
+
+**John Graham-Cumming:** \[44:48\] Yeah, sure. I had lived abroad for a long time, as I am now doing again. But in 2009 I came back to the U.K, and I think I saw [a tweet from probably Stephen Fry](https://twitter.com/stephenfry/status/2292753666) saying "It would have been Alan Turing's 90-something birthday today, if he hadn't killed himself." And I knew the story of Turing because first of all I'm a computer scientist, and I'm interested in computer security... So you end up -- Turing pops up all over the place.
+
+And I knew, of course, that he had killed himself, and I knew a lot about the history of the code-breaking stuff, and artificial intelligence stuff... And I was at home, and I just got really annoyed about it. I honestly thought to myself, you know, the problem here is that we don't talk about Alan Turing in general, because it's a kind of shameful thing which is sort of like "Well, he killed himself because he was gay, and because we prosecuted him and we treated him very badly..." So in a very typically British way we sort of forgot to talk about him.
+
+I thought, "Well, if we could get this out once and for all, we could all talk about it, and admit to ourselves it was a bad thing." Then we could celebrate what he did. So I wrote a blog post on my blog saying "This is terrible. Britain should apologize", and somebody in the comments said "You know, you could start a petition on the number ten website." So I immediately went there and created this petition. Now, I had to wait about a month for it to be approved, and I honestly thought maybe 500 people would sign it. You know, "Who cares about this intersection of gay rights, and computer science?" and I was like "There's probably 500 people in the world who care about this..." Well, in Britain anyway, because it's only open to residents.
+
+So sure enough, 500 people did sign quite quickly, and I thought "Well, I'm gonna give it a go. I'm gonna try and get the press to talk about it", and I just worked on it by myself, just trying to get people to write about it. Eventually, The Manchester Guardian wrote about it. Then The Independent, and so on, and it kind of snowballed a little bit.
+
+One of the first famous people who signed it was Richard Dawkins. The cool thing about Richard Dawkins signing it was that I could then go back to the press and say "I know I told you about this before, but now Richard Dawkins has signed it, so what you should write is "Richard Dawkins has signed this." So I did that, and eventually I wrote this cool postscript -- at the time, the signature names were public, and my father was actually reading them every day and saying "I think this is so-and-so (famous person)." And eventually, I automated my father.
+
+So what I did was I wrote this ugly perl script that took the names and then searched on Wikipedia to see if that person had a Wikipedia page, and if on that Wikipedia page it said something like "So-and-so is a British/Scottish/English/Welsh blah-blah-blah...", so trying to see if they were a notable person from the U.K. And if they were, then it would email me and then I could get a hold of them and ask them...
+
+I did that with Ian McEwan, the writer; his name was on there, and I found his email address. I emailed him and said "Are you the person who signed this?" He's like "Yes." And then I was like "Can I tell the press?" "Yes." I did stuff like that, and it grew and grew and grew.
+
+\[48:00\] Then just before the bank holiday at the end of August I emailed somebody I had met from the BBC, a journalist called Zoe Kleinman. I'd met her because I'd written this book before, and she's written a little bit about my book... And I sent her this really cheeky email which said "This is a really important story. You should write about this." So I was telling the BBC, "Get on with it." And she very kindly wrote back and said "You know what - I'm gonna write something. I'm going away for the bank holiday weekend; I'm gonna submit it to the editor. I have no idea if it's gonna get published." So she did, and I never saw it.
+
+I went to sleep on the Sunday night, bank holiday Monday, I wake up... I have a graph of the number of signatures, and there's almost this vertical line in the number of signatures, because it's suddenly thousands and thousands of people signing... Because of course, BBC has this incredible authority. If the BBC says this is something, people read it, it gets copied around the world... After that I was on the TV and all over the place, talking about it. At that point, 30,000 people signed it.
+
+**Mat Ryer:** I was one of those, John.
+
+**John Graham-Cumming:** Thank you. Thank you for signing it.
+
+**Mat Ryer:** I didn't get flagged up by your automated father as a notable person.
+
+**John Graham-Cumming:** Yes, I'm sorry...
+
+**Mat Ryer:** None taken. I hope you've deleted that program though... \[laughter\]
+
+**John Graham-Cumming:** I actually published everything about this on my blog.
+
+**Mat Ryer:** I'll submit a pull request, thank you... \[laughter\] Add myself to it.
+
+**John Graham-Cumming:** Add yourself to it. \[laughs\] I'm not sure it's gonna work now; it's a bit late. But... And then after that I managed to get the flu, and I was sick as a dog...
+
+**Mat Ryer:** Oh, man...
+
+**John Graham-Cumming:** And I was lying in bed and I thought "I've gotta look at my email. Maybe something's going on at work." I checked my personal email, and there's this email from this woman (I had no idea who she was) who says "Can you call number 10 Downing Street? Here's the phone number."
+
+**Mat Ryer:** Oh, wow.
+
+**John Graham-Cumming:** And I thought "oh it had to been wrong", So anyway, I googled the phone number, and sure enough, it's the switchboard of number 10 Downing Street. So imagine me - I'm lying in bed, groggy as anything, so I call this number and I'm like "Okay, this is who I am", and immediately I get put through to this woman, and she's like "The apology is going out tonight. We've already placed it in the Daily Telegraph." That's where it was gonna be published. "I just need to read it to you and I need you to approve it."
+
+So she reads it to me over the phone, and I thought it was great... It's a text that everyone can read. And then she goes "Gordon wants a word." Now, Gordon was Gordon Brown, the Prime Minister...
+
+**Mat Ryer:** Really?
+
+**John Graham-Cumming:** Yes... So she's like "Gordon wants a word." I was like "Okay..." She said "He'll call you." \[laughter\] So I hang up and I'm sort of sitting there... I'm now awake, and I'm thinking "My goodness, is this really gonna happen?" And my mobile phone rings, and it's Gordon Brown. There's no ceremony; nobody calls up and says "It's the Prime Minister. Are you ready for it?" Suddenly, it's like "Hello, John. It's Gordon..." And you can imagine, Gordon Brown is not a very chatty person, if you recall, and I had flu... \[laughter\] So the two of us were on the phone and not really wanting to talk to each other very much...
+
+The first thing he said to me - and I'm not gonna forget this - he said "Hello, John. It's Gordon. I think you know why I'm calling..." And I thought "Bloody hell I'm glad I know why you're calling! I definitely paid all my taxes and I definitely have never had a parking ticket..." And then we had this very stilted conversation, because I felt terrible, and he's quite a serious man... And that was it. There you go. So that's my connection to Bletchley Park. After that, obviously, Turing got more recognition, and we were able to celebrate him, and I think it helped give Bletchley a leg up...
+
+**Mat Ryer:** \[51:54\] Yeah. And your apology - what you petitioned for an apology, and I feel like that was appropriate. They later went on to pardon him. And for me, that implies that there was something they did wrong somewhere. So a pardon I didn't feel appropriate, but the apology for sure.
+
+**John Graham-Cumming:** Me neither. I didn't support the pardon campaign. Two things were important. One was I felt that the apology was good, because they apologized to Turing and everybody else who was convicted. That was important. The second thing is the coalition government - Cameron's government introduced... What was it called? It was something of freedom; it was the Protection of Freedom Act. And what this did is it decriminalized these homosexual acts... Because the problem is there are people alive today who had been convicted of this thing - it had never been expunged - because it was illegal at the time. So the Protection of Freedom Act fixed that. I felt that was all good, and I didn't feel like I needed to do more. I'd done more than I ever imagined.
+
+Then there was this idea of the pardon, and actually I was upset about the pardon, because Turing alone got pardoned... And there were people who were arguing for the pardon on the grounds that he was a genius, and that he somehow deserved it... And I felt that was like "No, I think either you do everybody--" Now, ultimately they did do everybody, so that's fine, but it just made me uneasy, this idea of first of all the pardon, because I don't think he did something wrong, and then to single him out seemed like a mistake. I thought that Gordon Brown's apology did a fantastic job, and I was glad I did it.
+
+**Mat Ryer:** Yeah, well I was really glad he did it. Thank you for doing that. I didn't realize you worked so hard on it. I kind of had the impression it was a petition that just went viral, so... That's even better. Thanks for sharing that story.
+
+**Jaana Dogan:** Yeah, it's amazing - in the last ten years there was a lot of representation of Alan Turing in the popular culture, and it's all because of you. Thank you.
+
+**John Graham-Cumming:** I think so. It's kind of hard to take credit for it, but if you look at it, it really did take off after the apology.
+
+**Jaana Dogan:** Exactly, yeah.
+
+**Mat Ryer:** I give you credit for that, John. I mean, I'm not a notable person, as your automated father made clear, but you still do get -- in fact, I mention it in my book, Go Programming Blueprints. Still available.
+
+Actually, John, there's another link we have... I won a copy of your Geek Atlas book by guessing -- you had some quiz, it was a reference to the prisoner. I don't know if you remember... You said whenever a website asks for your date of birth, you give the same fake date of birth, right?
+
+**John Graham-Cumming:** \[laughs\] Okay, so you're a giant nerd is what you're saying
+
+**Mat Ryer:** \[laughs\] Yeah.
+
+**John Graham-Cumming:** On Twitter -- Twitter thinks my date of birth is March 19th, 1928, which is the birth date of the prisoner in the classic Prisoner TV series, and also the actor who played him... So yes, well done.
+
+**Mat Ryer:** Thank you. Thank you very much. \[laughter\] Okay, well that's all the time we have. John, thank you so much for joining us and telling us all about Cloudflare and your experience with Go there. It was awesome.
+
+**John Graham-Cumming:** Thank you very much.
+
+**Mat Ryer:** Thanks to Jaana and Jon also, it's been great. We'll see you next time.

Go at Heroku_transcript.txt

@@ -0,0 +1,203 @@
+[0.00 --> 8.04]  what is up everyone adam stakowiak here editor-in-chief of changelog we teamed up with some
+[8.04 --> 13.12]  friends of ours over at heroku to promote their podcast called codish you can check it out at
+[13.12 --> 19.66]  heroku.com slash podcasts slash codish and today we're dropping a full-length episode of codish
+[19.66 --> 26.04]  into the go time feed this episode features johnny boricico panelists on go time on the mic with
+[26.04 --> 30.86]  guests ed moller and rasheed wasson talking about go at heroku here we go
+[30.86 --> 44.72]  hello and welcome to codish an exploration of the lives of modern developers join us as we dive
+[44.72 --> 50.44]  into topics like languages and frameworks data and event-driven architectures and individual and
+[50.44 --> 56.28]  productivity all tailored to developers and engineering leaders this episode is part of our
+[56.28 --> 63.08]  heroku in the wild series hello and welcome to codish my name is johnny boricico and i will be
+[63.08 --> 70.16]  your host for this episode of codish today we will be talking about go at heroku i am pleased to be
+[70.16 --> 75.46]  joined by two of my colleagues ed moller and richard wasson who will share their experiences with the
+[75.46 --> 81.48]  language and add some color to how they've seen it used within the organization um so let's begin
+[81.48 --> 87.66]  with you ed how long have you been at heroku and in what capacity uh let's say i've been at heroku now
+[87.66 --> 99.18]  for uh eight years uh and during that time i've been on well pretty much all the teams um i started uh on our
+[99.18 --> 106.40]  database team and transitioned through some infrastructure uh teams uh that worked on aws
+[106.40 --> 114.28]  and eventually metrics and and logging after that i wanted a little bit of challenge and uh became the
+[114.28 --> 122.84]  go language owner and uh eventually though found my way back to metrics and now into sre land uh where
+[122.84 --> 130.66]  i'm working on uh go tooling uh some open telemetry stuff and uh generally observability uh but rishab
+[130.66 --> 137.22]  um i know even less about you and what you've been up to at heroku so how long have you been at heroku
+[137.22 --> 143.80]  yeah so i've been at heroku for about a year and a half um i'm pretty new to the industry this is
+[143.80 --> 151.60]  actually like my first job and um a bit of a brief history about myself i recently graduated from
+[151.60 --> 158.46]  uc berkeley about two years ago um after studying some computer science and prior to that i was
+[158.46 --> 164.58]  actually a future force intern on one of the other clouds at salesforce called salesforce iot cloud
+[164.58 --> 171.16]  and most recently i've started as a software engineer on the runtime infrastructure team
+[171.16 --> 177.76]  very cool so as for me um i'm relatively new into heroku still i'd say at least you know
+[177.76 --> 183.70]  relative to the two of you uh it's uh been what maybe my sixth i think i'm entering my sixth month
+[183.70 --> 191.14]  at heroku uh i joined as an as an sre um and and ed and i are also part of the sre org within within
+[191.14 --> 197.40]  heroku and this is this has been an awesome journey for me so ed i know you've been at heroku the longest
+[197.40 --> 203.12]  so i know i know internally right now that you're definitely one of the people who looks after the
+[203.12 --> 207.76]  build pack so whenever there's a there's a new version of go that comes out um sort of uh you're
+[207.76 --> 212.48]  sort of one of one of the sort of the first people the org looks to to make sure that we're running the
+[212.48 --> 216.54]  latest and greatest we're patching and we're basically you know keeping up with the the releases
+[216.54 --> 221.22]  of go like can you can give us a little bit of insight into sort of what the process is like is it
+[221.22 --> 226.50]  something that's difficult to do hard to manage like you know is is you have automation around it how
+[226.50 --> 231.90]  how easy is it to keep the build pack fresh and serving our customers keeping go up to date on the
+[231.90 --> 238.18]  build pack itself is relatively straightforward we have a little script inside the build pack repo
+[238.18 --> 245.84]  itself that we can use to to do the work to bump versions and then things like ci and cd for the most
+[245.84 --> 253.12]  part you know validate that the existing tests and everything are are good um there are challenges
+[253.12 --> 260.14]  supporting every new major release so like go 112 to go 113
+[260.14 --> 269.64]  uh etc etc etc um for instance like the default is currently still uh one of the go 112 lines um
+[269.64 --> 276.06]  because there are a whole bunch of implications with uh turning to modules being the default on
+[276.06 --> 283.80]  and yeah we're not yet sure we understand all of the implications of it from the standpoint
+[283.80 --> 290.64]  of the build pack anyway and how that affects all the users of the build pack but the general process
+[290.64 --> 295.78]  of like bumping for point versions and things like that is is really simple pretty much since i've been
+[295.78 --> 301.38]  in heroku every new project it's it sounds like it's it's being written in go um but we still have a lot
+[301.38 --> 306.22]  of projects that have been written in ruby um some new projects even are either ruby and we have some
+[306.22 --> 314.16]  elixir as well um so we're truly a polyglot sort of organization um the we have a set of sort of go-to
+[314.16 --> 319.72]  languages that we use um but go is increasingly being a a a sort of a one of the first ones we look at
+[319.72 --> 324.18]  especially for for for back-end kind of systems and services that need to talk to each other that
+[324.18 --> 330.46]  kind of thing so this is how we do metrics and how we do sort of uh um when folks hit the api.heroku.com
+[330.46 --> 336.82]  um that stuff is is most of that stuff is in go private spaces are run um that most of that stuff
+[336.82 --> 342.60]  is controlled in in go so we have a lot a lot of go so does that mean do you think does that mean we
+[342.60 --> 348.94]  have a lot of go expertise in house and along those lines perhaps rishab you can sort of chime into this
+[348.94 --> 355.04]  one how have you how easy has it been for you right coming out of school and this being your first job
+[355.04 --> 360.08]  were you doing go you know at school or did you learn go on the job how easy has it been for you to
+[360.08 --> 366.06]  sort of learn go and be productive with the language at heroku uh at college actually i never
+[366.06 --> 373.16]  even heard of go um the main languages that i used was actually java and python and coming to heroku
+[373.16 --> 378.92]  and starting my my first job uh it was quite a bit of a surprise just hearing that um i would be
+[378.92 --> 386.12]  working in go and ruby and learning like a new language so specifically there were like a couple of
+[386.12 --> 392.04]  resources that really helped me um gain a bit of an understanding about the go ecosystem and learn
+[392.04 --> 399.46]  how i can actually like start writing um idiomatic go um for example the runtime team like specifically
+[399.46 --> 406.70]  uh we have a program called runtime university and it has a huge section on go which it also references
+[406.70 --> 413.26]  go by example and other resources where an engineer who is like relatively new to go or wants a refresher
+[413.26 --> 419.42]  can actually go in and do some practical exercises and learn for example syntax about the language
+[419.42 --> 426.94]  or like how to actually write um good go also for myself um one of the biggest things that has really
+[426.94 --> 435.56]  um helped me learn go at a at a at a good rate is um reading a lot of go code so taking like an
+[435.56 --> 441.92]  existing microservice that we have and then reading it end to end um really helped me understand how to
+[441.92 --> 449.72]  build a service and go and then also when i'm making changes to an existing service um reading that
+[449.72 --> 459.14]  whole like code base um was very helpful so i think overall um as like a new engineer uh we have some
+[459.14 --> 466.18]  patterns at heroku such as like runtime university um that really support a new engineer um to learn
+[466.18 --> 474.30]  some go so one of the things that that you as a go developer start to sort of uh hear as a new go
+[474.30 --> 479.10]  developer you start to hear more and more of the more go you do is this concept of idiomatic go right
+[479.10 --> 485.02]  so it's you could write go that looks kind of like java a little bit or some other language that you're
+[485.02 --> 490.86]  familiar with you know but you wouldn't be sort of um sort of doing it the way most go developers do
+[490.86 --> 494.84]  right so and this is something that basically is more subjective than it is sort of a you know a
+[494.84 --> 499.04]  specific set of sort of a um you know if you don't write it a certain way then your code will compile
+[499.04 --> 503.02]  kind of thing but it's more along the lines of this is the expectation right out in the go community
+[503.02 --> 507.62]  and this is you know when you look at a go code base and ed i know you've been sort of the the
+[507.62 --> 513.00]  leading sort of person internally behind our go design guide maybe you can sort of add some flavor
+[513.00 --> 518.54]  there as to basically how much of sort of the the community's sort of concept right of idiomatic go
+[518.54 --> 524.74]  how much of that basically is impacting or influencing uh how we approach go development internally
+[524.74 --> 534.34]  i like to think of as our go design guide first of all it's a it's a living doc right so i uh expect it
+[534.34 --> 545.02]  to evolve over time it's an attempt to basically try to get all the engineers on the same page about
+[545.02 --> 554.66]  um how to structure projects and how to write um heroku's version of idiomatic go
+[554.66 --> 565.94]  i do feel that a lot of things captured in that guide are inspired by the things you see in the
+[565.94 --> 573.36]  community at large and for instance the guide has many call outs to everything from like dave cheney's
+[573.36 --> 584.18]  blog um to yana's blog to the go wiki and things like that and usually has text um for
+[584.18 --> 595.20]  purposes of clarification or to uh you know better relate the contents that are linked out to to the
+[595.20 --> 604.04]  context of heroku and the types of services and tools that that that we need to write and like i said
+[604.04 --> 612.24]  some of the the main the main motivation for me uh starting it several years ago was um as the go
+[612.24 --> 620.02]  language owner then uh i would read a lot of go code internally and different projects were trying
+[620.02 --> 628.00]  different things which is something you know we want to continue to encourage but as the code bases um
+[628.00 --> 636.66]  go from experimentation phase uh into production and things like that in order to have that code base
+[636.66 --> 645.30]  uh easily maintainable and modifiable by a large swath of engineers it my my opinion is uh that
+[645.30 --> 655.32]  it's better if we do thing x in the same way um across as much of our code our go code bases as possible
+[655.32 --> 663.90]  and the intent the attempt for the design guide is to say is to state what those ways are um and the
+[663.90 --> 672.46]  reason uh and motivation behind why we feel that that is the right way for us to do it and to provide
+[672.46 --> 679.86]  examples and all that context uh for uh for posterity uh and again since it's a living doc though it doesn't
+[679.86 --> 685.46]  mean it's going to all like if we say do this for these reasons it doesn't mean in six months from
+[685.46 --> 691.54]  now it might not give different advice for the same thing because we've learned something in the in the
+[691.54 --> 701.84]  interim um and we have a process in place um through an internal rfc or request for comment um process uh
+[701.84 --> 708.18]  to modify that doc and rishabh uh i know you want to add some flavor to that yeah um so i this is actually
+[708.18 --> 713.04]  something i forgot to mention earlier the go design guide for myself has actually been like a fantastic
+[713.04 --> 719.96]  resource as a new engineer um i think for example when i start working in a new code base um with like
+[719.96 --> 725.90]  go like learning the learning the language as a beginner um you're always looking and asking around
+[725.90 --> 732.64]  for patterns or questions around is the way that i'm writing this correct and does it make sense and
+[732.64 --> 737.36]  the go design guide being there as kind of like an artifact that you can look at and
+[737.36 --> 745.30]  see patterns and standards which to most people make sense um provides kind of like a sense of
+[745.30 --> 752.56]  proof or like prior art to say hey this code makes sense because the go design guide references this
+[752.56 --> 759.98]  pattern so i'm going to implement it this way and also um when doing code reviews for example um i also
+[759.98 --> 767.10]  reference like the go design guide as a reason um as to why certain code should be written in a certain way
+[767.10 --> 775.50]  so it provides a lot of examples and provides a lot of explanation as to writing um a good standard of go
+[775.50 --> 785.36]  so as as someone who says sort of learn go on the job i do want to sort of ask though like as sort of a
+[785.36 --> 792.62]  newcomer to go what have you found to be the hardest thing right to learn about the language what what have
+[792.62 --> 796.78]  been sort of your stumbling blocks what things you keep coming back to and over and over that has
+[796.78 --> 803.36]  taken some time to sort of sink in for you yeah so i think there were like a few things that i found
+[803.36 --> 810.32]  difficult along the way when learning go um one of them was actually dependency management so knowing
+[810.32 --> 816.64]  so different code bases used different tools to manage their dependencies some code bases that we have
+[816.64 --> 825.10]  used depth others uses used go mod or modules so understanding like initially when i was starting
+[825.10 --> 831.28]  i had no idea how like dependency management work can go so knowing when to use either was a big issue
+[831.28 --> 838.62]  also a difficulty along the way that i had was knowing that the code that i was writing was idiomatic
+[838.62 --> 846.56]  so i would look to other resources such as the go design guide or dave cheney's blog to really validate
+[846.56 --> 853.00]  whether or not the code that i was right writing was idiomatic and another issue or another difficulty
+[853.00 --> 859.76]  along the way was knowing if the pattern that i was using to solve a problem uh was appropriate at the
+[859.76 --> 867.74]  time and it's very similar to writing idiomatic go but knowing if the solution that i had
+[867.74 --> 875.88]  uh was appropriate and have you found if you found that uh basically team members to be sort of a
+[875.88 --> 883.56]  go-to resources as well um so how much how much sort of a um face-to-face you have you had have you had
+[883.56 --> 889.70]  to to to to do right and then how much of it do you think you spend in sort of a um pull request going
+[889.70 --> 894.12]  back and forth kind of thing like did you find have you thought i guess what i'm asking what has the
+[894.12 --> 899.26]  experience overall been like for for somebody who's didn't know go before and has has sort of been
+[899.26 --> 905.58]  learning on the job yeah definitely um so i think being very new to go um just asking a lot of
+[905.58 --> 912.70]  questions and always questioning like why something was done the way it was done um so just being open
+[912.70 --> 917.54]  to like asking a lot of questions especially to like senior engineers who've been like writing go for
+[917.54 --> 925.76]  a multitude of years and specifically in pull review or pull requests um asking targeted questions to
+[925.76 --> 931.52]  the reviewers and saying hey does this section of code make sense based on this pattern that i found
+[931.52 --> 939.62]  here and really soliciting specific feedback um so yeah i also definitely took advantage of like
+[939.62 --> 945.00]  pairing uh while doing code reviews so we would hop on a call together uh look at the code together and
+[945.00 --> 952.30]  i would explain and do some kind of like a walkthrough um explaining the reasons why it took um when
+[952.30 --> 958.34]  writing a certain piece of code and really got a chance to add some more flavor to the code review
+[958.34 --> 965.34]  process so ed i'm curious do you do you think your journey i mean you've been doing go for many many
+[965.34 --> 972.86]  years um what do you where do you see yourself at this point in your sort of a career in terms of you
+[972.86 --> 977.36]  know your your your expertise with go like do you do you still think the language has a lot to offer
+[977.36 --> 983.56]  you you still you still think you there's a lot to learn like is it the has the way you've written
+[983.56 --> 989.52]  go along those lines has the way you you write go today has that changed over the years oh it's totally
+[989.52 --> 999.16]  changed um goes not my first language um it's maybe my seventh eighth sixth something like that language
+[999.16 --> 1005.74]  professionally i did what everybody does or at least i assume what everybody does when they first
+[1005.74 --> 1017.62]  learn a new language and that is mix in you know styles picked up and or used in their uh in the
+[1017.62 --> 1024.76]  language that they used previously most previous to ruby for instance i had used python a lot so when i
+[1024.76 --> 1031.68]  first came to ruby i wrote my ruby like python um and was quickly yelled at by a bunch of rubyists
+[1031.68 --> 1044.40]  um and uh so when i came to go i wrote what i currently call um gooby which is the mishmash of go and ruby
+[1044.40 --> 1053.48]  which i've seen plenty of i've also seen plenty a guython i've seen plenty gava um i i've seen all of
+[1053.48 --> 1059.50]  these and i and i'm not and i don't say that to like cast derision on the people doing that but i i
+[1059.50 --> 1068.04]  believe it's also important to recognize that that's part of the process of getting like transitioning and
+[1068.04 --> 1074.86]  and and and learning uh a new language uh as far as uh new people coming into go though i think like
+[1074.86 --> 1080.00]  uh i think the best thing we can do for new engineers is pair with them uh and pair with them
+[1080.00 --> 1092.16]  on um specific uh tasks i have offered and continue to offer pairing um with any any engineer on any
+[1092.16 --> 1100.36]  project and uh those who have taken me up on it the sessions have been most productive i i believe for
+[1100.36 --> 1109.38]  myself as well kind of experiencing the beginner mindset uh which i'm so removed from now so uh you
+[1109.38 --> 1117.04]  know experiencing that firsthand and then uh i hope for them as well because i i you know through that
+[1117.04 --> 1126.22]  process of pairing we can explore like the design philosophies that i think are enigmatic go where
+[1126.22 --> 1132.60]  some early attempts at this people i was pairing with attempted to use me as like an as i go encyclopedia
+[1132.60 --> 1140.72]  and i i feel like neither one of us uh you know neither party came away from that experience uh
+[1140.72 --> 1148.14]  feeling like they had gotten good use of that time uh so that's something else i i recommend for more
+[1148.14 --> 1155.70]  senior go engineers when you when you offer something or do pairing with somebody uh try to do it on a
+[1155.70 --> 1163.38]  like have a specific focus um and a goal to accomplish you know refactor uh new implement
+[1163.38 --> 1170.08]  something new uh both of those whatever not just like hey look at this code and tell me how it could
+[1170.08 --> 1177.34]  be better because there's no there's no absolute right answer to that it it involves too many contexts
+[1177.34 --> 1187.18]  right that are specific to the the project how the deployment uh the engineers involved etc etc where
+[1187.18 --> 1195.38]  where do we see go sort of uh going at heroku the obviously we've already mentioned that a lot of our
+[1195.38 --> 1202.20]  a lot of our control plane um software is written in go and you know everywhere i turn i think you know
+[1202.20 --> 1208.14]  within the organization i see more and more go do you think that's that's sort of the the the language
+[1208.14 --> 1213.80]  the the bulk of the of the engineering teams at heroku are going to be writing software for the
+[1213.80 --> 1223.90]  foreseeable future um in your mind i'd certainly like that um i i suspect um a lot of rubyists might
+[1223.90 --> 1230.20]  uh be more immediately attracted to something like elixir that's a theory not something i know
+[1230.20 --> 1238.36]  uh but i think elixir can give uh like the rubyist some of the advantages that you would get from go
+[1238.36 --> 1243.88]  while keeping some of the things that they like about ruby and those are the same a lot of those
+[1243.88 --> 1250.12]  things are the same things that kind of drive me crazy about ruby um and you know ruby and go are
+[1250.12 --> 1260.14]  different um with that said uh i do think like go is embedded into our infrastructure um to
+[1260.14 --> 1271.10]  the extent that it will likely continue to be uh one of one to four primarily primary languages that
+[1271.10 --> 1278.88]  haroku engineering ends up using um especially while things like docker and kubernetes uh and stuff like
+[1278.88 --> 1287.18]  that continue to um kind of like dominate mindshare uh in the industry so rishab what what are you most
+[1287.18 --> 1293.22]  looking forward to um on your end with regards to working with go are you do any project you get
+[1293.22 --> 1299.06]  to work on or that that involves go you're happy with or do you have a specific kind of project that
+[1299.06 --> 1303.98]  you find the most fun to write in go like where where where your head where's your head at so
+[1303.98 --> 1309.42]  currently i have the pleasure of actually writing a new service that actually deploys to our kubernetes
+[1309.42 --> 1317.70]  clusters in a secure and scalable fashion and that service is where didn't go so it's been kind of
+[1317.70 --> 1322.00]  like a journey for myself in the sense that it's like a foundational project where i get to actually
+[1322.00 --> 1329.60]  like bootstrap a new service and go and like really figure out for myself um good design patterns and
+[1329.60 --> 1337.34]  writing good go code so it gives me the space and the time to actually experiment and learn
+[1337.34 --> 1344.92]  um which i'm pretty happy about well i am glad we're able to to have this chat and and i personally
+[1344.92 --> 1351.74]  learned a bit more about how provoked users go and and knowing that there are folks who are sort of a
+[1351.74 --> 1356.90]  brand new to go in the organization as well um that are sort of uh learning how to how to write go
+[1356.90 --> 1363.70]  um and how to write idiomatic go and and folks like uh rishab and also folks like ed who have been
+[1363.70 --> 1369.50]  doing it for a while i'm providing some guidance and and you know me having joined the organization
+[1369.50 --> 1374.82]  and knowing knowing that go was a was one of the determining factors for me joining the organization
+[1374.82 --> 1380.00]  i'm glad to see that there's there's there's a lot more um to the org a lot more to the services and
+[1380.00 --> 1385.98]  everything else that we're doing that that's written in go and that makes me happy so i am uh yeah i'm
+[1385.98 --> 1390.88]  grateful that we're able to sort of get on the chat have this chat uh and thank you for being i guess
+[1390.88 --> 1394.52]  i'm sure today yeah yeah thank you thanks for hosting thank you johnny for having me
+[1394.52 --> 1402.14]  thanks for joining us for this episode of the codish podcast codish is produced by heroku the
+[1402.14 --> 1406.74]  easiest way to deploy manage and scale your applications in the cloud if you'd like to
+[1406.74 --> 1412.78]  learn more about codish or any of heroku's podcasts please visit heroku.com slash podcasts
+[1412.78 --> 1413.28]  you

Go in other spoken languages_transcript.txt

@@ -0,0 +1,815 @@
+**Mat Ryer:** Hello there, and welcome to Go Time. I only read what's on the \[unintelligible 00:02:49.05\] I'm Mat Ryer, and today we're discussing talking about Go in a not-English language - which is, if you think about it, most languages.
+
+We're joined today by Natalie Pistunovich. Hello, Natalie.
+
+**Natalie Pistunovich:** Hello, Mat Ryer.
+
+**Mat Ryer:** Welcome!
+
+**Natalie Pistunovich:** Thank you. It's great to be on the panelist thing.
+
+**Mat Ryer:** Yeah, good. Welcome to the thing.
+
+**Natalie Pistunovich:** \[laughs\]
+
+**Mat Ryer:** We're also joined by Kris Brandow. Hello, Kris.
+
+**Kris Brandow:** Hello, Mat.
+
+**Mat Ryer:** Welcome back. I got your name right the first time.
+
+**Kris Brandow:** Yeah, totally. Definitely.
+
+**Mat Ryer:** After the edit... And we have a very special guest too, it's only Ellen Körbes. Hello, Ellen.
+
+**L Körbes:** Hi, Mat Ryer. How are you doing? \[laughter\]
+
+**Mat Ryer:** Welcome to the show!
+
+**L Körbes:** Thank you.
+
+**Mat Ryer:** Yeah. If I get trolled, I don't take it well. I can give it, but I can't take it. So just bear that in mind, please.
+
+**L Körbes:** You're gonna get used to it as the show goes on, so don't worry.
+
+**Mat Ryer:** Okay. This is gotta be a tough one for me, isn't it? I've got a feeling...
+
+**Natalie Pistunovich:** This is very casual, especially with a drink in the hand.
+
+**Mat Ryer:** Yes. Okay, so maybe we could just get started by a bit of intro for each of us... And specifically, I'm interested in how you got into computers, and specifically Go. Maybe we could start there. Ellen, maybe you could kick us off.
+
+**L Körbes:** \[04:09\] Yeah, so my story actually is kind of boring...
+
+**Mat Ryer:** Okay. We'll go to someone else then. We'll go to someone else if it's boring, yeah. Natalie. How did you get into computers?
+
+**Natalie Pistunovich:** I clicked on the link and then I saw my picture. It's pretty cool. Now I'm in the computer.
+
+**Mat Ryer:** You're IN the computer? \[They're IN the computer?\]
+
+**Natalie Pistunovich:** That's what I see. It's crazy, I'm telling you. This is the future.
+
+**Mat Ryer:** You look like you're in my computer at the moment.
+
+**Natalie Pistunovich:** \[laughs\]
+
+**L Körbes:** Are you in the computer, or like just in the monitor?
+
+**Mat Ryer:** I've got an iMac, so it's the same thing.
+
+**Natalie Pistunovich:** It's the same thing. I don't have a monitor.
+
+**L Körbes:** Oh, wow...
+
+**Natalie Pistunovich:** So I'm in the computer.
+
+**Mat Ryer:** Kris, are you gonna give a sensible answer if I ask why you're into computers? \[laughter\]
+
+**Kris Brandow:** I mean, you already asked me that on the -- not last one, but the Go Time before that I was on.
+
+**Mat Ryer:** Okay, forget it. \[laughter\]
+
+**Natalie Pistunovich:** Mat, how did you get into computers?
+
+**Mat Ryer:** Actually, when I was little, we got a computer and we used to type in things from computer magazines, with my dad, and make games, and things like that. So that was really how. And I kind of fell in love with the fact that you can make computers do things. Once I had that going, I got really kind of hooked on that process. That's how you do a proper answer, okay? So, proper answers from now on, everyone. Serious faces, okay? \[laughter\]
+
+**Natalie Pistunovich:** Mat, we would like one more demonstration.
+
+**Mat Ryer:** Of what?
+
+**Natalie Pistunovich:** What got you not into just computers, but what got you to be interested in Go?
+
+**Mat Ryer:** Well, yes, I wanted to build a project on Google App Engine, which is kind of one of my first serverless experiences really, where you write the code and you upload the code, and then it does the rest for you, kind of magically. It feels magic. And the choices that you could use were either Java, so no, Python, so no, and then it had this little experimental one that said Go. And it was that little experimental badge that I sort of like a magpie couldn't resist.
+
+And then I loved the simplicity of it, and the minimalist design, and I was able to learn it actually quite quickly... Which is nice, because the point of programming is so that you can actually do things. We have some great teachers on this panel, so we'll definitely talk more about learning and teaching as well.
+
+If you can actually do something with it, then that's really the important thing... So that was it. And I really then stayed with it and just have been working with it ever since. I loved it. What about you then, Natalie?
+
+**Natalie Pistunovich:** When I moved to Berlin and had my first proper job as a developer after uni, after university... In Germany, you say uni, because we're efficient like that... \[laughs\]
+
+**Mat Ryer:** Yeah. In English.
+
+**Natalie Pistunovich:** \[laughs\] The language of the company was Go, and basically it was a "Welcome onboard! We're using Go, so good luck!" That's how I started using Go.
+
+**Mat Ryer:** Wow, that must have been quite early then.
+
+**Natalie Pistunovich:** That was 2014, I wanna say. End of 2014.
+
+**Mat Ryer:** Yeah, what I meant was they must have picked up Go to use it quite early, that company.
+
+**Natalie Pistunovich:** Yeah, yeah, it was pretty early. It was a young company at the time, but it was already using it for one or two years since they started; I forget when it was. They've been early adopters to use Go in production; they took a bet, and they took a good bet...
+
+It's interesting to see how -- I joined when it was already pretty good at scale, and was still scaling more, and Go was just handling this very nicely.
+
+**L Körbes:** I mean, if it's a German company, when you say that they started using Go very early, like that it happened very early, it was probably very early in the morning when they told that, Natalie, to you... The Germans...
+
+**Mat Ryer:** Yeah. Did they get up early?
+
+**Natalie Pistunovich:** \[08:12\] \[laughs\] If you work not in a startup, for sure. If you work in a startup, I think a stand up at 11 o'clock is normal.
+
+**Mat Ryer:** What time do you get up, Kris?
+
+**Kris Brandow:** It depends on the day. This morning I got up at 8. It was nice. Nice and peaceful.
+
+**Mat Ryer:** Oh. What did you have for breakfast.
+
+**Kris Brandow:** I really don't eat breakfast, so I didn't really have any breakfast.
+
+**Mat Ryer:** Okay.
+
+**Kris Brandow:** I should, but I don't.
+
+**Natalie Pistunovich:** Are you calling this intermediate fasting?
+
+**Kris Brandow:** No, I just forget to eat. That's a fun quirk about me, I can go a whole day, it'll be like 10 PM, and I'll be like "Did I eat today?" And then I'll remember "I haven't since like yesterday at 2 PM", and I'm like "I should probably eat", and then I just go to bed instead...
+
+**Mat Ryer:** Wow... Is that just because you get so engrossed in the work, or in what you're doing?
+
+**Kris Brandow:** Sometimes yeah. And sometimes "Oh, I'll make food", but then I'll be like "Oh, I have to do this one thing", and then I'll do that, and then forget that I was gonna make food, and then keep doing that 4-5 more times...
+
+**Mat Ryer:** \[laughs\] So I'm interested -- by the way, I'm always very impressed with people that speak multiple languages... And it's always interesting to me, because I speak U.K. English. We just call it English. But I still code in U.S. English. If I write the word "color", I will omit the u. We have a u in our spelling of the word color. But in code, I tend to use U.S. English, because it's sort of like the standard language. But there aren't many examples of that, so it's a very easy thing to do... But that must be different if English isn't your first language, because you have to learn -- I mean, what do the Go keywords look like to somebody where English isn't their first language?
+
+**L Körbes:** I think Go is cow in Sanskrit.
+
+**Mat Ryer:** Is it? So it means a cow...
+
+**L Körbes:** I know in yoga there is a thing called Gomukha, or something like that, that means cow face. I'm not sure which part is the cow and which part is the face, but... I mean, Go is either cow or face, in Sanskrit.
+
+**Mat Ryer:** Yeah. Well, that's interesting. So what about func? What's func? That's my question. I'll stand by it.
+
+**L Körbes:** Who are you asking?
+
+**Mat Ryer:** Anyone that can deal with that question...
+
+**Kris Brandow:** I don't think func is even an English word... It's an abbreviation.
+
+**Mat Ryer:** Right... So that's what I wonder - is it just that everyone has to just learn the keywords and that's it?
+
+**L Körbes:** So Aditya has a -- you all know Aditya, right? Aditya Mukerjee, I hope I'm pronouncing his name right... So he has a talk that I've seen a couple times, where he talks about translating the keywords of the language. I'm not sure if it's a framework/tool; there's something that does that, and then you can write Go in your first language, and it basically compiles that code into regular Go. I don't recall though what were his conclusions about how effective that is; is that a very big struggle?
+
+What I find when teaching people is that sometimes they don't know what something does when they first see a word... For example while. We don't have while in Go, but just as an example. What does while mean? I have no idea, I don't know. But once you explain the concept to the person, they pick it up pretty quick.
+
+I think the real problem with that is really just the beginning, where - okay, let's say it's a more complicated language and there's 50 keywords here; that's a lot of stuff to learn. If you speak English, you just look at them and it's like "Oh yeah, this is this." if you don't, then it's gonna take a bit longer. But I don't think it's like a huge impediment, especially in Go, when we don't have too many keywords.
+
+**Mat Ryer:** Yeah, I feel like Go helps there because of its minimalism. And then what about comments? I once worked on a codebase that was developed originally in France, and so all the comments were in French. Now, the code - I could read; I was basically using the code to figure out what the comments meant. So what do you think? What language would you write comments in?
+
+**L Körbes:** \[12:21\] Ukrainian. \[laughter\]
+
+**Mat Ryer:** Yeah. It makes sense.
+
+**L Körbes:** It's really pretty.
+
+**Kris Brandow:** I feel like that's one of the interesting thing about Go - since it's UTF8, you could really write the comments in whatever language; you're not gonna run into some weird character encoding problems. That seems like a nice benefit there. You could just write comments in your first language, and it's probably gonna be fine.
+
+**Mat Ryer:** Yeah, that's true. That's quite nice. But should you? Because it depends, I guess. If it's an open source project... I mean, U.S. English I think is the standard for computer science. I've read that somewhere once. Probably in America. So if that's the case, then we probably should even write comments in U.S. English. The French comments certainly didn't help me very much, unfortunately.
+
+**L Körbes:** Yeah, I think that gets complicated, because not everyone speaks English. Yeah, if you're from a rich country, you've probably learned some English at some point... But back in Brazil - no, that's simply not the case. So if you have a bunch of comments written in English and your team is mostly Brazilians, a good chunk of the people are just not gonna be able to understand it then. I mean, there's always Google Translate and stuff like that that can help... But I think what that causes is it creates silos, whereas like -- "Okay, this was written by the Brazilian team, so it's all in Portuguese." If you speak English you are not gonna be able to read it. And if we take code that is commented in English, are we gonna be able to understand this? We can't count on all the team being able to understand. That creates some barriers that -- again, I'm not sure in practice just how bad this can get, but it's an issue. It's not very straightforward.
+
+**Natalie Pistunovich:** And it's not just the comments, it's also function names. They also describe what you do, and you need to know the verbs, you need to know the words, you need to know the context of what it is; if you are in a domain-specific codebase, then you need to know all the terminology of that codebase. You say "Let's process this transaction." What is that transaction? What is the context of it?
+
+**Mat Ryer:** Yeah. I suppose that is a tricky problem really to solve; even when everyone speaks the same language, it's hard to name functions and things like this...
+
+**Natalie Pistunovich:** And variables.
+
+**Mat Ryer:** And variable names, yeah. So L, you did the Aprenda Go, right?
+
+**L Körbes:** \[laughs\] Yes, that's exactly how it should be pronounced...
+
+**Mat Ryer:** Should be, but isn't. How is it pronounced?
+
+**L Körbes:** Aprenda Go. The r is a lot softer.
+
+**Mat Ryer:** Literally what I said.
+
+**L Körbes:** Yeah, I know. I'm sorry.
+
+**Mat Ryer:** What's Aprenda Go? Tell us about that. What is that?
+
+**L Körbes:** Okay, so Aprenda Go is a --
+
+**Mat Ryer:** You said it wrong.
+
+**L Körbes:** What?
+
+**Mat Ryer:** You said it wrong... Carry on, it's fine. \[laughter\]
+
+**L Körbes:** Oh, I'm sorry. Look, I'm doing my best, man, okay?
+
+**Mat Ryer:** Yeah, yeah. You're doing great.
+
+**L Körbes:** So when I was first learning Go, I was learning from Todd McLeod course. I'm sure some of you know him... And his course was really good. Well, it was not that good actually, because every five minutes I would email him, "Hey, Todd, here's a better solution for that exercise", or "There was an error in that thing", or something-something.
+
+**Mat Ryer:** A nightmare.
+
+**L Körbes:** And I was just trying to help.
+
+**Mat Ryer:** It sounds like a nightmare. \[laughs\]
+
+**L Körbes:** Yeah, I know, I'm sorry. I mean, he's a teacher; he's probably into this stuff. I don't know.
+
+**Mat Ryer:** Yeah.
+
+**L Körbes:** So I was in touch with him a lot. And then at some point somehow in the conversation he said that if I wanted to record that course in Portuguese, that that would be really cool, because he wants the knowledge to spread, basically, and a lot of people just don't speak English. So that's what I did. So my first thing I did with Go, when I barely knew what Go was about, was to record a course, teaching everything about Go. So you know...
+
+\[16:13\] So I did that, and I put that course first on a closed platform, something like Udemy, and a couple years later I was like "Yeah, the money is not the matter here. I don't want people to have to pay for it." And the conversion from Brazilian currency to Euros - I'm not making any real money anyway, so I might as well open it on YouTube and make it free for everyone.
+
+So I called the YouTube channel Aprenda Go - it means "Learn Go." That's the exact translation. And then I put the whole course there. It goes from way in the beginning, like what's a variable, what are functions, what's a for loop, all of that, all the way down to "Okay, here's a fan-in and a fan-out pattern for concurrency", and all that stuff.
+
+So it doesn't get to web development - that's a bit more advanced - but all the way from the very first Hello World that you see, to the end of the primitives of the language, I'm covering all of that.
+
+Yeah, it got pretty popular, and I'm very happy to have done that.
+
+**Mat Ryer:** So how long did it take you to do? Because it's a big course, isn't it?
+
+**L Körbes:** Oh yeah, it's almost 200 videos, or something.
+
+**Natalie Pistunovich:** Wow.
+
+**L Körbes:** I mean, I had just quit my job, and I wanted to get a career in software... So I quit my job and I was doing that full-time. I think it took like a couple months.
+
+**Mat Ryer:** Wow.
+
+**L Körbes:** The hard part really was editing. I think I released only the first half of the course at first, and then it took me like a year to add the second half, or maybe more even... Because I had no experience with video stuff at all. So I would do a bunch of bad takes, and then have to redo it, and then in editing you end up with 500 files, how do you put all of this together... So that was a ton of work, but now that's done.
+
+**Mat Ryer:** Yeah. So what about explaining the concepts then in a different language? Was it just kind of a trivial thing? Is it just you have to understand it and then you can just explain it? Or are there concepts that are difficult, for whatever reason?
+
+**L Körbes:** So there are some metaphors that you can't use in a different language, because they just make no sense. For example, the first thing that just popped in my mind - there's a metaphor in Portuguese that was like "Water's soft, rock is hard, knock-knock and it breaks", something like that. Like, yeah, sure, if you think about it, you can understand, but I wouldn't use this idiom to an English-speaking audience... And likewise, there's a bunch of English idioms that just do not translate to Portuguese.
+
+Sometimes you learn something with a metaphor like that, but you can't use the same metaphor, so you need to find different things. Most of the time, that's easy, but some concepts are really hard, no matter the language.
+
+Teaching interfaces, polymorphism, pointer methods - that stuff was hard. And not necessarily because of the language, but just because it's the kind of stuff that once you understand, you understand, but until you do, it makes zero sense... And I find it hard -- so I climb the stairs and make easily-digestible steps; it's the kind of thing you either understand the whole thing, or it makes no sense.
+
+I think interfaces are probably the most difficult thing to explain, especially because they don't map to other languages; in Python or Java, what interfaces do is kind of different, so you can't really catch that other knowledge and use it. You have to kind of do it from scratch. Yeah, that gets a bit tricky.
+
+**Mat Ryer:** What about words like polymorphism, which you mentioned then? Do you have that word?
+
+**L Körbes:** \[20:05\] That's fun... So if you go to university in Brazil, you learn all of those concepts with Portuguese names. I didn't go to university, I learned all this stuff on the internet. So half of my videos are like "Oh yeah, what's this called in Portuguese again? I don't know, let me just make up a name", and then I make up a name and I go the rest of the video with some made-up name... And in the comments, everyone's like "Oh yeah, this is called such-and-such, you idiot", and I'm like "Oh. Right."
+
+**Mat Ryer:** "Oh. I thought just guessing would be fine."
+
+**Natalie Pistunovich:** But then what makes more sense, usually? Would you say that your translation makes more sense, because it's more modern maybe?
+
+**L Körbes:** Sometimes. I'm not gonna claim much credit here; I'm not that creative with naming. Often I would just see the correct name, and I'd be like "Okay, this is a much better way to convey the whole concept than this mess that I just did." So that's how it is.
+
+**Natalie Pistunovich:** Would you say that you learned the other programming languages that you know also by teaching?
+
+**L Körbes:** No, not really... Well, I learned C by using it, Bash (and that kind of stuff) by using it... Python - I just google whatever I need, whenever I need... So not really. And JavaScript was also by using... So yeah.
+
+**Natalie Pistunovich:** It's interesting... If you can compare a little bit learning a language by teaching it right away, and learning a language by copy-pasting it, and learning a language by using it. It's actually three different ways of learning it. Which one do you like, which one don't you like, what works best for you?
+
+**L Körbes:** That depends on your goal. If you're doing a thing that's gonna be just a few days, and it's in a new language and you don't really care for the language, you just wanna get your hacks to work, then you just Stack Overflow copy-paste. That is the best option, because it's the fastest.
+
+**Mat Ryer:** We'll just remind people, don't rush it. Don't copy from the top, because that's the question, and that code definitely doesn't work. You have to scroll down a little bit. It's worth remembering, because I've done that, and I still do.
+
+**L Körbes:** Yeah. Look for the checkmark. That usually tells you which code works.
+
+**Mat Ryer:** Yeah, good tip.
+
+**L Körbes:** But if you wanna really master something, I think the best way is to teach it; even if you don't actually teach it, learn it as if you're gonna teach it. So there's a bunch of things that -- I have tons of ideas of stuff to do on YouTube, and I often don't take the time to actually record the video for YouTube etc. But as I'm learning, I'm taking notes as if I need to explain this to someone else. It's like "Okay, I got this, but how do I convey this concept?" And then I write down, "Okay, I should say X metaphor, and this etc." And I create a little script of how to teach a thing. After I do that, I'm never gonna forget that ever, and it's gonna be crystal clear to me, because I force myself to make it crystal clear. I might be wrong, but I haven't found a way to learn something that crystallizes knowledge that sharply.
+
+**Mat Ryer:** That's a good tip.
+
+**Kris Brandow:** I feel like that's similar if you wanna teach through writing as well, going through the process of writing a blog post, or an article, or like a book; it will really help solidify a lot of the ideas.
+
+I think a lot of it too has to do with how you have to rearrange the knowledge in your mind to convey it to someone else; it forces you to learn it at a deeper level. It's like a little bit of a hack, because -- when you just learn something, you kind of piece it together from a whole bunch of different places. You might have gaps here or there, but then when you have to actually convey it to another person, it has to make sense; it has to be a story that you're telling them. And rearranging that knowledge helps you find your gaps, helps you learn more... I've read a bunch about that. I've always found that to be a really interesting process, of like teaching something to somebody in order for you to learn it better.
+
+**L Körbes:** I think there's some elements of different parts of the brain being used, something like that. You can read about riding bicycles, but if you just ride the bicycle, you're gonna learn so much more... I'm not sure how that maps into languages and stuff, but I think there's something to it where if you actually pretend to teach it, it probably activates different stuff in your brain, and that makes you learn more. I don't know, but I feel like there's something to it.
+
+**Mat Ryer:** \[24:14\] Yeah, I definitely learn by doing something. I find that to be the quickest way for me... And if people want to -- if they're doing that trick of like "I'm gonna imagine I'm gonna teach this", a nice way to do that (a low-commitment way) is to give a talk at a meetup, or something; or even at a conference. Because that is also -- I mean, it is the same thing; it's a great way to solidify your knowledge and get that across to an audience. I find that to be quite nice, and quite low-commitment, if it's just a local meetup and you're gonna do ten minutes and just explain a thing... I think that's quite a nice little tip. I've seen some people do that, and they did it brilliantly, so it really worked. So I always encourage that.
+
+**Kris Brandow:** Yeah, I feel like that's one of the things I always think of when I see people giving talks. It's like "Oh my God, they just know so much. They know everything about what they're talking about." Having given several talks myself, it's like, no, you just have ideas that you've thought about a lot, and you get up and you give a presentation... But you're making all sorts of mistakes that the audience can't pick up on, so it's not like you're perfect, or whatever.
+
+So yeah, I definitely think encouraging people to give, especially meetup talks, is a really good idea for like if you wanna learn something better; forcing yourself to put down all of that information into a 10-minute or 20-minute talk.
+
+**Mat Ryer:** Yeah.
+
+**Natalie Pistunovich:** Or even a 5-minute talk. Sometimes, as a meetup organizer, I find it really hard to convince people to submit the talk, because they feel this is a huge commitment, and this is way above something that they can do, even though they've been developing/using a concept or a language for years... Convincing them to give a 20-minute talk, a meetup talk is a lot. A 5-minute talk is also great. This is something that I think everybody should talk about more, as meetup organizers, as attendees, and as speakers.
+
+**Break:** \[26:08\]
+
+**L Körbes:** Hey Natalie, I have a question for you. Since you have a lot of experience organizing meetups and events, is there a shortcut to the fact that people are often comfortable with the subject matter of their talk, but they're just terrified of being on stage? Is there some hack to that?
+
+**Natalie Pistunovich:** Some. I'm sure there's a bunch of hacks that I cannot think of, and I wouldn't mention, and actually I would love to hear if anybody is listening and wants to share, whether it's on Twitter, whether it's on Slack; please share more tips. My biggest thing - the thing that I do that seems to help the most is to say "No questions." And then this is so much less intimidating.
+
+And then sometimes people would say "No, I'm actually fine with questions", and then this is okay. But when you ask somebody new to give a talk, I think one of the bigger fears is like "They're gonna know that I have no idea what am I doing."
+
+So if you create this safer space by saying you only talk and do exactly what you have rehearsed, and nobody will ask you about anything else, that's completely fine. And this helps people relax a little bit.
+
+**Kris Brandow:** \[28:10\] I just think Q&A sessions are usually a pretty terrible idea overall, because it's like, someone has rehearsed and prepared this material, and now you're just like throwing whatever question out of left field at them, and it's like "Oh! I haven't thought about this at all, so I don't have a good, coherent response to this, so I'm just gonna babble for two minutes until I get the next question." So I think that not having questions is good in a multitude of ways.
+
+**Mat Ryer:** Yeah. And it's nice to think of that, remember that from the other side as well, if you are asking a question. Sometimes -- and I think it just happens by accident, if I'm honest... But sometimes the question is either loaded, it encodes somehow something that either wasn't covered, or is a gap, they've noticed a gap in the speaker's knowledge... Which I think is quite natural in a conversation, that's sort of normal, but that can be quite detrimental.
+
+We have to be careful and bear that in mind when we're asking questions as well. In fact, the U.K. GopherCon (GopherCon U.K. it's called; I call it the U.K. GopherCon), they don't take questions. It's like, no questions in the whole conference; it's just a thing. And if you want to have a conversation, you sort of use the networking time to go and find people to do that. And it's quite nice, because it does encourage people.
+
+I also see a lot of people giving talks where English isn't their first language as well... And this impresses me tremendously, because I can't even really get my head around what that would be like for people. But are there any other tips for people in that situation, where you're giving not only a complicated technical presentation, but you're doing it in a different language?
+
+**L Körbes:** As a violinist, what I'm gonna say is rehearse, rehearse, rehearse... Play the talk. Do a pretend presentation at home. Don't just look at the slides and think about what you're gonna say, because performance is all about muscle memory. So say the words out loud. Go over the talk many times, saying the words out loud. Maybe if you have the stomach for it - and some people don't - record it and then look back and see "Oh, this part I got my ideas all jumbled together. How do I explain this better?" And just rehearse. Because at the end of the day, you can give a talk from a natural standpoint if you're comfortable with that language. If you're not, then it's gonna be a performance, and then you should rehearse for it like any performance.
+
+Just do the thing a few times until you're physically comfortable doing it. After a few times, your body is gonna have developed the muscle memory. It's just like playing an instrument. And then you're gonna be able to do it on command. It's gonna become second-nature. And you just do that for your first three talks, for example, in English let's say, and then the fourth one you're gonna be like "Yeah, I've kind of got my way around this." But the first few times you need to rehearse.
+
+**Mat Ryer:** Yeah. You can't just turn up to the orchestra with your violin and like \[31:19\] I can't play. I should have learned this. Sorry, everyone."
+
+**L Körbes:** That happens way more often than you'd think. I've played in an orchestra... \[laughs\]
+
+**Mat Ryer:** It sounds like Natalie's cat. Sorry.
+
+**Kris Brandow:** \[laughs\] I feel like that advice is applicable even to people where English is their first language... Especially the first couple of times you give a talk. I think people don't -- it takes a while to acclimate to how much you can fit into a talk and how you should structure it.
+
+\[31:57\] I remember when I first started speaking, the company I was working for had this speaking coach that would help everybody with their talks... And one of the things that he said to me that's stuck with me is that in a 45-minute talk you have enough time to really cover three things. So pick the three things that you're gonna cover, and form the whole talk around those things.
+
+I think what usually winds up happening with people in their first talk is they just try and shove too much stuff into it, and because you know it so well, you're just like "Well, it shouldn't take that long to discuss this topic, so I'm just gonna fly through it." And it's like, no, you have to convey all of this information to another person, so you have to actually build up a lot of stuff around it.
+
+I definitely feel with my GopherCon talks this year -- I think I would have given it differently if it was on stage. I think I packed a little bit too much stuff into it. But I remember just watching it and being like "Oh my God, I'm talking so slow. People are gonna get so bored." And there were all of these messages flying in Discord being like "Oh my God, I have to watch this again, because I think I missed some things."
+
+So I think that's a really important thing to remember - your audience is hearing this for the first time, and processing it as you're talking... So you who has practiced and knows this subject well, you're going to understand things much quicker than your audience is... So make sure to give space.
+
+**Mat Ryer:** Yeah. That's great.
+
+**L Körbes:** I've made that mistake a hundred times. Most of my most popular talks were about "Oh, the state of Kubernetes development tooling, where I talk about 20 different tools." And then you have to explain in nuance, like okay, why is this different from the other one etc. By the end of the talk, everyone's exhausted. People like the content because it's useful, but it's very exhausting.
+
+I've also given a talk that was like "How to write a neural network from scratch", and I could have made it a lightning talk, where I just go on stage, don't say anything, and just type out the code. That would have been fine. But explaining all the concepts and why they work - it's too many things. I think I've presented that talk five times, and only on the last time I felt that most people in the audience actually got it... Because you're just bombarding people with information, and... It's like you said, Kris - we know the subject, we have rehearsed this; for us it's perfectly clear. But when you need to make someone understand it, that takes so much energy.
+
+I think everyone has tried to explain more things than they should in a talk. I certainly have.
+
+**Mat Ryer:** I understand how it happens, because you want to give such value, really. And you also want it to be complete, ideally, because then it stands alone as a thing that is very comprehensive. But I think it's great advice, and it's kind of like almost -- kind of less is more sort of philosophy as well. So I've seen the same, and I've also done it for sure myself.
+
+Sometimes when I -- and I don't know if you have any tips for this... Sometimes when I give a talk, I'll do a joke, and then I think it's out of respect, the audience is just silent. \[crickets\] How do you fix that? Any tips?
+
+**L Körbes:** Tell better jokes...
+
+**Mat Ryer:** Right.
+
+**Kris Brandow:** I kind of feel like you have to, at the start of your talk or in the beginning, make it known that you're going to be expecting that type of interaction... Because I think in talks, people are really just trying to concentrate on what you're saying, and absorb all the information/knowledge and maybe take notes, so they're not really sure if what you said or anything you're saying is meant to be funny... So like "Oh, we don't wanna laugh at something and then potentially insult the person." But if you start right off the bat, and you're like -- you start off telling a joke from the beginning, then it kind of adjusts the audience to be like "Oh, this dude's funny. Now we can all feel like we can laugh."
+
+**Mat Ryer:** That's really interesting, and I've definitely experienced that. And Kris, you're a writer, so that kind of fits naturally for you... But you're dead right; you find at conferences -- if someone says a joke at a conference, the sorts of jokes that get the best laughs are not really that funny. They usually are just sort of safe, and everyone knows it's humor.
+
+\[36:23\] They'll say "Oh, don't worry about this code. I probably wrote it on a Monday." That will get a big laugh, because everyone's like "Yeah, I remember Mondays! They're rubbish! You write bad code on Mondays, because you're tired, or hung-over still." But yeah, if you say other things that people aren't expecting, it's exactly that experience; people don't know what to do. That is the story of my life.
+
+**L Körbes:** I think the secrets to telling jokes is they have to be really bad jokes...
+
+**Mat Ryer:** I thought I had that nailed.
+
+**L Körbes:** I mean, your jokes are medium, that's the problem. You can't have medium jokes. The thing is, if you do a really smart joke, half the people are not gonna get it, and some of the people will. So if you're gonna do a really smart joke, then just keep the same voice tone. Don't do the whole jokey thing. Just say the things, and people who get it are gonna laugh, people who don't get it - they're not gonna say anything.
+
+If you want everyone to be engaged with the joke, it needs to be a really bad, a really obvious joke, and you need to be kind of a clown about it...
+
+**Mat Ryer:** Yeah, that's true.
+
+**L Körbes:** ...or else, people are just gonna be confused.
+
+**Mat Ryer:** It's inclusive, isn't it?
+
+**L Körbes:** Exactly.
+
+**Mat Ryer:** Bad jokes are kind of more inclusive.
+
+**L Körbes:** Exactly.
+
+**Mat Ryer:** We have these festive Christmas crackers in the U.K. I don't know if you have them elsewhere... And they're basically like -- they're weird, if you've never seen them... But what we do is we pull them at Christmas and they make a little bang, and they just rip in half. And they've got little toys and things and all sorts in there... And someone will win when you break it in half. And they always come with a little joke, and they're always terrible jokes. And there's a theory that it's deliberately so, and you can almost unite together against this terrible joke.
+
+**Kris Brandow:** I'm trying to imagine what it is you're describing. It sounds like a mix between a piñata and a fortune cookie...
+
+**Mat Ryer:** Yeah, it kind of is, yeah.
+
+**Natalie Pistunovich:** I was thinking of piñatas as well.
+
+**Kris Brandow:** At first I thought you meant like crackers, and I was like "Oh, he's British. He probably means something else."
+
+**Mat Ryer:** Yeah. We do have crackers though.
+
+**L Körbes:** I thought of Kinder eggs.
+
+**Mat Ryer:** Yeah, Kinder eggs... Which are illegal in America, I think. But guns are fine. So back to the different languages, and things... No, but Kinder eggs are -- you know, it's meant to be a Kinder surprise; every time it's a little plastic toy. Do you know what I mean?
+
+One time I just wanna open it and be like "Huh!" And have an actual surprise. I thought you said bad jokes were \[unintelligible 00:39:02.02\]
+
+**L Körbes:** See, this is not bad enough. This was a medium joke.
+
+**Mat Ryer:** Oh, that is a bad insult, that your jokes are medium.
+
+**L Körbes:** The Monday one was perfect.
+
+**Mat Ryer:** Yeah, the Monday one, exactly. I'm not gonna tell that; I'm not gonna say that Monday joke. It's office humor.
+
+**Natalie Pistunovich:** And it's not inclusive. Not everybody starts their week on Monday.
+
+**Mat Ryer:** Good point, yeah. Some people -- well, when do they start it?
+
+**Natalie Pistunovich:** Sunday.
+
+**Mat Ryer:** So they're like "I'm already warmed up by Monday."
+
+**Natalie Pistunovich:** Yeah, Monday is like a Tuesday.
+
+**Mat Ryer:** Yeah.
+
+**Kris Brandow:** Some people have weekends that are Friday-Saturday.
+
+**L Körbes:** Wow...
+
+**Mat Ryer:** Hm. Okay. No, it's fair; it's all good stuff, good learning for me. I'm gonna go back and watch my talks now and I'm gonna have a lot more clarity on what's happening, and it's gonna change things for me.
+
+**Natalie Pistunovich:** When you do those rehearsals, you can also practice on the jokes.
+
+**Mat Ryer:** Yeah.
+
+**Natalie Pistunovich:** I think once you heard something several times, you can start building jokes around it in your mind.
+
+**Mat Ryer:** Yeah.
+
+**Kris Brandow:** Comedy is like a whole other discipline, too. So you shouldn't expect to be a really great comedian and a good software engineer and a good public speaker... Unless you train and practice at them a lot.
+
+**Mat Ryer:** I do. Kris -- no, obviously not... I would say good things, if that happened... Kris, you were talking about your first talk, and stuff... It's interesting... Natalie, do you remember the first talk you gave, and what language it was in?
+
+**Natalie Pistunovich:** Yes, the first conference talk I ever gave was at GopherCon U.K.
+
+**Mat Ryer:** When was that?
+
+**Natalie Pistunovich:** Golang U.K. it was called at the time.
+
+**Mat Ryer:** Yeah... Was it? I saw that talk.
+
+**Natalie Pistunovich:** I guess you did, yeah.
+
+**Mat Ryer:** It was in the U.K, wasn't it?
+
+**Natalie Pistunovich:** It was in the U.K. \[laughter\]
+
+**Mat Ryer:** That's right, I remember it. How did it go, from your point of view?
+
+**Natalie Pistunovich:** A crazy rush of adrenaline. That's all I remember from it. But obviously, I liked it, so I kept doing it... In a good way.
+
+**Mat Ryer:** Yeah. It was very good, and it is still available on the internet, for anyone...
+
+**Natalie Pistunovich:** It's not very good, it's \[unintelligible 00:41:07.23\] let's be honest.
+
+**Mat Ryer:** I thought it was good. L, do you remember your first talk?
+
+**L Körbes:** Yeah, I went to GopherCon Brazil at some point, and there I met someone who was running the Python meetups in my city... And I don't remember how the conversation went, but she was like "Hey, do you wanna give a talk at my meetup?" and I was like "What?" And she was like "Oh yeah, you sound fun. Do you wanna just give a talk about whatever?" I was like "Oh yeah, okay. I can do that."
+
+So I gave a talk about getting started in tech, because that's what I was doing, so I just compiled all the best tips and tricks that I had come across at that time. I told a little bit about my story, and I was about to start on my first internship, writing code on Kube Control with Kubernetes. I think that was kind of impressive, and that's why people liked it so much.
+
+**Mat Ryer:** I thought it was just medium, but...
+
+**L Körbes:** \[laughs\]
+
+**Natalie Pistunovich:** No, Medium is where she published the content. \[laughter\] See, that was a bad joke. Everybody laughs.
+
+**Mat Ryer:** No, that's definitely a medium joke. That's literally a medium joke. \[laughter\] Brilliant.
+
+**L Körbes:** Yeah. No, I presented that talk, it was very well-received. After that, I was thinking "Hm. I wish there was a job I could do where I just do talks all the time." And I think it was six months later I got a job to do that all the time, so that was fun.
+
+**Mat Ryer:** Nice! Wow. That's quick. Have you ever done a talk in not-English?
+
+**L Körbes:** Yes, the first few talks I gave were in Portuguese.
+
+**Mat Ryer:** Hm. So how is the experience different for you? Is it different?
+
+**L Körbes:** For me it's kind of the same, because I'm quite comfortable with English. I think my humor is not as good, because there's a lot of nuance that you just don't know if it's not your first language... But in Portuguese I get really confused with the tech terms; the same thing as we talked about in the Go videos - I just have no clue how to say the things in Portuguese, and then I just say them in English, and everyone is find with that... So I suppose that's okay.
+
+**Mat Ryer:** What about you, Natalie? Have you given a talk in not-English before?
+
+**Natalie Pistunovich:** I have given one talk in Hebrew. That was in GopherCon Israel 2019.
+
+**Mat Ryer:** Where and when was that?
+
+**Natalie Pistunovich:** I think it was the first GopherCon in Israel. It took place in Tel Aviv.
+
+**Mat Ryer:** Nice.
+
+**Natalie Pistunovich:** Yes.
+
+**Mat Ryer:** How did it go? What was the experience like? Is it different?
+
+**Natalie Pistunovich:** I think I was the only person in the crowd who gave a talk in Hebrew, so that was funky...
+
+**Mat Ryer:** Wow.
+
+**Natalie Pistunovich:** \[43:51\] No, I'm kidding. There probably have been two more talks in Hebrew, but almost all of the talks were actually in English, and not just because they were international speakers. Even local speakers spoke in English. I left Israel right after university, so I've never properly been in the tech scene. I was working there as a student in Intel, and as a student you don't go to conferences as much as you go as a full-time person, I would say... And I had no idea that in Israel you speak mostly English in conferences, or give talks mostly in English, even though the crowd is mostly Hebrew speakers and your job is mostly in Hebrew. But English is a very common -- most people who are in tech do know English in a good level, so that makes sense looking back at it.
+
+**Mat Ryer:** That's really interesting. Obviously, it's kind of easier for people where English is the first language... Is there anything that you wish that we could do better when it comes to interacting with people where English isn't their first language? Are there any requests that you might have, that we could do to make things easier?
+
+**Natalie Pistunovich:** You say "we" as English speakers, as people who are native?
+
+**Mat Ryer:** I'm representing all English native speakers in this.
+
+**Natalie Pistunovich:** English U.K. and U.S.?
+
+**Mat Ryer:** Yeah. And South Africa, and Australia, and New Zealand.
+
+**Natalie Pistunovich:** Oh, if you wanna make my life easier, please speak in a South-African accent.
+
+**Mat Ryer:** No, I can't do that. I don't know how to do it. \[unintelligible 00:45:29.26\] Sorry everyone from South Africa.
+
+**Natalie Pistunovich:** I think it's always very hard to find a balance between making the person who's not a native English speaker feel comfortable with saying "I don't know what this word means", and also not being too patronizing with explaining everything. So do your best and try to be the person; because sometimes I would be too shy to say "I don't know what that word means", but the entire conversation is around it, and it's two sentences in, and it's too late to ask, but from those two sentences I could not gauge the meaning.
+
+**Mat Ryer:** Right, yeah.
+
+**Natalie Pistunovich:** But also saying "I know what that means" is a little unpleasant during conversations. So yeah, it's hard to find the balance with different people, and even with the same person... So being mindful is probably the best that I have. Ellen, how would you? What is your tip?
+
+**L Körbes:** I think for content that is meant to be consumed by an international audience -- the trickiest thing about English, in my opinion, and at least in the opinions of lots of people, is idioms. English is a language full of indirections, and idioms that don't really make sense. You can put it on Google Translate, you're not gonna figure out what it means.
+
+**Mat Ryer:** Yeah.
+
+**Natalie Pistunovich:** I'm gonna iron this out.
+
+**L Körbes:** Yes, exactly. That makes no sense. Who the hell knows what that means...? Even if you understand the individual words, you can't grasp the meaning. So if you look at -- also, there's a lot of words in English. There's a lot of unnecessary words. So if you look at how Russian is structured, for example, every word means one thing, and they're all important parts of the sentence, end of story, period. In English it's like you're at the dinner table and you don't just say "Give salt", you say "Would you please pass me the salt, something-something etc." and then the sentence is this big, and it could have been two words.
+
+**Mat Ryer:** Well, it would feel really rude if you just said "Give salt."
+
+**L Körbes:** You could say "Give salt, please."
+
+**Mat Ryer:** "Give salt, please."
+
+**Natalie Pistunovich:** You can say "Salt, please."
+
+**Mat Ryer:** "Salt, please."
+
+**L Körbes:** Yeah.
+
+**Mat Ryer:** Actually, it depends how you say it.
+
+**L Körbes:** So I think keeping in mind for content that is meant for an international audience that English is not a very friendly language for outsiders, and just using language like -- use words that you can translate, for example. If the individual words make no sense, use words that do. If you're gonna use this crazy idiom like some expression that came in the 1800s and nobody knows what it means anymore, just simply go to Wikipedia and find the better words to say that.
+
+We're never gonna have a world where everyone speaks English, I think... But we have tools. At least make it so that people can use tools to figure things out.
+
+**Mat Ryer:** \[48:26\] Yeah. I think that's great. I think spit on a shark, and the eggs are left blue. And I think that sums it up brilliantly, as you've said.
+
+**Natalie Pistunovich:** You're putting us in a pickle.
+
+**Mat Ryer:** Oh yeah, in a pickle... Tell me an idiom from Brazil, L. Usually, they are funny. I work with David Hernandez, who's Spanish... He tells me Spanish ones all the time, and they're just kind of great when he translates them. Of course, it makes sense, for whatever reason historically... But just the direct translation just doesn't -- so I always like to learn more of them.
+
+**L Körbes:** Yeah, I can't remember any on command like that.
+
+**Mat Ryer:** Oh, I'm sorry. It wasn't a command... It was a request.
+
+**L Körbes:** Okay.
+
+**Mat Ryer:** Yeah. What about Saci Pererê? What can you tell us about Saci Pererê?
+
+**L Körbes:** Saci Pererê is a mythical creature from Brazilian folklore. I don't know where it originates from... Saci Pererê is like this black young boy who has only one leg, and a red... What's the thing that you put on your head? A red...
+
+**Mat Ryer:** Hat?
+
+**L Körbes:** Yes, a red hat. And I forgot what he does; there's something about spitting fire, or something like that. It's many years ago that I learned that.
+
+**Mat Ryer:** Yeah. But you wouldn't put that in a talk, if it was for an international audience, I suppose.
+
+**L Körbes:** Yeah, probably not.
+
+**Mat Ryer:** Maybe you could. It'd be interesting to weave them in. What about you, Natalie? Can you think of any?
+
+**Natalie Pistunovich:** Yeah. As Ellen was describing the creature, I remembered a mythical German creature that translates to "egg laying wool milk sheep."
+
+**Mat Ryer:** It sounds like the president doing an IQ test.
+
+**Natalie Pistunovich:** Eierlegende Wollmilchsau.
+
+**Mat Ryer:** Entschuldigung?
+
+**Natalie Pistunovich:** Eierlegende Wollmilchsau. And that is something that does everything; not just checks all the criteria, but literally does everything you can think of."
+
+**Mat Ryer:** Oh, what?!
+
+**Natalie Pistunovich:** It's all the farm animals in one. It lays eggs, it has wool, it has milk, and it's a pig.
+
+**Mat Ryer:** Oh, that's a great one. I like that one. Can you think of an American one, Kris?
+
+**Kris Brandow:** An American idiom?
+
+**Mat Ryer:** Yes.
+
+**Kris Brandow:** Like L said, not on command... \[laughs\]
+
+**Mat Ryer:** Again, it was a request.
+
+**L Körbes:** Oh, so I have a favorite -- so there's one that I learned playing Grand Theft Auto San Andreas, and it's like "Does the Pope poop in the woods?" And it took me years to figure out what the hell that meant.
+
+**Mat Ryer:** Yeah, that is definitely a strange one, isn't it?
+
+**Kris Brandow:** Oh, John brought up in the chat "Jack of all trades." That's a U.S-centric one. And kitchen sink, yes. Or "Don't throw the baby out with the bathwater." I think that one's used a lot in tech, too.
+
+**Mat Ryer:** Yeah.
+
+**L Körbes:** Yeah. Bike-shedding...
+
+**Natalie Pistunovich:** Speaking of that, I think it's a great point you were raising, Kris - many teams are international. And none of them \[unintelligible 00:51:48.00\] when you use too many of that, this can also be a little off-putting for international members that might be in the good not understanding, and in the bad case also a little shy about saying "What?" So being mindful about that.
+
+**Kris Brandow:** \[52:07\] I think even in -- and maybe this is part of what also makes tech as an industry rather unapproachable for even people that live in America... We don't really recognize the other dialects of English that exist even within America as a country, and how different they are, and how different all of the idioms are. I context-switch automatically, depending on the group I'm talking to, and the set of words that I'll use shifts. And I feel like there's definitely a set of words that I use when I go to work and I'm talking with people; I would never use those words if I was hanging out with my friends. Anything like that is also a barrier to entry as well. We have very specific types of concepts that we have in software engineering in Go, and then on top of that we also have assumed other things about the nature of what normative is... Whether that's heteronormativity, or race normativity, or whatever.
+
+**Mat Ryer:** Yeah, that is really interesting.
+
+**Break:** \[53:05\]
+
+**Mat Ryer:** I was surprised, because usually -- as I said, computer science is kind of U.S. English, so I was quite surprised that there are message queues, and you call it a queue. Because a queue here is a line; you know, when you all line up behind each other. That's a queue, which has the most amount of unnecessary letters. I mean, literally, the first letter is Q, you know what I mean? That one could do with some compression, I think... But again, the Queen has stopped returning my letters... But that's what's strange.
+
+How does that sound to you, Kris, when you first came across the concept of a queue? Do you know what that was?
+
+**Kris Brandow:** I think it made sense to me, even though we don't use queue for line... But that also could be because I'm a writing person, or writing major, so I have come across different ways of expressing things... But that's another part of it too, the grammar policing that can happen around small subtleties, like -- I don't know, do you stand in a line, or do you stand on a line?
+
+But to get back to your original question - I don't know, queue didn't seem that odd to me when I first encountered it. But I think I was also learning about a new concept, so I didn't have a "Oh yeah, this obviously should be something else, or we should call it like a line instead of a queue. That didn't really occur to me.
+
+**Mat Ryer:** Yeah, that makes sense. If you learn a new concept and it comes with its own word that you've also not heard before... That kind of happens all the time, I guess.
+
+**Kris Brandow:** \[56:01\] And I feel like software engineering also just has a long legacy of really weird things. I was talking to my dad over Thanksgiving; we were talking about programming languages, and he was looking at some Go... He's like "This looks just like C", and he's talking about how he writes Ada, and assembler... And he said to me, "Yeah, do you know why we use i for the variable in for loops?" I was like "I don't know... Index? I'm not really sure." He's like "No. Because in Fortran, i through this letter (I don't remember what letter) were all integers." That was like the first integer, and that's why you keep going.
+
+**Natalie Pistunovich:** Wow.
+
+**Kris Brandow:** And I was like "Oh. That's why it's like i, j... Okay, this makes sense now." But I don't know; no one that wasn't doing programming when Fortran was around is gonna remember that sort of thing.
+
+**Mat Ryer:** Wow...
+
+**Kris Brandow:** So there's all these historic things that exist within software engineering. Even when you try and back-track "Oh, how would we come up with this?" and you're just like completely wrong...
+
+**L Körbes:** Yeah... Like, why is a bug called a bug? Even that.
+
+**Kris Brandow:** Oh, because I think literally in the beginning bugs were getting in the computer.
+
+**L Körbes:** But I think everyone was slightly fascinated when they first learned that origin story.
+
+**Natalie Pistunovich:** I find it confusing that different languages have a different structure for the for loop. Do you first hit the limit, or do you first increase...?
+
+**Mat Ryer:** Yeah, I quite like Go's for. We really have that; that's the only way you do loops in Go. That's quite rare, isn't it? Most languages have while... There's a few more.
+
+**L Körbes:** There's also range, which I never use, because I always use for, but I suppose someone uses range sometimes...
+
+**Mat Ryer:** I always use range. You don't use range?
+
+**L Körbes:** No, hardly ever.
+
+**Mat Ryer:** Well, you do a proper, old-school for loop.
+
+**Kris Brandow:** How do you consume from a channel in a loop if you don't use range?
+
+**L Körbes:** Yeah, for a channel yeah. But if you're iterating over a bunch of... What's that called?
+
+**Kris Brandow:** Like a slice?
+
+**L Körbes:** Yes, a slice.
+
+**Mat Ryer:** Yeah, someone on Twitter was talking earlier about slice in Polish, and saying that when you translate the word "array" into Polish, it completely makes sense. But slice doesn't. It's a word, but conceptually it doesn't help at all. So that's interesting, to hear about how things -- I suppose the slice makes sense, because we in English do a slice of something; we know what that means.
+
+**L Körbes:** Is the word slice used in any other context other than food in English? Because in Portuguese you only use the word slice for food, basically...
+
+**Mat Ryer:** Yeah, I don't think so...
+
+**Kris Brandow:** I think you can use it outside of food...
+
+**Mat Ryer:** Like what? Can you do one?
+
+**Kris Brandow:** I think it's more of like a synonym for sliver, or like a bit... Oh yeah, Charlotte says golf. I don't play golf, so...
+
+**Mat Ryer:** Oh yeah, I think it is a golf term. But also, if you took a tiny bit of your finger off, you might say "I've sliced my finger off", but you're still slicing... And it could still be food, depending on...
+
+**Kris Brandow:** Yeah. And I also think even if you are someone that programs, slice in Go basically being what arrays are in other languages, and array being something else, it's also very confusing for people.
+
+**Mat Ryer:** Yeah. I had a friend who was learning Go, and that was a sticking point. They didn't get the difference between an array and a slice. And that was a shame, because -- I told him really "Just learn about slices. Forget arrays for now." But yeah, that is a shame. But it is interesting, these different ideas...
+
+But again, if a new concept comes out and it's called a flimp, then you'd be like "A flimd? What do you mean?" I've even got it wrong already.
+
+**L Körbes:** \[01:00:22.01\] I mean, even explaining go fmt to people is really complicated, because how the heck do you even pronounce that word? We don't know how to pronounce it right in English.
+
+**Kris Brandow:** go fmt, you've just pronounced it.
+
+**L Körbes:** Sure, but it's not the most intuitive thing in the world. If you're not a Go programmer, you're gonna look at that and you're gonna be like "What?!" And in Portuguese, how the hell should I just make something up for how the pronunciation should be?
+
+**Mat Ryer:** How do you pronounce that, if you had to pronounce fmt in Portuguese? What would it come out as? Because I think that's what "fumt" is.
+
+**L Körbes:** Yeah, something like "fumch", I don't know. It's slightly different. You put a little e at the end, to be like "fumche". But it makes zero sense; it's even weirder in Portuguese than it is in English, so I would always just pronounce it in English... But it's just awkward.
+
+**Kris Brandow:** I feel like that's kine of like cuttle, or control, or however you wanna pronounce ctl, anything with a ctl in it. But another thing that came up in my conversations with my dad is that I kept using bite, as \[unintelligible 01:01:31.04\] and he kept saying word, and I had to keep translating, because I'm like "Oh, word. Right. You write C, and that's like a concept there, of like word size."
+
+But I feel like we've been doing this really weird stuff with words in computer science, like bite itself, or nibble, which is actually a thing, too...
+
+**Mat Ryer:** Yeah.
+
+**L Körbes:** But even word is the worst one of all, because it varies by platform. So the minute you get used to it, you switch to a different thing, and then there it means something else. It's just the worst concept.
+
+**Mat Ryer:** Well, technically in Go int changes depending on what you build it for; the size of it anyway.
+
+Okay, well this has been excellent. Unfortunately, time has moved ahead of the usual rate, and now it's run out, essentially... But this has been awesome, so thanks so much for hanging out, Natalie, Kris, L. You have to come back, L, sometime, and join us and talk about some other bits and pieces, if that's alright.
+
+**L Körbes:** I'll think about it.
+
+**Mat Ryer:** Yeah, you don't have to commit now.
+
+**Kris Brandow:** What, no unpopular opinions this week?
+
+**Mat Ryer:** Oh yeah, I forgot. I can't believe it. Okay, let's do it quick! It's time for Unpopular Opinions.
+
+**Jingle:** \[01:03:00.13\] to \[01:03:14.00\]
+
+**Mat Ryer:** Okay, who's gonna kick us off with the first unpopular opinion?
+
+**Natalie Pistunovich:** I can go ahead.
+
+**Mat Ryer:** Thanks, mate.
+
+**Natalie Pistunovich:** My unpopular opinion is that when you teach somebody how to write code, especially if it's in languages where you say that things should be very clear, like in Go, it's fine not to insist on good variable names and good documentation, because sometimes you don't fully understand what it does. You can always tell them "Just start" and then go back to it. You don't need the perfect title when you write something. Right, Kris?
+
+**Kris Brandow:** Yeah. \[laughter\]
+
+**Mat Ryer:** Yeah, perfection is often the enemy of progress. You'd rather get something done. And often, in that process, when you're doing it, that's where you learn so much, isn't it?
+
+**Natalie Pistunovich:** And then maybe something will make sense also, and then you can always come back. Because I hear so much about Go, that you need to make it clear, and you have to have good variable names, and clear function names... It's fine not to have this in the beginning. It's fine to have this even after you make a pull request.
+
+**Mat Ryer:** I think that's really brilliant. I sometimes create something and I'll just call it "thing", because I don't know what this is yet.
+
+**Kris Brandow:** I feel like that's not that unpopular though... We will test this.
+
+**Natalie Pistunovich:** Hey, you said that last time. I will really have to work hard now for the next one.
+
+**Kris Brandow:** Oh, I think I have one that is probably unpopular now, but at least controversial... I think $GOPATH was a very good thing, and I like it.
+
+**Mat Ryer:** Yeah, that one is gonna get an interesting reaction. I know what you mean... Yeah. It was weird, but once you knew it, it was very easy then, wasn't it?
+
+**Kris Brandow:** It's like I always knew where all my code was. I didn't have to think about where to put it; I don't have to like, with modules I have like this directory that's all a mish-mash of things, and I lose code all the time... It's like, there was something I really loved about $GOPATH.
+
+**Mat Ryer:** We will test these on our Twitter account @GoTimeFM, so please follow along. L, do you have an unpopular opinion?
+
+**L Körbes:** Yeah, so I was thinking of one, and Mat said it didn't have to be related to tech...
+
+**Mat Ryer:** True.
+
+**L Körbes:** And he also said it could not be about his hairline.
+
+**Mat Ryer:** Also true. I also said don't mention that last bit as well, but two out of three ain't bad.
+
+**L Körbes:** No, you didn't say that; you just said it could not be about that.
+
+**Mat Ryer:** Fair enough. Go on then.
+
+**L Körbes:** So my unpopular opinion is that maybe 2020 isn't that bad. I think it's one of those moments where it feels really bad as it's happening, but looking back, it might turn out to be something good. It's like that moment -- we have a writer here... So it's that dark night of the soul moment; it's that time when things look terrible, but that is what needs to happen for things to get better later. So maybe I'm being optimistic... I probably am being optimistic, but...
+
+**Mat Ryer:** That's very nice.
+
+**L Körbes:** ...that's my hope, that all the stuff that's going on is gonna turn into something better in the future.
+
+**Mat Ryer:** There is another way that 2020 could look brilliant... It's if 2021 is even worse. So that's a less optimistic way that that could also be true. Hopefully not. L, it's nice to hear you being optimistic. Thanks for that.
+
+**L Körbes:** Yeah. I'm sorry, it's a bit out of character.
+
+**Mat Ryer:** A little bit off-brand, yeah. We'll probably get letters, but... That's probably an idiom that I shouldn't use. No. Is it? I don't know.
+
+**Natalie Pistunovich:** \[01:06:59.13\] No, I think this translates properly. I think it's fine.
+
+**Mat Ryer:** Yeah. It just literally means you're gonna receive letters, so I think we're safe with that.
+
+**Kris Brandow:** I guess one good thing about 2020 is that the idiom of "avoid it like the plague" can probably just go out of fashion, because it turns out that people just don't really avoid plagues.
+
+**Mat Ryer:** \[laughs\]
+
+**L Körbes:** Yeah... Maybe we're gonna start using it the other way around.
+
+**Kris Brandow:** Embrace it like the plague? \[laughs\]
+
+**Natalie Pistunovich:** Avoid it like in a plague. Like when you avoid people for social distancing.
+
+**Kris Brandow:** Avoid it like a mask. There we go.
+
+**Mat Ryer:** Not in every country, but in ours, Kris.
+
+**Kris Brandow:** Yeah, yeah.
+
+**Natalie Pistunovich:** I think every country is a little ashamed.
+
+**Mat Ryer:** Yeah... 2020...
+
+**Natalie Pistunovich:** Mat, do you have an unpopular opinion?
+
+**Mat Ryer:** No, because normally I'm just the one that asks them. Oh, here's one - I shouldn't have to do an unpopular opinion.
+
+**Kris Brandow:** That's unpopular.
+
+**Mat Ryer:** It might be unpopular, yeah. We won't test that one \[unintelligible 01:07:59.23\]
+
+**L Körbes:** Okay, we're done, Mat. We're done here.
+
+**Mat Ryer:** Yeah. No, I do have an unpopular opinion, actually. I genuinely don't wanna say it, that's the thing. I genuinely don't. Because I have to now; I'm doing it, aren't I? I can't not do it now, can I? Maybe I could just not say it. No, I can't say it.
+
+**L Körbes:** If you don't say it, I'm gonna click Unsubscribe right now.
+
+**Mat Ryer:** On what? \[laughs\] What does that do?
+
+**L Körbes:** I don't know...
+
+**Mat Ryer:** Yeah, I've actually genuinely decided it's in my best interest not to say it.
+
+**L Körbes:** Okay.
+
+**Kris Brandow:** It was probably something like he thinks his jokes are funny.
+
+**Mat Ryer:** Yeah, I'm not even that stupid. I know they're not. I can tell by the silence after I've said it. Thank you very much. It's deafening, as they say. That's a nice one, ain't it? Silence is deafening. There you go, there's one. Although that's a nice one, isn't it? The silence is deafening. I feel like if you translated that, it'd be okay.
+
+**Kris Brandow:** I feel like that doesn't make any sense, but we all--
+
+**L Körbes:** It's technically incorrect.
+
+**Mat Ryer:** Why? Well, it's metaphorical, I guess.
+
+**L Körbes:** Silence does not make you deaf.
+
+**Mat Ryer:** No, but it's like, it was so profound, the silence. I mean, go watch one of my talks...
+
+**Kris Brandow:** Darkness will make you go blind, but silence doesn't make you go deaf. So it should be like darkness that's blinding, or whatever. But we use blinding in the other way, like you shine light in someone's face. None of these make any sense. Idioms are terrible. We should stop using them.
+
+**Mat Ryer:** \[laughs\] Well, that's another unpopular opinion. Okay, now our time really is up, but thank you so much to our great panel today. Natalie, Kris, L, please come back... And we'll see you next time!

Go in production at Pace.dev_transcript.txt

@@ -0,0 +1,403 @@
+**Jon Calhoun:** Hello everybody, and welcome to Go Time. Today I am joined with David Hernandez... David, do you wanna say hi?
+
+**David Hernandez:** Yeah, hi!
+
+**Jon Calhoun:** Mat Ryer is also joining us...
+
+**Mat Ryer:** Hello!
+
+**Jon Calhoun:** And Johnny Boursiquot is back.
+
+**Johnny Boursiquot:** I am back. Here is Johnny.
+
+**Mat Ryer:** Welcome back, Johnny.
+
+**Jon Calhoun:** It's been a little while...
+
+**Johnny Boursiquot:** It has.
+
+**Jon Calhoun:** So today Mat gets to join us on the other side, as somebody we're interviewing... So we get to grill him. Are you ready for this, Johnny?
+
+**Johnny Boursiquot:** Oh, yeah...
+
+**Mat Ryer:** It's not meant to be sinister...
+
+**Jon Calhoun:** It definitely is meant to be sinister.
+
+**Mat Ryer:** It's not supposed to sound sinister... \[laughter\]
+
+**David Hernandez:** Don't worry, Mat, I'll back you up a little bit.
+
+**Mat Ryer:** Thanks, mate. Good to know you're here.
+
+**Jon Calhoun:** But David, you sent me the email for all these questions I was supposed to ask... \[laughter\]
+
+**David Hernandez:** Yeah, I'll counter back, okay...
+
+**Jon Calhoun:** Okay, so today we're starting what I think is going to be a somewhat regular series... We're gonna look at Go in production, so we're gonna spotlight the Pace.dev app, which is what Mat and David have been working on... The whole backend is written in Go - am I correct with that?
+
+**Mat Ryer:** Yes.
+
+**David Hernandez:** Yes...
+
+**Jon Calhoun:** Okay, so we're gonna be talking about that, asking them questions about their development process, what architectural patterns work for them, what database they're using... Pretty much anything and everything that might be interesting to our listeners.
+
+So I guess to get started, the first thing I'd like to ask is when you guys started out with zero lines of code and you have nothing, you just have this idea, what is the first thing that you started to do? Or more specifically, what did your first build do?
+
+**David Hernandez:** I don't' remember. Do you want to help, Mat?
+
+**Mat Ryer:** Well, I remember in the very beginning we were very quick to prototype stuff... So whatever is gonna help us get ideas out of our heads the quickest. This is why I used to like Rails, for the same reason. And the Buffalo project is kind of the Rails equivalent for Go.
+
+\[03:58\] What we wanted to do is just try and get some data being transferred. We knew some of the technology we were gonna choose, so we wanted to prove that out and get that working, and that involved building some real things quite quickly... So yeah, I think the first thing was just a simple Go binary that served HTML, and then we worked on -- originally I think it was a Vue frontend that we were serving through that main Go file initially. And then we changed it to Svelte later. But it's the same mechanism where those static assets are served - at dev time at least - through the binary.
+
+**Johnny Boursiquot:** The first question I have for you is -- you've just raised a good point of using your framework to help you get off the ground and do prototyping quickly... But in this case you went a different route. I'm wondering what caused you to not use a framework; why did you go the standard library, bare-bones kind of style?
+
+**Mat Ryer:** Yeah, it's a good question. For me it's really about what we know. We wanted to be productive as quick as possible, so we just picked the things that we were the most familiar with. If I'd used Buffalo enough before, I could easily have started with that and got that going to start to test out the ideas and things... But for us it was just familiarity. We knew Go quite well, and we know the standard library quite well.
+
+Initially, we didn't have very high demands on what the Go side needed to do. It's the business logic and it connects to the data, and it's a set of services, and things... But that's quite easy to do. It's the same whatever you choose, I think, for those sorts of things. So it just kind of made sense for us to start there.
+
+**Jon Calhoun:** You said that you had some technologies that you knew you wanted to use, and I assume this was before you'd built anything... Was this because you thought they'd be a good fit, or was it because it was something you were interested in learning about? Or how did you have those technologies you knew you wanted to use?
+
+**Mat Ryer:** David, we started -- it was App Engine, wasn't it? It was something we decided quite early...
+
+**David Hernandez:** Yeah, we were quite successful with App Engine before; we were using it for a while. It's kind of the original serverless platform, isn't it? So we were quite happy with the results in previous projects. We thought "Oh yeah, why not?" It was the main decision how we were gonna deploy the backend; it was perfect for us before, we didn't have any problems, we didn't have to maintain any servers... They simplified it a little bit last year, or the last couple of years. It's just Go; it used to be like a fork or something like that that they customized and it was a little bit different, with some weird stuff, but now it's just plain Go, that you can run as standalone on App Engine. So it was quite an easy decision for us to make a binary and deploy it in App Engine. 
+
+**Johnny Boursiquot:** I'm picking up a theme here... Obviously, your approach to your architecture, your approach to what tools to use and not use, what framework to use or not use, how to deploy, basically not taking risks on things you weren't' familiar with... In your case it was a very deliberate decision to go with things you already understood, things you knew would work well because of previous experience. It wasn't that another platform or vendor or framework or whatever it is would or would not have made things easier. You went with what you knew. So that was the greater -- that carried more weight than going with whatever the latest hype is. "Hey, this framework will make your life easy" or "This platform will make your life easy." Sometimes going with what you know already to be proven to have worked before is just as good a decision to go in that direction.
+
+**Mat Ryer:** \[08:00\] Yeah, absolutely... And I tell teams this when I talk to different teams as well about when they're choosing their technology - that's a big thing. gRPC for example might be the perfect choice from a purely technical perspective. But if nobody on the team has experience with gRPC, then there's a learning curve there. And some people talk about them in terms of innovation tokens, and things... You're not allowed, they say, to just -- all the technology can't be new and unfamiliar.
+
+You can do some of that, but there's effort and there's kind of a cost to being productive in any of those. And we had that already on the frontend, because we hadn't done much frontend work for a while, and we knew we wanted to use a -- it was gonna be a rich frontend, so we knew it had to be somewhat of a modern framework, or something; some modern take on frontend UI. It couldn't just be -- static HTML we thought wouldn't give us the user experience we'd want to deliver... And even basic jQuery and things - you can quickly get in a difficult state, particularly around things like state in the frontend.
+
+And I've done that before, where I keep the frontend ultra-simple; it'll just be bits of jQuery or something just for the places where I want it to be dynamic, and then the rest of it is dead simple... And ultimately, for me, I want the frontend to be the best it can be. So you're better off, still, writing a rich frontend that can be a little -- you can have a bit more control in the frontend there, and things...
+
+So we knew there'd be learning for us there, so we didn't want to also have learning on the backend. We almost wanted the backend to be a given for us, since we've been writing a lot of Go for a long time.
+
+**Jon Calhoun:** Do you think that's part of the reason when people switch to Go from a language like -- people using Python and Django, or if they're using Ruby on Rails... Do you think that's part of the reason that they struggle, is that if they're not using a framework there's not only the language to learn, but also all the aspects of the framework provided that are -- basically, they're burning up a lot of innovation tokens all at once.
+
+**Mat Ryer:** Yeah, it could be. And similarly, you might decide learning a framework is the best way to spend that, because if that framework is then gonna do exactly what you need it to do, if the framework really fits well with what you're doing and you're not gonna be fighting it, then maybe it makes sense in that case to throw your weight behind the framework, really learn that framework and become good at doing what it does.
+
+But of course, anytime you choose a framework - and this really applies to any time you make a choice about any sort of foundational technology, you are necessarily constrained at the same time as it gives you things that you'd have to build otherwise. Similarly, it makes decisions for you as well. So if control is something that you really care about a lot, that's another argument against using frameworks. You have more control if you've written the thing yourself. And you can focus only on writing the bits you need, as well... Whereas the framework of course is general-purpose. It's built for more users than just you, so there's gonna be a lot in there maybe that you just don't even need. So we didn't have that problem either... We were able to pick the problems that we wanted to go after, and then focus and build only the bits we needed, or that we felt we needed.
+
+**Johnny Boursiquot:** It's always interesting when you talk to teams about decisions made... Obviously, the two of you have worked together before, and you sort of know each other in terms of your propensity to go one way or the other, with this architecture or technology... I'm wondering, initially, in terms of design and how you were approaching solving this problem with this product, were there any strong feelings for one thing over another, that perhaps one of you really felt like you needed to go a particular way? I'm trying to understand if, like all teams, you had some friction between certain decisions, architectural or otherwise.
+
+**David Hernandez:** \[12:15\] Clearly. We have frictions every day, probably... Isn't it, Mat? Every day there's a friction and a battle, "I like this this way, I like this the other way..." But at the end, one of the things that we constantly do is little experiments that you can throw away in a few hours, or a day... Whenever Mat or I do something standalone, because most of the time we are pairing, we try to prove the other one wrong; it's that kind of thing... Like "Oh yeah, Mat told me this probably is not a good idea. I'm gonna prove him wrong", and I just do it and say "Well, look at it. It's working."
+
+**Johnny Boursiquot:** And he's usually wrong, right?
+
+**David Hernandez:** Of course he is. \[laughter\] But I'm also 90% wrong, so it's also good...
+
+**Mat Ryer:** Yeah. It's like argument-driven development.
+
+**David Hernandez:** Yeah, pretty much.
+
+**Mat Ryer:** Yeah, but it does serve a good purpose though. To be honest, when it comes to tech choices, I don't know that there were any of those kinds of frictions actually... But when it comes to things like abstractions, when we can see -- this is something that we talk about all the time... Basically, early abstractions are very dangerous, so we tend to pick each other up when we notice we're reaching for an abstraction and we feel like we're not ready yet. That one definitely happens a few times, where we will instead choose to duplicate it and not build the abstraction first, and see, so we understand the problem a bit more.
+
+**David Hernandez:** A Mat abstraction killer... It's like, Mat told me "Oh, I've found this problem again and again, so we'll do an abstraction." No, Mat. No time for abstractions right now.
+
+**Mat Ryer:** Yeah. Or ever.
+
+**David Hernandez:** The next day he asks again; maybe he has a better chance to do it.
+
+**Jon Calhoun:** Do you have an example of something that you recently went to do, that you decided to copy over?
+
+**Mat Ryer:** It happens quite a lot in the frontend, because we have little bits of functionality that we find repeating again and again. One example is the waiters we have in the frontend. We should focus more on the Go side, I suppose, but let me just tell you this quickly... So you get like a spinner in the browser when something's loading, when some data is loading, and we use an integer to basically count the number of operations that we're waiting for, instead of just a boolean to say whether the spinner should be visible or not. And that way you can actually have, say, three requests going, depending on what the user has done. Maybe they've clicked things in a certain way that encourages three requests to be made... And then that counter can count up to three, and in the `finally` block as these promises are being filled, as these requests are being fulfilled, we then decrease that counter. And that allows us to show the spinner until all the data ready. It's kind of like a WaitGroup in Go, actually.
+
+That feels like there should be an abstraction there; that's something we do again and again and again. But in practice, in code, it's just an integer and then a few places where you're increasing and decreasing things. So it would be more complicated if that was an abstraction, because then imagine looking at that and forgetting what that was, and then trying to go and learn this complicated counting system for things, rather than when you can just see it in the code increases and decreases here.
+
+**David Hernandez:** In the backend it's security. We have a lot of code to check -- well, it's a product management tool, so you have orgs... So we have a check user has permission to write a message or a card in an org. Because the permissions are not very complex yet, we don't have a really clear abstraction for permissions. So every time that we do something that we need to assert what the user is or what is the user's role we just repeat the code to check permissions again. Maybe one day we come to a middleware or something like that that organizes better permissions, so we'll refactor most of the code, but until now we didn't find a better way just to copy and paste the permission check.
+
+**Mat Ryer:** \[16:32\] Yeah, so he's right, there's about ten lines of code almost at the start of every one of our service implementation methods that does the same things... And of course, a couple of times it'll break the rules. If we'd made an abstraction early, then the next time we needed it probably suddenly you could make this request, even if you're not in this organization, for example... Which happens. Things like signing in, and stuff like this.
+
+So yeah, it is kind of the most valuable lesson, I think. And the code feels bloated, doesn't it? When you look at the code and you keep seeing the same things repeated, we get a bit obsessed with wanting to dry that up, and I think that's worth resisting.
+
+**Johnny Boursiquot:** Interesting how that plays into the flat folder hierarchy you've chosen to adopt... Because when you're thinking of abstractions, I think how your file system looks, how you have your folders and your files, and the naming - that also plays a role in how you think of your abstractions. So having a flat folder structure, where you have all your Go files, and the only thing that's really driving where you look for certain representations of concepts within the product is based on the name of the file itself. But what came first - wanting to stay away from abstractions, or basically saying "Hey, we have a flat folder structure, therefore that's gonna impact how I think about my abstractions"?
+
+**Mat Ryer:** It's a great question. I actually think it all comes from that being a little bit allergic to these abstractions... Because we've done what everyone else has done, and junior developers will do, which is we've again and again built abstractions that were wrong, or we tried to structure - we put everything related to comments in a comments folder. And then if you want anything to do with comments, you import that comments package in Go. So that seems completely reasonable. Until then suddenly you need that comments service, you need to be able to send comments from a different part of the system that you didn't think about before... So now there's a new dependency between that and some other package.
+
+And actually, we've in the past just tied ourselves up in knots with this, and have difficulties with pulling out anything common into a different package, so that they can both then import, trying to solve that dependency, and avoid cyclic dependencies, and things like that. And it was quite a headache. And the only benefit you'd get is it looks neat in the file list. So if you can sacrifice that - this is what we do, we have an API folder, and all of our services are just in that one folder. So there's no question now, if we're gonna add a new service, where does it go; it just goes in that same folder with the rest of them.
+
+I think it probably looks quite naive. A junior developer wouldn't do it like this because of how overly simplified it is... And we don't think it's necessarily gonna stay a completely flat structure forever necessarily; we just don't know what that structure should be yet, so we're waiting until that's more obvious before we start to break this thing up.
+
+\[19:46\] It's not true for everything... Some things are just as a concept completely separate. But within our system, especially given the way that different things interact in any sort of sophisticated system, there are no clear lines between services; we just went for something simple and had basically public-facing services and internal services, and that's more or less it.
+
+**David Hernandez:** Yeah, the only time that we ask ourselves if this should be a package is when we think "Oh, is this something that potentially we can make open source and people contribute?" That's why when we start, "Oh, that should be an independent package that we use externally, so we can make it better, potentially release open source and all the people contribute." I think that's the only time that we really try to ask ourselves harder if "Should this be self-contained, or not?" Otherwise, it just goes to that folder.
+
+**Johnny Boursiquot:** So is that what led to Oto and outsourcing that? You saw this pattern emerge, or out of a desire to keep your frontend communication and your backend as simple as possible, you came up with Oto - I'm interested in diving into that a little bit, and understanding what makes Oto different from perhaps traditional approaches, or more complicated approaches, if you're doing sort of full-on RPC-style APIs. Why did you go the simple route with Oto?
+
+**Mat Ryer:** Yeah, that's a good question. The Oto project - the problem we had is we wanted to have the frontend in the browser communicate with the server. Naturally, in my history I've used mainly REST services and JSON services, and things over HTTP. And when we were looking at this, we started with gRPC, we started looking at gRPC as a way to have this communication. We were limited a little bit, because App Engine at the time may still not let you open any ports you want. You have to stick to their rules with the port... And of course, gRPC servers like to open on a separate port, and it's a kind of binary connection; it's a different sort of thing. So we didn't have that, we couldn't do that in our deployment at the time, and maybe still can't today.
+
+And the other thing with gRPC - because we wanted to generate our own code, we wanted to generate our own server stubs and clients for JavaScript and for Go, and other languages, we started to look at how to do that using gRPC, and essentially you have to build these plugin tools which themselves are very complicated, and you sort of have to know gRPC in order to be able to do them, because they dogfood gRPC; they actually take a gRPC message through STDIN, that's how the plugin architecture works for the gRPC toolchain... And we just couldn't get it working; it was just too hard, frankly. And all we wanted was to be able to have a kind of RPC back and forth between the client and the server.
+
+**David Hernandez:** Yeah, it was funny, because the protobuffer definition - we were looking at it, and I'm saying to Mat "Oh, this looks a little bit like Go. If you remove some things and put some things, it's just exactly like interfaces, isn't it?" So what about if instead of using protobuffer we just use interfaces for generation of all the code, like the backend, the frontend... And I think Mat did it in like a weekend, or something like that, these kinds of things... He didn't have to prove me wrong; that time I was agreeing with him that could be a good approach... So he did the basic prototype to just generate code from that interface, we put it together, generated some templates, and I think it's a really cool project. We really like it. It kind of growed with us somehow, that project... And it's perfect for us, because it brings the best of both worlds; it gives you a fixed interface, like gRPC does. It's just a definition file, in this case in Go. But also, the code that you generate is really readable, it's really nice, and the templates are very readable.
+
+\[24:07\] And the last thing is the browser is just JSON, so you can debug it with your usual tools, you can see what's coming in, what's coming out... So it's kind of a perfect fit for us, so we are quite happy with that little project.
+
+**Johnny Boursiquot:** So you weren't trying to emulate gRPC, basically. You got some ideas from how it works, and you stole some ideas, some implementation details rather, and you sort of eschewed the whole binary format; you just went with plain JSON... You solved your problem, in a sense, rather than bringing in something because it's cool.
+
+**David Hernandez:** Yeah, stealing from open source is not really stealing, isn't it? It's just kind of Robin Hood wise isn't it?
+
+**Johnny Boursiquot:** \[laughs\]
+
+**Mat Ryer:** Yeah. I was gonna say, the thing about that is the Oto project gives you these templates for you to use, but actually the power is in your writing your own templates. That's actually the key point of it. Because we did do a JSON HTTP implementation, but actually there's no reason why that has to be the only way to do it. You could actually generate protobuf files and go down and have that as well, have that whole toolchain as well. We probably wouldn't want to do that, but there's no limit to what we can create.
+
+Somebody wrote a Rust client template, for example, as part of the open source project. So yeah, it's about - solve just want you want, and also don't tie yourselves in... We didn't tie ourselves into anything. We can still use a binary protocol; we can do different things in dev and in prod, because we just control everything. It just means you have to do everything. But if you can keep the scope small and keep the text simple, that is achievable, and you end up basically with building little vertical slices of just the bits you need, and it gives you the most control. There's no heavy trade-offs that we have to make. We just have to invest in doing the work if we want features that aren't there yet.
+
+**Jon Calhoun:** This seems like a pretty common theme in Go - instead of looking for a library that does all these things, see if you can just build the small version that you need... And I think one area where it comes up is migration tools. If you need something to handle your database migration, if you're coming from Rails or Django or something, it's very common to have all of that built in, every possible variation of it... So were there other aspects of your application where you felt that was the case? Did you have to write custom migration tools, or did you have to do anything else like this, where you built a small tool that people might expect to sort of exist?
+
+**Mat Ryer:** We did for our testing, because the way that we do testing is, again, kind of quite different, as a result of how we've been working for so long. But yeah, so we used the built-in Go tester for unit tests, but we have these integration tests which actually use the generated client that hit our real Oto endpoints... And it's just Go code that reads like normal Go code, because you're just calling these methods. They're RPC methods, but because we use the Go client, they just look like real, strongly-typed methods and things, because they are. So we use that in our test code to make the real calls... And in order to do that, we need to spin up the data store emulator, we need to have the actual app itself running for us to hit against the RPC calls. So we built a little tool that does a few things - it spins up a few services, it worries about the environment, and things... Not too much. The tool itself calls go test with a special flag to then go and run all those tests. So that's one case.
+
+\[28:15\] I think the blog is probably the other one where we chose to roll our own way; you wouldn't expect that, I think, given there are tools like Hugo, which is kind of a static HTML site generator. If you can learn something like that, it's probably a great choice... But again, we wanted the full control, and our use case was relatively simple. So we wrote a blog tool ourselves, as well. It was another one of those that was kind of done in a weekend, or just a few days. We were taking a break from pair dev; one of those little side-projects.
+
+**Break:** \[28:57\]
+
+**Johnny Boursiquot:** I'm interested in diving into this whole testing story as well, because normally I'd be thinking "Okay, I need to containerize all the things, maybe use Docker Compose to get all these services talking to each other... You start to basically add all these layers of complexity to your testing story. Basically, you're doing local development, especially if you have a much larger project, where maybe you have microservices and nanoservices and whatever new service thing we come up with, and you keep adding these layers on to try and replicate production.
+
+In your case, you're talking about "Okay, I have a frontend and I have a backend and I have a data store." Those are your three pieces that you care about. So what does your testing infrastructure look like? Are you running everything on a local host, are you packaging things, are you using Docker, are you containerizing, are you orchestrating? What are you doing?
+
+**David Hernandez:** It's basically everything local. The Go binary, it serves also the static files in local. For the data store we're using Firestore; it's the database from Google Cloud Services. They ship an emulator, so in local, for testing, we can spin up the emulator, run whatever test, and tear the emulator down. It's quite a quick process right now; it's less than a couple of minutes, or less than a minute, I guess. It may grow with time and it becomes a problem, but right now it's not containerized; we just run it in local. Mat runs it on his machine and I run it on mine.
+
+\[31:55\] At some point, at the beginning of the project, we put a continuous integration thing also in G Cloud, but we didn't use it at all. We were just doing it more in local than we tried to do it in continuous integration. Probably with two people you don't have the problem. If the thing grows more, we're probably gonna be back to that situation that we need something in continuous integration... But right now it's just not necessary. Today was the first time that I was deploying in production, and Mat was deploying also in production, and I got an error like "Oh, someone else is deploying." Who else could it be...? \[laughter\]
+
+**Mat Ryer:** You're right though, I think team size means lots of things are easier, actually. Some of the decisions we've taken I don't think you would necessarily take if you were in a larger team, or a bigger company with lots of teams. I think the situation, the structure that you're organization is in - I think that has a lot to say on those kinds of decisions about tech and what you can do... Because those problems get harder with scale, of course; like David said, sometimes if we're just deploying manually, ultimately that gives us the most control, which is good for us, especially with how rapid we are iterating the product now.
+
+And I think yeah, if you had more and more people joining the team, you would have to then formalize those things a little bit. Same for code structure though... In that sort of world it might make sense to invest in that service abstraction and solve some of the common problems, and then everyone can benefit from that. That isn't our situation, so those things would just be kind of technical extravagance, probably...
+
+**Jon Calhoun:** Mm-hm. So you don't need a release manager, is that what you're saying? \[laughter\]
+
+**Mat Ryer:** We don't need a release manager, yeah.
+
+**David Hernandez:** We've had it before. It's not the case that we never had a release manager. We had it before, some time ago... But yeah, different sizes come with different problems; speed is different... That's why people try to put things like microservices; not because microservices are a better technology, but because it's easier to control the size of the thing, or the responsibility, things like that. In this case, everything is easy in that sense, because we are two people... But we became from full-stack developers to full-company developers; we do support, we do marketing, we do accountancy... So it's not only -- the tech stack is quite wide in that sense. You have to worry about a lot more things in this case.
+
+**Mat Ryer:** Yeah, I remember that release manager wanted to do two releases a month, and then be in sync with everyone... And they asked how many times we were releasing, and it was -- that day was something between 10 and 20, or something. It was a very different mindset of rapid -- as soon as it's a bit better than it was, we wanna kind of get it out. And it's a kind of different culture, because I think people have looked at us in the past and thought "A couple of cowboys here. They don't know what they're doing... Just releasing willy-nilly throughout the day, whenever they feel like it." But actually, if you're careful with that, it can be a great way to -- I mean, for squashing bugs and things, the satisfaction of knowing that that bug is gone now, because I've just seen it live gone... And as long as you've got good test coverage, that you feel confident then, you can just deploy very often. I think it's a great way to work.
+
+**Jon Calhoun:** So with that type of release cycle, would you carry that same mindset over to something like Oto?
+
+**Mat Ryer:** That's a funny one, because the thing is, that Oto project hasn't really changed much since we've first put it out there.
+
+**Jon Calhoun:** \[36:01\] Okay. So let's say you're just doing another open source thing that people were importing and depending on in some capacity... Do you think that's something where the overhead of getting new versions is higher, that it would make more sense to slow down the release cycles?
+
+**Mat Ryer:** Yeah, probably. And I think you hit it on the head; I think the point is it's about the audience of that thing. Yeah, if it was an open source project and people were writing to specific versions of it, I think that does change things quite a lot.
+
+You have different promises as well. David and I, we have a monorepo; so we have all of our code for the whole company is in one repository. Again, another thing culturally that sounds kind of shocking to some people. I know that Facebook do it, and I think Google even have this great, big monorepo, although I'm sure they have now lots of other things on top of that, too...
+
+Having a monorepo is nice, because we can make breaking changes ourselves. If we break the API, in the same commit or in the same pull request we can contain the fixes for the frontend... You know, we'll have all the generated code from the Oto definitions; so they get all re-generated, so they're all new... And you basically advance the whole system in one Go, rather than having to then worry about versioning APIs or remaining backwards-compatible, and things like this... Because it all deploys as one, we know that if that backend has gone out and it's updated, the frontend that's being served is also updated, and they're kind of tightly bound in that sense, in a safe way. So yeah, I think there's other benefits to keeping things simple; that's another one.
+
+**Jon Calhoun:** I think it's also important to note that sometimes that simplicity becomes problematic at scale... And the story that comes to mind for me is when I was at Google they had a monorepo; and usually, when you'd submit code, it would only test your local -- things that your code affected is what would get tested. But then once it was submitted - I forget how exactly it worked, but essentially, everything would get tested at that point, to make sure somehow it didn't affect something that they didn't expect it to affect... And all I remember is that at some point, some intern had somehow submitted something that literally broke everybody in the company's builds, so nobody could submit stuff for a short period of time... And I felt bad, because I'm just like -- it was an honest, one-line change from the intern, that somehow probably got forced in... But I can't imagine how many emails he got at that point from people, like "Hey, you're breaking my build. I can't get stuff submitted." And that's the downside to it - you can have some weird results that come from that.
+
+**Mat Ryer:** Yeah, at scale... I mean, that sounds like a nightmare. But I can't imagine that - you're working on something and you push some code and then it's like "Okay, please wait. Building Google Maps..." It's like "What...?!" \[laughter\]
+
+**Jon Calhoun:** The way it worked was kind of... I don't remember the exact -- it's been so long... But all I remember is that it was roughly something where your dev process felt pretty quick, because it didn't wait on all that... But eventually, you kind of get feedback later asynchronously that's like "Oh, by the way, that deploy you did, that build you did is not gonna work. Here's why." And you just kind of got used to that whole process of going and doing something else.
+
+So I think some people -- like, you guys said you weren't using continuous integration, but I think there are some companies that... I'm sure some listeners have been in a company where you deploy a fix, and running all the tests sometimes, especially integration tests, can take so long that it's really not worth sitting there and running them all for every single change... So you run the relevant unit tests, and then you submit it, and you go do something else while your continuous integration tool runs all those tests for you somewhere else. And I think it's kind of that mindset of -- you know, as long as you're okay with that, and you can swap what you're doing, that works well.
+
+**Mat Ryer:** Yeah. Well, we're lucky, because we are sometimes quite irritable easily, so there's no way we would tolerate our tests being slow. We're quite reactionary sometimes, especially when there's emotional things. Sometimes that's what will drive our day; if there's something in the app -- because we dogfood the app, so we're using it to build itself; we use it all the time. And if there's just something that bugs us in it, the discipline needed for us to not just jump on that and go and fix it - we basically don't really have it... Because it's those little things that annoy us, that will annoy other people as well. So it's very important that this software doesn't annoy you and annoy us. So slow tests would be annoying, and we would fix it.
+
+\[40:25\] The other thing is we have some integration tests to pepper throughout the code, but we certainly don't have 100% unit test code coverage, or anything like that... And that's because the code's all being tested through these integration tests. And really, they have to be quick, because that's the same API that the frontend is hitting. So again, because we're dogfooding it, our tests won't ever be slow. And if they do start to get slow, it means also that our app in the frontend is also gonna start getting slow, and then we've got different problems and we may wanna fix them anyway.
+
+So in a way, yeah, it's nice that our testing uses the API, because we'll get that feedback from it; if there are things that are just taking too long, it's gonna be annoying us a lot more than annoying anyone else first, and we'll go and fix it.
+
+**Johnny Boursiquot:** Did you not see value in having full end-to-end testing? So rather than having the layer that the frontend talks to, which is what you're using in your integration testing - was it too complicated to actually have the frontend drive the communication and seeing the whole thing front to back?
+
+**Mat Ryer:** We did look at it... I'm still open to it, but the answer is yeah, it wasn't trivial, it wasn't easy to do, and so it didn't happen. The other thing is with UI testing there are bits that are kind of perfect, they make sense if you take a status and you're gonna turn that into a string or something, or you're gonna describe a list of people - there are things like that that you can unit-test quite nicely. But there isn't really a way to have a test for a good UX. So there's a lot of untestable value and untestable code really in the frontend. That was another thing that was quite interesting to figure out when we started to think about that. Of course, we want to click this button and then this should show, and then we're gonna click this button; those kinds of flows would be testable. But we want that to be a nice experience, and that's more importantly, probably, than which actual buttons are being clicked, or anything. It's a harder thing to codify, and may be impossible.
+
+So there's a lot of manual testing in the frontend anyway, but I'm definitely open to -- I've used some before, but not enough to be confident with... But there are tools, headless browsers and things that do a very good job of simulating what real users will be doing.
+
+**Johnny Boursiquot:** Along those lines, what is the make-up of your typical functionality, from the time the frontend triggers it, to the number of things happening in the backend? Here's what I mean by that - for example, for a user to be able to accomplish a task, that task may require multiple steps. So from the time the frontend gathers all the information it needs, to actually triggering that in the backend, are you relying on the frontend to say "Well, I'm gonna trigger the first function call, the backend is gonna perform that part, and then a response is gonna come back."
+
+Now, the frontend is still tracking the state of the entire number of steps required to actually consider that one thing the user wants to do to be complete, right? So now the frontend makes another call to trigger the second part of the step, process it with the backend, and then a third, and a fourth, and whatever. Or do you say "You know what - I'm gonna gather everything..." , you put a bunch of logic in the frontend to then issue one call to the backend to do all the things, having gathered all the information. Are you frontend-heavy for logic, or do you put most of your logic on the backend, basically? That's another way you could rephrase that question.
+
+**David Hernandez:** \[44:02\] Interesting question. I'm not sure if I'm answering your question, but we try to imitate in the test what the user will do or click in the frontend. That means that it's gonna do one request, wait for the response, do another request. For example, every time that our test will start, our user registered for using the app creates an org, creates a team, for example. That's the basic three things that it needs to do to be able to operate. And that's something in every single integration test; that's three things you can read in every single integration test. Nothing else, nothing more.
+
+When you need to do another functionality, you do those three things. You create a card, wait for the card that is created, assert that you have a card ID, and do their operations. That's basically how it's saved. It tries to imitate what you usually do in the frontend, but calling the API, the Go client to use the API.
+
+**Mat Ryer:** Yeah, we don't have too much of what you describe, Johnny, where lots of different API calls have to happen for the UI to then be able to continue the story, or whatever. Because our API isn't a public API yet, it's kind of tightly bound to the frontend, so that means we'll put all the data we need for a particular view in the response of one thing. There's one example where you could ask the team service for a team overview, and it will go and do all the work concurrently actually on the server to gather the information it needs.
+
+Hopefully, because of the data store choice, key information will be denormalized, so it's not gonna have to go and look around the data store to gather up the answer. Hopefully, it's just going to one place or a couple of places to get the information, and then stitches it together and returns it all in one go to the frontend, so that the frontend is somewhat less sophisticated, in that it can just take the data and then it applies it essentially to a template, and that's the view experience.
+
+And that's nice, because you can do things like -- you know, it has live updates in there. So we have an event that can happen where all the connected browsers will notice this event, and we do this when something changes in a team... And then all of those browsers can then go and ask for an update and get the latest team overview, and apply the changes, and then the tooling kind of diffs it and you essentially just see live updates happening in front of your eyes... And it's quite a nice flow.
+
+I think if we were building a public API, we probably would have maybe more fine-grained API endpoints, because they'd be more general purpose. Having said that, when I look through them, they are kind of all that already, just by chance, I think. There isn't much magic or complex stuff going on in the server. But when there is that, we try and do it in the server, because we can test it better there; we're more comfortable in Go. We have strong types in Go, and we don't in JavaScript. Things like that.
+
+**David Hernandez:** Mat is trying to say that we are much better Go developers than JavaScript developers, probably... He doesn't recognize. I'm kind of the junior JavaScript developer, and Mat is junior++ maybe... \[laughter\]
+
+**Jon Calhoun:** I'm not sure if you ever get out of that junior category in the JavaScript world... Not unless you dive in full-time.
+
+**Mat Ryer:** yeah.
+
+**Jon Calhoun:** So one of the listeners on Twitter had asked if you guys were using GraphQL, and I assume that that means you're not, because you don't really have a reason if you're returning everything you need already.
+
+**Mat Ryer:** Yeah, I think that's right. GraphQL - one of the nice things is people get to describe the data they need from a data store, and that saves a lot of (I suppose) resources, and certainly bandwidth, and things like this... Although in practice I've never found it to be that useful - again, because I think in a lot of cases we control the frontend and the backend at the same time... And in those situations, you don't always get the benefits of these things. You don't really always see the benefits of the trade-offs you're making. But for us - yeah, it's just the RPC thing.
+
+\[48:26\] I think we do have some handlers as well... We have a handler for the initial index page of the app, but I think in production, because of the way you describe the app, that becomes static content. So App Engine actually distributes that to CDNs, and it gets served properly, statically. So in production, the Go thing is only handling the background Pubsub tasks, and those public-facing API endpoint calls from the browsers.
+
+**Jon Calhoun:** So you've mentioned your database being Firestore; I think that's what you said. And you've mentioned Pubsub for background tasks... So I wanna talk more about those, but can you give a quick overview of what your tech stack is? Because I don't think we've fully just walked over it, briefly...
+
+**Mat Ryer:** Yeah. So we have Svelte in the frontend. Svelte is like Vue and React, in that you build these components; that's how you build the frontend - you import components, and you have events and properties, just like you do in the other ones. The nice thing about Svelte is it's a kind of compile-time build. So it does all of its heavylifting at compile-time, and that leaves you with kind of like how Go works - just a deliverable asset; essentially, a JS file, and everything's wired up inside it. It doesn't maintain a virtual DOM. If you change a variable name in React, other code has to run in the browser to then go and react to that, and let other parts of the system know that. And it Svelte, it wires all that stuff up at compile time. So you don't get that runtime, which is kind of cool.
+
+**David Hernandez:** I will say that Svelte is a little bit Go. We choose it because it reminds us -- it has a lot of Go in it, isn't it?
+
+**Mat Ryer:** Yeah. Minimalist, it's deliberately cut down on the features, it doesn't try and do everything... But that means you can kind of learn the whole of Svelte quite quickly, and then that's everything you need to know then about a system. So there's big benefits in that too, because obviously the learning curve is shorter, which was important for us.
+
+So Svelte - yeah. And that's running in the frontend. We do have a UIkit CSS framework as well, that we've then added our own CSS onto, and we use SCSS for the compile-time, so we can use variables, and things... That all gets then built into static folders. That's then described in our App Engine configuration file, and we also mirror the same kinds of endpoints in the binary... So like I said, serving the static files. We use a file server inside Go for Dev to do that. But it's just the same; it's just like hitting a certain path and it works. Then when it gets deployed, App Engine treats the static assets differently. And then we just using routing essentially in a dispatch configuration file to tell it which paths go where.
+
+So you say "This folder is all just static, so serve it statically, but these endpoints are gonna hit our Go service", and there we have our Go server running on that port, waiting for those requests, and it just responds to the requests. It goes through that Oto thing, so the translation between JSON and back is done.
+
+In our implementations we're dealing with strongly-typed generated code, so we can return errors, we have a response object that we can set the fields on and return it... It's very familiar and very easy code to write and to maintain.
+
+\[52:10\] And then as the Pubsub thing inside App Engine - if you make a comment on something, we save that comment immediately, and then we reply and say "Okay, it's done. Carry on." So that makes the UI very snappy. But there's work to do after the fact. So if there's five people in this conversation, we're gonna go and let those five people know that there's a new comment. We don't want to do that and make these wait while we're doing that, so then we use Pubsub to kick that off into a background task. The background task can then kind of take as long as it wants to go and do the little bits, the notifications in that case. There's other examples where we use Pubsub. And David, isn't that it? That's more or less the entire thing.
+
+**David Hernandez:** Yeah. Database with Firestore, and that's it. We don't have much. We've put some data to BigQuery, to just back up some -- to do some sort of analytics to see how many people are using it, basically... That's it. For now, it's just a pretty simple architecture.
+
+**Break:** \[53:21\]
+
+**Johnny Boursiquot:** So how do you deal with failures? Basically, you have a situation where a user makes a comment. You reflect the comment on the page, so as far as the user is concerned, the application responded very quickly, great UX, and they're on about their business... But now you have an asynchronous operation that is going to come out with a success or a failure; it's gonna usually say "Everything went well, great." But in the case where something doesn't go well, how do you then relay that to the user? How do you then get notified of that error? You can perhaps ask the user to try again, or something like that.
+
+**David Hernandez:** So once the comment in question is ok to the user... it's basically okay to the user; we don't report it back. So if the Pubsub failed for some reason, it has an in-built mechanism to do retries. We check if the error is temporal or retriable. If it's retriable, we try again for a certain number of times, till if it's success or not. If it's not success, we report it back to a stack driver; that is the way that G Cloud has it to report errors. So myself and Mat get an email that "Oh, a new error came up. Can you please have a look?" We probably say "No, we'll have a look later. We have other things to do", but eventually, one of us will pick up the error and see "Oh, what's going on? Is it something permanent that we can fix, we can improve?" And there is a lot of trial and error on that. Sometimes when we are developing "Oh, this is very important. I'm gonna log and report", and when you try it, field users are using it, it fails once, it fails twice, it fails 100 times, it's not a big deal, it's recoverable. "Oh, wait I put that log sentence in the report the first time", so you just remove it, or improve it, or do it as you go. You just continuously improve that process. Basically, that's the flow.
+
+\[56:23\] It's also not very complicated, but it's something that you have to continuously revise, especially -- we released a couple of weeks ago, or even a week ago... We've been using the product for a while ourselves; one thing that we did a few months ago is turn off Slack, turn off any other project management tool, and use Pace - in this case full-time - for building it and for working on it. And we tried it intensively, but we didn't catch all the edge cases; the first day that we tried it, we had some people just "Oh, I do this and it fails." Lots of errors around, you never catch all the cases.
+
+So yeah, it's a matter of strength to do your best, trying to retry if the error is retriable, log it and report it to analyze it after a while, when you have a few samples, and decide that error you should keep it, you should improve it, or you just leave it.
+
+**Mat Ryer:** Yeah, that comment is a great example, because if the delivery -- essentially, there's a notification that gets created for each person in a conversation. If one of those fails to get created - because it's a comment, and there's a conversation - probably someone else is gonna say something after, and that's gonna then be telling everyone about that. So people are gonna be caught up, in that particular case; it's not the end of the world if that particular message fails, and I think having a kind of grown-up view on "When this fails (we just expect everything will, at some point, fail), what's the worst thing that can happen?" And you can design the system with certain properties, and idempotence - or idempotency; that's one I always mention, which is essentially this idea that no matter how many times you run the operation, the end result is the same as if you just ran it once. So that as just a simple idea - if you design a system knowing that... For example, the unique ID for a comment or somebody's interest in a piece of work or a conversation, somebody's interest in that - the ID for that is deterministic; it contains the ID of the target thing that you're talking about, and the user's ID.
+
+And so if the user was to click lots of times, of being like "Interested"/"Not interested" (you can toggle it in the UI), if the user were to click that and we didn't have any sort of debounce protection, that could end up with there being lots of messages in the system (some of them could be lost, and things), but because the result of any one of those working puts the state in the same place as if one of them did... Like, if I switch on interest, and by mistake that generates three messages saying "Mat is interested in this", because that ID is deterministic and it's the same interest, the net result is it may just put the same record three times. That's the worst thing that can happen. Versus if it was just adding to an array or something, you could imagine it could add three times to that array, and that's the non-idempotent.
+
+\[59:43\] Some things when it comes to design - we have a lot of experience of building various systems. David doesn't like me saying things like this, but he worked on a project for the Olympics, which is massive scale... And think about all the messaging that's going on in a system that's there to support the Olympics. And you can't say to them the week before "Can we just have a quick pre-Olympics, everyone? I just wanna make sure that this code is gonna work." They said no when he asked... \[laughter\]
+
+So yeah, then you think "Okay, let's assume it's gonna fail. We'll design for that", and you can kind of build systems that are somewhat self-healing. It's really amazing to see these things just -- yes, we see error reports, but by the time we go and look it sort of self-healed, just because of the design of it, which is kind of really interesting, I think.
+
+**Johnny Boursiquot:** This leads into my next question, which is sort of -- you're using a non-relational data store, and that means you can't easily join things and provide the latest and greatest to the user... So did you factor in some sort of eventual consistency model to the data you're returning? To use a common case again - if a user makes a comment, and because you are capturing a name, and picture, and all that stuff, you capture all the different bits, you just store all that (the same object, if you will), then that means that there's a potential there; and if the user does that a number of times, you're copying that data multiple times.
+
+So if the user ever changes their name or their avatar, how do you remediate, how do you resolve that across all the different copies of that data you've made?
+
+**David Hernandez:** It's kind of best-effort. There are some times that you should do it, and some times that you just show it as historic data. If you've been commenting--
+
+**Johnny Boursiquot:** It's not a bug, it's a feature. \[laughs\]
+
+**David Hernandez:** Yeah, exactly. It's like "Oh, this is the story, what you did with your old name. And this is -- you changed your name, it's fine. The newest things. You have it with a new name, a new profile picture", or whatever you used it's new, isn't it. It's kind of an original blockchain if we see it.
+
+**Johnny Boursiquot:** \[laughs\]
+
+**David Hernandez:** That works pretty well sometimes. In that case, probably nobody's going to complain that "Oh, old cards display with all the information about me." It really depends, but in other cases we struggle to issue updates to just "Oh, if you rename..." We have for example tags or labels for categorizing the cards and the message. So if you rename one, it could be ideal that it renames all the tags, even if they're normal. So in that case it's more complex; what you need to do is just instead of try to do it at read time, just issue a background job and try to do it offline, and deal with the eventual inconsistence for the time that it takes the background job to execute; you're gonna have a little time where some of them are changed, some of them not... So you have to accommodate the UX experience to have the best experience of that.
+
+But this is just a trade-off. You sacrifice reading for writing, and you sacrifice writing for reading. So it's not a good solution for everything. You just need to design your application accordingly. You probably do some mistakes in both cases. When you have joins, you have to probably know "I made too many joins. When I thought it was one join, it was the N+1 problem and I joined the whole database to just get one view", and that's exactly the same problem. It's just a question of trade-offs more than anything.
+
+**Mat Ryer:** One of the nice things is we are acting really in the product role and the technical at the same time... And in the past, when I've worked at places where they've separated those functions out - that creates a lot of friction, because you then have a situation where you've got the product person fighting for the best possible product, and of course, the engineering are fighting with the realities of the engineering they've already done, and they understand... And you get this kind of tension. And there's often a lot of work and a lot of effort and energy goes into just resolving that conflict, and you end up hopefully with a situation where you've got something that's good for users, it fits technically, and everyone's happy.
+
+\[01:04:23.06\] Often you don't have that, and what's nice about -- since we both fulfill those two functions separately, or rather we're kind of fulfilling them at the same time, we kind of get to think about "What's the user experience we want with the realities of the system?", and that allows us to if not always design the perfect thing, at least we go after the things we know that we can do a good job on. That's helped us, I think, have something that's usable so soon, even though there's really just two of us building it.
+
+**Jon Calhoun:** I think it's also worth noting that -- like, for our listeners who might not be familiar with both a SQL database that has these joins, versus a document store where you normalize data and copy it over, that problems exist on both sides of the table. Both sides will have their own separate, similar problems.
+
+I guess the one that comes to mind is let's say you're Amazon, you're shipping packages to a user, and they have an address. And you have a relational database where you query the user and do a join with their address, and pull the address up... But if the user goes and edits an address and changes it, depending on how you stored the database, you need your historical records to show the correct address you shipped to. So at that point you need to -- you have to think about the same type of problems of "Okay, now the user can't change their address. They have to create a new one", and is that the use case you want. So I definitely think that there's that problem on both sides of it, it's just a matter of finding the right balance.
+
+**Mat Ryer:** Yeah, it's true. I think you're at risk of not being as rapid. If you use just a SQL data store, you can prototype and you can sort of throw data in and query in different ways. You can do a lot more of that. Having said that, I haven't found that our creativity has in any way been stifled by having a schemaless data store... But it is quite strict. With Firestore, if you want to do a query - even if you're doing an order by one field and filtering on another field, you need a dedicated index for that.
+
+So you need a dedicated index for every kind of query you're gonna be doing, essentially. It creates by default single-field ones for you, and your job then is to go and exclude any that you're not gonna need. But having that - that's quite strange initially, because if you wanna just rapidly prototype in the browser and build things, to some extent you have to know ahead of the time what it is you're gonna get out of the data store, more so than with other technologies.
+
+But honestly, it hasn't slowed us down at all, and of course, the trade-off is if you go to Pace.dev, if you actually play with it and use it, it's lightning fast, and it'll stay fast, because of the nature of that choice. The reason why that data store is so limited is so that if you use it properly, you can deliver massive power. So that's it, you're right, it's just about the trade-off. There isn't a technology that just solves all your problems. There's always gonna be decisions that you have to make, and you're making them at the wrong time; you make them at the beginning, which is the wrong time. You're much better off making decisions at the end, if you can... But of course, you've got to do something meanwhile.
+
+**Jingle:** \[01:07:50.09\]
+
+**Jon Calhoun:** David, Mat, do either of you have an unpopular opinion you'd like to share?
+
+**Johnny Boursiquot:** Is pairing with Mat an unpopular opinion?
+
+**Mat Ryer:** \[laughs\]
+
+**David Hernandez:** Well, it's one of the things that we usually do while we are pairing. Mat gets the guitar and we try to do something completely different and build a little song, or something like that.
+
+**Mat Ryer:** Yeah, just do some songwriting instead. \[laughter\] But actually, an unpopular opinion I have is you should try and work in tiny teams. A lot of the problems when it comes to software engineering come at scale. And that's not just code scale, but people scale. So if you can have tiny little teams working on a problem - and you can do this within bigger teams. Literally, two or three people - you are now a new little team.
+
+You can be so effective in such a small group, because you cut out a lot of the work needed really to marshal the team. You can't always do it, and it sounds a little bit anti-social, but that would be my unpopular opinion. Tiny teams. Do you have an unpopular opinion, David?
+
+**David Hernandez:** I don't have an unpopular opinion. I'm very populist, probably. \[laughter\] Going with the flow.
+
+**Jon Calhoun:** You'd probably have to be "Go with the flow" to work with Mat that much...
+
+**Johnny Boursiquot:** Oh, man...
+
+**Mat Ryer:** I'm getting a grilling on this.
+
+**Jon Calhoun:** I'm just kidding, Mat.
+
+**Johnny Boursiquot:** Yeah, we kind of warned you though. You knew this was coming.
+
+**Mat Ryer:** Yeah, but if you knew David...
+
+**David Hernandez:** \[laughs\] Yeah, that's what I was about to say. You don't know me. I'm congratulating Mat for working with me for a while also. \[laughter\]
+
+**Jon Calhoun:** So you're telling us that if he turned his camera we'd see a blank wall right now, but really if he turns it the other way it's a bunch of unpopular (I don't know) posters, or something? \[laughter\]
+
+**Mat Ryer:** No, he's alright.
+
+**David Hernandez:** I usually tell Mat that he's a terrible developer. His ideas are bad, so I'm not sure why he's working with me...
+
+**Mat Ryer:** Yeah, I thought it was a language thing when you would say that my ideas are a terrible idea. I thought that was just the language, but no; it turns out he does think that, and he will tell you. \[laughter\]
+
+**Jon Calhoun:** Is that why you spend so much time trying to prove him wrong? \[laughter\]
+
+**David Hernandez:** Pretty much...
+
+**Mat Ryer:** Actually, the serious point of just honesty about things is a big shortened when you want to be rapid as well. No one wants to hurt anyone's feelings, but if you can very quickly just have a very open and honest discussion about things, I think it does save a lot of time.
+
+**Jon Calhoun:** I suspect a lot of that comes with working -- like, small team helps, but partially in the sense that you're probably working with similar people every time, so you build a good relationship and trust, and when somebody says "I don't agree with this", I guess it almost feels like you don't have to worry about hurting their feelings as much, or worry about -- like, people trust each other more in those senses.
+
+**Mat Ryer:** Yeah. Well, I've worked in places where that's how I've always been, just completely honest about it... Because that's the idea - we'll all just put the best ideas out, and we can all figure it out as a team. And a few times in my career that's hurt me, where I've just been doing that and I've been kind of politically tone-deaf to other things going, and I've just been quite naive about it, just sort of getting on with it.
+
+So yeah, I then got a bit sensitive about that, because it is important - you want to make sure that the ideas are there, but there are people that hold their ideas very personally, and will feel personally attacked if you disagree with their ideas, and stuff... So it's definitely worth watching out for, but it is nice being in a situation where that's less of a consideration and you can sort of just focus on the important bits. You're right, it's about the trust thing and getting to know each other, getting to know your teammates.
+
+**David Hernandez:** Yeah, it's definitely about trust. I kind of trust Mat, and I can say "This looks terrible. I dislike this completely", or something like that; he doesn't get hurt. But don't forget that when you communicate in open source, when you're war mates, you try to be exactly the opposite, isn't it? You try to just be polite, try to be respectful, and that's kind of the healthy way to do it.
+
+We get so much reports -- I don't know about you, but you can feel the tone in the words, in written words, in many issues... And you know when something doesn't smell right in the tone, in the messages, when something is wrong. Trying to be honest, we try to be very clear and very respectful in that way. Every time that we write to public communication or to each other, we try to maintain that. Apart from joking or healthy behaviors in a team, that is also quite fun to do, we also try to do it up to the best of standards, like we do it in a normal team, or with people that we didn't work together yet. Or if we're gonna hire someone, we'll probably do the same.
+
+**Jon Calhoun:** Alright. Well, Mat, David, thank you for sharing about Pace.dev with us. For anybody who wants to check it out, you can find it pace.dev. If you wanna hear more about why Mat named it that - or Mat and David both, I suppose - you can message them on Twitter as well. Johnny, also, thank you for joining...
+
+**Johnny Boursiquot:** My pleasure.
+
+**Jon Calhoun:** ...and we'll see you guys next time.
+
+**Outro:** \[01:13:48.27\]
+
+**Jerod Santo:** Well, Google blew the world away, because the search engines before that kind of were like steam engines, they were so bad...
+
+**Mat Ryer:** Yeah. I remember Jeeves. I would ask Jeeves all kinds of things and it'd never know any of it.
+
+**Jerod Santo:** Oh, yeah. Altavista...
+
+**Mat Ryer:** Yeah.
+
+**Jerod Santo:** Web -- what was the Web one? It had an actual spider as its logo. WebCrawler?
+
+**Mat Ryer:** Oh yeah, I don't remember... It was a dog?
+
+**Jerod Santo:** I don't know, I just remember the spider.
+
+**Jon Calhoun:** Ask Jeeves just confused me, because -- first off, search was already bad, and they were like "Now we're gonna try to process natural language from questions, and use that for search", and I'm like "How is this gonna get any better? You just made the problem twice as hard."
+
+**Mat Ryer:** Yeah.
+
+**Jerod Santo:** That's why portals were so big, because search was so bad. So you'd just go to Yahoo.com and they would be your portal to the internet, because you could find it all.
+
+**Mat Ryer:** Yeah, and you could browse it. Remember when you could browse just a directory of the internet?
+
+**Johnny Boursiquot:** A whole directory of the internet.
+
+**Jerod Santo:** Yeah, pretty much. It was like one page of links.
+
+**Mat Ryer:** Yeah... \[laughs\]
+
+**Johnny Boursiquot:** Back in the days when I used to be called a webmaster...
+
+**Mat Ryer:** Oh, webmaster, yeah...
+
+**Johnny Boursiquot:** Yeah, don't judge me...
+
+**Jon Calhoun:** I still kind of want that to be my email... Just webmaster@...
+
+**Johnny Boursiquot:** \[laughs\]
+
+**Jerod Santo:** Webmaster... That's funny.
+
+**Johnny Boursiquot:** Good times.
+
+**Jerod Santo:** The good old days...

Grokking Go.dev_transcript.txt

@@ -0,0 +1,569 @@
+**Mat Ryer:** Hello, and welcome to GoTime! I'm Mat Ryer. Today we're talking about Go.dev. It's a user-friendly hub of curated resources for Go, and we're lucky enough to have three of the brains behind it joining us today: Carmen Andoh, Steve Francia (also known as @spf13) and Julie Qiu. Hello, everybody!
+
+**Steve Francia:** Hi!
+
+**Julie Qiu:** Hi!
+
+**Mat Ryer:** We're also joined by Jon Calhoun, a regular. Hi, Jon. How are you?
+
+**Jon Calhoun:** Hi, Mat. I'm doing well.
+
+**Mat Ryer:** What's been going on?
+
+**Jon Calhoun:** Not too much.
+
+**Mat Ryer:** Okay, well let's get started. I wanna hear all about Go.dev, so who wants to give me an intro? What is it, and what's it for?
+
+**Steve Francia:** I can share that. It is a new official Go website for our Go community. It's intended to be a single destination, where the entire community can gather. Today it contains learning resources, it contains a package discovery or a module discovery component, and it also contains different resources to be able to help evangelize Go within your companies, or to new companies.
+
+**Mat Ryer:** And this is not gonna replace Golang.org, is it?
+
+**Steve Francia:** No.
+
+**Mat Ryer:** So they're gonna coexist, those two.
+
+**Steve Francia:** They're gonna coexist, yeah. That's right.
+
+**Mat Ryer:** And the difference then being, I guess, the Golang.org is the kind of open source home, it's the technical home of the project itself.
+
+**Steve Francia:** We struggled a lot... Creating two websites is not always the best thing. Two websites could create more confusion, and we initially started trying to expand Golang.org to contain all these things, and we really struggled with that because the intent of Golang.org - which if you don't know, actually gets distributed with the Go release; most of the content of Golang.org - is to be an official place for the Go project (the language, the compilers, the tools). And what we wanna do is expand it to be something that encompassed all of this stuff for the Go ecosystem.
+
+Eventually, after a few months of trying to figure out how to shoehorn these competing goals into this existing site, we realized that maybe they're best as two separate sites. So the Golang.org is gonna remain the way it is, and the second site is really curated content that's created by the Go community and for the Go community.
+
+**Mat Ryer:** Right. So how did it get started then? Was this something that you internally saw that there was something missing, or was this something that the community asked for?
+
+**Steve Francia:** It's an interesting answer to that question, and I like the way you framed it.
+
+**Mat Ryer:** I'll be the judge of whether the answer is interesting or not, Steve, if you don't mind... \[laughter\]
+
+**Steve Francia:** So we actually started this project over four years ago.
+
+**Mat Ryer:** Really?
+
+**Steve Francia:** As a community member, I was talking to the Go leadership at Google about potentially joining, and it was in a new role that they'd never really had someone do before... So as part of that, I wrote a document that was from my community member perspective of the things that were missing, that we needed to address. And if I was gonna join Google, I wanted to join to build those things.
+
+And on that document, it had - and I'll read it to you - it was "Provide educational resources for Go adoption and best practices, working with partners to create and provide the best training materials. Write the story of the value of Go and communicate that story broadly. Solve the problem of discoverability of libraries and packages."
+
+The idea behind this wasn't necessarily one project at the time, but those were all things that started at that point, over four years ago. Fast-forward -- and I'll also say, that document also had other things; the Go user surveys came out of that document, which we've just finished doing our fourth one of those. Improved IDE experience, dependency management - those were all things in that document. If you look at the things the Go team has done for the past few years, a lot of them were in there.
+
+So now we fast-forward to me joining Google... I worked with Russ and Samir to obtain the right staffing, by presenting these concepts to the Google leadership. And they did, they gave us the opportunity to expand the team, which led to hiring Carmen and Julie, who then did most of the heavy-lifting of bringing the project forward.
+
+**Mat Ryer:** Great. Well, thank you very much for doing all that excellent work. The thing you mentioned about package discovery is quite an interesting one, because for a while in the beginning of Go there weren't that many packages; over time, of course, it's grown, and so now people actually have a lot of different choices if they're gonna bring in some dependencies and things. And it is a little bit like the Wild West... So do you see this as being a way to tame that a little bit and find some of the more trustable open source packages out there?
+
+**Steve Francia:** This is not a Go problem. We think this is a modern software development problem. As you kind of get critical mass, the problem gets bigger, as you're saying... But it's a problem that every language is struggling with right now. Some to an extreme, and some -- I think Go is actually on the smaller end of that. But we do see this as an opportunity to solve this problem for Go.
+
+**Mat Ryer:** And will it take into account things like if packages have stopped being maintained, or if they do things that are generally accepted to not be great patterns, and things like that? Is it gonna be opinionated in that way?
+
+**Julie Qiu:** Yeah, I definitely think it's something that we've talked about and considered adding for the future. Right now, if you think about the Go ecosystem, the thing that's really out there is GoDoc.org, which provides package documentation... But it's missing a lot of this information about "Is this package still being maintained? Is this something I actually wanna integrate into my site?" So that's definitely part of what we wanna do, especially at [pkg.go.dev](https://pkg.go.dev/), now that it's launched, we've laid the foundation... That's all stuff we're thinking about.
+
+**Mat Ryer:** Julie, you spoke at -- I've seen it a couple of times, actually... You do a talk about how to select good dependencies, and what to look for in things. Anyone interested in that should definitely have a search around. It's a great talk, and you talk about things like "Are there tests in this package?" What are some of the other things that are important when it comes to choosing dependencies, and why is that important?
+
+**Julie Qiu:** If I had to bucket it, I would say that the three main things to think about are popularity, quality and stability. By popularity, what we're really wondering is "Are other people using this package?" The reason that that's really important is because it essentially gives you a little bit of a heuristic of "Are there other people out there that care about this?"
+
+Say the author suddenly decides that they don't really want anything to do with this package; what are the chances that someone else is gonna step up? What are the chances that they're gonna look for bugs? What are the chances that if I make a big part of my codebase depend on this, I can count on this being there in the future?
+
+I think quality is things like "Is it a well-documented package? Does the code have tests? Does it look essentially like idiomatic Go code to me?" And you want that, because it gives you a sense of how familiar the package author is maybe with creating Go packages. And essentially what you're evaluating when you look at third-party packages - is this code that I actually wanna put into my codebase? So if, you know, on a Friday afternoon I'm trying to deploy - which we all admit we do - you wanna know that you're not gonna suddenly look at code that looks like something you've never seen before.
+
+And then the third thing is essentially stability. Obviously, technology changes, the Go ecosystem changes, and so as things are changing, can I count on this author to think about where I am with my project? Something you don't wanna see is someone had an exported function in one version, and then all of a sudden in the next version it's not a major release, but that function got taken away... Because that is gonna create a lot of work for you as you are upgrading your codebase. That's how I would go about thinking about it.
+
+**Jon Calhoun:** One of our actual listeners had asked the question "Will any of this data that you guys are using to figure out what packages are there - I'm assuming you'll be gathering some data - be made publicly available?" The specific question they were asking was "When you're looking at packages, one of the ways you can tell if it's a good package to use is if bigger projects are using it. So if Docker is using a package, there's a good chance that they're gonna step up and make sure it doesn't die, whereas if it's a bunch of really small applications, then popularity there might not matter as much as one really big entity." If you're planning on exposing that sort of data and making it possible for people to explore what big projects are using these packages...
+
+**Julie Qiu:** It sounds like that information is -- they're looking for what a package is importing, and then what a package is being imported by. All of that information actually is already available today on pkg.go.dev.
+
+What you would do is for any package that you're interested, you can click on the Imports tab to see what other packages it's using; or on the flip side, you can look at the Imported By tab and see who else is using this.
+
+**Mat Ryer:** That's cool. And that's different to GoDoc, isn't it? That new pkg.go.dev.
+
+**Julie Qiu:** Mm-hm.
+
+**Steve Francia:** GoDoc is just documentation, and the intent for pkg.go.dev is to be much more encompassing of these more rich information around each package.
+
+**Mat Ryer:** Right, yeah.
+
+**Julie Qiu:** Yeah, there is some information about imports and importers on GoDoc.org, but our goal is to take this information and bucket it in a way that's something that people care about a lot more.
+
+One thing that you might notice on the Imported By tab is that we tend to group the importers and count them depending on not just like -- say Kubernetes has one package, and they're importing this one thing 1,000 times. We wanna be thoughtful about how much that weighs into the popularity of the package.
+
+**Mat Ryer:** Oh, I see. Can you explain that a bit more?
+
+**Julie Qiu:** Yeah. So if you think about it like "Why do I care about popularity?", essentially what it's answering is telling me how many people have decided to use this package and integrate it into their codebase. One thing is that you might notice there's one repository - let's call it julieqiu.org/foo - and I have a million packages in there, and they're all importing this one package. So does this third-party package now have a million importers, or should it really have one importer? It's something that we don't actually quite have an answer to yet, but that's the kind of level that we wanna be thinking about it.
+
+You can think about this in sort of like "Maybe I'll count it as a million importers, but the way that users might wanna see it is that all of it is grouped under one thing", so that you can answer questions like "Does one big organization use it, or does one really important module use it", without seeing just a giant list of things that essentially are all telling you the same thing.
+
+**Mat Ryer:** Right, that makes sense. Something else you mention in your talk, Julie, is about looking at dependencies for packages - how many dependencies another dependency has. So if you're gonna important a package - say you've got two options. It might be sensible to choose the one that has fewer dependencies... But why really is that? Why do we care about that? Because it's not our problem, is it?
+
+**Julie Qiu:** Well, the funny thing is that a lot of times it feels like it isn't our problem, until it is. In my talk I gave this example about this package called pad-left; completely hypothetical package... \[laughter\] And it might be transitively depended on by millions of different packages. And you personally never end up using pad-left, but if it got deleted all of a sudden from the Go ecosystem and all that code disappeared and you can never find it again, it could actually end up breaking a lot of your stuff and then you now suddenly have to be like "Oh no, how is this happening? I have to find the tree of things. I was depending on this..." So it can cause you a lot of problems, even though on a day-to-day basis it might not seem like this is something that actually matters.
+
+Obviously, getting deleted is a really bad situation, but there are also other things, like security vulnerabilities, or maybe licensing issues, and other things along those lines that you think a lot about when you're thinking about your own package, but you should really think about it with all of your dependencies, and your transitive dependencies as well.
+
+**Steve Francia:** I think there's an interesting thing here, too... We often reduce things to things that we currently have. The question you asked, Mat, was "Is it better to have more dependencies or less?" and there's not an answer to that question. It's really an "It depends." If you have more dependencies, but they're stable and they're well-tested, and they're high-quality dependencies, I will take that every day over a fewer number of lower-quality, less stable dependencies.
+
+So I think really the question is not "How big is the dependency tree?" but "What is the quality of your dependency tree?" And are we really evaluating -- anytime you import a dependency, it really becomes a part of your project, and I don't think we really think about that as much as we should... That as soon as you do this, you're inviting that code into your home, and now you're responsible for it. So if you're gonna invite them in, you wanna make sure that that's something that you feel comfortable with.
+
+So I don't think quantity is necessarily the right measurement of that. I think quality is really the right measurement, and then multiplied by quantity.
+
+**Jon Calhoun:** That's a good point, because the standard library really is just a set of third-party libraries, it's just one we know that is maintained at a very high standard... Whereas anything on GitHub or whatever - we don't know really know what their standards of quality are... So it's just kind of a gamble at that point. But if you know it's an organization that has those same levels of standards, then you can kind of import those with a lot more -- you can import more of them without as much fear of something breaking.
+
+**Mat Ryer:** Yeah. This also helps package developers, if we think about what criteria people are gonna use, what they're gonna look for in packages... If someone out there wants to roll their own package for something, then there's kind of now a nice little -- I mean, Julie, your talk is great for people that also want to do packages. There are some standards, there are some things that the community now is starting to expect... But I wonder, could this make it more difficult for new packages to emerge, if we start to really shine a light on the packages that are tried and tested? Can you see a world where we actually end up making it difficult for new packages to emerge, or is this sort of selection process gonna just be good for everybody?
+
+**Steve Francia:** I'll take that one. I'll start by saying, if a package solves the problem really well, then there isn't much of a need for another package. The standard library is a great model. When I first started learning Go, I looked at that as the pillar of excellence, and I tried to strive for that in the packages I wrote... And there's a reason there's not a lot of competitive string packages, because the Strings package does a really good job. But when there are ones, it's because they fill gaps that that package doesn't address.
+
+So if a package is stable and well-tested, and does the job well, then we don't have a need for an alternative. We need an alternative when there might be a fundamental shift in design philosophy, or there's things that it doesn't accomplish, and that's a very natural thing that happens. We have a similar debate about companies; the larger a company is and the more established, does it stifle new companies being able to emerge?
+
+There's a big political and philosophical debate around that, but the reality is that over time we've seen that startups do emerge, and they're there to fill gaps, and the larger a company is, and I will translate to packages or libraries - the more established it is, the less it has a chance to migrate and move, and it can't adapt to maybe new requirements that come through... So when the requirements change, which over time requirements always do, it really opens the door for new packages to emerge and do that. I think it's a very healthy thing for an ecosystem to have.
+
+**Jon Calhoun:** I think the JSON package is a good example of this, where the one in the standard library is great, but there have been ones that emerged that sort of solved a slightly different problem. If you don't wanna build a struct to get something that's like six levels nested, there's a couple packages out there that make it really easy to dive six levels in and get one specific piece of information and that's it, from the JSON. And there's other ones that maybe they try to be faster... There's different goals, depending on what you're looking for, and even though you think "Oh, the standard library has it. Nobody's ever gonna compete with that", the truth is people do compete with them because there's specific goals that they're going for.
+
+And even if you look at front-end frameworks for JavaScript, you'd think at some point one of them would have just won out and everybody would have stopped... But they all solve different goals, and if something comes along and it's unique enough and solves a different enough problem, I think we see that that tends to happen, it still tends to get some traction and move up.
+
+**Break:** \[19:59\]
+
+**Mat Ryer:** The Context package, when that happened - that was kind of one of those shifts you talked about, Steve, where suddenly now people expected to be able to cancel things that they couldn't cancel before... For example like copying. Is there a context-aware copy operation in the standard library? If you do io.Copy(), it just copies the whole thing until the end of the file, right? Is there a cancelable one, or do you have to write that on your own?
+
+**Steve Francia:** I am not aware of it in the standard library.
+
+**Mat Ryer:** So use cases like that are -- there's still plenty of opportunity for anyone really to contribute, that's the thing... So I'd hate for people to be put off because standards are going up; that doesn't mean you can't contribute something... And especially like -- Steve, your point about "find the thing that isn't already solved, and that's where you can innovate." I think that's great, and I agree that raising the standards is only gonna be good for -- for most people it's gonna be great, because that is what we care about; we do care about good dependencies that we can rely on, and that are gonna stick around.
+
+**Jon Calhoun:** One of the goals you guys had mentioned was helping companies understand what other people are doing with Go, and I guess learning from that, deciding whether or not to adopt it. I guess this is something that a lot of us probably don't see; at least me, because while I want people do adopt Go, I don't generally go out and talk to big companies and say "Here's why you should use it." So is that something that the Go team is actively pushing? Is that a big goal, to make it easier for people to understand that? Can you elaborate on what people are looking for, how you can help them? And if somebody is looking to maybe talk their company into using Go, what you'd recommend?
+
+**Steve Francia:** We've spent a lot of time meeting with different companies, trying to understand what their needs are and what they're struggling with. I'll also say, one of our big goals as the Go team is adoption. We want Go to be used as widely as possible. And as part of the adoption journey, which we're all familiar with that graph that starts slowly, and then there's a chasm, which is early adopters, and then there's the big chasm at the end of that, which is kind of mainstream... There's different stages to it. And as a language growth, the enterprise is really that next big-cap for Go to be able to tap into adoption more.
+
+We really got lots of adoption on startups and hobbyist very early, and so we recognize that the enterprise is important to get Go adoption to really fulfill its potential. As part of that we're doing lots of research, and we've been talking to lots of companies, from all different walks - from retailers, to banks, to multimedia companies... From every different style of company and in lots of different continents.
+
+And through those conversations we've heard two things that almost every company asked us, two questions almost every company asked. These are companies that are either thinking about using Go, or they've done a prototype with Go, or they've adopted it for some projects... So the two questions that everybody asked was "Who else is using Go?" and "What are they using it for?" As we talked to them, we heard lots of different stories which really got us excited. They talked about prototypes that they wrote in Go, and because they tried to talk to their management about it, they had challenges, and they thought Go would be a good solution for those, their leadership, the technical decision-makers wanted to know "Well, who else is doing this, and what are they doing with Go?" and they didn't have good answers to those questions.
+
+So a lot of those stories stop there. But some of those stories continued, and we had people that felt confident enough that it was a good decision that they kind of went rogue, and they built a prototype in Go, and it ended up being phenomenally successful.
+
+So those pioneers in those companies really provide the things that they wanted before, which were these stories... So we're really happy that we're able to tell some of these stories. We've been working with these companies for many months to be able to capture these stories that they shared with us, and make them publicly available, and share them broadly. And hopefully, the next round of people answers to those questions, so that when they wanna go to their leadership and answer these questions "Who else is using it? What are they using it for? Are they in our demographic of company or vertical of company?" they'll have answers to those questions.
+
+It wasn't just in these internal meetings, we also did different surveys, and this was one of the top things that people asked for in our surveys. Now, if you're coming from a small company or you might be an individual consultant, this probably sounds foreign to you. But if you're working at a big company... You know, on our website we've launched several case studies from big companies: American Express, PayPal, MercadoLibre... You can go on the website and see them all there. These are larger companies that have more hoops to jump through, more permissions to obtain, and this was an overwhelming thing that they've been asking for.
+
+**Mat Ryer:** Yeah. It's interesting, if you go to Go.dev, there are a lot of logos, but they're not just -- sometimes you see these on websites and you feel like they're sort of brags, or something... But these are actually -- you can click these and go and read about the actual ways in which people have used it, and what they're doing there, too.
+
+**Carmen Andoh:** You sure can.
+
+**Mat Ryer:** You can, you can.
+
+**Steve Francia:** You definitely can.
+
+**Carmen Andoh:** Yup.
+
+**Steve Francia:** I think they're very relatable stories, and I think they're very inspiring. As you click through and read these articles and these different stories... I remember back -- I first adopted Go when I was working at MongoDB, very early on. And there was not a lot of these stories that existed yet, but in using it I fell in love with it. It lit a spark, and showed me potential of what a programming language could be... Which led me to work at Docker, which was a big Go user, and then eventually on the Go team at Google.
+
+These stories shared a similar thing. If you've been using Go a while, read them; it'll remind you why. It'll take you back to that moment when you had the first a-ha and fell in love with the language.
+
+**Carmen Andoh:** But the other thing that I wanted to point out about these case studies that I think are really useful is when I learn how to code, I am not taught how to influence my manager, or influence my higher-ups on choosing Go. That's just not something that they teach you in school... And this is just the perfect thing that I can point to people, saying "I wanna choose this technology, and here's why."
+
+I think so many people, when I had been in contact with them over the years, like "How can I convince my manager?" or "How can I convince my CTO?" or whoever in the organization, to use Go. Certainly, there's the rogue tactic that Steve mentioned, but this now doesn't -- you don't have to worry about going rogue, or taking yourself out of the critical path in order to get it adopted. You can simply go "Here, read the testimonials" and then make a commitment to trying maybe one service at a time, or refactoring one corner of your codebase here and there.
+
+And what's great about these is that some of them are more detailed than others, and they can give you a blueprint for how you might wanna do it, to certain extents... Whether you wanna go whole hog with Go, or whether you wanna maybe just instrument some of your observability toolset with Go, or you wanna do some of your automation.
+
+So I really, really liked that, and I tell people now that certainly learning how to influence is a vital skill that they don't teach in school, but this is super-helpful.
+
+**Mat Ryer:** Yeah, I spoke to somebody at the last London Go meetup who -- they were reluctant, because learning a new language sounds like quite a bit thing, especially for people that maybe only know one language; that was this case. But learning Go, especially if you are already familiar with the kind of C base language. Learning Go I think is easier than some other languages, actually, because of its minimalist nature... So I always kind of encourage that.
+
+And I also like this idea of using it to solve a real problem you have, even if you're not 100% certain of what it's gonna turn out like. When you learn the context that you have, when you're trying to learn something, if you've got context around a real problem you're trying to solve, it really focuses the mind. You don't go down the rabbit holes learning the details of how channels work, or how to do struct packing for optimum memory use. You focus really on the bits that are important in your case... And that was what this person was telling me - their experience. They were surprised how easy it was to pick up and to apply when solving some little problem they had. And I think it was a tiny little problem, but they loved that experience, and then they did a presentation to the rest of the people - anyone that would listen, really - in the company, to show it off. That's quite a nice way to also evangelize for it, too.
+
+**Carmen Andoh:** And to add further to that context, I also like the idea -- when we talk about adoption, there are different mindsets, depending on where you're at within your company and where that company is at. Sometimes what I often heard was "I don't wanna just learn about theory" or "I wanna see what it looks like in practice, from soup to nuts." And I think that's the other thing that these case studies also -- they have some level of detail that can tell you, hold your hand from end-to-end.
+
+Some people like to just go around and play and find what works for them, but others just need to see what it looks like in practice, and see how that matured over time. I think the case studies is one of my favorite pieces. Well, I like the whole site, but I really sing the case studies praises, for lots of reasons.
+
+**Steve Francia:** Moving forward with these case studies - we launched with a handful of case studies and articles that were published on external websites as well, and we're excited to tell these stories more; to tell additional stories, to tell deeper stories, and we're hoping that some of the listeners today are from companies that wanna show these stories. We'll give two pieces of advice for that.
+
+One is you don't need us to tell the story. A lot of these companies, like Capital One, published several stories on their own blog about their journey to using Go. And we link to them here. So feel free to tell your own story; we'd love to link to it and give more exposure to it. Also, we're happy to meet with you and to learn more about your experience and your story, and the best way to get in touch with us is if you go to Go.dev, on the very bottom, in the footer, there's a Share Feedback link. Please fill out that link and tell us if you're interested in working with us on a case study, or writing an article.
+
+The most important thing - that's an anonymous button, so if you want us to actually get a hold of you, you have to put some identifying information in there, so that we have a way to reach back out to you, or else \[unintelligible 00:35:24.27\] Which we've had a few companies do. "Oh, we're really interested in the case study, and doing something" and then Submit.
+
+**Mat Ryer:** Okay... \[laughs\] So put your name and email in, or something.
+
+**Steve Francia:** Do it someway.
+
+**Julie Qiu:** And also, if you wanna just send us a non-confidential email, you can feel free to email go-discovery-feedback@google.com. In case you can't remember all of this, we have an About page on Go.dev. Go.dev/about has all of this information.
+
+**Mat Ryer:** Thank you. We'll also put some information in the show notes, too.
+
+Will you have videos on there, and talks, and things from conferences, do you think?
+
+**Steve Francia:** As we said in our blog post when we launched this - it was really us getting what we felt was like a minimum viable product out to the community, as early as possible. You'll notice on the website we say it's for the Go ecosystem, by the Go ecosystem, but it doesn't have a ton of community resources on there yet. There is intent to add those items and to make it more of a place where the whole community can come together and learn more about conferences, and meetups, and talks, and really be a resource for them.
+
+**Mat Ryer:** There's an Event section on the homepage, isn't there?
+
+**Steve Francia:** Yeah. And it only gives you those three events... And I think they're only meetups. It's not conferences yet. So it's a start, and we're really happy that it's there, but we have so many meetups around the world that -- if you look on it today, it says January 11th, it has three meetups, but there are actually more than three meetups on January 11th. So it does give a taste to it, but we really wanna expand that further and really make it so that whether you're in Melbourne, or Bangalore, or Eugen, Oregon - which happen to be the three on the website today - or anywhere else in the world, it's gonna give you when your upcoming thing is, when call for papers or participation is due etc.
+
+**Mat Ryer:** Yeah, that's great. That's gonna be really great to have that, because it's difficult even for those of us that have been in the community for a while to know what's going on. I think it's not just gonna be a good resource for new people; I think for everybody.
+
+**Steve Francia:** It's also partly solving the problem that kind of crept up on us... If we look back two years, there was -- I don't remember the exact number, but around a dozen (maybe a little more) Go conferences. And this year, in 2020, there's scheduled to be over 30 conferences throughout the year. That's three a month. That's pretty overwhelming. That means there'll be more weeks with conferences than not this year. So with all of the excitement that's happening around the community, it's really helpful to have a place that organizes and gathers that.
+
+We're not gonna have that delivered in the next month or two, but it is on our roadmap and it's things that we're intending to do.
+
+**Jon Calhoun:** One of the questions that I've seen people ask a lot, either on Twitter, or even on our GoTimeFM Slack - this is sort of intended to eventually be a resource for the community, so they're all asking "Are there plans to make it open source, or to give the community a bigger role in the project?" I get early on it's usually easier to not do that with a lot of things, but I didn't know if you could share plans for the future with that sort of thing.
+
+**Steve Francia:** It's something we're looking into. We definitely wanna make sure that whatever we do, it's the best thing for the community and our users, and serves their needs the best way. We're trying to make sure that we're doing that in the best way possible. So there's ongoing discussion; it's actually part of our Go developer survey, we ask specific questions around this... So we're ongoing doing research to try to better understand our users' needs, and making sure that we're satisfying them as best as possible.
+
+**Julie Qiu:** Yeah, and something that we also are planning on doing pretty soon is opening up the Go issue tracker to accepting feedback. We already described that we have two different channels to get feedback - the email, and also sharing feedback in the footer, but obviously, these are all private topics, and we've heard everyone sort of saying they wanted a more public forum to be able to have discussions... So we're actually working on a process for that, and are gonna be sharing that pretty soon. That's another way that we really wanna be clear that we want people to share their feedback of us, like this is something that we wanna build for the Go ecosystem.
+
+**Mat Ryer:** That's great. In your pkg.go.dev, how do you decide what's a popular package, and how do you decide which packages you're gonna feature on there?
+
+**Steve Francia:** So you're saying on specifically that page...
+
+**Mat Ryer:** Yes. Because that's gonna be essentially -- I mean, packages listed there are gonna be the ones people are using probably, right? ...eventually, if not already.
+
+**Steve Francia:** We hope so. To some degree, they're already packages people are using quite often. That's why they're there.
+
+**Mat Ryer:** That's why they're called "Popular Packages", is it?
+
+**Steve Francia:** Yeah, that's how we got the name. That's how we came up with it.
+
+**Mat Ryer:** It makes sense.
+
+**Steve Francia:** Believe it or not, there were several meetings to -- no, I'm just kidding. We just did that.
+
+**Mat Ryer:** \[laughs\]
+
+**Steve Francia:** So Featured Packages is a little bit of a curated one. They're largely popular packages as well, but they're ones that we thought fit a niche, or address needs that people were looking for... And then Popular Packages is just the popular packages from the database, based on import count.
+
+**Jon Calhoun:** When you're looking at those curated lists, I know there's always these -- they're like Awesome Go, or Awesome... You'll see different lists like that on GitHub, that list a bunch of packages grouped by what they're for. Some will be graphical user interfaces, others will be like database packages... And generally speaking, they do a little bit of curation, but I feel like sometimes they just throw anything and everything in there.
+
+I guess one of the questions I'd ask is how do you guys draw that line between -- you know, you don't wanna reject people or be a gatekeeper, but at the same time you need to... Like, just having everything listed in one place isn't necessarily useful... You know, just listing every single package that can connect to a database or something might just be overwhelming to users. So how do you navigate that middle ground?
+
+**Steve Francia:** If we look back to the beginning of search engines, the early days of the internet, early '90s, mid-'90s, you might recall that Yahoo! was one of the leading at the time, and they did it by doing the website directory, that was human-curated. And it worked well for a time. In fact, the reason everyone else struggled is you couldn't really compete with the quality of that.
+
+Then Altavista came out. And Altavista had accuracy and quality, and it was fast. Well, it was fast... It was fast until people started using it; there was a time when it was blazing fast, and everyone was excited...
+
+**Mat Ryer:** Yeah, but Steve, all software is fast until people start using it. It's the people using it that ruins it. That's why we have to care about making things work...
+
+**Steve Francia:** Well, then Google came around and figured out how to solve that problem too, of making it fast, and quality, and accurate. But I think there's a lot to learn from that experience.
+
+Awesome Go was and is still a great resource, but at the beginning, when packages were smaller, and there was less of a list, I think it was easier to maintain it and to keep track of it. As that list grows, it's harder and harder for humans to keep on top of it. So what Julie talked about earlier was these signals, these visual indicators on packages -- I think the solution here is not to maintain manual lists, but really to have dynamic things.
+
+If you're searching for -- you know, sometimes you're searching for things that Awesome Go curated predefined categories, but sometimes you're searching for things that aren't in those predefined categories. And no matter what you're searching for, you really want to know quality... And I think it comes back to those indicators, to really help us. That's the discovery portion of the site, that is still ongoing, that we're working on.
+
+**Mat Ryer:** I once wrote a blog post and made a little repo alongside it to show the code... And I made some changes to it at some point a couple years later, and I started getting people opening issues, saying "You broke our build." This was just a repo to show off some ideas, it wasn't ever meant to be imported by anybody, so I was shocked that that happened. Is there a way that package writers and package authors and maintainers can indicate to the tools Go.dev and to pkg.go.dev - is there a way that we can indicate that these things are deprecated, or perhaps shouldn't be imported, or even to say "If you wanna solve this problem, there's a better package over here, so go and use that one"?
+
+**Julie Qiu:** I think to indicate that something isn't working anymore, you can email us. We've gotten requests for people to take down their packages from pkg.go.dev. So that's something that we do support doing.
+
+I think in the future -- some things that we've been discussing are like, say when someone archives their repository, for example, or deletes their repository, even if we might have the code for that to provide some kind of flag, so that people know about it.
+
+We don't currently have anything right now on the side for authors to say "You should use this package instead." I think that that feature would require some sort of thought about what that user experience would actually look like, or if that's something that we even want people to be able to do... So those are kind of the avenues right now.
+
+**Mat Ryer:** Yeah. I mean, I would imagine something like a .go.dev file in the repo root, or something, where it could potentially have some metadata in there where we could communicate that. There are a few examples of that working quite well, where the tooling can notice those things. It's probably just that - a few things like "Check out this package, these alternatives..." Because even though it might be nice for my ego that people are using some package I've written, if for whatever reason there are better packages, which believe it or not, does happen, I just want people to use those packages. I don't want them to just use mine for the sake of it. So I would be happy to go and put a bit of effort in if that meant that the whole experience quality goes up for everybody.
+
+**Julie Qiu:** Yeah, I think we've also discussed very early on during the brainstorming phase about other types of metadata that would help with that problem, too. For example, keywords would be great. If you could tag this package as like "It's a logging package", and so maybe that's what your package is, and that way you don't have to say "This is exactly the package you're looking for", but you can just have keywords for people to look in the ecosystem... But these are all things we've been thinking about and are still brainstorming about, and don't quite know what it will look like yet.
+
+**Mat Ryer:** Cool. No, but it's exciting though.
+
+**Julie Qiu:** Yeah. There's a lot of cool places that we can go.
+
+**Jon Calhoun:** It sounds like the experience you've described, Mat -- it's one I'm familiar with too, where you're kind of doing educational material, and you want them to have something that compiles and runs, but at the same time by making it something that compiles and runs, it also means somebody can import it and use it, and it's like "Well, this was clearly meant to teach, not to necessarily be the thing that does that." So you might show somebody how to make an HTTP router, but that doesn't mean that they should go use that one. There's some way more stable ones out there that you should check out. So that becomes a little bit trickier, but... Maybe just having a way of tagging things is like "This is a learning resource, rather than something else that could actually help with that."
+
+**Steve Francia:** For this specific one -- I think you bring up a good example of different needs that we might have, that expand beyond just the static metadata, of like readme and license files that we currently use. For this specific one, if you don't want someone to import it, change the license to something that is dot-something-that's-very-importable. It might be a solution to this specific problem.
+
+**Mat Ryer:** But the tools won't guard against that, will they? ...assuming that everyone checks their license before they import a package...
+
+**Jon Calhoun:** The other issue I have with that is let's say I'm teaching somebody how to do something... A lot of the times I want them to have the freedom to take chunks of that code and use it. And if I say "This is some really restricted license", then all of a sudden they're like "Well, I can't use what I learned here without fearing that I'm gonna get sued later for using this code." And you kind of have to worry about that, where I know most instructors or people who do educational stuff basically just open source everything with an MIT license, because they don't want people to feel like everything they learned is something that they're worried about using that code.
+
+**Mat Ryer:** Yeah. On the other hand, Jon, you've just thought up an excellent scam.
+
+**Carmen Andoh:** Don't give ideas.
+
+**Mat Ryer:** \[laughs\]
+
+**Steve Francia:** So the license question - I give that answer mostly tongue-in-cheek, because I think everything you said, Jon, is right. But we've heard from a lot of -- while maybe not every individual is doing this, companies are very concerned with this. License compliance is a huge deal, because it can really get you into a lot of hot water. Every company and individual should be concerned with it, but the larger the company is, the bigger the defensive legal staff needs to be, and the more concerned they are with this. All of the Go tooling certainly doesn't do this today, but we did build in more license awareness into Go.dev.
+
+**Mat Ryer:** Yeah, it reports the licensing on all the packages, doesn't it? And does it exclude things where you don't recognize the license?
+
+**Julie Qiu:** Yeah, so it excludes certain content, but not exactly the repository itself. So the way that we make that distinction is based off of whether or not it's factual information about this repository, as opposed to content that we are taking and editing in some way. For example, what that package imports is not something that we would exclude, or something that imports it... Because that's just factual information about it. Or the last time it was released. But things like its readme, and its documentation are things that we consider content that we are not able to redistribute.
+
+If you look on pkg.go.dev and you're thinking about whether or not you should use a package, and you go to the documentation page and we don't think that its license is something that is redistributable, it will become really obvious really quickly.
+
+**Mat Ryer:** So therefore on the other hand if you do want things including, then pick a license that is gonna allow this.
+
+**Julie Qiu:** Yeah. That's something that we had gotten a lot of feedback about right after launch, because our license policy had been a bit strict... And also, I think we haven't provided a lot of information about exactly what information you need for your license... But we actually updated our license policy very recently. So if that's something that you want us to know more information about, we've shared that we do license detection by using the license check library, and also a list of licenses and a copy of that content in case you wanted to pick one of the ones that would just fit on the site.
+
+**Mat Ryer:** Great, yeah. There is actually a Go tool - because I've used it - which will check the licenses of all your repos as well. So I'll dig that out and put it in the show notes too, because that can be quite useful. And it's actually worth bearing in mind from the beginning, because it's all very well -- usually, what happens is you build the thing, you get it working, and then towards the end of the process someone from the legal team will say "Oh, just make sure the licenses make this all okay." And then if it doesn't, you kind of get yourself into a little bit of trouble... Or at least there's work to do to go and either find an alternative, or you sometimes have to rewrite the little bits and pieces yourself. So yeah, it is worth checking the license, as Steve was alluding to earlier. Check the license before you import.
+
+**Jon Calhoun:** I'm surprised somebody hasn't come up with something like goreturns, that you can customize for each company, that basically just does that when you're saving... Because every company has a different set of policies around what they'll allow, and it would be kind of nice to just have your code flag it as like "Hey, you can't import this. It's not gonna work."
+
+**Mat Ryer:** Like a compiler-time error.
+
+**Jon Calhoun:** Yeah. Because if you just had something, it would just build it into that; it'd be kind of nice to have.
+
+**Mat Ryer:** Hm. Nice idea.
+
+**Carmen Andoh:** That sounds like a slick idea, for sure... Especially now, with software engineering in this day and age, and all this reuse, and the risks that carry with that.
+
+**Jon Calhoun:** Generally, I think people look on GitHub and they think "Oh, it's open source", but that doesn't necessarily -- people don't always put licenses there, and something being open source doesn't necessarily mean you can use it for what you wanna use it, especially if it's commercial.
+
+**Carmen Andoh:** Right. And the average software developer is not a lawyer... So they're just pulling it in, saying "Hey, let's see if it works." Yeah... \[laughs\]
+
+**Jon Calhoun:** I'm sometimes not even sure if the lawyers know for sure.
+
+**Carmen Andoh:** That's accurately stating the state of affairs in this day and age.
+
+**Mat Ryer:** What language is Go.dev written in? I caution you to be very careful here...
+
+**Steve Francia:** Elixir.
+
+**Carmen Andoh:** \[laughs\] Steve...!
+
+**Steve Francia:** What? Oh, are we not saying that?
+
+**Carmen Andoh:** Redact... Redact...! No... Troll.
+
+**Julie Qiu:** It's Ruby on Rails.
+
+**Carmen Andoh:** \[laughs\] I thought it was Haskell... It was Haskell, right?! \[laughter\]
+
+**Mat Ryer:** So what's the real answer?
+
+**Julie Qiu:** It's Go.
+
+**Mat Ryer:** It's Go! We'll play some celebratory music there.
+
+**Carmen Andoh:** Of course it's Go!
+
+**Jon Calhoun:** Can you talk about more of the technology? Is it an API, or are you using templates on the back-end? Can you talk a little bit about what you're using to build it all?
+
+**Julie Qiu:** Yeah, so the entire back-end is written in Go, and then the front-end is just all Go templates. The majority of it is HTML and CSS. I think for a really long time we had absolutely no JavaScript. And even now, it's pretty limited.
+
+Then the site itself is hosted on the Google Cloud Platform. The high-level architecture is we have a system that we call our data ingestion system, which essentially extracts data from the module mirror, and then transforms it, and then puts it into a Postgres database, which is hosted on Google Cloud SQL. Then the front-end - we also just have a service that is pulling data out of this Postgres database, serving requests... We have Redis, which we use for caching, and... Yeah, that's a pretty high-level overview of it.
+
+**Mat Ryer:** Is it Google App Engine, or...?
+
+**Julie Qiu:** It is, yeah.
+
+**Mat Ryer:** Yeah. I use App Engine all the time, I love it.
+
+**Julie Qiu:** It's very easy for deploying and scaling and all of that... So it's been kind of nice, especially given that we had a small engineering team working on it.
+
+**Mat Ryer:** Yeah. And it will scale significantly as well, won't it? That's something that's nice, especially when you're not really into the operational side of things. You can sort of not worry about it, so yeah, I'm all over that... It sounds great to hear.
+
+**Break:** \[54:53\]
+
+**Mat Ryer:** So, actually we have a new regular part of our show, and it's gonna get its own jingle as well... It's Unpopular Opinion.
+
+**Jingle:** \[56:20\]
+
+**Mat Ryer:** So we're gonna ask "Do you have an unpopular opinion that you'd like to share?" Anyone?
+
+**Julie Qiu:** I can start...
+
+**Mat Ryer:** Please.
+
+**Julie Qiu:** This just comes to mind, because people on the Go team make fun of me for this a lot, but... My unpopular opinion is that the New York City buses are the best way to commute across Manhattan.
+
+**Mat Ryer:** Oh, that does sound controversial.
+
+**Carmen Andoh:** Better than cabs, better than subway. Take the busses.
+
+**Julie Qiu:** It's so great!
+
+**Mat Ryer:** Really?
+
+**Julie Qiu:** Yeah. It's basically an Uber Black car. It's like a giant car, it's come here, it's picked you up, it's got Wi-Fi, it's got views...
+
+**Carmen Andoh:** \[laughs\] We're waiting for New York City mass transit to disrupt...
+
+**Julie Qiu:** There are new seats on the M14 now...
+
+**Mat Ryer:** You're joking!
+
+**Carmen Andoh:** Nice.
+
+**Julie Qiu:** It's so great!
+
+**Mat Ryer:** That's a great one. Steve, do you have one?
+
+**Steve Francia:** My unpopular opinion is that I think Windows is the best operating system... And it was proven unpopular in preparing for this podcast. \[laughter\]
+
+**Jon Calhoun:** So for anybody who isn't familiar, when we do these episodes, every guest records their own audio, just so we have a little bit better quality... And I think Steve is our first guest with Windows -- or at least our first guest with Windows who made me help him set up the recording, so I didn't know how to do it, and had to figure that out.
+
+**Mat Ryer:** Steve's the first modern-day programmer I've ever met that uses Windows, actually... So Steve, yes, that is an unpopular opinion.
+
+**Steve Francia:** I use the other operating systems, too. I'm not exclusive to Windows. But I really like Windows 10, I think they've done a really good job with it. I like Windows Subsystem for Linux, and I've got Bash in my Windows and I really feel very comfortable with it... I do develop on it, it's my primary development environment, but I also -- it's nice that I can do photography work, and video editing...
+
+**Mat Ryer:** Minecraft...
+
+**Steve Francia:** I don't do Minecraft on it, but I might play the occasional game, and Windows is quite good at that, as well.
+
+**Mat Ryer:** What's that little game with the grid, where you find the bombs?
+
+**Steve Francia:** Oh, Minesweeper.
+
+**Mat Ryer:** Minesweeper. Let me do that again, and we can edit that in. Minesweeper... \[laughter\]
+
+**Steve Francia:** Oh, I like the way it was. I thought that was better.
+
+**Mat Ryer:** No, but Minecraft's on everything; I think Minesweeper is only on Windows. And I miss it. That's the thing I miss from Windows.
+
+**Steve Francia:** Honestly, I don't even know if it still is on Windows. Let's find out.
+
+**Carmen Andoh:** 1997 is calling Mat, and it wants you back there.
+
+**Mat Ryer:** I liked XP...
+
+**Steve Francia:** No, it does not ship with Windows anymore, apparently... I just searched for Minesweeper and it did not...
+
+**Carmen Andoh:** It wants to party like it's in 1999.
+
+**Jon Calhoun:** Apparently, people weren't productive enough at work, so they had to get rid of it.
+
+**Mat Ryer:** Yeah. I once dual booted my Mac so I could play the game of Minesweeper.
+
+**Carmen Andoh:** Wow...
+
+**Jon Calhoun:** You couldn't find one online somewhere?
+
+**Mat Ryer:** No, I didn't have the internet. \[laughter\] Yeah, XP was alright... But actually, I know that they've put a lot of effort into Windows recently, for developers and stuff. And of course, you can be successful with Go on Windows, can't you?
+
+**Steve Francia:** Yeah, Go honestly was the language that let me shift to Windows full-time...
+
+**Mat Ryer:** Said nobody ever? Steve, you're the only person in the world where that's happened...
+
+**Jon Calhoun:** He's not the only one.
+
+**Steve Francia:** I'm not the only one.
+
+**Mat Ryer:** Guaranteed.
+
+**Jon Calhoun:** I think Brian Ketelsen uses Windows every once in a while...
+
+**Carmen Andoh:** Yeah.
+
+**Mat Ryer:** No, but Go led you. Go was your gateway drug to Windows, you know what I mean? \[laughter\]
+
+**Steve Francia:** From my personal experience, other dynamic languages and other languages were a little more cumbersome. And I'm not a Windows, Visual C++ programmer, or .NET programmer. So using the more dynamic open source languages, I always found it was jumping through hoops, and you'd find edge cases that nobody else was hitting... And then Go just worked. And I could cross-compile from my Windows machine for all the Linux, and Mac...
+
+**Carmen Andoh:** You're like a walking advertisement for Windows. We'd better call them to sponsor...
+
+**Mat Ryer:** I can't believe you just said it just worked. That's the Apple slogan. \[laughter\]
+
+**Steve Francia:** That's the Apple slogan?
+
+**Mat Ryer:** It's the Apple slogan, yeah.
+
+**Steve Francia:** Well, they should live up to it more.
+
+**Mat Ryer:** Yeah. \[laughs\]
+
+**Jon Calhoun:** So did you ever use Java before then?
+
+**Steve Francia:** I have spent my entire career avoiding Java.
+
+**Mat Ryer:** Congratulations.
+
+**Jon Calhoun:** Okay. I was gonna say, Java was one of the few languages where I didn't have much issue using different operating systems.
+
+**Steve Francia:** You had the same issue across all operating systems, you mean...
+
+**Jon Calhoun:** Exactly. \[laughter\] But that was part of the reason why I learned Java when I was in college, and it was kind of the language I stuck with then... But then later I learned Ruby, and that one would have been terrible. Because I tried it on Windows and I was just like "Nope. This is not working."
+
+**Mat Ryer:** Yeah, met too, actually. Ruby was the reason I got a Mac. I bought a Mac so I could do Ruby on Rails. Visual Studio though, I have to say, was - and I think still is - very good for if you're doing C\# or any of the .NET flavors of things. Visual Studio was just amazing, really... And of course, VS Code, which is Microsoft - I think that is the most popular editor for Go on the Mac for people, still.
+
+**Jon Calhoun:** Okay, so we have a little bit of time left... Do you guys wanna talk about the last aspect of Go.dev - the learning side?
+
+**Steve Francia:** I was hoping we'd get to it, because that's Carmen's show.
+
+**Carmen Andoh:** Is it my show?
+
+**Steve Francia:** Oh, yeah.
+
+**Carmen Andoh:** Alright...
+
+**Steve Francia:** And it's also kind of your show.
+
+**Carmen Andoh:** It's also kind of my show. I am both a guest and a host today. Pretty weird. Learn.go.dev - yes, it is what I would call the part of go.dev that I am really excited about, and is ripe for the most community collaboration, contribution and ownership. Some of the original feedback when it first came out was "Why didn't my site/my YouTube channel make it in?" and to that, I say "Let's talk." Because I really wanna be able to do this.
+
+The one thing that we're finding when we're doing -- I did a lot of research and put a lot of work into wondering "Okay, how can we make this useful?" is finding who are the people that we're forgetting about... What we've found there was 1) zero coding experience. We're not even starting with Go as a second or third language... So we partnered with Codeacademy to provide that... Meaning "I know nothing about coding at all, and I wanna try Go as my first language." And Codeacademy - that's their bread and butter, so we partnered with them, and it was a really good partnership. There's probably -- the data that I can share is that we have roughly 70,000 people since they launched that course, and we made it available free, through a sponsorship.
+
+**Steve Francia:** That's a huge number, 70,000 people.
+
+**Carmen Andoh:** It really is. I'll share more later, but yeah... I get weekly reports on that. And there's four modules for free; the whole course is eight modules. You can go to codeacademy.com.
+
+The other gap that was missing was people working in companies that they just -- for whatever reason, they didn't want... The tour of Go wasn't working for them, or other self-learning sites - Jon, Gophercises - and also other things... They wanted to be handed the problem that they wanted to solve. Like, "How do I do X in Go?" Or "From Java to Go for why?" So we just looked at some of the things that many people were using Go - the specialties, verticals, if you will - and we chose the four most common based on the data that we have via surveys and other research, and we decided to find and provide curated learning journeys for those learners in particular, and we got really good feedback for that. Thank you for taking me step by step.
+
+Because what we found was there was two different psychological mindsets for adopters. If you've heard of Crossing the Chasm book - there's early majority/early adopters, late majority/late adopters, and there's a different psychology with each one of those. And the psychology that we're finding with Crossing the Chasm is "I wanna just explore, I wanna learn, I wanna be given the space to kind of take the time to learn it..." Versus the late majority adopter, what we call enterprise, and that is "Just show me how to do it in Go."
+
+So that's what the idea behind the curated learning journey is, based on these very specific "I have things that I wanna do." So plans in 2020 are to continue to partner with more of the community to help find more gaps for those people for whom all the existing things aren't working. We wanna keep it free, and we wanna be able to make sure that all the different -- what we're finding is that you ask ten people "How do you learn?" and you're gonna get ten different answers. There's so many different learning modalities, and we wanna be able to have a variety of learning modalities that appeal to a lot of people.
+
+But the last thing is - and this touched on the events that we talked about earlier - the best outcomes happen when you learn together in-person, in a group. That's really hard to do, but we're hoping to leverage maybe meetups or online meetups. In-person doesn't mean that I'm right next to you, it's also leveraging tools like VS Code for pairing, and then learning together or going through together and hand-over-hand testing things out... So we're trying to look into that a little bit more.
+
+So I've asked a lot of opinions... Jon has a learning site, and I actually wrote to Jon; I didn't know that I was doing it in that capacity, because we hadn't make Go.dev public at the time... But I just wanna ask people's feedback; I've tried to ask a lot of different people's feedback for various things, and I'm gonna continue to do so, so that we can continue to make learn.go.dev really what we envision it to be, which is a collaborative work of art that is quality, and serves a variety of learners to bring in the next two million.
+
+**Jon Calhoun:** Can I just add...? I know one of the things, at least from my perspective, that got me excited about seeing that site is that one of the things that's at least hard for me is to reach people who have limited accessibility. Let's say they speak another language, English isn't their first language, or they need transcripts if they're doing videos, or something like that... I know at least for independent creators, that's a challenging thing, but I know that as the language grows, sometimes people will learn Go better if they can learn it in Spanish or whatever language is native to them... And I like seeing something with Google behind it - or it seems like Google's behind it at least - in the sense that I feel like it opens more doors for that reaching a much larger audience, that it would be much harder for somebody like myself or anybody else to really reach.
+
+**Carmen Andoh:** That internationalization is in our future, and it's also been the thing that I've seen at some of the bigger conferences. We have people like Friends of Go, a company based in Spain, that wrote back and said "Hey, we have this training for Spanish speakers", and we also have some trainers in a variety of countries, including India, parts of Asia, and then in Europe, that have also said "Can we collaborate?"
+
+So if you wanna go fast, go alone; if you wanna go far, go together... So the name of the game for learn.go.dev is really seeking the feedback that we need to seek, and making sure that we get both representation, as well as quality.
+
+**Mat Ryer:** Yeah, that's great. And do you see that being a community aspect of this, too? Will it stay curated completely, or do you ever imagine people being able to vote up things?
+
+**Carmen Andoh:** We go round and round on that. Some people say "Oh gosh, if we could vote up, vote down, that is gonna be its own quality indicator of itself." The problem is everything can be gamed... So right now the only thing that can't be gamed is people whom we can trust to be ethical about curating things, and also making sure that we constantly go back and feed into what the global community needs in terms of learning gaps, whether that's content gaps, modality gaps, whatnot.
+
+So for the forseeable future, it's going to be curated. If at such time we could find a way to do voting that we don't feel can be gamed, or turns into -- you know, the thing I wanted to stop and prevent was someone saying "Go vote on my thing, because you're my friend", versus "Go vote on my thing because you personally learned from it, or you've found it to be very beneficial for you."
+
+But we've gone round and round on it. I remember having a working group or a roundtable at GopherCon in July in San Diego, and then we had about 15 people show up to two sessions, and this was an idea that came up... Someone really wanted to push forward on that. I continued to research it, and look into it, and ultimately said "Not right now", until we can answer some of the gamifying questions.
+
+**Mat Ryer:** Yeah. I mean, of course, even if there isn't the mechanisms automatically or programmatically to vote, people do still have a voice, of course, in any of the communities they're in. There is a great Go community on Twitter, and there are other communities. Of course, there's a Gopher Slack... So yeah, I think -- I've seen a few examples really in the community where although there isn't an official way for people to share ideas and things yet, having the conversation out in public really does influence things, doesn't it? So people's views are of course heard, so therefore you should say it, if people do want to contribute...
+
+**Carmen Andoh:** I think I will. The one thing that I want to start talking about more publicly is in order for a site to be useful, you're navigating two things that pull up against each other. One is keeping a 30,000-foot view, to make sure that you aren't having any blind spots, but you also need to dive deep down into the actual needs of a particular subgroup. So coming up and down and up and down is an incredibly challenging thing, and it's one I hope to help with.
+
+**Steve Francia:** Mat, you also mentioned their Twitter, and Gopher Slack... It's important to recognize that the intent of this site is not to displace those great resources that already exist. So we say it's "by the ecosystem, for the ecosystem", but it doesn't mean it's gonna replace all the existing ecosystem solutions. The intent here was to fill some gaps that we saw were there. It's largely a curation site; it's actually to reference those existing things that are there.
+
+As I heard you describing these voting mechanisms, it sounded to me a lot like Reddit. And the Reddit channel I think is great. I subscribe to the Golang Reddit channel, I read it every day. I always see good news, and new articles, and new talks on it, and I think that's a great mechanism to get the voice out... And of course, the other things you referenced already. But if you want a voting thing, we have it. It's the Reddit r/golang. Feel free to use it, it's a really good resource.
+
+**Mat Ryer:** Yeah, great. There's also the Go weekly newsletter, and there's a Changelog newsletter as well, which is the home of this podcast. So anyone that wants to sign up to that... You really can keep your finger on the pulse of what's going on that way; it's great.
+
+**Jon Calhoun:** I will say, the only thing I differentiate between Reddit and the voting thing is just that Reddit is kind of -- it's not real-time, but it's time-boxed, in some sense... Whereas I could see some value in voting for learning resources, but I completely agree that the way voting and everything works, it'd be very hard to do that correctly... So I completely get why you're not doing it. But I do see that as slightly different from Reddit. Because I agree, it's very similar, but I've seen tons of cases where -- you know, I have free resources that I give out, and I've seen people post them; even though they've been posted on Reddit before, they'll post them again and people are like "I've never seen this before", and it's like "Okay, clearly nobody is going back and searching these things", or something's happening... So there is some difference there.
+
+**Steve Francia:** It's also important to recognize, most of the content on the site is static. As Carmen said, we've talked about doing internationalization; we're using a tool that lets us do internationalization with it... And there's opportunities to -- like, we do not have plans to do this yet, but we've made sure that the options are there later, if we wanted to add additional curators, maybe localized curators from different areas to help us with this as well.
+
+So there's an element there where we can open it to some degree and get support from the community, but also keep it tightly curated and a high-quality bar up there.
+
+**Mat Ryer:** Well, I think that's all the time we have for today. Thank you very much to our guests, Julie Qiu, Steve Francia, and our regular panelists, Jon Calhoun and Carmen Andoh. We'll see you next time on Go Time!
+
+**Break:** \[01:14:14.21\]
+
+**Mat Ryer:** Hello, and welcome to Go Time! I'm Mat Ryer. Today we're talking about Go.dev. It's a user-friendly hub of curated resources for Go, and we're joined by two of the brains behind it - Steve Francia (also known as @spf13) and Julie Qiu are joining us. Hello!
+
+**Steve Francia:** Hi!
+
+**Julie Qiu:** Hi!
+
+**Mat Ryer:** And we're also joined by Carmen Andoh and Jon Calhoun. Hello, you two!
+
+**Carmen Andoh:** Hi!
+
+**Mat Ryer:** How's it going?
+
+**Jon Calhoun:** Mat, I think you kind of lied. I think Carmen is partially behind Go.dev as well.
+
+**Steve Francia:** Yeah, we're actually joined by three of them. You just happen to be regularly joined by one of them.
+
+**Carmen Andoh:** I'm stealth. I'm stealth.
+
+**Mat Ryer:** Okay, I'll do it again. So it's three of the brains behind GoDev.
+
+**Steve Francia:** And it's Go.dev.
+
+**Mat Ryer:** Okay, Go.dev. This is why we do it. So for anyone listening, this is how the sausages are made... \[laughter\] I have to do it again now, and make it sound like it was the first time. That's the hard bit, you all know it...
+
+**Steve Francia:** Yeah, you don't call it like GoogleCom. \[laughter\]
+
+**Jon Calhoun:** Actually... \[laughter\]
+
+**Steve Francia:** \[01:17:15.14\]
+
+**Mat Ryer:** Go.dev, okay. It's clever, because it's also the domain, isn't it?
+
+**Steve Francia:** Yeah.
+
+**Mat Ryer:** Got it, thank you. Good. Alright, let's do it again then, everyone. It's fine, I'm not embarrassed.
+
+**Steve Francia:** I thought you did a really good job, by the way.
+
+**Carmen Andoh:** Same here.
+
+**Steve Francia:** Other than the obvious mistakes, it was really well done.
+
+**Mat Ryer:** \[laughs\] Yeah... Yeah. Okay, well there we go. This is why we do iterative development.

Hits of the Summer_transcript.txt

@@ -0,0 +1,824 @@
+[0.00 --> 8.94]  what up gophers jared here this episode is different than what you're used to hearing
+[8.94 --> 13.38]  from go time we've been clipping highlights of the show for a while now to share on twitter
+[13.38 --> 18.00]  and youtube and a side effect of that effort is i have a bunch of awesome clips just sitting on
+[18.00 --> 24.66]  my hard drive collecting digital dust so here's a beta test of a best of style clips show covering
+[24.66 --> 29.46]  the summer months i hope you enjoy it you might if you missed a few episodes or if you listen
+[29.46 --> 34.12]  distracted sometimes like i do please let me know what you think in the comments if people dig it
+[34.12 --> 39.94]  we'll probably do this more often if not well i'll just pipe the whole thing to dev knowl oh yeah and
+[39.94 --> 46.34]  since it's a special kind of episode we have a very special intro song for you here we go
+[46.34 --> 57.28]  band with her changelog is provided by fastly learn more at fastly.com we move fast and fix things here
+[57.28 --> 62.20]  changelog because of rollbar check them out at rollbar.com and we're hosted on linode cloud
+[62.20 --> 72.16]  servers head to leno.com changelog this episode is brought to you by digital ocean droplets managed
+[72.16 --> 79.04]  kubernetes managed databases spaces object storage volume block storage advanced networking like
+[79.04 --> 84.82]  virtual private clouds and cloud firewalls developer tooling like the robust api and cli
+[84.82 --> 89.76]  to make sure you can interact with your infrastructure the way you want to digital ocean is designed for
+[89.76 --> 97.24]  developers and built for businesses join over 150 000 businesses that develop manage and scale their
+[97.24 --> 102.60]  applications with digital ocean head to do.co slash changelog to get started with a 100 credit
+[102.60 --> 105.12]  again do.co slash changelog
+[105.12 --> 118.58]  i see bars and kilobytes kilobytes kilobytes kilobytes
+[118.58 --> 145.08]  alenom
+[145.08 --> 147.08]  Kilobytes.
+[154.08 --> 155.08]  R's.
+[155.08 --> 156.08]  Kilobytes.
+[157.08 --> 158.08]  R's.
+[159.08 --> 160.08]  Kilobytes.
+[164.08 --> 165.08]  R.
+[165.08 --> 166.08]  Kilobytes.
+[166.08 --> 167.08]  R.
+[167.08 --> 168.08]  Kilobytes.
+[168.08 --> 169.08]  R.
+[169.08 --> 170.08]  R.
+[170.08 --> 171.08]  R.
+[171.08 --> 172.08]  R.
+[172.08 --> 173.08]  R.
+[173.08 --> 174.08]  R.
+[174.08 --> 175.08]  R.
+[175.08 --> 177.08]  First up, we have a panel-only show.
+[177.08 --> 183.08]  Yana, Matt, and John discuss how to effectively work with databases in episode 132, The Trouble
+[183.08 --> 184.08]  with Databases.
+[184.08 --> 188.08]  It just so happens this was the most popular episode of the summer.
+[188.08 --> 195.08]  The way the database works or the way it models things is a lot of things in common with the
+[195.08 --> 196.08]  storage engine.
+[196.08 --> 201.08]  So the way you store, the way you shard, the way you really partition the data, there's
+[201.08 --> 206.08]  a lot to do with the type of capabilities it provides to query.
+[206.08 --> 212.08]  So from a high-level perspective, it's always important, I think, for a user to understand
+[212.08 --> 217.08]  how at some sort of lower layer things are stored.
+[217.08 --> 223.08]  So you can estimate what is feasible, what kind of use cases are actually a good fit for
+[223.08 --> 224.08]  that type of database.
+[224.08 --> 229.08]  Even though it sounds like a bit of work, I really suggest people to take a look at
+[229.08 --> 232.08]  what type of use cases make sense.
+[232.08 --> 239.08]  And in the end of the day, at the storage level, what do they do before evaluating anything?
+[240.08 --> 244.08]  The classic example of that that I've heard is that I've been told that at Stripe, one
+[244.08 --> 247.08]  of the common things they've done is that they have a NoSQL database that they're using
+[247.08 --> 250.08]  for all the really high-speed transactions.
+[250.08 --> 253.08]  But then on the back end, when they want to run analytics and do all these other things,
+[253.08 --> 255.08]  it's really hard to do that.
+[255.08 --> 259.08]  And a lot of times people want SQL, they want to be able to use some tools that use SQL for
+[259.08 --> 260.08]  that.
+[260.08 --> 262.08]  So they actually take a lot of that data and translate it into a SQL database.
+[262.08 --> 265.08]  And while it's delayed, it's only used internally, so that's okay.
+[265.08 --> 270.08]  So they're taking that trade-off and deciding it's useful to have this data in both formats.
+[270.08 --> 272.08]  And it's like you said, they didn't switch from one to the other.
+[272.08 --> 277.08]  It's more of a, this makes sense for this use case, and we port it over to this for another use case.
+[277.08 --> 283.08]  Yeah, in my experience, I'm seeing always like two or three databases in a system.
+[283.08 --> 285.08]  You can't really fight the trade-offs.
+[285.08 --> 288.08]  You get benefit from them differently.
+[288.08 --> 293.08]  So, basically, there's usually a relational database and other database for warehousing reasons,
+[293.08 --> 295.08]  like analytics and so on.
+[295.08 --> 301.08]  And then there's usually a database like, or something like elastic for, you know, for search reasons.
+[301.08 --> 306.08]  So, you know, you can at least like list three core data resources.
+[306.08 --> 307.08]  Yeah.
+[307.08 --> 311.08]  And then of course, backup could even be a different one where you're taking, taking backup
+[311.08 --> 316.08]  and putting it in some kind of cold storage or just less active place.
+[316.08 --> 321.08]  It's common, I think, for developers to want to get the perfect solution from the beginning
+[321.08 --> 323.08]  and just build that.
+[323.08 --> 328.08]  But probably a better strategy is to just start with something, one thing, simple,
+[328.08 --> 330.08]  do what you're going to do with it.
+[330.08 --> 337.08]  There are like three things you can have, you know, in distributed systems, you can't have
+[337.08 --> 339.08]  three of these things you have to pick to.
+[339.08 --> 343.08]  And those three things are represented by CAP, which is C-A-P.
+[343.08 --> 346.08]  CAP means consistency, sorry.
+[346.08 --> 348.08]  A means availability.
+[348.08 --> 353.08]  P means partition tolerance, like, you know, network partitioning tolerance.
+[353.08 --> 363.08]  And what he says is, if you want 100% consistency and 100% network failure partitioning tolerance,
+[363.08 --> 366.08]  you can't have 100% availability.
+[366.08 --> 369.08]  You always have to make that sacrifice.
+[369.08 --> 374.08]  You know, we're talking about relational databases as well as like, Scimulus, NoSQL type of databases.
+[374.08 --> 379.08]  Actually, relational databases are more like CP systems.
+[379.08 --> 381.08]  They have higher consistency.
+[381.08 --> 383.08]  They are more tolerant to partitioning.
+[383.08 --> 389.08]  On the other hand, NoSQL databases are, you know, compromising from consistency.
+[389.08 --> 394.08]  They're eventually consistent, but they provide higher availability.
+[394.08 --> 396.08]  So they're AP systems.
+[396.08 --> 401.08]  So if you have this like mental model, I think databases are becoming easier to understand because, you know,
+[401.08 --> 408.08]  that like there's a limit, like there's like physical limits to the world and you can't have it all.
+[408.08 --> 413.08]  And I've worked for some project managers that really just don't agree with this.
+[413.08 --> 415.08]  They want all three.
+[415.08 --> 416.08]  Yes.
+[416.08 --> 418.08]  Because, you know, it's just hard to explain.
+[418.08 --> 419.08]  It's almost like.
+[419.08 --> 422.08]  Spanner is actually a typical CP system.
+[422.08 --> 432.08]  It has 100% consistency and it's very tolerant to partitioning, but its availability is significantly higher than any other relational database,
+[432.08 --> 438.08]  which is provides five nines of availability, which means like five minutes downtime a year.
+[438.08 --> 440.08]  You know, that's like amazing.
+[440.08 --> 447.08]  Like most of the relational databases require 10 minutes or whatever a month for maintenance and so on.
+[447.08 --> 455.08]  Or if you want to upgrade the, you know, the schema, it requires downtime or the failover requires downtime.
+[455.08 --> 457.08]  So how did this happen?
+[457.08 --> 463.08]  Like, you know, the Spanner team kind of says that they're beating the cap theorem because they provide this like high availability.
+[463.08 --> 474.08]  And it has a lot to do the way how the internals of this distributed system is working, plus our good networking infrastructure.
+[474.08 --> 481.08]  So we're just kind of like, you know, improving the availability, not to 100%.
+[481.08 --> 486.08]  We're still talking about five nines, but five nines is actually a lot in practice.
+[486.08 --> 493.08]  So, you know, our goal is like, maybe you shouldn't make as many as compromises.
+[493.08 --> 500.08]  We will try to, you know, provide you a higher availability, but you will still have the like transactional relational database.
+[500.08 --> 509.08]  But at the same time, we have a lot of limitations around like, you know, the type of the schema limitations, for example, some SQL limitations.
+[509.08 --> 516.08]  I actually was reading Martin Kleppman's book, maybe like two months ago.
+[516.08 --> 522.08]  And then in my dream, I saw myself writing that blog post.
+[522.08 --> 530.08]  And as soon as I woke up, I took notes, like, I think I dropped the 10 items on the cover of that book.
+[530.08 --> 537.08]  So that blog post actually like probably came from some of the ideas that I got from his book and so on.
+[537.08 --> 544.08]  So that's really funny that like, you know, I saw that like the article was being very useful in my dream.
+[544.08 --> 545.08]  And it turned out to be true.
+[545.08 --> 546.08]  It's so funny.
+[546.08 --> 553.08]  Well, if people didn't feel stupid before, the fact that Yana's coming up with this stuff in her sleep is certainly going to do that.
+[553.08 --> 555.08]  That's how Paul McCartney wrote yesterday, by the way.
+[555.08 --> 558.08]  He just woke up and had the song.
+[558.08 --> 559.08]  My dreams are way less productive.
+[559.08 --> 563.08]  Yeah, I just dreamed my legs were made of jelly.
+[563.08 --> 564.08]  That's not helping anyone.
+[564.08 --> 566.08]  I can't turn that into a blog post.
+[566.08 --> 567.08]  Probably could.
+[567.08 --> 568.08]  I might.
+[574.08 --> 583.08]  On episode 135, John was joined by Chris Brando, Ben Johnson and Aaron Schlesinger to discuss some of their mistakes and how they learn from them.
+[583.08 --> 593.08]  One of the mistakes I made when I first was jumping into Go was that I just felt like I over planned or I tried to like over optimize for getting things perfect.
+[593.08 --> 598.08]  So you'd read about how you shouldn't use MVC and you shouldn't do all these different things.
+[598.08 --> 601.08]  And I sat down and I'm like, all right, I'm going to write this side project.
+[601.08 --> 602.08]  It was a side project.
+[602.08 --> 603.08]  It wasn't like my main work stuff.
+[603.08 --> 608.08]  So I was like, I'm going to build this thing and it's going to be a good go code.
+[608.08 --> 613.08]  And I think I spent so much time rewriting some stuff because I was like, oh, this is a bad way.
+[613.08 --> 614.08]  And I realized why.
+[614.08 --> 616.08]  And then I'd go back and rewrite it.
+[616.08 --> 622.08]  And in the end, I'm pretty sure what I ended up doing was just using a simple MVC model and just got it done.
+[622.08 --> 625.08]  And then later I was able to come back and tweak things and adjust.
+[625.08 --> 633.08]  But it just I wasted so much time trying to like meet the expectations of everybody, I guess, is how I'd put it.
+[633.08 --> 642.08]  And to me, that was a big mistake because I feel like you learn more by just kind of jumping in and doing stuff rather than like trying to find the optimal path.
+[642.08 --> 653.08]  I guess kind of a long ongoing mistake in my career is just not understanding the underlying technologies that I use, especially early on, like using like an HTTP framework, like web framework.
+[653.08 --> 655.08]  You know, just assume that that's what you do.
+[655.08 --> 666.08]  But like understanding that framework and maybe even like layers on top of that framework is a lot of times more complicated than just understanding HTTP or whatever the underlying technology is.
+[666.08 --> 674.08]  So I feel like I've done that for a long time in my career, but more recently just trying to step back and just understand, like, you know, what are those frameworks actually give me?
+[674.08 --> 675.08]  What do they add?
+[675.08 --> 679.08]  And just a lot of times the net HTTP is, you know, enough with a router.
+[679.08 --> 686.08]  So I guess first off, are you talking about going all the way down to actually understanding how like TCP works and going even below that or?
+[686.08 --> 691.08]  I mean, I think it's a trade off of like, what do you get for that abstraction that you're working with?
+[691.08 --> 705.08]  Like, I'm, I mean, I'm not pushing bits across an ethernet accord, but if I can, you know, understand like, you know, just basic headers and what they do, rather than having some other library on top of that to actually, you know, set certain parameters.
+[705.08 --> 714.08]  I feel like just understanding that kind of underlying HTTP RFC or some aspects about it helps me just to write more direct code.
+[714.08 --> 731.08]  I feel like if you go on GitHub and you look at some popular open source projects, I feel like that's sort of the Instagram of programming and that you see the perfect result after everyone's, you know, put on the makeup and rolled camera and everything.
+[731.08 --> 734.08]  And, you know, we put that on ourselves.
+[734.08 --> 741.08]  Whereas, you know, like Ben, you just said, you could just go in and write a for loop and it'll be good for a while.
+[741.08 --> 744.08]  It's just, you got to start somewhere.
+[744.08 --> 751.08]  And there's always the telltale sign of like initial commit has like 20 source files and 4,000 lines of code.
+[751.08 --> 753.08]  And you're like, that was your initial commit.
+[753.08 --> 756.08]  It's like that clearly was not your initial commit.
+[756.08 --> 763.08]  I like to just kill the dot get repo or the folder and just restart after probably a couple hundred commits in.
+[763.08 --> 764.08]  Oh, I'm the same way.
+[764.08 --> 769.08]  It's like I've got the initial version and then that eventually gets to something and I'm like, okay, I'm okay with sharing this.
+[769.08 --> 772.08]  And then delete, you know, and just go from scratch.
+[772.08 --> 776.08]  On one hand, I get why you do it because you don't necessarily want that bad history there.
+[776.08 --> 784.08]  But on the other hand, it is kind of hard for somebody jumping in who actually thinks that might be the initial commit.
+[784.08 --> 790.08]  Next up, 136.
+[790.08 --> 799.08]  This is when Matt joined the show as a guest with his Pace co-founder David Hernandez to talk about the technical decisions they made while building Pace with Go.
+[799.08 --> 802.08]  Yeah, absolutely.
+[802.08 --> 807.08]  And I tell teams this when I talk to different teams as well about that when they're choosing their technology.
+[808.08 --> 809.08]  That's a big thing.
+[809.08 --> 815.08]  You know, GRPC, for example, might be the perfect choice from a purely technical perspective.
+[815.08 --> 821.08]  But if nobody on the team has experience with GRPC, then there's a learning curve there.
+[821.08 --> 830.08]  And some people talk about them in terms of innovation tokens and things, you know, you're not allowed, they say, to just all the technology can't be new and unfamiliar.
+[830.08 --> 836.08]  You can do some of that, but there's effort and there's kind of a cost to being productive in any of those.
+[836.08 --> 842.08]  And we had that already on the front end because we hadn't done much front end work for a while.
+[842.08 --> 847.08]  And we knew we wanted to use a, it was going to be a rich front end.
+[847.08 --> 850.08]  So we knew it had to be somewhat of a modern.
+[850.08 --> 863.08]  So you weren't trying to emulate GRPC, basically, you wanted to, you got some ideas from how sort of it works and you stole some, and I use that very generously.
+[863.08 --> 866.08]  You stole some ideas, some implementation details, rather.
+[866.08 --> 867.08]  Sure.
+[867.08 --> 871.08]  And you sort of issued the whole binary format.
+[871.08 --> 880.08]  You just went, you're playing JSON and basically you solved your own problem in a sense, and rather than sort of bringing in something for the sake of, because it's cool.
+[880.08 --> 881.08]  Yeah.
+[881.08 --> 884.08]  Stealing from open source is not really stealing, isn't it?
+[884.08 --> 887.08]  It's just kind of Robin Hood.
+[887.08 --> 888.08]  Why?
+[888.08 --> 889.08]  Yeah.
+[889.08 --> 893.08]  Different sizes comes with different problems.
+[893.08 --> 895.08]  Speed is different.
+[895.08 --> 906.08]  And that's why people try to put things like microservices, not because the microservices are better technology, because it's easier to control the size of the team or the responsibility, things like that.
+[906.08 --> 917.08]  In this case, everything is easy in that sense, because we are two people, but we became from full stack developers to full company developers.
+[917.08 --> 922.08]  We do support, we do marketing, we do accountancy.
+[922.08 --> 926.08]  So it's not only the tech stack is quite wide in that sense.
+[926.08 --> 930.08]  You have to worry about a lot of more things in this case.
+[930.08 --> 931.08]  Yeah.
+[931.08 --> 938.08]  I remember that release manager wanted to do like two releases a month and then be in sync with everyone.
+[938.08 --> 940.08]  And they asked how many times we were releasing.
+[940.08 --> 947.08]  And it was that day was something between 10 and 20 or something, you know, it was a very different mindset of rapid.
+[947.08 --> 953.08]  As soon as it's a bit better than it was, you know, we want to kind of get it out.
+[953.08 --> 962.08]  On episode 137 databases, we're back on the table.
+[962.08 --> 963.08]  See what I did there.
+[963.08 --> 970.08]  This time, Johan Braunhorst joined Johnny, Matt and John to focus in on Postgres.
+[970.08 --> 977.08]  Big thing that Postgres has over SQLite in Go specifically is really good library support.
+[977.08 --> 982.08]  Unfortunately, the SQLite driver that everyone uses is a C Go driver.
+[982.08 --> 986.08]  And as most of you probably know, C Go means building with C.
+[986.08 --> 987.08]  It means longer build times.
+[987.08 --> 990.08]  It means less portable binaries and stuff like that.
+[990.08 --> 997.08]  So in Go with Postgres, we actually have several different pure Go libraries to speak with Postgres, which is really great.
+[997.08 --> 999.08]  So that's just one reason to use Postgres.
+[999.08 --> 1005.08]  But other things such as the stability of the software, for example, as you say, it's over 20 years old.
+[1005.08 --> 1008.08]  It's been used by thousands of companies worldwide.
+[1008.08 --> 1015.08]  You know, it's not going to just, you know, corrupt your files because those bugs have been ironed out by now.
+[1015.08 --> 1024.08]  So within databases, I think you often say you don't want to use something that's less than 10 years old, because like this data needs to live for a long time.
+[1024.08 --> 1028.08]  You want to make sure it doesn't, you know, corrupt on the disk or whatever.
+[1028.08 --> 1032.08]  And Postgres is one such stable, mature solution.
+[1032.08 --> 1035.08]  It's also very fast because it's written in C and C is fast.
+[1035.08 --> 1044.08]  Yeah, no, I was sitting here nodding along to what you were saying there that start with Postgres, because I think a lot of users come into the space knowing what to use.
+[1044.08 --> 1054.08]  And that was kind of the hope with my talk, just giving you an introduction to something you can use, some opinionated tips on what libraries to use and so on, giving you like a head start.
+[1054.08 --> 1062.08]  Just getting started with actually like moving away from the problem of choosing a technology and actually using the technology to solve your problems.
+[1062.08 --> 1077.08]  I really agree with that sentiment that a lot of people will try and, you know, Google, what should I use to store my data or even like read some popular programmers blog where they have tested out the latest technology and found it to be really good for their very specialized use case.
+[1077.08 --> 1081.08]  And then apply that to like all of their problems, because that's all they know.
+[1081.08 --> 1090.08]  And, you know, starting with something well tested, well established, like you say, good to develop mindshare like Postgres will take you very far before you need to change.
+[1090.08 --> 1094.08]  How do you actually use Postgres?
+[1094.08 --> 1104.08]  And I'm even thinking like in the context of, say, a website that is hosting a blog, at what point would you make a connection to Postgres?
+[1104.08 --> 1112.08]  Do you tend to make one connection per instance of your code running and then create sessions off that?
+[1112.08 --> 1116.08]  Or does each handler, would each handler make its own connection?
+[1116.08 --> 1120.08]  How does it actually work, you know, from a Go developer's point of view?
+[1120.08 --> 1128.08]  Yeah, so if we take a step back and look at the standard library database SQL package, that actually has a connection pool built in.
+[1128.08 --> 1141.08]  So if you come from another language like Python or Ruby, you might be familiar with putting something, I think there's a Postgres bouncer or something like that, which does connection pool in between your database and your client.
+[1141.08 --> 1145.08]  In Go, you don't need to do that sort of thing because it's already built into the standard library.
+[1145.08 --> 1151.08]  So what you would normally do when you connect to the database is just create a single SQL.db handle.
+[1151.08 --> 1153.08]  And then that's safe for concurrent use.
+[1153.08 --> 1158.08]  So you can use that in all of your handlers, even though they're being called from different Go routines from different clients.
+[1158.08 --> 1165.08]  And you can also configure things such as max connections on the SQL connection.
+[1165.08 --> 1168.08]  But normally that's all handled by the Go standard libraries.
+[1168.08 --> 1171.08]  You don't really have to worry about it, which is really nice.
+[1171.08 --> 1174.08]  Actually, I have never had the pleasure of using an ORM.
+[1174.08 --> 1177.08]  So what am I doing on this show talking about databases?
+[1177.08 --> 1186.08]  I would very early on in my career, I was kind of persuaded against using an ORM because naturally, as a beginner programmer, I was like, oh, this looks cool.
+[1186.08 --> 1191.08]  But I never really had to use it because I was told by someone who knew better than me that that was a bad idea.
+[1191.08 --> 1202.08]  So here I am saying to other beginners like myself, Bonds was, that you should try not to use the ORM, even though it looks really appealing at first.
+[1202.08 --> 1206.08]  You should probably just learn to use SQL and it's really not that bad.
+[1206.08 --> 1209.08]  And you'll learn to love it, I think, like I do.
+[1209.08 --> 1217.08]  I would agree with that, having come from the other side, because everything specific to Rails that I learned is useless to me now.
+[1217.08 --> 1222.08]  And everything I learned about SQL along the way when I couldn't get that to work is much more useful to me.
+[1222.08 --> 1224.08]  And it'll carry over to any language.
+[1224.08 --> 1225.08]  That's right.
+[1225.08 --> 1230.08]  Well, that would have been an unpopular opinion section, but everyone agrees.
+[1230.08 --> 1232.08]  It's not that unpopular, I guess.
+[1232.08 --> 1234.08]  It's not happened before where we all agree.
+[1234.08 --> 1237.08]  I need to ask, have you met Bobby Tables?
+[1237.08 --> 1238.08]  I have met Bobby Tables.
+[1238.08 --> 1240.08]  This is a great one, actually.
+[1240.08 --> 1263.08]  So the reference that Johnny is making there is to an XKCD comic, which has a very illustrative way of showing just exactly what SQL injection means, where there's a school principal who's making a call to a concerned parent, I suppose, asking about their son Bobby Tables, dash dash, drop table students or something like that.
+[1263.08 --> 1266.08]  The parent says, yes, we can't hit Bobby Tables.
+[1266.08 --> 1270.08]  And then the principal says, well, I hope you're happy we've lost this year's student records.
+[1270.08 --> 1279.08]  It's really funny because obviously the implication there is that they had to enter their kid's name somewhere in some sort of form or something.
+[1279.08 --> 1282.08]  And they thought, hey, it would be fun to see if this is vulnerable to SQL injection.
+[1282.08 --> 1288.08]  And then they put in that command that would, if it was vulnerable to SQL injection, drop the table, call students.
+[1288.08 --> 1291.08]  And of course, the joke then is that it actually did.
+[1291.08 --> 1293.08]  And principal is furious about it.
+[1293.08 --> 1298.08]  And the kind of lesson, I guess, is that you shouldn't have been vulnerable to SQL injection.
+[1298.08 --> 1301.08]  Now he's got no job, though.
+[1301.08 --> 1304.08]  So that's no good telling him that, is there?
+[1304.08 --> 1306.08]  His life's devastated by that.
+[1306.08 --> 1307.08]  But yeah.
+[1313.08 --> 1317.08]  Episode 138 was quite a departure and inspiring to many listeners.
+[1317.08 --> 1319.08]  Inspiring to many listeners so we hear.
+[1319.08 --> 1325.08]  Jackie Grinrod and Deshaun Carter joined John to talk about their first week with the Go programming language.
+[1327.08 --> 1329.08]  You know, if somebody is trying to learn Go, what do you suggest?
+[1329.08 --> 1330.08]  I think, Jackie, you said exorcism.
+[1330.08 --> 1332.08]  Is that where you would suggest people start?
+[1332.08 --> 1340.08]  So I found it worked really well for me because you'd put your solution in, you could see how other people solved it, but you'd also have mentors come and comment.
+[1340.08 --> 1345.08]  And the really nice thing for me about the mentors was that they're really good at giving like iterative feedback.
+[1345.08 --> 1349.08]  So give me like some feedback and be like, here's how you can solve this a little bit more elegantly.
+[1349.08 --> 1350.08]  And then I would do that.
+[1350.08 --> 1354.08]  And then they'd come back and be like, OK, so now that you've done that, here's like another step you can do.
+[1354.08 --> 1356.08]  And so we got to like that final iteration.
+[1356.08 --> 1362.08]  And I do have all those commits like tracked so I can kind of watch the way that I grew as I was learning and the feedback I got.
+[1362.08 --> 1365.08]  So I would actually recommend that because it was a very nice experience for me.
+[1365.08 --> 1371.08]  You got to do it and you also got like professional feedback from people who do this, which you might not always get if you're just learning a language on your own.
+[1372.08 --> 1383.08]  On the front of learning things, an advantage that I didn't really realize, like I knew I had it, but I didn't really consider it until a conversation I had, I think last week or this week, was about concepts like garbage collection.
+[1383.08 --> 1386.08]  I would have learned that in my computer science classes.
+[1386.08 --> 1393.08]  And there are things that I just kind of knew about going into this that I didn't really think about as in terms of like what I'm learning it on the job in an ops role.
+[1393.08 --> 1396.08]  Maybe that's not something that just gets presented to you.
+[1396.08 --> 1400.08]  It might be like those kind of concepts that you need to learn outside of just the code.
+[1400.08 --> 1409.08]  And that's kind of been interesting to figure out, like, how do you review either like best practices across different languages or different architecture patterns, which is not my strength.
+[1409.08 --> 1415.08]  But trying to figure that out while learning the language can be pretty overwhelming and just figuring out where to start.
+[1415.08 --> 1424.08]  But yeah, there was people that were consistently there and it was nice because it was kind of like having this little support circle that I really didn't expect this.
+[1424.08 --> 1430.08]  Like I expected streaming to be more like people coming in and being like, hey, you're doing that wrong.
+[1430.08 --> 1432.08]  Like, hey, why didn't you do this?
+[1432.08 --> 1435.08]  So it was it was really nice.
+[1435.08 --> 1437.08]  I'm going to paint a picture.
+[1437.08 --> 1438.08]  This is very similar.
+[1438.08 --> 1440.08]  Like I said, every rep counts.
+[1440.08 --> 1441.08]  I've been to the gym.
+[1441.08 --> 1445.08]  You guys been to the gym, maybe not lately, but it's kind of the same experience.
+[1445.08 --> 1449.08]  Like you definitely went out and you kind of said, hey, I'm going to go to the gym.
+[1449.08 --> 1453.08]  I put it on a schedule and I'm going and maybe you have a gym buddy.
+[1453.08 --> 1454.08]  I think everything goes better.
+[1454.08 --> 1455.08]  You know, you're pairing.
+[1455.08 --> 1457.08]  You had some accountability.
+[1457.08 --> 1462.08]  And I basically said, I'm going to buy some weights and I'm going to put them out here.
+[1462.08 --> 1463.08]  And this was my path.
+[1463.08 --> 1468.08]  I didn't put any like deadline or I didn't have any severe accountability.
+[1468.08 --> 1471.08]  I was kind of fitting it in, but it was my number one focus.
+[1471.08 --> 1476.08]  But still, not only is it going to be stickier for you, I think that accountability.
+[1476.08 --> 1479.08]  Yeah, I think you probably went further faster.
+[1479.08 --> 1487.08]  How much time does your team spend building and maintaining internal tooling?
+[1487.08 --> 1489.08]  I'm talking about those behind the scenes apps.
+[1489.08 --> 1491.08]  The ones no one else sees.
+[1491.08 --> 1494.08]  The S3 uploader you built last year for the marketing team.
+[1494.08 --> 1498.08]  That quick Firebase admin panel that lets you monitor key KPIs.
+[1498.08 --> 1503.08]  Maybe even the tool your data science team hacked together so they can provide custom ad spend analytics.
+[1503.08 --> 1505.08]  Now, these are tools you need so you build them.
+[1505.08 --> 1507.08]  And that makes sense.
+[1507.08 --> 1514.08]  But the question is, could you have built them in less time, with less effort and less overhead and maintenance required?
+[1514.08 --> 1516.08]  And the answer to that question is, yes.
+[1516.08 --> 1518.08]  That's where Retool comes in.
+[1518.08 --> 1522.08]  Rohan Chopra, engineering director at DoorDash, has this to say about Retool.
+[1522.08 --> 1531.08]  Quote, the tools we've been able to quickly build with Retool have allowed us to empower and scale our local operators, all while reducing the dependency on engineering.
+[1531.08 --> 1532.08]  End quote.
+[1532.08 --> 1562.08] 
+[1562.08 --> 1565.08]  Again, retool.com slash change load.
+[1565.08 --> 1585.08]  Next up, Testify's maintainer, Boyan Subachov joins Matt and Mark Bates to talk about testing frameworks in Go.
+[1585.08 --> 1590.08]  Because you literally put those things as strings into the test code.
+[1590.08 --> 1591.08]  Yeah.
+[1591.08 --> 1595.08]  And then from that you can generate some quite nice looking failures.
+[1595.08 --> 1599.08]  If something fails, you can, you know, it reads quite nicely.
+[1599.08 --> 1601.08]  But I found it to be too verbose, actually.
+[1601.08 --> 1608.08]  And just saying, you know, not equal and then showing you the two values or something was just easier to see.
+[1608.08 --> 1616.08]  The thing I like about BDD versus unit testing actually has to do with those kind of the names of the tests.
+[1616.08 --> 1617.08]  Hmm.
+[1617.08 --> 1621.08]  You know, when you're writing a simple test, right, you know, test that create does something, right?
+[1621.08 --> 1623.08]  That's a pretty simple test name.
+[1623.08 --> 1631.08]  But when you start having all those weird variants, right, then the string based text names become really, really useful.
+[1631.08 --> 1637.08]  You know, when you can, when you just need a little bit more description as to what it is you're trying to test.
+[1637.08 --> 1642.08]  And that's harder to do in a unit type of a test where you have a function name.
+[1642.08 --> 1643.08]  Hmm.
+[1643.08 --> 1644.08]  Yeah.
+[1644.08 --> 1647.08]  Have you seen property based testing as well?
+[1647.08 --> 1650.08]  This is another kind of style.
+[1650.08 --> 1663.08]  Yeah, it's almost like fuzzing for your functions where you specify not what values to test with, but what types of values your function takes as input.
+[1663.08 --> 1669.08]  And obviously also what kind of output you expect.
+[1669.08 --> 1688.08]  And then the property based testing framework of which one example is copter will then just generate, you know, whether it's random or in some specific sequence will generate plenty of values to then test your function and try and find edge cases for you that do not conform to a specification.
+[1688.08 --> 1690.08]  That's really cool, isn't it?
+[1690.08 --> 1691.08]  Yeah.
+[1691.08 --> 1696.08]  The first exposure I had to that was with hypothesis in Python, and it was pretty awesome.
+[1696.08 --> 1705.08]  I mean, an obvious problem there is, you know, have, especially if your functions are a bit more complex, that it can take a while for tests to run.
+[1705.08 --> 1711.08]  Because you're now running the same function 10,000 times rather than once or five times.
+[1711.08 --> 1717.08]  But, you know, for pure functions where, you know, where your code doesn't have side effects.
+[1717.08 --> 1718.08]  Yeah.
+[1718.08 --> 1725.08]  It's very good at finding edge cases and little behaviors that you didn't anticipate.
+[1725.08 --> 1728.08]  In Ruby, we had a great thing called Time Cop.
+[1728.08 --> 1737.08]  And it was such a random Ruby thing where it would override basically time.now to be whatever it wanted it to be.
+[1737.08 --> 1739.08]  Because you could just override anything.
+[1739.08 --> 1744.08]  So you could say, like, I want to be three weeks into the future and time.now would return three weeks in the future.
+[1744.08 --> 1746.08]  Great Scott.
+[1746.08 --> 1747.08]  Yeah.
+[1747.08 --> 1752.08]  It was spectacularly awful and good and fun.
+[1752.08 --> 1753.08]  Yeah.
+[1753.08 --> 1756.08]  It was one of the things you can only do in a dynamic language though.
+[1756.08 --> 1757.08]  Yeah.
+[1757.08 --> 1759.08]  It's also the reason I'm not using a dynamic language.
+[1759.08 --> 1769.08]  No, the way I look at it at least is if it's something that's, how likely is it to change?
+[1769.08 --> 1771.08]  Basically, how likely is your database state to change?
+[1771.08 --> 1774.08]  How likely is your time to change?
+[1774.08 --> 1781.08]  If it's anything other than low, take it out, mock it out, make it as a dependency that's injected.
+[1781.08 --> 1784.08]  But that's a rule of thumb that I follow anyways.
+[1784.08 --> 1794.08]  Episode 140 was our most anticipated show of the summer.
+[1794.08 --> 1799.08]  The latest draft proposal for Generics was posted in late June and it took us a few weeks.
+[1799.08 --> 1807.08]  But we managed to get Robert Griesemer and Ian Lance Taylor on the show to answer many of the community's nagging questions about the latest proposal.
+[1807.08 --> 1814.08]  Robert and I released the updated design draft for moving forward with Generics.
+[1814.08 --> 1835.08]  The biggest change was that we dropped the idea of contract and just decided that instead of having a separate syntactic construct, which was a contract, that we could just use interface types to describe the contract between the type argument and the type parameter.
+[1835.08 --> 1845.08]  A lot of people looking at contracts had seen that they seemed a lot like interfaces and people had trouble separating out exactly when you would use a contract and when you would use an interface.
+[1845.08 --> 1847.08]  So we simplified this.
+[1847.08 --> 1850.08]  And this was, I should add, almost entirely due to Robert.
+[1850.08 --> 1854.08]  We simplified this to just use interface types.
+[1854.08 --> 1860.08]  And then the second big step we made was we've released a translation tool and a type checker.
+[1860.08 --> 1866.08]  So we have a type checker that works for the design draft, the description of generics in the design draft.
+[1866.08 --> 1871.08]  So that gives us, you know, some confidence that what we have written about can actually work.
+[1871.08 --> 1876.08]  And we have a translation tool which translates code into ordinary Go.
+[1876.08 --> 1879.08]  The translation tool is not, by any means, a final thing.
+[1879.08 --> 1881.08]  There's cases it doesn't handle.
+[1881.08 --> 1886.08]  It's just an experimental tool, but it lets people actually write code that can actually run using generics.
+[1886.08 --> 1895.08]  So we can get a feel for whether generics actually works for people and whether it actually addresses the issues that they have.
+[1895.08 --> 1912.08]  So more practically for the Go community, when do you think that you're going to get enough feedback to move forward with moving from a draft proposal to actually putting it forth as a proposal to change in the language?
+[1912.08 --> 1915.08]  Yeah, we don't have any timelines in mind, I'd say.
+[1915.08 --> 1923.08]  As we mentioned earlier, we're still trying to pin down some of the precise semantics, which I don't think is going to affect any existing code.
+[1923.08 --> 1925.08]  In fact, I'm sure it's not going to affect any existing code.
+[1925.08 --> 1927.08]  We want to make sure that we understand it.
+[1927.08 --> 1932.08]  We want to make sure that, you know, the multiple Go compilers will implement the same thing.
+[1932.08 --> 1935.08]  We're going to have to have some sense of how to add to the language spec.
+[1935.08 --> 1938.08]  So those are the steps we're looking at now.
+[1938.08 --> 1943.08]  I mean, we're certainly going to move forward as fast as we can toward making a formal proposal.
+[1943.08 --> 1945.08]  Of course, at that time, none of it will be a surprise.
+[1945.08 --> 1950.08]  People will have seen all of the ideas already, and we'll just have to see how it flies.
+[1950.08 --> 1954.08]  So far, I feel like the reaction has been largely positive, which is encouraging.
+[1954.08 --> 1957.08]  But I don't know exactly what the timeline is going to be.
+[1957.08 --> 1964.08]  My experience with generics was maybe C++ with templates and probably the highest,
+[1964.08 --> 1970.08]  the highest point there was when I was able to, as Ian alluded to before, it's Turing complete.
+[1970.08 --> 1977.08]  I was able to write a program using C++ templates that would decide whether a constant was a prime number or not.
+[1977.08 --> 1980.08]  And the compiler would decide it at compile time.
+[1980.08 --> 1984.08]  So that's not the kind of thing we would like to support.
+[1984.08 --> 1993.08]  With respect to, you know, what I'd like to see or not see is, honestly, I'm worried about the kind of code that people are going to write.
+[1993.08 --> 1995.08]  I mean, there's no question about that.
+[1995.08 --> 2005.08]  And we see some of the examples that people send us that cause crashes in the prototype, and they're just unbelievably convoluted and really, really hard to decipher.
+[2005.08 --> 2010.08]  But as other people have pointed out, those people are really pushing the envelope.
+[2010.08 --> 2012.08]  They're trying to just see what can I do with this thing.
+[2012.08 --> 2017.08]  And I hope this is not going to be, you know, the kind of code that people are going to write down the road.
+[2017.08 --> 2033.08]  I think one of the first things we need to do, if we have this for real, we need to come up with a kind of best practices guide that guides everybody a little bit as to how you should use generics and when you should use them and when you should not use them.
+[2033.08 --> 2034.08]  I think there's also very important으로 to desarrollo.
+[2034.08 --> 2038.08]  In these two things I've been working with community of learning to implement humanity, and I may not do that for l administrations, but I won't add that quickly.
+[2038.08 --> 2039.08]  I'll divide my together to our影片 away.
+[2039.08 --> 2040.08]  You can just add this.
+[2040.08 --> 2045.08]  Many of our друга statistics are interesting, but if we don't know, we have arms behind or perpendicular to normal analysis and provide tools for it to the field.
+[2045.08 --> 2048.08]  And those are» kinda important.
+[2048.08 --> 2052.78]  The experimental tool has no similarity whatsoever to any real implementation.
+[2053.24 --> 2057.70]  So we know it's slow, and it's going to be slow, and that's just inevitable.
+[2058.42 --> 2068.40]  If this does move forward to become a proposal and it gets accepted, then most likely the implementation will be to start with a branch of the main Go tool chain,
+[2068.40 --> 2078.16]  and we'll start adding generic support on that branch, which will involve changing the compiler mainly and any other changes to other tools that are required.
+[2078.64 --> 2083.42]  And so that'll be the time to start giving feedback about changes to build speed.
+[2083.74 --> 2087.94]  We've talked about it with some of the compiler developers, like Keith Randall especially,
+[2088.40 --> 2093.06]  and we think we can do it without a significant increase in build speed.
+[2093.22 --> 2094.56]  I mean, there will be some increase in build speed.
+[2094.56 --> 2099.90]  We don't think it's going to be a huge increase, but, you know, this is really speculative at this point.
+[2100.14 --> 2103.28]  So the time to give that feedback is when we're able to start doing development,
+[2103.40 --> 2108.32]  and hopefully people will also be able to contribute work when we start doing that work on the public branch.
+[2109.54 --> 2114.48]  Programming language evolution is really a social process.
+[2115.02 --> 2120.80]  It doesn't actually matter if you have seen the light and you know exactly the perfect language,
+[2120.80 --> 2125.04]  you know, and you would just put it out there and, you know, maybe it's 20 years ahead.
+[2125.14 --> 2130.28]  Nobody would even buy it because people would not see the reasoning why you got to that point.
+[2130.42 --> 2134.14]  And so you really have to get everybody along.
+[2134.48 --> 2137.98]  And some people may already be where you are and some people may not,
+[2138.10 --> 2140.92]  but you have to get everybody along in little steps.
+[2140.92 --> 2144.54]  And that's how we eventually end up where we want to be.
+[2144.80 --> 2147.80]  And we can see this with all kinds of things like a garbage collection.
+[2147.88 --> 2152.88]  A garbage collection was invented, you know, 1950 something with Lisp.
+[2153.04 --> 2155.90]  You know, the first Lisp had garbage collection, 1958, I believe.
+[2156.56 --> 2162.26]  And it's taken forever before it became accepted as something that the programming language should,
+[2162.70 --> 2164.62]  you know, a mainstream programming language should have.
+[2165.24 --> 2168.24]  Maybe Java was the first one that really made it mainstream.
+[2168.24 --> 2174.00]  And now this is not something that is, I mean, still disputed or debated, I should say,
+[2174.06 --> 2176.62]  but it's not as outrageous anymore.
+[2176.74 --> 2178.44]  And so I think that's true for other things.
+[2184.76 --> 2192.52]  On episode 141, guest Daniel Marti helped Matt and Johnny explore Go's encoding JSON package.
+[2194.40 --> 2197.26]  Yeah, and it's JavaScript object notation.
+[2197.26 --> 2199.50]  So it comes out of JavaScript.
+[2200.20 --> 2203.94]  But it turns out to be really kind of useful across a lot.
+[2204.12 --> 2207.18]  Every language really has now some kind of JSON support.
+[2207.82 --> 2209.50]  It's practically everywhere.
+[2210.08 --> 2214.44]  Practically every language out there that's modern today has to have JSON support because you just do.
+[2214.76 --> 2218.62]  And your computer, you might not see it, but it definitely is running JSON at some level.
+[2218.62 --> 2219.26]  Hmm.
+[2219.58 --> 2220.02]  Yeah.
+[2220.12 --> 2225.44]  And so there's like, it's an object and it has fields and those fields have some types.
+[2225.80 --> 2230.98]  And it's the types that we're used to as well in Go, like strings and numbers and booleans.
+[2231.26 --> 2232.04]  Any others?
+[2232.18 --> 2234.60]  Other objects, arrays, those kinds of things.
+[2234.66 --> 2236.22]  I think that might be the whole list.
+[2236.66 --> 2240.06]  And why did it get such popular use on the web?
+[2240.06 --> 2244.40]  I mean, it kind of is kind of perfect, isn't it, for web technologies?
+[2244.92 --> 2249.96]  I would say it came from all the success that browsers had, you know, the modern web had.
+[2250.36 --> 2258.10]  And, you know, suddenly HTTP, HTML, CSS and JavaScript and JSON, all these technologies kind of took everybody by surprise.
+[2258.10 --> 2260.20]  Initially, everybody thought they were just toys.
+[2260.74 --> 2263.76]  But now suddenly people are building real companies on top of them.
+[2264.52 --> 2267.58]  And JSON is just, you know, has too much momentum.
+[2268.26 --> 2271.20]  I don't think anything is ever going to replace it at this point, honestly.
+[2272.06 --> 2278.20]  And I have mixed opinions and feelings about all the third-party JSON re-implementations out there.
+[2278.86 --> 2280.22]  I think some of them do make sense.
+[2280.22 --> 2288.58]  For example, one use case is you do absolutely want the most performance you can get because maybe this is a bottleneck for you.
+[2288.80 --> 2297.52]  And you don't mind Go generating some code to then, you know, write, generate automatically a decoder for you for JSON.
+[2298.02 --> 2300.70]  So you can use packages like easyJSON for that, which is pretty popular.
+[2301.00 --> 2307.12]  And the trade-off there is you have to run Go generate and your binary is going to weigh quite a little bit more because it has quite a lot of extra code.
+[2307.12 --> 2312.72]  But that extra code, it just encodes all the logic directly in binary code, in machine code.
+[2313.04 --> 2317.72]  So there's no reflect, there's no dereferences, there's no extra work involved.
+[2318.50 --> 2322.16]  So I think that's clearly one of the cases where it might make sense for a use case.
+[2322.56 --> 2323.90]  I like how you framed that as well.
+[2323.98 --> 2326.46]  You're saying maybe it's a bottleneck in your case.
+[2326.70 --> 2327.42]  And that's the thing.
+[2327.48 --> 2332.72]  It's like once you've seen that this is a place where an improvement is going to make a difference for you,
+[2332.72 --> 2338.94]  then it's worth taking on the extra pain, whether it's complexity or learning a new API or whatever it is.
+[2339.26 --> 2343.76]  I like that approach because, well, I think it's what we should always be doing.
+[2343.90 --> 2350.34]  You know, as you alluded to, Johnny, we kind of can get a bit obsessed with why wouldn't we want the fastest possible thing?
+[2350.38 --> 2354.28]  And the answer is it might be good enough just using the standard library.
+[2354.28 --> 2366.52]  There are some bugs, for example, there's one that I would say affects most code bases out there, which is the standard, you know, you have an HTTP endpoint and the body is JSON.
+[2366.68 --> 2367.54]  So you want to decode it.
+[2367.90 --> 2376.90]  So what you do is you take the R.body and you do JSON.newdecoder.decode with the body and then into some structure.
+[2377.44 --> 2378.78]  And if you do that, it's buggy.
+[2379.08 --> 2380.12]  If you just do that.
+[2380.54 --> 2381.52]  I've just got to go.
+[2382.50 --> 2383.82]  What do you mean it's buggy?
+[2383.82 --> 2385.04]  Tell me why, please.
+[2385.42 --> 2390.38]  So this was found by Joe, one of the maintainers, I want to say about a year ago.
+[2390.62 --> 2397.40]  And the bug is the decoder is meant to be useful for streams of JSON values.
+[2398.04 --> 2407.10]  And that is, for example, when you do go test with the JSON flag, it's going to give you a new line separated stream of JSON values of JSON objects.
+[2407.40 --> 2410.10]  Yeah, that's kind of how I was using it in those tools I was talking about.
+[2410.30 --> 2410.84]  Yep, exactly.
+[2410.84 --> 2414.18]  In a way, it is kind of streaming in a way.
+[2414.40 --> 2415.34]  Like it takes the reader.
+[2415.90 --> 2418.72]  For each object, it buffers it, I guess.
+[2418.86 --> 2421.22]  But it discards that previous object, doesn't it?
+[2421.40 --> 2421.56]  Yeah.
+[2421.82 --> 2422.56]  Yeah, next time.
+[2422.62 --> 2422.80]  Right.
+[2422.88 --> 2424.70]  So in a sense, it's streaming.
+[2424.98 --> 2428.18]  It appears to you as if it's streaming, but internally, that's not what it's doing.
+[2428.18 --> 2432.62]  Well, it's still doing it only one object at a time, which you could say is a stream.
+[2432.74 --> 2435.10]  It's just if it's a great big fat object, then.
+[2435.50 --> 2435.86]  Exactly.
+[2436.40 --> 2436.88]  In trouble.
+[2437.24 --> 2437.34]  Yeah.
+[2437.34 --> 2449.82]  Episode 142 was all about that infrastructure with special guest Shabekshah Jalan.
+[2449.82 --> 2460.54]  Maybe we should bust some of this jargon because, you know, you hear infra, I hear systems engineering, and I hear DevOps.
+[2461.04 --> 2463.48]  Do we agree at least on what these terms mean?
+[2463.88 --> 2464.12]  No.
+[2464.12 --> 2464.62]  No.
+[2466.04 --> 2466.44]  No.
+[2466.90 --> 2468.92]  I mean, it's kind of funny.
+[2469.10 --> 2472.78]  I mean, let's take, for example, one of the hot new titles out there, right?
+[2472.82 --> 2473.70]  SRE, right?
+[2473.80 --> 2474.02]  Yeah.
+[2474.20 --> 2479.88]  And you'd think that, like, that would carry some sort of consistency, right, from organization to organization.
+[2480.40 --> 2482.04]  That is entirely not the case.
+[2482.14 --> 2482.34]  Yeah.
+[2482.38 --> 2483.02]  Like at all, right?
+[2483.68 --> 2488.96]  An SRE at Google is going to be very different from an SRE at Salesforce, which is going to be very different from an SRE at Microsoft.
+[2489.44 --> 2493.36]  Yes, there is a through line between these things, and that goes for the other titles as well.
+[2493.36 --> 2496.60]  Systems engineering, you know, DevOps engineer.
+[2496.80 --> 2497.78]  I cringe a little bit when I say that.
+[2497.90 --> 2498.08]  Yeah.
+[2498.84 --> 2502.38]  But, like, basically, these things are going to mean different things in different organizations.
+[2502.58 --> 2511.40]  Even in different sort of a – and over the lifespan of an engineering team, right, the definition of that role may also change, right?
+[2511.44 --> 2521.24]  So it's not – I don't think there is one sort of solid definition, right, of what, you know, software engineer, database engineer, infrastructure engineer, ops, whatever it is.
+[2521.24 --> 2523.26]  I think it's going to be different everywhere you go.
+[2523.50 --> 2523.76]  Yeah.
+[2524.00 --> 2526.38]  I thought SRE was text speak for sorry.
+[2527.00 --> 2527.94]  That's how I was thought.
+[2527.94 --> 2532.60]  That's exactly what I ran into as well.
+[2532.68 --> 2533.82]  It's a chicken and egg problem.
+[2533.96 --> 2538.32]  You need experience to get a job, but to get a job, you need experience.
+[2538.32 --> 2540.18]  So, like, where do you actually start?
+[2540.26 --> 2548.62]  Because there is a very hard limit to, like, how much you can do and learn on your own when it comes to a job of this type, especially if you want to work at scale.
+[2548.80 --> 2552.96]  Like, you simply cannot replicate the kind of things you'll be doing every day at home.
+[2552.96 --> 2554.40]  So why is that, though?
+[2554.54 --> 2556.36]  And it's such a big missed opportunity.
+[2556.92 --> 2564.20]  Like, I'm seeing a lot of, like, people without a lot of experience coming into this field and, like, looking things from a very, like, new perspective.
+[2564.60 --> 2564.76]  Yeah.
+[2564.76 --> 2573.30]  In my experience, they've been really much more effective in terms of, like, pointing out the, you know, the core friction points than the experienced engineers.
+[2573.44 --> 2577.72]  I think experienced people have a lot of, like, you know, they're accepting the current status.
+[2577.72 --> 2584.74]  And they always assume that, like, there's all this, like, layers of layers of things that, you know, you have to satisfy in order to kind of, like, provide.
+[2584.86 --> 2590.54]  But is a new inexperienced engineer just coming in, like, questioning some of these, like, things more carefully?
+[2591.24 --> 2602.30]  And that's sort of, like, the perspective that we are missing in infrastructure in general because, you know, there's no good connectivity and we're just doing a bad job in terms of hiring people into this area.
+[2602.68 --> 2603.70]  Yeah, completely.
+[2607.72 --> 2621.18]  Last but not least, on episode 143, Francesc Campoy and Isabella Rettelmeier join Matt and Yana for a deep dive on Go's context package.
+[2623.18 --> 2631.20]  Like, there's a very straightforward way of cutting down the latency, like, the tail latency in your requests.
+[2631.20 --> 2644.80]  So if you have a request, let's say that you have a request that you send into a server and it takes five milliseconds 99% of the time, but then there's 1% of the time that it takes one minute, which is not good.
+[2645.16 --> 2651.46]  So what you could do is make that call multiple times and you're going to do it with cancellation.
+[2651.46 --> 2655.12]  So you're going to do context with cancel of your original context.
+[2655.82 --> 2663.60]  And then what you're going to do is you're going to use the same context for all of those calls and have defer cancel at the top of your function.
+[2663.88 --> 2670.34]  And then as soon as any of those values returns and you return from that function, the rest will be canceled.
+[2670.34 --> 2676.06]  And that's going to take down your 99th percentile from like one minute down to five milliseconds.
+[2676.76 --> 2684.86]  So that's like small things that you can get a lot of performance, especially when you're using a server that is not something you manage.
+[2685.02 --> 2690.54]  So you cannot go and complain to them about like, hey, your 99th percentile latency is awful.
+[2690.96 --> 2693.36]  You can still fix it by doing this little hack.
+[2694.10 --> 2696.58]  Why is there context.to do?
+[2696.80 --> 2697.20]  Hmm.
+[2697.94 --> 2698.78]  Great question.
+[2699.40 --> 2699.74]  Yeah.
+[2699.74 --> 2700.62]  I don't know.
+[2702.08 --> 2708.24]  So context.to do and context.background do exactly the same thing.
+[2708.76 --> 2710.30]  They return an empty context.
+[2710.78 --> 2716.32]  In an empty context, that doesn't have values, doesn't have timeouts, it never gets canceled.
+[2716.90 --> 2718.88]  So it's literally an empty struct.
+[2719.04 --> 2719.68]  That's what it is.
+[2719.68 --> 2727.88]  So the interesting thing is that when you return a background, what you're saying is that, oh, this is something that I'm starting from scratch.
+[2727.88 --> 2731.02]  So you're basically saying there's no previous context.
+[2731.02 --> 2732.62]  This is something that I'm creating, right?
+[2732.68 --> 2736.96]  So for instance, in the example of the CLI, you're running your CLI.
+[2736.96 --> 2739.58]  And at the beginning, there's no previous context or anything.
+[2739.58 --> 2743.28]  Like maybe at one point, we'll have actually a context coming from signal.
+[2743.72 --> 2744.70]  That would be an interesting thing.
+[2744.74 --> 2746.08]  But otherwise, we don't have anything.
+[2746.08 --> 2748.70]  So you would call background.
+[2748.70 --> 2761.04]  Context.to do is actually was added just so as different functions, like you need to start, like you're creating a tree of functions that are calling and passing context around.
+[2761.04 --> 2768.24]  So how do you do it if you want to add it to all of them, but little by little?
+[2768.66 --> 2774.42]  Like if you start from the top, it's going to be, you cannot pass functions until they're accepted.
+[2774.80 --> 2781.46]  But if you do it the other way around, like you build a function that starts by saying, oh, I accept a context now.
+[2781.60 --> 2783.12]  And you can pass a context to me.
+[2783.26 --> 2788.66]  Then the caller could say, oh, okay, so I should have a context, but I do not have it yet.
+[2788.66 --> 2795.34]  So instead of calling context.background, which implicitly says, I do not have a context and I will never will.
+[2796.02 --> 2800.04]  So to do it just, hey, I do not have it yet, but let's fix it later.
+[2800.16 --> 2805.72]  So it's literally just so when you grab to do, you can find where you need to still do more work.
+[2806.24 --> 2807.22]  I think that's kind of cool.
+[2807.28 --> 2809.60]  Like the fact that they thought about these.
+[2810.20 --> 2817.92]  Otherwise, I mean, you could have done the same, calling context.background and then having on top like a comment doing to do has a real context.
+[2817.92 --> 2819.58]  But they did it this way.
+[2819.64 --> 2824.68]  So it's more explicit and you actually could do code analysis and look like, hey, this is not done yet.
+[2828.74 --> 2829.80]  What's up, gophers?
+[2829.96 --> 2836.82]  Are you looking for a way to instantly debug and troubleshoot your applications and services running in production on Kubernetes?
+[2837.00 --> 2837.68]  That's a mouthful.
+[2838.00 --> 2841.84]  Well, Pixie gives you a magical API to get instant debug data.
+[2841.84 --> 2845.20]  And the best part is this doesn't involve changing code.
+[2845.54 --> 2849.22]  There are no manual UIs and all this lives inside Kubernetes.
+[2849.90 --> 2859.42]  Pixie is an API which lives inside your platform, harvests all of your data that you need and exposes a bunch of interfaces that you can ping to get data you need.
+[2859.82 --> 2862.66]  Pixie is essentially like a decentralized Splunk.
+[2862.66 --> 2869.70]  It's a programmable edge intelligence platform which captures metrics, traces, logs and events without any code changes.
+[2870.02 --> 2875.02]  And the team behind Pixie is working hard to bring it to market for broad use by the end of 2020.
+[2875.58 --> 2878.08]  But I'm here to tell you how you can get your hands on the beta today.
+[2878.58 --> 2883.08]  Links are in the show notes, so check them out so you can click through to the beta and their Slack community.
+[2883.40 --> 2887.14]  Once again, links from the show notes, check them out and look forward to Pixie Day coming soon.
+[2892.66 --> 2922.64]  Pixie Day coming soon.
+[2922.66 --> 2924.34]  Rolling Unpop Rockapella.
+[2924.68 --> 2925.16]  Enjoy.
+[2928.16 --> 2930.32]  Unpopular opinion.
+[2936.22 --> 2938.02]  Unpopular opinion.
+[2941.86 --> 2942.72]  I don't know.
+[2942.86 --> 2948.28]  I feel like most new technologies are just not necessary, I guess, would be my unpopular opinion.
+[2949.16 --> 2950.22]  Docker, honestly.
+[2950.22 --> 2951.22]  Like Kubernetes.
+[2951.22 --> 2951.90]  Like Kubernetes.
+[2951.90 --> 2952.90]  A lot of those.
+[2952.90 --> 2955.30]  I feel like you can run a stack.
+[2955.30 --> 2956.56]  Not the same stack.
+[2956.56 --> 2962.64]  But you can run basically like what we used to run back in the 90s or whatnot or early 2000s where it's just like here's a web server running.
+[2962.90 --> 2964.34]  You can run a business off of that.
+[2964.46 --> 2965.28]  Probably be fine.
+[2965.28 --> 2968.62]  I mean, obviously, back up your data, but that would be my unpopular opinion.
+[2968.62 --> 2983.76]  I think for me, an unpopular opinion I have is, I guess, once your organization or your, I guess, the engineers in your organization reach a certain level that you shouldn't really just take software from other companies.
+[2983.76 --> 2994.10]  Like, I think GRPC is like a big one for me where it's just like once you have a group of people that understands how to like build things with TCP and HTTP, you could probably just build it yourself.
+[2994.28 --> 3000.76]  And you should do that because your organization's needs are going to be very different than like what Google needed for when they built GRPC.
+[3000.76 --> 3006.94]  So I think it's like I fall closer to that what used to be called not invented here syndrome.
+[3007.10 --> 3010.20]  And I think like that's probably where we should be edging back to.
+[3010.36 --> 3013.42]  But I realize that that is wildly unpopular with a lot of people.
+[3013.84 --> 3017.34]  And they usually say, you know, just use whatever is out there because that's better.
+[3018.16 --> 3020.42]  Well, I mentioned REST APIs earlier.
+[3020.96 --> 3022.62]  I don't think they're good.
+[3023.18 --> 3024.78]  That's my unpopular opinion.
+[3025.56 --> 3030.54]  I think they cause more confusion than problems they solve.
+[3030.76 --> 3035.94]  Actually, an unpopular opinion I have is, you know, you should try and work in tiny teams.
+[3036.26 --> 3042.04]  A lot of the problems when it comes to software engineering come at scale.
+[3042.30 --> 3045.02]  And that's not just code scale, but people scale.
+[3045.34 --> 3056.78]  So if you can have tiny little teams working on a problem, and you can do this within bigger teams, like just literally as, you know, two or three people, you are now a new little team.
+[3056.78 --> 3065.18]  You can be so effective in such a small group because you cut out a lot of the work needed really to marshal the team.
+[3065.50 --> 3066.54]  You can't always do it.
+[3066.76 --> 3069.78]  And it sounds a little bit antisocial, but that would be my unpopular opinion.
+[3069.92 --> 3070.72]  Tiny teams.
+[3071.34 --> 3077.28]  I have this library that I like to use, which is called Squirrel, and it's a query builder.
+[3077.74 --> 3079.90]  And it uses the builder pattern.
+[3080.98 --> 3081.22]  Aha!
+[3081.52 --> 3083.42]  Everyone hates the builder pattern in Go, right?
+[3083.48 --> 3090.82]  And for a good reason, because the builder pattern doesn't work well with the static typing that Go provides because we don't have generics.
+[3091.46 --> 3093.24]  And Squirrel suffers from this problem as well.
+[3093.24 --> 3097.18]  But it also provides a lot of power at the same time.
+[3097.58 --> 3101.78]  So it's like the one exception to the rule of don't use the builder pattern anywhere
+[3101.78 --> 3105.34]  is for query building, use the squirrel package
+[3105.34 --> 3109.82]  because it's really easy to use and constructing queries with.
+[3110.94 --> 3113.58]  So the builder package then, just for anyone not familiar,
+[3113.74 --> 3115.64]  this is where you get these fluent APIs
+[3115.64 --> 3119.48]  where every method returns the main object itself
+[3119.48 --> 3121.72]  and then lets you chain them.
+[3121.72 --> 3122.48]  Yeah, right.
+[3122.48 --> 3124.36]  And I should say I hate these
+[3124.36 --> 3127.68]  because you can't define interfaces that work with them at all.
+[3128.34 --> 3129.14]  It's just a nightmare.
+[3129.94 --> 3131.38]  So I do have one question, Johan.
+[3132.20 --> 3135.90]  Could you reconstruct the squirrel package using functional options?
+[3137.22 --> 3137.66]  Potentially.
+[3138.92 --> 3141.42]  So do either of you have something you'd like to share?
+[3141.78 --> 3142.34]  I have one.
+[3142.62 --> 3145.80]  I think that pair programming is an unpopular opinion.
+[3146.30 --> 3148.58]  I did it early in my career, took a lot of time off.
+[3149.58 --> 3151.68]  A lot of companies are afraid to adopt it.
+[3151.68 --> 3154.64]  I work for a company that I think does it really well.
+[3155.00 --> 3158.60]  But I still think it's an unpopular opinion that pair programming, mod programming,
+[3159.16 --> 3162.80]  I think that's the way things are going to be done if you want to be successful in the future.
+[3162.80 --> 3167.98]  I think the closest I have is that hopefully it's not controversial here because that'd make me sad.
+[3168.22 --> 3175.60]  But the documentation contribution is contribution that is as important and sometimes more important than just the actual code.
+[3176.00 --> 3178.40]  And I see a lot of that where people always kind of drop the docs.
+[3178.56 --> 3179.72]  That's the closest thing I have right now.
+[3179.72 --> 3181.38]  But, you know, it's important.
+[3181.54 --> 3187.84]  It builds up that how do we bring people into our circle that's building this tech and expand and then be able to build better things together.
+[3188.14 --> 3190.88]  So hopefully not too much of a hot take here.
+[3191.12 --> 3192.52]  But I think it is.
+[3192.58 --> 3193.36]  I think it's a hot take.
+[3193.68 --> 3195.68]  And yeah, let's fight the good fight.
+[3195.88 --> 3196.62]  Keep preaching it.
+[3196.62 --> 3201.46]  I don't know if it's an unpopular opinion, but, you know, I like short identifiers.
+[3202.06 --> 3202.86]  I do.
+[3203.34 --> 3209.92]  And I feel like the closer they are to where you use them, the shorter they can be.
+[3210.76 --> 3216.00]  And the further away they are from where you use them, the longer they should be.
+[3216.56 --> 3224.30]  And then there's some exceptions, like when an identifier is really, really, really important in your package and prevalent,
+[3224.30 --> 3227.82]  then it can be one letter, even if it's a global.
+[3228.70 --> 3232.96]  And, you know, the most prominent example for that is perhaps testing.t.
+[3233.42 --> 3235.46]  I'm not sure if you're allowed to be a teacher now.
+[3236.44 --> 3237.08]  Uh-oh.
+[3237.72 --> 3238.72]  Why, Don?
+[3239.18 --> 3247.20]  Every teacher, like, expects you to write really long, like, self-explanatory variable names, regardless of where you use them or when you're using them.
+[3247.20 --> 3248.20]  At least that was my experience.
+[3248.30 --> 3250.42]  I felt like every teacher wanted long variable names.
+[3250.42 --> 3263.10]  So I will comment on your change list if it uses, you know, in a simple for loop, if the iteration variable is called index, I will, you know, probably comment on that.
+[3263.34 --> 3265.64]  So, you know, call it I or J or whatever.
+[3265.64 --> 3270.56]  Okay, I don't know if this opinion is unpopular, but I feel like I write it a lot.
+[3270.74 --> 3274.80]  So there's certainly people who don't seem to grasp it.
+[3274.88 --> 3281.64]  And that's that the language is not perfect, but every change to the language carries a heavy cost.
+[3281.64 --> 3289.96]  So when you want to come and argue for why the language should be changed, and, you know, we see that a lot.
+[3290.04 --> 3295.00]  I'd say that there's probably one a day suggestion for some way to change the Go language.
+[3295.56 --> 3303.64]  Don't just talk about how it makes the language better, but also spend some time to talk about how it makes the language worse.
+[3303.64 --> 3308.72]  Because there's no such thing as a 100% good change to the language.
+[3308.82 --> 3309.82]  I shouldn't say there's no such thing.
+[3309.90 --> 3310.70]  Maybe it's out there.
+[3311.20 --> 3312.68]  Maybe no one has thought of it yet.
+[3313.28 --> 3319.88]  But probably it's a good bet that all the 100% good changes to the language have already been made.
+[3320.68 --> 3327.92]  And so when you want to change the language, spend some time to think about how it makes things worse as well as how it makes things better.
+[3328.24 --> 3331.62]  I think I just saw you drop a metaphorical mic.
+[3333.64 --> 3340.26]  We're going to go back now and think a lot about how generics makes things worse.
+[3341.20 --> 3346.90]  So my unpopular opinion is that encoding JSON is fast enough.
+[3348.28 --> 3349.74]  Oh, come on.
+[3349.96 --> 3350.30]  Wow.
+[3350.66 --> 3353.16]  This is the guy responsible for making it faster.
+[3355.84 --> 3357.52]  Well, I'm going to say generally.
+[3358.12 --> 3361.28]  Where generally means it most likely applies to you.
+[3361.28 --> 3368.12]  But it might not apply to the one person that's doing something completely esoteric, such as handling 20 gigabytes of JSON.
+[3368.64 --> 3369.96]  But most people don't do that.
+[3370.90 --> 3373.58]  And kind of my point goes back to the trade-offs, right?
+[3373.78 --> 3379.58]  Yes, if you pick another package, you can get maybe a 2x, 3x, maybe even 4x improvement.
+[3379.58 --> 3383.38]  But is it really worth sticking with JSON at that point?
+[3383.70 --> 3390.94]  The overlap between the people that are stuck with JSON, because they are, and the people that have to deal with a lot of data, is very small.
+[3391.06 --> 3396.70]  Because the people that have to deal with a lot of data, they generally pick better formats that are faster to decode.
+[3396.70 --> 3400.06]  I think that is a pretty solid argument, actually.
+[3400.52 --> 3402.52]  Yeah, that's not unpopular with me, that one.
+[3402.72 --> 3403.78]  I think you've nailed that.
+[3404.44 --> 3404.64]  Yeah.
+[3405.34 --> 3409.56]  Well, you would think that the amount of people yelling about encoding JSON being too slow would disagree.
+[3410.64 --> 3411.04]  Sure.
+[3411.98 --> 3414.84]  Well, but that's because we gave them the tools to benchmark things.
+[3414.92 --> 3416.02]  I don't know what you expect.
+[3416.40 --> 3417.92]  Of course, you're going to be moaning.
+[3418.20 --> 3418.94]  We should take them back.
+[3420.48 --> 3423.52]  JSON isn't always as bad as people make it out to be.
+[3424.52 --> 3424.92]  Hmm.
+[3425.26 --> 3426.02]  Tell me more.
+[3426.02 --> 3427.04]  Who is he?
+[3427.30 --> 3428.94]  Who is this Jason you talk about?
+[3429.60 --> 3432.10]  Yeah, that Jason who gets so much flack.
+[3432.24 --> 3433.58]  Why are you defending him all the time?
+[3433.96 --> 3434.22]  Yeah.
+[3434.36 --> 3439.66]  Well, I've seen a lot of people switch, in my opinion, prematurely to protobuffs in particular.
+[3440.06 --> 3445.20]  Sometimes to thrift, where you just change from one problem to another.
+[3445.20 --> 3450.14]  And especially, I think, for anything that is used externally to your company.
+[3450.36 --> 3451.88]  So, for example, open source code.
+[3452.24 --> 3455.92]  Protobuffs can get very complicated, especially if you're exposing something.
+[3456.02 --> 3458.76]  That is going to be used across multiple languages.
+[3458.76 --> 3461.26]  So, nice to use in Go.
+[3461.76 --> 3465.52]  Not necessarily as nice to use in Ruby, for example, or in PHP.
+[3466.14 --> 3467.70]  Yeah, or indeed the web browser.
+[3467.94 --> 3469.14]  Actually, yeah.
+[3469.36 --> 3470.02]  That's a big one.
+[3470.18 --> 3472.42]  We did an episode on this very recently.
+[3472.90 --> 3474.34]  We called it Encoding JSON.
+[3474.34 --> 3484.16]  And we actually spelled the episode title using JSON to see if any podcasting technology is vulnerable to JSON injection attacks.
+[3484.52 --> 3487.26]  So far, everything's just been fine, which is a shame.
+[3488.94 --> 3491.04]  Jana, do you have an unpopular opinion?
+[3491.40 --> 3492.88]  I have a controversial one.
+[3492.88 --> 3494.08]  Oh, let's do it.
+[3494.18 --> 3496.10]  Let me just re-record the theme tune then.
+[3500.10 --> 3502.04]  God, that's a controversial opinion.
+[3502.64 --> 3506.08]  I do think that I really like Go as a language.
+[3506.48 --> 3512.98]  You know, like the simplicity and verbosity-wise, it's just one of the best options that you have.
+[3512.98 --> 3520.34]  But all the proto-generated artifacts is just making everything just kind of like messed up.
+[3520.62 --> 3528.18]  Like each time I have to touch, you know, some proto-generated protos, it just doesn't look like Go anymore.
+[3528.32 --> 3529.72]  It's like so cryptic.
+[3529.82 --> 3533.98]  Like there's all these like types on top of the standard library I have to learn about.
+[3534.56 --> 3536.30]  You know, proto has its own like struct.
+[3536.30 --> 3544.46]  Like all of the like, you know, mess and even like the timestamp, for example, type is like a completely different representation.
+[3544.96 --> 3551.98]  So you basically have to adopt into that like verbose alternative universe.
+[3552.64 --> 3555.60]  And it's just like my main pain point.
+[3556.26 --> 3562.28]  And I've been like trying to collect all these like gotchas and tips and everything about protos for a long time.
+[3562.28 --> 3569.10]  And I can tell you there's like at least 20 pages of me putting some tips like here and there.
+[3569.18 --> 3579.90]  And I still need to go back to that document in reference to be able to kind of like take a look like, hey, this is what I'm supposed to do if I, you know, see a type like this, proto-generated type like this.
+[3580.24 --> 3583.10]  And that's just like a big struggle to me.
+[3583.62 --> 3588.78]  They've been trying to improve, you know, the generated artifacts, but it's just too late, I think.
+[3588.78 --> 3597.76]  I don't know if it's popular or unpopular, but I think that generics in Go are a good idea.
+[3598.58 --> 3601.36]  I would say that's unpopular with many people that I know.
+[3602.18 --> 3613.50]  But I do think that like I gave this talk long time ago around functional programming in Go and basically why not to do it.
+[3613.84 --> 3616.20]  And one of the biggest reasons, like there were two reasons.
+[3616.20 --> 3622.98]  The first one is because there's no tail precursor optimization, which means that your program is actually 10 times slower just because of it.
+[3623.40 --> 3626.16]  So that's, you know, like that's a small thing that maybe we should fix.
+[3626.50 --> 3633.94]  But the biggest thing was the fact that if you want to do any kind of like interesting composition of types without generics, you're out of luck.
+[3634.18 --> 3635.18]  You cannot really do it.
+[3635.22 --> 3636.84]  You need to do empty interfaces everywhere.
+[3637.50 --> 3637.62]  Right.
+[3637.62 --> 3641.02]  So generics, I'm very excited about seeing them.
+[3641.10 --> 3643.32]  Like I've been trying them and how they look.
+[3643.38 --> 3649.62]  And now that, you know, contracts are kind of like gone or at least like they make much more sense.
+[3649.68 --> 3651.42]  They're not as complicated as they used to be.
+[3651.90 --> 3653.58]  I'm pretty excited about getting to use it.
+[3653.70 --> 3657.40]  So I don't know when it's going to be released for real, but looking forward to that.
+[3657.40 --> 3663.50]  Okay, those were your summer hits.
+[3663.98 --> 3667.78]  I'd love to know if you find value in this style episode.
+[3667.98 --> 3669.84]  Please leave a comment one way or the other.
+[3670.20 --> 3674.62]  Open your show notes, click discuss on ChangeLog News, and let your voice be heard.
+[3675.00 --> 3677.12]  You can also tweet at me if that's easier.
+[3677.30 --> 3678.68]  I'm at JaredSanto.
+[3678.80 --> 3680.30]  That's J-E-R-O-D.
+[3680.86 --> 3681.64]  S-A-N-T-O.
+[3681.64 --> 3686.68]  Special thanks to BMC for all of the music and sounds you hear on our shows,
+[3686.82 --> 3691.08]  especially that Johnny Borzico I See Bars remix we played at the top.
+[3691.32 --> 3692.30]  I see bars.
+[3692.90 --> 3693.72]  10 kilobytes.
+[3694.00 --> 3695.16]  Pretty dope if you ask me.
+[3695.90 --> 3697.82]  Thanks also to our longtime sponsors.
+[3698.48 --> 3699.42]  Shout out to Fastly.
+[3699.80 --> 3700.94]  Shout out to Linode.
+[3701.24 --> 3702.22]  Shout out to Rollbar.
+[3703.18 --> 3704.38]  That's all for now.
+[3704.54 --> 3707.12]  Back to our regularly scheduled programming next week.
+[3707.12 --> 3711.06]  The panel did a Reddit AMA, so we're answering all of the community's questions.
+[3711.06 --> 3712.80]  with a couple special guests.
+[3713.34 --> 3714.34]  Stay tuned for that.
+[3714.58 --> 3715.60]  We appreciate you listening.
+[3716.00 --> 3717.02]  Talk to you again next time.
+[3741.06 --> 3771.04]  We'll be right back.

How Go helped save HealthCare.gov_transcript.txt

@@ -0,0 +1,543 @@
+**Mat Ryer:** Hello there, and welcome to Go Time. I'm Mat Ryer, and I've just pushed to production. Today on Go Time we're talking about caring about HealthCare.gov, and actually, I think why simplicity matters, especially so as the stakes get higher; apologies to any vegans.
+
+On today's show we have Johnny Boursiquot... Hello, Johnny.
+
+**Johnny Boursiquot:** Hello. I'm a carnivore.
+
+**Mat Ryer:** Fair enough. You don't have to state your preference, but you can; you're welcome to.
+
+**Johnny Boursiquot:** Oh, okay. Okay.
+
+**Mat Ryer:** We're also joined - it's Jerod Santo, from The Changelog. Hello, Jerod.
+
+**Jerod Santo:** That's correct. Omnivore... I'm an omnivore.
+
+**Mat Ryer:** Great. Does that mean you eat everything?
+
+**Jerod Santo:** Just anything, yeah.
+
+**Mat Ryer:** Yeah, great.
+
+**Jerod Santo:** Happy to be here. \[laughter\]
+
+**Mat Ryer:** You don't care how big the menu is, you will go to that place.
+
+**Jerod Santo:** \[04:04\] That's right. Supersize Me.
+
+**Mat Ryer:** Yeah, omnivore. And we're also joined by a special guest today, who you may remember from a lightning talk back at GopherCon 2015. It's Paul Smith. Hello, Paul.
+
+**Paul Smith:** Hi, everybody! Glad to be here.
+
+**Mat Ryer:** Welcome. thanks for joining us.
+
+**Paul Smith:** Yeah, I'm excited to talk with you all. Thanks for inviting me.
+
+**Mat Ryer:** So you've got a very interesting story, but maybe before we jump into it, you could just tell us a little bit about your technical background. How did you first get into computers in the first place?
+
+**Paul Smith:** Sure. Well, I think it's actually somewhat of a common story for boys growing up in the '80s - getting a Commodore VIC-20, a Commodore 64 plopped in your lap... Sadly, all too not common for girls in that time, which is something of a tragedy, but... Yeah, typing BASIC programs in, machine code programs out of magazines, spending a lot of time with my Commodores... I think we had an Amiga at one point, too... And then in high school I got an internship in the early '90s at a local laboratory. They were studying -- a biology laboratory, and they actually had a mathematical bend to it. So they were studying DNA protein binding sites, and the information conservation that occurs there when protein binds to DNA... And kind of molecular machines kind of thing. Anyway, that was my first exposure to Unix, and I wrote Perl, and C, and also the nascent web was just getting off the ground around that same time - '94, '95. So I've been basically typing into computers most of my life.
+
+**Mat Ryer:** How much of that's been doing Go? How much actual Go code do you write?
+
+**Paul Smith:** Well, I first learned about Go as soon as it launched in 2009, and it seemed immediately appealing to me. I had been writing Python primarily for work as my job. At that point I had been working professionally for about ten years, mostly web application development. Pretty standard stuff, of that era especially - so relational database-backed web applications; and I loved Python, and I still think Python's a great language, but I remember that Go felt really good right away. It felt like something -- and I remember I had worked with C at that laboratory, and it kind of like rekindled some of those feelings, too.
+
+I had also pushed up against some of the limits of Python in my work, especially with performance and scaling... Yeah, it just immediately felt pretty good. I didn't really have a chance to work professionally with Go until a few years after that, but I would say I've definitely been using Go and a fan of Go since the early days.
+
+**Mat Ryer:** So were you working in small startups then, originally? Because I think there's something interesting about the mindset of startups and what you have to do in a startup environment. It's quite different, it can be a very different situation than bigger enterprise companies, and things... And I think that probably plays a part a little bit in this story, doesn't it?
+
+**Paul Smith:** Yeah, it does. So my first professional web development job was working for a small non-profit here in Chicago... We were an environmental non-profit, and I was basically one of the few web developers there, so I had a lot of freedom to pick and choose technologies. At the time I remember using Cold Fusion, and PHP, even some early Ruby on Rails in the very early days of that stack... But I helped co-found a startup with the co-creator of Django, the Django web framework, Adrian Holovaty, in 2007, called EveryBlock. EveryBlock was a hyper-local news startups. So the idea that we would go out and collect information on the web, in different sources, about news that was happening near you, on your block, in your neighborhood - you wouldn't care about it if it was across town, but if it's happening on your block, you super-care about it a lot.
+
+\[08:17\] So since obviously Adrian was the creator of Django, we used Django for that. So that kind of made the choice easy, but I've definitely experienced in my time that it was an interesting set of factors that lead to you picking a different technology or a different stack... But for me, it's been about expressiveness, how productive I can be in it, and does it perform well enough. And Django, Python checked a lot of those boxes, for sure, and EveryBlock went on to be a pretty successful, although relatively short-lived startup.
+
+**Mat Ryer:** Yeah, because you sold it to MSNBC, right?
+
+**Paul Smith:** It was acquired by MSNBC in 2011. It was actually part of NBC News... Because we had that news angle. Nowadays people take for granted things like Nextdoor, and Facebook, local news about their neighborhood. So we were kind of one of the early pioneers of that... And we sold the company, and kept working on it for a little bit, but... We did some interesting things on EveryBlock. In fact, one of the things I'm most proud about is we built our own map stack.
+
+At the time, if you remember back in 2006-2007, JavaScript engines and browsers were starting to get faster, Google Maps popped on the scene, and it was suddenly like "Oh, you can do these native desktop app-like things in your browser for the first time." And in fact, EveryBlock came out of this idea of sort of like a Google Maps mash-up of taking Google Maps and then using its API and slapping data points on it.
+
+And when we started the company, we thought it would be great, since that's gonna be such a central part of this -- you wanna be able to look at a map of your neighborhood, drill in to your block, see where news is happening... And when I say news, I'm talking about maybe your block is mentioned in the news, or maybe a building license has been issued, or a restaurant inspection, or things like that. Public records. Crime information. We would aggregate all that and put that onto a map.
+
+So Google Maps was great, but we wanted to have control over the look and feel, and users experience... So we built a map stack from the ground-up using open geospatial tools at the time. OpenLayers, Mapnik, some other tools like that, and then kind of combine that with a Django app server we were using to pull the data out of the database, and then present that on the browser.
+
+We worked with a great designer, Wilson Miner, to kind of come up with our own palette and design for the maps themselves, which - I thought they looked really beautiful. So it was a way of visualizing the data. It was a pretty interesting accomplishment.
+
+Now you've got things like Mapbox, and there's just a lot more flexibility when it comes to the sort of in-browser custom map and geospatial experience.
+
+**Mat Ryer:** Yeah, there's loads of SDKs and things that we can just use... But I guess when you didn't have that, sometimes you do have to build things.
+
+**Paul Smith:** Yeah.
+
+**Mat Ryer:** That's a cool one.
+
+**Paul Smith:** We just kind of figured it out... And again, because we were a startup, we could experiment and help differentiate ourselves.
+
+**Jerod Santo:** I'm trying to map in my mind the path you would take from a startup to government contractor. I'm curious if you'd take us on that walk.
+
+**Paul Smith:** Yeah. So after EveryBlock I found myself working to support President Obama's reelection campaign in 2012, which was headquartered here in Chicago. There was a big technology effort around the campaign. For the first time -- well, technology had been a part of his original campaign for Office, but they really brought it in-house; we were gonna build a lot of our own tools, the software that we use not just for the website, but how we interact with our volunteers, how we reach out to potential voters, how we organize and coordinate the campaign... Writing custom software in-house.
+
+\[12:19\] So I was the deputy director for technology at the Democratic National Committee during his reelection. So we were supporting the campaign, and coming up with all these tools, and building the technology to run the campaign. Actually, that was one of the places where I first had an idea that Go could really do the job at scale. So this is leading to how this all wound up in the government, but we were building tools to support the final days of the election, when millions and millions of people were gonna turn out. At the time, early voting and mail-in ballots wasn't quite as popular as it is now, for obvious reasons... But we were building tools to help with that get out to vote effort, so mainly people looking up their polling place, "Where do I go to vote?"
+
+So that was a very popular page on the BarackObama.com website, and we decided to make a key component of that sort of back-end service that was looking up, kind of translating from your home address into the database of polling locations, where you actually go to vote... There was a key component there that we decided to use Go for to do the middle layer. Because we knew it was gonna be high-volume, we wanted low-latency, and it performed fantastic. I knew Go at that point was something that you could put into production on mission-critical services. It gave me a lot of confidence about the language itself.
+
+So the president was reelected, obviously, and sort of how I get involved in government technology is about a year later HealthCare.gov is about to launch... And just for your non-U.S. listeners, healthcare in the United States works a little bit differently than it does in a lot of countries. It's mainly about health insurance that your job provides you. That's the main way that most people get health insurance. And if you're older, you can get on something called Medicare, and if you're poor or have a disability, you can get something called Medicaid, Medicare and Medicaid being government programs... But by and large, most people get it through their employer.
+
+Well, the president and Congress passed a law called "The Affordable Care Act" that did two big things. One, it created a new marketplace for insurance, so people could go buy insurance on this marketplace. It had subsidies, so you could afford it, and there were rules about what the insurance could cover. So it made sure that it wasn't just junk insurance if you actually showed up and needed to get a procedure or something like that, to see your doctor and go to the hospital; it would actually cover those things. So it was a regulated market, and it expanded the Medicaid program, the program for the poor and people with disabilities.
+
+So it did those two big things, and then HealthCare.gov was the way that they were primarily going to deliver it to people. And the president talked about wanting to have this consumer, Amazon-like experience for getting health coverage through the website. So that was the aspiration.
+
+**Johnny Boursiquot:** But... \[laughs\]
+
+**Jerod Santo:** But... The plot thickens.
+
+**Paul Smith:** So October 2013 rolls around, and the site launches and it's immediately clear that it's not working. It's in the news, and people were talking about it. It's kind of all anybody is really talking about. And the folks that I worked with on the campaign, that technology team that I talked about, were texting each other, back-channeling, like "What's going on? How did we get it so right on the campaign side, but when it came to this really critical part of governing, how is it going so wrong?" And we were brainstorming "What could possibly be going wrong?"
+
+\[16:08\] We didn't really have visibility into it, nobody really did... So I get a call in a couple of days after that... This is like mid-October 2013. It's from Todd Park; he was at the time the CTO of the United States. So he works inside the White House as the chief technology officer of the United States... And they're putting together a team.
+
+Basically, they want to get some outside folks who have technology experience, and figure out what's going wrong, because they themselves didn't know what was wrong with the site. They were asking the people who were working on it, the contractors, the government agency, and they didn't know; they couldn't get that information up to the White House, believe it or not.
+
+So I said yes immediately, and there was a small group of us that joined Todd. I'm talking single-digits of people.
+
+**Jerod Santo:** This is like The Avengers.
+
+**Paul Smith:** You know, people called us the Tech Surge, because that's how it was characterized to the media...
+
+**Jerod Santo:** Did they call you on a red phone, like "We need you. Report immediately"?
+
+**Johnny Boursiquot:** And instead of Mjolnir you show up with a keyboard? \[laughter\]
+
+**Paul Smith:** I mean, honestly, there's the kind of cliché moment in movies where it's like your country -- I mean, it really felt like that, like your country needs you. It felt like that, because we knew what the stakes were. The stakes were very high, and we could see this thing kind of failing in real time.
+
+So yeah, I said immediately yes, and the very -- I think next day, or maybe it was the day after, I'm in front of the West Wing of the White House at six o'clock in the morning, meeting the other members of this team that's been put together... And it goes from there.
+
+**Jerod Santo:** So y'all didn't know each other.
+
+**Paul Smith:** Well, some of us knew each other from the campaign. That's kind of how the connection was made to "How are we gonna put this team together? Well, let's start with the people who did a good job on the technology of the campaign, and then we'll go from there." So I knew one other person on the team from the campaign, but we were all relatively new to each other. Our backgrounds were software engineers, or product managers in technology companies, or just kind of in this broader Silicon Valley startup - although I hadn't worked on a Silicon Valley startup myself, but just that idea of private sector Silicon Valley startups... That was the kind of tech talent and experience that was being drawn from.
+
+**Johnny Boursiquot:** So this team is brought in, the rescue team, The Avengers, if you will... What happened with the other team? Things were going wrong... So generally speaking -- I have this idea in my head (a crazy idea) that if something is going wrong with the project, you go to the team, and you start asking questions, "Hey, what's going on? Can you fill me in?" and you give a chance to the team to react, and come up with solutions etc. Things you might expect to do at any other organization. But it sounds like this team is brought in, and the team that actually built the tech just gets jettisoned, and they're gone... So you just get handed this thing, and they go like "Fix it"? What was that transition?
+
+**Paul Smith:** Well, we didn't really know... So this is a really important part - the team that built HealthCare.gov was still there, and from what we can understand... I think some important context here is -- just remember how much pressure there was every single day on this thing. This signature, political thing. It's literally on the news every single day. We're walking into the buildings where this is going on, and it's on CNN, in the lobby, on the big, flat panel screens.
+
+\[20:05\] The pressure was intense. And the people who were working on it, who had built it - because we didn't build it, we were just showing up there to kind of figure out what was going wrong - they're still there. The problem was -- well, in some ways they didn't know what they didn't know, and I'll get to that in a second, what I mean by that.
+
+So they didn't quite know how to fix it, and they weren't doing the things that they needed to do to get the right kind of information up to people like the president, people in the West Wing, the White House, who were trying to operationalize this and try to understand what was wrong and communicate, try to prioritize how it would get fixed. They weren't doing the right sort of things that -- so for example there wasn't monitoring. Like, there was, but it wasn't accessible; it was maybe hidden behind a VPN that some people had access to, but it was really hard to figure out "Is the site up or down? What parts are up or down? What is the performance degraded, what's the baseline?" So that didn't exist. That's kind of problem one.
+
+**Mat Ryer:** So they didn't have any visibility into really what was going on.
+
+**Paul Smith:** There was no visibility. Or there was, but it was so compartmented off, and for all intents and purpose inaccessible to people who were needed to make decisions from that information.
+
+**Johnny Boursiquot:** It sounds like a cultural aspect of things there too, but... Yeah, keep going.
+
+**Paul Smith:** Yeah.
+
+**Johnny Boursiquot:** So what was next?
+
+**Paul Smith:** What I was saying about not knowing what they didn't know - if I had to sum it up, the fundamental problem with HealthCare.gov as it originally launched was they built the wrong thing. So they had the wrong model of what they needed in their heads when they architected and designed and built the site. What I mean is what they needed to launch was high transaction, consumer-like web technology... Like Amazon, or like a piece of consumer technology - lots of people concurrently using it, you want low latency, you want a good user experience... It's transacting a lot of important data, so make sure you get that stuff right; good data integrity... All these sorts of things. But fundamentally, a good consumer experience, which is the site interacts with you well, responds well...
+
+But what they built was enterprise software. They architected a big, complex machine that had enterprise components that maybe work well if you've got like an analyst sitting at their desk, and maybe there's 12 concurrent users ever using this thing; maybe that works fine. But those were the building blocks... And then deploy that into a data center that didn't have elastic scaling. You couldn't add capacity easily.
+
+**Jerod Santo:** Was it merely scale that was the problem, or was it that it didn't actually work the way it needed to as well?
+
+**Paul Smith:** It both was the wrong conceptual model for a transactional website, like it needed to be; the wrong model. They architected the wrong house.
+
+**Jerod Santo:** Right.
+
+**Paul Smith:** And then it couldn't scale. So you could potentially use scale, you could throw resources at it to kind of overcome those limitations... But the design of it made that really hard. And some of the physical realities. We take for granted we can spin up a VM in AWS, or Google Cloud, or Azure, or whatever it is. The government was not ready for all that stuff in 2013. So HealthCare.gov was deployed into a data center that -- you know, they had VMware, they had some tools like that, but fundamentally, there were racks of servers that were like "These are the HealthCare.gov racks." And that's it. And like a SAND attached to it for network storage, and things like this... But like I said, when that traffic starts flowing in and the individual components are not architected in a way for low latency and responsiveness, you start to get these bottlenecks, these pile-ups dogpile, there's not good caching... So all those components get strained and stressed, and they sort of cascadingly fail.
+
+\[24:12\] And then on top of all of that, the team that was building it was - they were running through the tape. They were still building things. They were exhausted. They were not communicating well across teams. So they just had this big, big, complex thing that wasn't quite the right shape for what they needed, and it wasn't in a physical place where you could just turn up the horizontal scaling knob... And then there was just this lack of communication and coordination.
+
+So yeah, we walked into the situation on day one, honestly thinking "Oh, maybe we'll be here for a couple of days, give them some ideas of what to do next..." Little did we know, we were gonna spend the next 2,5-3 months of our lives basically seven days a week to get this thing turned around... Because we knew that's what it would take, given what we walked into.
+
+**Break:** \[25:09\]
+
+**Mat Ryer:** So were these people used to building government websites with relatively low traffic, and they usually just -- like, manual forms are turned into web... We have that same here local government especially; they don't feel very modern. Is it just literally that, that the experience of the people building it was just for those types of systems, and they'd never really encountered a situation like this sort of high-throughput situation?
+
+**Paul Smith:** Yeah, I think that's exactly right. Basically, when time came to build HealthCare.gov, the way government contracting works is you sort of work with government contractors. You don't really just go out and contract with Google... Although Google does have some government work; but that's not how it would normally work. You would normally reach out to these companies that have historically worked with the government, and government is their main customer.
+
+\[27:55\] And yes, so for the ten years or twenty years prior to HealthCare.gov, the kinds of companies that were sort of bidding on the HealthCare.gov work - their main experience was with building those more enterprise software stacks, and they didn't really have the experience of that consumer web, that at the time, in 2013, was becoming more common and more of a commodity. We were understanding about memcache, and how you scale up an application, how you deliver a good experience in AWS... It was becoming more and more common. That experience and expertise hadn't made its way over to government.
+
+**Mat Ryer:** Yeah. So there's something else about the way of working like that, in these older companies, or bigger companies, with all this architecture and hierarchies, and things - often, you end up isolating by functionality, don't you? So you end up having database people separate from application or business logic people; it's all kind of divided up like this. So that's hard, to have that kind of coherent idea about anything, I find anyway.
+
+And when you think about then having those requirements that are written in stone and written in law often, which you can't then deviate from, it kind of sucks out a lot of creativity. In the startups world, for sure, people are more used to being agile... And really, out of necessity, because we don't really know what we're doing; we just admit it. Whereas in enterprise you can't admit that you don't know what you're doing, so you have to sort of plan everything out in every detail, and then your hands are really tied. Did that play a role, do you think, in this?
+
+**Paul Smith:** Yeah, for sure. What I would say about that is - I think it's okay that government lags behind the private sector in the startup world when it comes to technology... Government is not where you wanna be taking a bunch of risk, and trying out the latest web framework... I mean, maybe you could play around with that a little bit, but in the main, you wanna be a little bit more conservative; let the startups take the risks and figure out what's the next hot tech stack, and then hopefully that trickles into everything else.
+
+Your point about the division of labor is a really important one, because this thing was huge... And for what it ultimately was, which was -- and maybe it's helpful if I just describe HealthCare.gov really quickly, like what it was actually meant to do... The idea is you first go to HealthCare.gov, you sign up for an account - which already should tell you something. Like, if you go to Amazon, you can browse and add things to your cart, and then if you need to create an account at the end -- it's the funnel; you wanna bring people in, and you don't wanna push them through the hardest part of the funnel, which is signing up. That can be a laborious, and kind of get you off the game. In this case, you just wanna look for health insurance.
+
+So we put you through the narrowest part of the funnel up top... So you sign up, and then you have to apply. Apply means with all my personal information about me and my household, am I eligible to buy this healthcare with this subsidy? Or maybe get Medicaid, the expanded version of medicaid. Okay, so there's this application part. And that involves -- there's some business logic there, looking up rules, database interactions a little bit... And then you get to the place where you can actually browser health plans. That's basically a database of plans, with information about their premiums, their copays, their deductibles, the things they cover, what regions of the country they cover, things like that.
+
+**Johnny Boursiquot:** All the things you should have seen first... The shopping part, the browsing part.
+
+**Paul Smith:** Yes, exactly. That's right. You have to fight your way through Mordor, and then you get to the Shire, instead of the other way around.
+
+**Jerod Santo:** Well, the Shire is here. We have socialized healthcare. \[laughter\]
+
+**Paul Smith:** Right, exactly. We used to joke on the rescue that if we were the healthcare.ca, it would just be like "You have healthcare", and that's it; static page. Anyway... \[laughter\]
+
+**Mat Ryer:** \[32:02\] Yeah... Much easier tech. That's a good reason to do it, if no other.
+
+**Paul Smith:** And I have to say, I personally believe we should have affordable, universal healthcare in this country; I think it's a right. I'm really proud of the Affordable Care Act for moving us closer to that goal. It expanded coverage tremendously. What's what was so important to us, and why it was critical that we worked so hard to turn it around, was because we didn't wanna go backwards. We didn't wanna lose 20 million people covered with healthcare; we wanted to lock that in, although now it's up for...
+
+**Mat Ryer:** So it had a kind of political imperative for you, along the thing that we all have about wanting to make the tech work.
+
+**Paul Smith:** Right.
+
+**Mat Ryer:** Did you also have that sort of personal political motivation as well?
+
+**Paul Smith:** I mean, absolutely... Just for myself -- and I don't think that this is a prerequisite for somebody who believes the government should work... Government as a function of something that we do collectively together. You don't have to believe that President Obama was a good president, or that you worked on his campaign as a prerequisite to have worked on the HealthCare.gov rescue. That was an important aspect for me, but I will say that we were hearing all the time -- so the Affordable Care Act was already the law; HealthCare.gov was sort of the delivery mechanism.
+
+But we are hearing all the time from people for whom the law had already made their lives better. They could stay on their parents' health insurance longer, until they were 26, or they couldn't be denied coverage because of a pre-existing condition. Those stories were filtering up to the White House, and then down into the team...
+
+So it's visceral. It's people's lives, and you have this almost direct connection to them... So yeah, when you're flagging a little bit, it's like -- you know, you've worked all day, it's nine o'clock on a Saturday and you'd rather just be done... It gave us all that extra little bit of like "Well, we can't really slack off here. We have to take this over the finish line." So yeah, that was definitely an important part.
+
+I wanted to go back and say that the team that we encountered -- so we were talking a little bit about how there was some team that built it, and this is a combination of government contractors and government agency folks... Our mission and our belief as a team was to have high EQ first... Bring our high IQ about the web stack, but we weren't there to blow anything up. There was six or seven of us at the beginning - we weren't gonna rewrite HealthCare.gov in a couple of weeks, or whatever. We needed them to succeed.
+
+So really, what we did more than anything was point the way to "Here's what this thing should be doing. Here's how you know you're on a path to success incrementally, by adding monitoring, by having a process by which we sort of prioritize bugs and defects, and tackle them in sort of a reverse order of their impact... And here's what the indicators of a successful, high-traffic website look like, and how we can move closer in that direction."
+
+So really, our innovation, if anything, on the rescue itself was bringing -- one of our team members was Mikey Dickerson, who was a site reliability engineer at Google... One of the early people at Google who kind of helped create that culture... Bringing some of those ideas to government. So having a daily stand-up where all the stakeholders could talk about the technical issues, and we could coordinate and communicate and prioritize and plan, which - none of that was happening before we showed up on the scene.
+
+So it created a sense of urgency, it created accountability, which was good... And not just like finger-pointing or blame, but "Hey, we really need you to do this thing, and it's really important, because we need this bug to be fixed", or whatever. And people really rallied to that. So we wrote very little code, although we did write some Go code, that turned out to be pretty load-bearing.
+
+**Johnny Boursiquot:** \[36:11\] Yeah, let's get into that a little bit. So what is the extent to which Go played a role here? It sounds like there was some immediate impact derived out of the involvement of Go, and I'm curious to hear all the different layers and where you got a chance to involve Go in the rescue.
+
+**Paul Smith:** Sure. So the kind of -- I'll try to put you in the mindset of where we were in late November of that year, which was... We had a deadline that we were working towards - end of December. So if you were an American and you wanted use HealthCare.gov, you needed to have signed up by December 23rd to be covered for the subsequent year. So that was sort of driving everything we were doing, that deadline... In the sense that people may have left HealthCare.gov, they tried to use it in the early days and it was a bad experience, they couldn't get on, they had problems, and they went away... But through the media and through other signals, and just the knowing that this is the deadline, a bunch of people were gonna come back in December, all at once. So we had better have this thing be able to handle that surge of traffic.
+
+So everything we were doing was sort of oriented around that. It's how we prioritized what we were gonna work on. So through November we had made a lot of improvements... I'm talking about things like database configuration tuning. Don't have long timeouts on your connections when you need to recycle them, so you can let more throughput through. Things like that.
+
+So we were doing a lot of that, there was a lot of application-level logic fixes, and the site had gotten a lot better. But we knew that when traffic really peaked, and for example the president would come out with a tweet, or something, or he would talk about it on the news, and there'd be this surge of traffic to HealthCare.gov, the site would fall over.
+
+We knew that we weren't quite there yet, so we started to think about "How do you manage that peak demand?" One of the ideas we had was just smoothing the curve of that peak demand. So the peak is in the middle of the day. If you can flatten the peak and then have it spread out over more hours of the day, you reserve some room at the top to keep the site operational. So our strategy was "Let's use some sort of mechanism by which we can essentially shift people in time." So if you're coming to the site and it's a little overloaded right now, we'll invite you back later when the load is less. That's where we came up with this email queue, essentially.
+
+**Mat Ryer:** I thought you meant transport them through time... Because that's easier than solving the scaling issues. \[laughter\]
+
+**Jerod Santo:** Fly them somewhere.
+
+**Paul Smith:** Probably, yeah. Probably a miscalculation on our part would have been an easier route to solve the Schrödinger equation, or something like that.
+
+**Mat Ryer:** Yeah... \[laughs\] So you'd say "We're busy now, but here's a ticket. Come back at this time or between this time", or something like that.
+
+**Paul Smith:** Pretty much exactly that. So - super-simple idea, but we were trying to think of creative ways to just keep everybody from trying to click Reload on the site at the same time in the middle of the day, and nobody have a good experience.
+
+**Jerod Santo:** What I like about that idea is it's pragmatic. It's not perfect, it's a compromise... I mean, it's not cool to be like "Hey, our website's busy. Come back later." That's not what you would want to have to do...
+
+**Paul Smith:** Super-not-cool.
+
+**Jerod Santo:** ...but it's way better than the alternative, which is like everybody at this time of day is just not getting what they need. So it's very pragmatic.
+
+**Mat Ryer:** Yeah, Amazon's never said "Come back this time." It just said "Put your credit card in here right now."
+
+**Jerod Santo:** That's right.
+
+**Johnny Boursiquot:** You can only pull off this whole ticket-based, come-back-later thing for something that people actually really need, and they don't have a choice about.
+
+**Jerod Santo:** Yeah. Captive audience.
+
+**Paul Smith:** \[39:57\] That's the thing... If you're trying to sign up for healthcare for your family, you're sufficiently motivated to keep trying. You're gonna come back. Yeah, so that's what we did. But here's the thing - we're still operating in this environment of this complex site and data center, which - I didn't even talk about how difficult it was to even just deploy code. That was a high-risk endeavor, just to do a deployment. Just to change the code or change the configuration was very high-risk.
+
+**Mat Ryer:** That's a terrible one, actually... And I even see some teams working on far less important tech fall into that same problem, where you're either too scared to change and deploy, or sometimes it is just really hard to do; there's lots of process or lots of things that have to happen. And yeah, there's something, again, about being able to be iterative and quick, because you can be sort of opportunistic and pounce on things; you can be more agile (the lower-case agile).
+
+**Paul Smith:** Right. It was a nervous-making event every time we changed the site, whether it was new code, a configuration change... We actually had static logic in a business rules engine; I don't know if anybody is familiar with these things, but they're like basically outboard brains with if, then, else statements, that had their own lifecycle of change, and very complex... So yeah, it was just not a good environment in which to introduce something like "Hey, this emergency email queue."
+
+So what we did was we just -- we made the case that "Hey, we're gonna requisition these two servers over here that have nothing to do with anything - they're not part of the data center - but they're within the same kind of like general security boundary as the rest of the thing... And we're gonna run our own code over here." And then at the CDN level we're gonna route everything past HealthCare.gov/emailqueue or whatever; we're gonna route that to those servers.
+
+So the rest of the infrastructure stays the same, we don't touch that, and if our thing blows up, you can just get rid of that route and it's fine. So we decided to write this thing... We had a couple of design decisions at the top, which was this thing had to be dead simple... Because we were gonna be the ones to develop it, and we were already sleep-deprived, and dealing with a hundred different things... So we didn't wanna add more complexity than we absolutely needed to. It needed to be dead-simple to operate, for some of the same reasons... And we wanted something that was going to be easy to deploy, easy to operate, and then easy to kind of get people back to the site. So what we came up with was basically just a loop, a Go thread or goroutine that would pull off a JSON request from the website - a simple form in the website that we injected with JavaScript, that grabs your email and a couple of other bits of metadata - and then we just wrote it to a file, and we did that atomically inside of a lock... So literally, all this traffic is just flowing into these files, just text files, that we're just like appending rows to... Because we didn't wanna mess with the database and separate processes... We just wanted a process, an OS process that we had control over, that we could use text processing tools on the backend to do the actual email send. So that's what we did.
+
+So we just collected people's emails all day, and then when we saw the load dip under the threshold that we thought it was safe, we would do these sends to invite them back, with a special code that let them by-pass the waiting room if that was still a thing... And we brought them back.
+
+I should mention that we had like this throttling mechanism that essentially dialed in whether you got the email waiting room, or if you could go straight in through the site. So it was this sort of like probabilistic thing that was like a function of the load on the site at the time.
+
+**Johnny Boursiquot:** That sounds really low-tech in a good way... No more than is needed.
+
+**Paul Smith:** \[44:04\] It was the least clever thing we could come up with. If I have to convey some life lessons here, going back to just that last point about deploying code, I think one of the things you wanna do as early on in an endeavor, a project, a startup, a new project, whatever it is - you wanna exercise that path to production as early as you possibly can... Even if it's just putting a Hello World out there, it exercises your DNS, it exercises your hosting, it exercises your CI/CD pipeline. You wanna do all of that early, instead of finding out when you're ready to have a big publicity campaign that you forgot to tell so-and-so to turn on auto-scaling, or something like this. So that's lesson one.
+
+The other lesson is like the higher the stakes, and the bigger the audience maybe, the less clever you wanna be... Because when things break, they break non-linearly. They don't break in just simple, straightforward ways at scale like that. They kind of catastrophically break. And then you have this added pressure to restore service... So you wanna make it as easy on yourself as the person who's in operations to recover... And the best way you do that is by not being too clever while you're building the software.
+
+**Mat Ryer:** Yeah, I think that's great advice.
+
+**Paul Smith:** Easier said than done, but that's kind of a good rule of thumb.
+
+**Mat Ryer:** Good goal, yeah. I like that things break at scale. That's a good lesson.
+
+**Paul Smith:** Right. They kind of splinter out in ways that are hard to predict... Especially when you're talking about a distributed system with a lot of components, cascading failure is a real failure mode that is hard to reason about in advance.
+
+**Break:** \[46:01\]
+
+**Jerod Santo:** So what was the total time that you were on this project? When did you feel like you could call it quits, like "Well, we're no longer needed here. Go back to regular life"?
+
+**Paul Smith:** Well, I'm looking at my watch... It's been seven years and...
+
+**Jerod Santo:** Oh, you're still on it? \[laughter\]
+
+**Paul Smith:** My initial involvement was through that first period of time, through that deadline of December 23rd, 2013 (I think it was). That rescue team kind of grew and contracted over the next several months, because there was a final deadline in 2014, which was March, something like this... And so I stepped away from that. But the experience was so searing... What I mean by that is having come from that campaign, having come from a startup community, and then seeing this piece of critical infrastructure... Because I think the right way to talk about HealthCare.gov, or any kind of government digital service, whether it's a website, or something you interact with to get either a service or a benefit - that's critical infrastructure. It's a form of infrastructure; it just happens to be through digital channels.
+
+**Jerod Santo:** For sure.
+
+**Paul Smith:** \[48:07\] It's unacceptable to me that somebody could not get their healthcare because a website didn't work. There's something so viscerally wrong about that. We know how to make websites work, right? We know how to make websites scale, we know how to have good user experiences... It is unacceptable, so I felt that really viscerally. And it's not just about the technology scaling too, from the technology hardware and software perspective. It's also about -- you know, user experience can be... An interface, the language of the site, the design can also be a way to kind of disenfranchise people, or keep them from achieving their goals. So that has to be a consideration as well. But that visceral feeling...
+
+So my co-rescue team partner, Greg Gershman, who I met outside the White House that first morning - he was also a software engineer, had been a presidential innovation fellow, so that's how he kind of came into the whole thing... We looked at each other and were like "We should really start a company. We should start a company that can bring the knowledge and experience that we have about developing modern digital services - web applications, websites - with great customer experience, great user experience, and offer that to government, and say "This is a better way of doing the things that you yourself are saying you wanna build, but you don't have the talent and experience to do it." So that's what we did.
+
+We called ourselves Ad Hoc, because we called ourselves the Ad Hoc team during the rescue... Because when you're in a meeting with a bunch of government agencies and contractors, you go around the room and announce who you're with... And since we were kind of assembled just Avengers style, like you said, one of us said "We're the Ad Hoc team", and that kind of stuck. So in an homage to that effort, we called the company Ad Hoc. Our first customer was CMS, who is the government agency responsible for HealthCare.gov.
+
+**Mat Ryer:** Around the table you're like "We're the Ad Hoc team", and no one's interested, and you're like "We're actually from the White House." \[laughter\]
+
+**Paul Smith:** Well, you know, that was the thing - we didn't wanna bigfoot our way into the situation like that, because that's a way to get people to cease up... And we wanted them to open up to us, and we wanted to show them we were in the fight with them. We weren't just gonna point fingers and be gone the next week. We were there...
+
+So yeah, people knew we were from the White House. Word gets around in an instant. But we did everything we could to show "Hey, we're just part of the team. We just wanna get this thing to work."
+
+**Mat Ryer:** Right. That's great.
+
+**Jerod Santo:** This is why Mat didn't get the call.
+
+**Mat Ryer:** Yeah, I'd be like "Hi. I'm just like you, although I did arrive in a motorcade.. \[laughter\]
+
+**Paul Smith:** The Chief of Staff at the White House told me not to screw this up, so... Don't screw it up. \[laughter\]
+
+**Johnny Boursiquot:** No pressure.
+
+**Jerod Santo:** Yeah...
+
+**Mat Ryer:** That's what Jerod says to me at the start of these shows...
+
+**Jerod Santo:** That's right.
+
+**Mat Ryer:** It doesn't work... Did it work for you though? It kind of worked out for you, didn't it? You didn't screw it up.
+
+**Paul Smith:** Well, I think we felt like -- you know, there was this question of "Should we scrap the site?" The question was being asked, "Is this thing recoverable?" and I think there was a sense that "Yeah, maybe they built the wrong thing, but we can make it good enough to get through this deadline..." But the challenge is really going to be that people problem of communication, and prioritizing, and knowing what the right fixes are from our experience of having worked on the high-traffic things... So yeah, all of that pressure was there to help keep us focused. It's hard to ever say failure is not an option. You just don't quite invite that into your head in the moment.
+
+**Mat Ryer:** Hm... Well, I usually don't like it when I hear managers of teams saying that... Because in a way, you need to be free to fail in an environment where you're building things... But yeah, sometimes it's just "Yeah, we actually can't fail on this one. It's just too big a deal." That's really interesting, to hear that perspective.
+
+Well, it's that time... We're gonna do Unpopular Opinions!
+
+**Jingle:** \[52:37\] to \[52:53\]
+
+**Mat Ryer:** So who would like to put forward our first unpopular opinion of the evening?
+
+**Paul Smith:** Can I go first? Because I don't know what the history of Unpopular Opinions on this show is... So I wanna make sure that mine is like -- you know, it's like when the figure skaters go first in the order, so that the judges are like "Yeah, yeah..." And the real skaters go after. \[laughter\]
+
+**Jerod Santo:** Please do. Go ahead.
+
+**Paul Smith:** Okay. Because I really don't know what the stakes are. So my unpopular opinion is that server-side generation of websites is superior than static single-page applications.
+
+**Mat Ryer:** I see. So you're talking about "Do all the rendering on the server, and just ship the HTML rendered", versus ship a big, thick JavaScript client and then use Ajax or something for the back-and-forth.
+
+**Paul Smith:** Yeah.
+
+**Mat Ryer:** That's a controversial one, I think...
+
+**Jerod Santo:** Yeah. I 'm with you on that one, but Mat is not. Are you, Mat?
+
+**Johnny Boursiquot:** I haven't heard that one in a while...
+
+**Jerod Santo:** Yeah.
+
+**Mat Ryer:** Well, it has the simplicity going for it, that's for sure. And in frontend dev -- frontend, especially if you have big frameworks working and there's lots of things going on, you can get into some strange situations. Some user will click this first, they open this drawer, and then they go and click something else, and suddenly that's a state that has never entered our minds. So you do kind of control a little bit more doing the server-side rendering. Why else do you like it, Paul?
+
+**Paul Smith:** Well, I think there's a couple of big wins... One is you can have a better user experience, especially over high latency and low throughput/low bandwidth links, because you can just push a minimal set of HTML, versus a big monolithic JavaScript application payload.
+
+Now, I know that there's splitting and there's been some innovation on that front, but that first interactive usability I think is still superior on the server-side sites. So that's one.
+
+Another one is accessibility... And I know that accessibility has come a long way on SPAs, but my experience has been that it's easier to kind of bake that in on the server-side HTML, because you're really leveraging everything that the browser is giving you by default, instead of having to essentially rebuild up a browser in JavaScript, for all intents and purposes, in your SPA. So those are the two big reasons that I think of...
+
+It definitely has some downsides. There's another flywheel to go wrong somewhere... So it's not all roses, but I like the trade-offs better. And I'm not saying no JavaScript at all, I'm just saying the primary rendering should happen on the server.
+
+**Jerod Santo:** Sure.
+
+**Mat Ryer:** What do you think of that one, Johnny?
+
+**Johnny Boursiquot:** I think I've seen this evolution take place over the years... Usually, those who start out being backend developers, where that's really what their bread and butter is, once they start doing a little bit of frontend development, they're like "Yeah, this is the natural progression. I'm gonna use my server-side code to push out the frontend code. Great." And eventually, they'll either make that transition to doing full-on frontend, all-there, stateful JavaScript development, or they'll sort of stick with those server-side rendered pages, because there's a comfort zone there.
+
+\[56:11\] Then you have people coming to it from the other side, saying "Hey, I'm a JavaScript developer, I'm all about the UI, I'm into the CSS, I'm into the HTML DOM and all that stuff..." They're coming at it from this other side, and then they get to the edge, where they're like "Okay, I don't really wanna go do that backendy stuff you're talking about... That Django, the Rails... Maybe not." And then Node comes along and you say "Oh yeah, I can do backend now. \[laughter\] I'm gonna take my JavaScript skills and go to some backend." So it depends on where you're coming from. You're gonna have a different posture but yeah, it's not right or wrong, it's just different.
+
+**Jerod Santo:** I really think it does depend on what you're building... And I know that's the moderate stance, the "it depends" stance...
+
+**Mat Ryer:** Yeah. You should get a bumper sticker for your car.
+
+**Jerod Santo:** I like to differentiate between a website and a web app. I think that's a useful distinction. I know a lot of people say there's no difference... But I think most websites should be server-side rendered... And I think most web apps or rich web apps - I would not server-side-render Gmail, I would not server-side-render Slack or Trello. Those are applications running in a web browser context... Especially if you have a multi-client situation. If you're building a startup that's gonna be multi-client from the start... Slack knew they needed an iOS client, Android client, web client... I think an API plus an SPA is a smart move. Now, most startups don't make it to that point. They fail far before they get to that point, so it really does depend. But I tend to be with you, Paul.
+
+**Mat Ryer:** Paul, you'll be pleased to know we test these unpopular opinions on our Twitter, @GoTimeFM. So we'll find out if that's unpopular or popular based on that.
+
+**Jerod Santo:** That's right.
+
+**Johnny Boursiquot:** Jerod, didn't you come with some unpopular opinions?
+
+**Jerod Santo:** I did, I brought one.
+
+**Mat Ryer:** Oh, here we go.
+
+**Johnny Boursiquot:** Lay it on.
+
+**Jerod Santo:** And let me just say that I'm a bit disappointed -- impressed, but still disappointed by the unpopular opinions that have been represented thus far, because to a one, on Twitter, they've all been actually popular.
+
+**Johnny Boursiquot:** \[laughs\]
+
+**Mat Ryer:** I think people on the show - they make a good case, and they're quite convincing. So then you put the clip out and ask people to vote, they're like "Yeah..."
+
+**Johnny Boursiquot:** They're like "Yeah, I can see that..."
+
+**Mat Ryer:** "Yeah, I think Johnny's right again." \[laughter\]
+
+**Jerod Santo:** Well, I'm here to break the streak, okay? I came up with what I truly believe will be an unpopular opinion.
+
+**Mat Ryer:** But is it a firmly, really honestly held opinion of yours? Or are you just trolling and you're just trying to find one that's the most unpopular?
+
+**Jerod Santo:** Well, let me state it and then you can decide at the end.
+
+**Mat Ryer:** Okay.
+
+**Jerod Santo:** So I'm not gonna come on a podcast about Go and say that JavaScript is a better programming language; I'm no fool. I wanna walk out of here alive. But. I will happily start a proxy war by saying that JS Party is a superior podcast to Go Time.
+
+**Mat Ryer:** Ooooh...
+
+**Johnny Boursiquot:** You're off our show. You're off our show.
+
+**Jerod Santo:** \[laughs\] Let me quantify this a bit, okay? I have some evidence. So more is better, okay? We have more panelists, we have more male panelists, we have more female panelists, we have more variety, we play game shows, we host formal debates, we write and rehearse poems, we explain things to each other like we're five... You guys don't explain anything to each other like you're five.
+
+Go Time records on Tuesdays, one of the worst days of the week. JS Party records on Thursdays. Thursday is closer to the weekend; obviously, better. We cover more topics... Go Time is about Go. JS Party is about JavaScript and the web. That's twice as many things.
+
+**Mat Ryer:** That's cheating. That's cheating.
+
+**Jerod Santo:** That's twice as many things. We know the web is huge, so... Tons of variety.
+
+**Mat Ryer:** You can't take HTTP to a JS Party. \[laughter\]
+
+**Jerod Santo:** So in review...
+
+**Mat Ryer:** See, we do poetry...
+
+**Jerod Santo:** ...we have more awesome panelists, we have more variety, it's on a better day... And this is the big finale point. You're gonna like this one. JS Party has 100% less Mat Ryer, which means we really cut down on those awkward silences. \[laughter\]
+
+**Johnny Boursiquot:** Wow...!
+
+**Mat Ryer:** That is quite the pitch. That was the first time an unpopular opinion has been used to advertise things. \[laughter\] Johnny, have you got any products you wanna push while we're doing Unpopular Opinions? \[laughter\] My book's really unpopular. Let me just get a copy and show you all... \[laughter\]
+
+**Johnny Boursiquot:** \[01:00:16.26\] Wow...!
+
+**Paul Smith:** It was more of an alienating opinion, I'd say...
+
+**Johnny Boursiquot:** I know... \[laughter\]
+
+**Mat Ryer:** But more is better. It doesn't sound like you were listening to Paul on his message there of simplicity...
+
+**Jerod Santo:** Well, I ended up with the "Less is better. Less Mat Ryer." So I went on both sides of the equation.
+
+**Mat Ryer:** Good point.
+
+**Paul Smith:** Goodness...
+
+**Mat Ryer:** I think it's fewer Mat Ryer. \[laughter\]
+
+**Johnny Boursiquot:** You got me clutching my pearls. \[laughs\]
+
+**Jerod Santo:** I think maybe I've offended everybody here, but that's the unpopular--
+
+**Mat Ryer:** Yeah, it's a challenge. You've thrown down a gauntlet and we're probably not gonna pick it up. We're quite happy with this show... \[laughter\] We're not gonna mess around. We're all gonna do some game shows, and things... Mess around a little bit.
+
+**Jerod Santo:** Okay.
+
+**Johnny Boursiquot:** Yeah, I guess we need to add some game shows, and...
+
+**Jerod Santo:** Spice it up, guys.
+
+**Johnny Boursiquot:** ...have Mat do a little dance, or something... I don't know.
+
+**Mat Ryer:** Yeah. On a podcast. We could all pretend it was good. No one would think otherwise... \[laughter\]
+
+**Jerod Santo:** Mat does do impressions. We're gonna get those going.
+
+**Mat Ryer:** Yeah... I was gonna do a series of videos reading the Go documentation as Jack Sparrow, if you'd like a preview of that...
+
+**Jerod Santo:** Please do. Give us a taste.
+
+**Mat Ryer:** So here's Jack Sparrow reading filepath.Walk. \[in Jack Sparrow voice\] "Walk walks the file tree rooted at root, mate, calling walkFn for each file or directory in the tree, including root. All errors that arise visiting files and directories are filtered by walkFn. The files are walked in lexical order, mate, which makes the output deterministic, but means that for very large directories Walk can be inefficient. Walk does not follow symbolic links. Savvy?" \[laughter\]
+
+**Jerod Santo:** Okay, I take it back. Go Time is better.
+
+**Johnny Boursiquot:** Oh, wow...
+
+**Mat Ryer:** That will get cut out though, don't worry...
+
+**Jerod Santo:** Oh, no.
+
+**Johnny Boursiquot:** No, that's going in.
+
+**Jerod Santo:** That's in there, baby. It might be the cold open. \[laughter\]
+
+**Mat Ryer:** I might do that entire standard library as an audiobook...
+
+**Jerod Santo:** I like that last line, because you made it sound very eerie and dangerous.
+
+**Mat Ryer:** I was running out of breath, and then halfway through I thought "Why am I doing this?! It's being broadcast..." \[laughter\] So there were a few things going on there...
+
+**Johnny Boursiquot:** My goodness.
+
+**Mat Ryer:** Well, that's all the time we've got today...
+
+**Johnny Boursiquot:** \[laughs\] Yeah, it is...
+
+**Jerod Santo:** it is.
+
+**Mat Ryer:** But Paul, thank you so much for coming on and sharing your story with us. It's such an interesting one. It's nice to hear Go making a difference. Thanks for all the work you're doing; it seems very important... So yeah, please come back anytime and hang out.
+
+**Paul Smith:** Yeah.
+
+**Mat Ryer:** We'll see you next time!
+
+**Paul Smith:** I appreciate it, thank you. Thanks, everybody.
+
+**Outro:** \[01:03:02.13\]
+
+**Jerod Santo:** Good show, guys!
+
+**Mat Ryer:** Yeah, very interesting. Really good show.
+
+**Paul Smith:** That was fun. Thank you so much.
+
+**Jerod Santo:** Thanks for sharing that story.
+
+**Mat Ryer:** Yeah, really good. We're still live, by the way, for a bit... But that's the end of the recording.
+
+**Jerod Santo:** Yeah. The official show.
+
+**Paul Smith:** Can I plug my company?
+
+**Jerod Santo:** Oh yeah, man.
+
+**Johnny Boursiquot:** Plug it. Plug it. Plug it.
+
+**Paul Smith:** So we're still working on HealthCare.gov, and the story there is we've basically rebuilt it from the inside out, for the most part, over the last N number of years... So it's basically substantially new, or changed, than what we found... But that's not our only work; we're working on Medicare... So if your parents are picking a plan for Medicare, they're probably using our Plan Finder tool... We also work with the Department of Veterans Affairs, and build services for veterans... And we're getting into more and more stuff.
+
+So if you like the idea of working on high-impact digital services that increase the general welfare, I'd encourage you to take a look at Ad Hoc, because we have a lot of open positions. We are hiring product managers, we're hiring software engineers, designers, user researchers... We actually have more than one former anthropology Ph.D. on staff, doing user research. So they brought their anthropology skills to conducting user research to make better user experiences and better interfaces for veterans and other people. So it's really interesting...
+
+I'd like to say that we're kind of like coming up with what product management in government can and should be; the synthesis of the best of the consumer world, consumer technology and what government needs... It's a really interesting space, and we're still growing... So check us out.
+
+**Mat Ryer:** And what's your website?
+
+**Johnny Boursiquot:** How big is the team? Whoops...
+
+**Paul Smith:** Sorry, I heard two questions...
+
+**Jerod Santo:** Pick your favorite.
+
+**Paul Smith:** Yeah...
+
+**Mat Ryer:** How big is your website? \[laughter\]
+
+**Paul Smith:** How big is the website? ...
+
+**Mat Ryer:** Yeah, if you could blend the two questions together...
+
+**Paul Smith:** 75 pages...
+
+**Jerod Santo:** Where can we find your team? \[laughter\]
+
+**Paul Smith:** The website is adhoc.team. Also resolves at adhocteam.us. I think we're about 300 people actually, so we're not a small enterprise anymore. We're not a small startup, we are a going concern. That's what you transition into after being a small startup, is a going concern.
+
+But we use Go in a lot of our projects still, and... Yeah, if I had another hour, I would tell you about how we've used Go in rebuilding HealthCare.gov. So yeah, we're still trying to bring the best of startup technology and proven web technology to the government.
+
+**Mat Ryer:** Great.
+
+**Jerod Santo:** Awesome.
+
+**Mat Ryer:** It sounds good. We'll post the link also in the show notes, for anybody listening that's interested there.
+
+**Paul Smith:** Cool.
+
+**Mat Ryer:** Yeah, great stuff.

Immediate mode GUIs_transcript.txt

@@ -0,0 +1,395 @@
+**Mat Ryer:** Hello, and welcome to Go Time. I'm Mat Ryer. Today we're talking about immediate mode GUIs. Hm, what does that mean...? And specifically, we're gonna be taking a look at the Gio project, which you can check out one gioui.org. We're gonna find out what that's all about.
+
+Joining me today -- well, it's only Johnny Boursiquot, isn't it? Hello, Johnny.
+
+**Johnny Boursiquot:** Hello. It's just me. Nothing special.
+
+**Mat Ryer:** It's just you, mate. Very special, I think. And also special - Jon Calhoun.
+
+**Jon Calhoun:** Hey, Mat.
+
+**Mat Ryer:** Jon, I did a compliment for you in your intro.
+
+**Jon Calhoun:** "Also special", is that what you said? \[laughter\]
+
+**Johnny Boursiquot:** He had to call that out.
+
+**Mat Ryer:** Yeah... Well, last time I think I said you should be tolerated, so I wanted to be nice.
+
+**Jon Calhoun:** I guess I'm moving up. \[laugher\]
+
+**Mat Ryer:** Well, don't worry though, we're also joined by somebody who I've seen speak and does some great presentations. You can check them out online. It's Elias Naur. Hello, Elias!
+
+**Elias Naur:** Hello, Mat.
+
+**Mat Ryer:** Welcome to the show.
+
+**Elias Naur:** Thank you very much!
+
+**Mat Ryer:** I'm really excited about this episode, because I saw your talk at GopherCon last year, 2019, and it was excellent. Could you just kick off and tell us, what do we mean by immediate mode GUIs? What does that mean?
+
+**Elias Naur:** So immediate mode is, in contrast to retained mode - and retained mode is the design or the way to structure regular or most other user interface libraries. So retained mode is where you have a lot (if not all) of your visible user interface state in the user interface library. The example I like to use is the browser, because almost everyone knows a little bit of JavaScript programming and the browser DOM... And in the DOM, you store all that is visible on the browser page, and also of course many things that are not visible... But for our case it's the visible thing that we're interested in.
+
+\[04:15\] And the thing about retained mode is that you usually have (at a very high level) your program state duplicated. So if you have, say, a list of contacts or something else you want to display in your browser page, you typically have that list twice. You have some representation in your program that you got from a database, or from the network, or somewhere else... And then you have it -- in the case of Go, you have Go structures to represent those contexts.
+
+But on the other hand, you also have that information stored in the DOM in the case of the browser, because you have a DOM element for the label of the name of the contact, you have a DOM element that represents the profile image of the contact, and so on. If you have other information on that page, well that's also represented by DOM elements.
+
+So in contrast, immediate mode gets rid of all that state from the user interface library, and replaces that with something that - if you've ever done any video game programming, you would know the structure of a video game is essentially a giant for loop. Not a giant for loop, but a top-level for loop that simply loops and asks whether there's an input from the users, say from the mouse or keyboard or whatever your device is, and updates your state - the game state, so to speak - and then draws it, and then on and on it goes.
+
+For immediate mode user library it's almost the same thing, except that you don't want to do this all the time, because that would be a waste of resources... So instead of the for loop that goes on and one, you have a for loop that waits for events at the very top level, and that's very neatly modeled with a go select statement of channels.
+
+The interesting thing about immediate mode is that one of those events that you're waiting for is an event that essentially says "Redraw everything." And that's what immediate mode is; you draw your entire program state -- not your entire program state, but the state that is relevant for the screen, for what the user should be seeing in the window.
+
+**Mat Ryer:** It draws it every time. I mean, it draws it in full.
+
+**Elias Naur:** It draws it every time. It's basically drawing everything. Because what you can do - and we can come back to that later - is you can cache what you drew in the previous frames. And that is done for you automatically in libraries like Gio. So even though the program behaves and uses the library as if it were drawing everything from scratch, every frame, what will actually be recomputed and drawn onto the screen is more or less what differs from the previous frames, or for previous five or ten frames, or something like that.
+
+And the great thing about that is that you don't have this duplication of state. That is not just the duplication for, say, the labels -- let's take an obvious example. If you have a window and one label inside it, in the browser DOM you will bring that to the screen by constructing a new element, and putting (say) a paragraph element, or a div element, and then put the text inside that, and then add it to the DOM. In immediate mode you don't do that; you end your update function that redraws what is relevant. It will notice, somehow, that the program wants to display this label, so it just draws that label, and that's it.
+
+And again, back to the retained mode example, if you say you have this label that represents a name, and an update to the profile comes from the network, to say "Well, the user changed his name", or something, then you have to change that name in your structure, which is unavoidable, but you also have to have an existing reference, or obtain a reference to the DOM element that represents the names. And sometimes the name is displayed several times, in the same user interface; you have to update all those duplicate state somehow.
+
+\[08:12\] And again, if you want to remove a label, or remove a contact, or something like that, then you have to go in and take the element out. So not just duplicating the actual state, which is the username, you're also duplicating so to speak the metadata of that label, which - does it exist at all? Where is it? Where is it positioned? And so on.
+
+And again, in immediate mode you get an update, "Okay, I'll redraw/respecify my entire interface, and draw whatever is relevant." And if you don't need the label, you just don't draw it.
+
+So that's the drawing example, and I like to say that a user interface has three major tasks. The first one is drawing things - draw the button, draw the text... And that is actually very complicated, because it has to be done -- especially text is very complicated to draw. But at a high-level, it's not that interesting, in the sense that it's more or less a solved problem. It's not trivial to do, but it's a solved problem.
+
+What is even more interesting is the second task, which is layout. Layout is simply the job of positioning your things. So instead of using "I'd like this label to be positioned at this X and Y coordinate, and I want it to be this wide and this high", and do that for every element, and make sure that they don't collide, and they react to the window size changes or anything else - that's very difficult to do for anything other than trivial programs.
+
+So the layout is simply a way for your user interface library to give you tools to automatically place -- or maybe not automatically, but place things relative to other things. And a common thing is, for example, centering that label inside the window. If you want to center it, then again, from the browser example, you have to add more state. In the browser you could probably set a CSS property on the paragraph or div elements, for some other libraries you need to construct a container element, and especially if your layout operation is more complicated than just centering... You have to construct a container widget, which is not even visible, and put the label inside of that container, and tell the container "center whatever there is inside you." And then again, you have another part, another state introduced twice, because you have the knowledge that you want to center it in your program, which might just be implicit from your program, and then you have to encode that state in the browser, in the form of a container or properties.
+
+And again, back to the immediate mode - as you draw the label, in Gio that's simply something that could be condensed just to a function that says "center this thing", and in Gio it's implemented as a function, but takes another function, and everything you draw or specify inside that inner function - well, it's just centered. And then that's it. There's no knowledge from the library that you have centered something, and there's not something that you should update the next frame to say "Well, now it's not centered. It has to be right-aligned", or something like that. So the state, again - it's implicit and it's gone as soon as you used it, so to speak.
+
+And the last task for any user interface, which I think is the most interesting, is event handling. That's where - at least in my opinion - retained mode libraries, or the design of retained mode leads to the most unmanageable programs and applications, because what they do, what most libraries do is that they force you... If you want to, say, have a button in your interface, and react to that being clicked, then you need to register something, typically a callback. In Go it could be a function or a closure, where you say "Okay, to this button set the event list to this function." And that, again, means that you have state hidden -- not exactly hidden, but you have state duplicated in your user interface library.
+
+In the DOM you say addEventListener() and you call it 'click', and then you give it a JavaScript closure or a function.
+
+\[12:10\] And that of course leads to "When do you remove this callback?" What if this callback is invoked? Because you can't really -- a lot of bugs are hidden, in the sense that they only trigger when you click the button at the wrong time. That's a very typical example of something that can be very difficult to deal with in traditional user interfaces. What if the user clicks it on the wrong time? And that almost never happens, because the user almost always behaves nicely, if you can say that. He/she clicks the button when it's time to click the button, and don't click it when it's not relevant to the user interface. But sometimes they do anyway, because of delays, or something like that, and then you get a weird crash and you have to debug that.
+
+So again, back to the immediate mode design - you don't want state in the user interface, in the library itself, so you don't have callbacks. What you do instead is as you specify the interface, you at some convenient time (convenient for you, the programmer), you ask the button "Have you been clicked since the last frame?" And that does an if check and if you have something like a toggle bar, then you could do a for loop, say "As long as you've been clicked", if you have a user that very quickly clicks a toggle, a checkbox twice, then you need to update the state twice. But other than that, it's more or less just an if statement. Say "if button clicked...", and then update your state. Do whatever you want. Show something, print something to the screen, initiate a network request, whatever you want to do. And that is it.
+
+And again, if you don't want to handle events because you're not ready to do it, or the program does not expect the clicks to happen from this button, then you just don't check it. You don't execute that if statement somewhere in your program. So adding and removing the callback is done automatically, because there are no callbacks... And you get to handle this button click where you want it. At this point in my program I'm ready to handle the click - then that's where you write that if statement. If you have a callback, then you don't know really -- it's not that it's just multi-threaded, so you don't have a data race as such, but your program is not in a well-defined state when you get a callback from the user interface library... Because it detects the click, and then it calls the function, and who knows where your program is -- whether it's really ready to accept these input events. So that's three wide and large tasks that user interface libraries handle, and where I think that immediate mode really makes a difference.
+
+**Johnny Boursiquot:** As you were explaining how this mechanism works, I was picking up on some of your terminology. You were talking about frames... And that sort of took me back a little bit to when I was exposed to folks doing work with -- maybe I'm gonna age myself a little bit here, but Macromedia Director...
+
+**Mat Ryer:** Casper?
+
+**Johnny Boursiquot:** \[laughs\] Casper... Technologies like Macromedia Director, Flash, Flash authoring, Flex, and that whole category of tooling. And there are some more sophisticated things out there as well that have come out, but all these things operate sort of on a timeline, if you will. It's almost like there's an expectation that you're gonna be moving through the "frames", on a timeline, at a given rate, at a given frames per second kind of rate.
+
+So is there a loop within (in this case) Gio, where basically you have the opportunity at the next iteration of the loop? Are you redrawing everything? Or why are you using some of that terminology? How does that connect back to this kind of timeline-based animation and creative tooling?
+
+**Elias Naur:** \[15:52\] It connects back because that's exactly what happens. You have a for loop at some top level. In a Gio program you have a for loop at the top level of your program, typically... And that for loop typically contains a select statement, which waits for events from (say) the network, but also from the window, which feeds the event channel with frame events. A frame event is simply a request from the system for any reason to redraw you. And in essence, you go through the whole user interface as it is right now. Not your whole program state, as I said before, but you go through the relevant program state necessary to respecify, redraw the window as it is now.
+
+And you say framing and timing - there's not an explicit timeline as such, but the interesting thing about what you're saying is that to animate things with immediate mode and Gio. There's a most simple thing you can imagine, and that is simply asking for the frame events at regular intervals, which is typically when your monitor will refresh (say) 60 times per second if it's a 60hz monitor, or even more if it's on a phone... But you ask the Gio system for "I'd like to animate things." More concretely, what you do is say "My state is changing all the time." So you get these frame events at regular intervals, and you simply update your state according to the current time, and that's it. That's how you animate.
+
+So if you want something to fly from one part of the screen to the other side, you simply start with positioning it at the beginning, as for these frame events as long as you're animating, and then slowly move that button along the path. And it could be a spline, it could be a linear path, it could be whatever you want.
+
+The cool thing about this is that you still have to program the animation, which is not trivial, but it doesn't have to be built into the user interface library. As far as I know, most other retained mode libraries need some way for you as a programmer to specify "I'd like to animate this property." Say a button going from green to red; if you want to animate it going from some state to another. And you have to tell it "Okay, I'm starting with red, I'm ending with green, and it has to go through this path, follow this curve, and it has to take (let's say) half a second, or a hundred milliseconds" and so on. And then say "Go."
+
+So you describe this animation typically to the user interface library and apply it to a property or a set of properties, and then make it do the animation... Which is very convenient, of course, but is also very inflexible, because you can't do anything. Well, you can do many things, but you're not free to change your user interface libraries in any way you like. You're only free to do it in the way that the library expects you to do, and have added support for doing.
+
+So I think that's one of the many advantages - something like animation is taken completely out of the picture; you can just not talk about animations if you like to, because it's not really relevant to the library.
+
+**Mat Ryer:** it's interesting. So in having this immediate mode, presumably you draw the whole thing from scratch. So you kind of start with clearing the screen. And if you're doing that very quickly, then drawing has to be very fast. So how does Gio draw things quickly? Is it just that it doesn't take that long to draw things, or is there something else that Gio does to help keep it quick?
+
+**Elias Naur:** Two things. The first thing is that you don't explicitly say "clear screen" and then "draw this, draw that". What you do as a program developer is you actually specify your interface as a list of operations. So you say "Okay, I want an area here which is wide, and on top of that I want this button shaped blue, and on top of that I want some text" and so on. So what you end up with is a list of operations that describe at a quite low level what your screen is going to look like, and you pass that to the georenderer... Which you don't know anything about, but you just pass it along and say "Okay, do whatever you want with that."
+
+\[19:57\] And what Gio can do at that point is that -- it doesn't do that much yet, because it's pretty fast already, but what it can do is that it can recognize a diffing algorithm. It can take "Okay, what did you draw in the last frame, and what are you drawing now? What are the differences between these two operation lists?" And it can do that very efficiently, because I decide in the API what is the format of this operation list. It can be constructed so that it is very easy to find the differences, and then just draw that difference. That's one reason.
+
+The second reason, just very shortly, is that it uses the GPU to draw, and the GPU is screamingly fast, even on phones.
+
+**Break:** \[20:44\]
+
+**Jon Calhoun:** Assuming I'm understanding this correctly, whenever an immediate mode program, something like Gio is rendering, and you say it's like this for loop that continues iterating every time, I assume that that means that every single time one of the ways that you're avoiding bugs is you're essentially getting like a snapshot of all the data, because like you said, you're not duplicating it. So you don't have to worry about that "Well, I was in the middle of changing something, but it's not quite done." Because this is the data every time, it presumably should make things easier to replicate. If you give it this set of data and tell it to render, you should see the same thing on screen every single time. Is that correct?
+
+**Elias Naur:** Yes. A Gio set of functions or methods is basically an iterator that goes through your program state, however you have defined it, and then outputs Gio operations to draw it. That's a very high-level view of it. So yes, it is stateless, in the sense that the same input in your program should give the same results, at all times.
+
+**Jon Calhoun:** So would it be fair to say that React was kind of trying to make immediate mode sort of available to people working in the DOM?
+
+**Elias Naur:** Yes.
+
+**Jon Calhoun:** Okay. I was just trying to make sure I understood that all correctly... Because it seemed like a lot of the sales pitch React had was the same things you're pitching, of "You've got the state, you don't have to worry about all this. It does all that for you", and you're not trying to alter all these things.
+
+**Elias Naur:** Yes. I haven't used React, but what I've seen from it is that they still have an explicit representation of state. And I don't know why they have it. Perhaps it's a limitation of trying to put an immediate mode library such as React on top of something that is fundamentally retained, such as the browser DOM.
+
+But as far as I can tell, they still have some kind of explicit state tracking. So you decide what the state is, but you still have to give that state to the React library somehow. And again, I haven't used it. And then the React library will, in some automatic/magic way sort of diff between the previous state and the updated state and makes sure that the update happens efficiently.
+
+**Mat Ryer:** Yeah, it uses a virtual DOM.
+
+**Elias Naur:** Yeah.
+
+**Mat Ryer:** We're actually using Svelte... You built the same kinds of things -- as a JavaScript framework you built the same kinds of things, but all the processing happens at compile time. So there isn't a lot of runtime in the browser with Svelte. It's just -- they do all that work at runtime. So I think that's quite interesting... But still, conceptually, I'm not sure where it fits in with those... But it's really interesting to hear about that.
+
+\[24:19\] I was gonna ask - so you mentioned clicking a button. I had an Amiga, and I used to love building UIs on this Amiga. And you actually didn't have any frameworks then. You could draw rectangles and lines and pixels and circles, I think, and you could fill, and not much else. So to do things like even change the state of a button on hover, it was a case of catching the mouse events, and then comparing through some -- the way I used to do it was just kind of a global X/Y on the screen, positioning. So I would just check "Is the mouse kind of greater than the X, but not greater than the X plus the width of the button?" and that told you if it was in this part of the screen... And same for the Y axis. And then you could know... So does Gio have to do things like that, at that low level? Does it deal with any other kinds of abstractions, or is it literally kind of from the ground up, it's dealing with those low-level problems?
+
+**Elias Naur:** Gio as a library is from the ground up. When you consider what to base your UI library on, it's very tempting to use something else... Say for Android and iPhone, for example, they have a very rich set of widgets, and they have a lot of behavior already encoded in those widgets... So many frameworks take the approach of reusing -- they say "Okay, we'll just reuse whatever is available on the platforms", and then try to make that behave the same across platforms... Which, of course, works pretty well in the beginning, but then (at least in my experience) breaks down. The devil is in the details, essentially.
+
+So Gio went the same route as Flutter. It's actually inspired by Flutter in the sense that -- Flutter is Google's Dart library for doing cross-platform user interfaces... And they took - in my eyes at least - a very bold step of saying "We'll never use (as far as I know) the native widgets and tool kits. We'll only require somehow to draw, some access to your GPU - probably rendered software if you wanted to - and some way to access the keyboard and mouse events."
+
+So yes, in essence, Flutter as well as Gio does at the lowest level handle events like that. So the button has been programmed to say -- it will register and input areas, and say "I'm interested in pointer events in this area." And then whenever it's inside that area, then it will hover or react to it if it's been pressed and then released, and then only if it's pressed and released inside that area, and so on. So all those low-level checks, so to speak, and pattern matching on the input stream is done at some level. But of course, it all boils down to when you're using a Gio button in your program, you simply say -- you have a button, you draw the button, and ask for what it clicked. That's it.
+
+**Mat Ryer:** Great.
+
+**Elias Naur:** But it's done in a way -- so there's nothing magical about that button, or the editor, or all the other widgets in Gio. There's no tricks up the sleeve, so to speak. So you can actually take the button and modify it as you want, and you can do that in practical terms, because the source is unlicensed... And modify it however you want to, and it will render in the same way as the built-in (so to speak) widget. So there's nothing magical about the built-in widgets; it's just the widgets that I've deemed that they're generally useful, and then add it to the standard library, so to speak... To the standard Gio library.
+
+**Mat Ryer:** So what's the code look like for a Go programmer then? If you're gonna describe, say, a container with three buttons in there, maybe a dialogue box with some label, and then two buttons, how do you describe that in Go code? Or are you talking solo level there's literally a draw or an update interface or something, and you just have to implement that?
+
+**Elias Naur:** I can show a program, but that would not be very useful on a podcast... So what you're asking for is what is the structure of a typical Gio program...
+
+**Mat Ryer:** \[28:07\] Yeah. Would you create a struct that describes a kind of view, and have that contain sort of child nodes of the things that make up that page? Do you build it in that kind of --
+
+**Elias Naur:** No, we will do it -- actually, it depends on the program, of course. If you have something in the program that needs to be dynamic... Say you have a list of users; then you need somewhere a slice of user objects, I suppose, which is then filled in from somewhere else, and then you need that slice to represent a dynamic number of users.
+
+But say you have one button and a label beside it, and that's it, then it's actually -- and that's one of the great things, and perhaps also a bit surprising... What you do is you simply say "draw button", "draw label" and that's it. It's a little bit more complicated, because you want the button and the label to be beside each other, and not on top of each other, and so on. So you typically have a layout object around it. But that's not an object that you construct and then save. You actually construct a value object in Go, which is very easy to do... Like a struct constructor, setting the margin or the alignment, whether it's right, left, or center... You construct a value object and you call a function on that, which takes a closure, that is another function, that you define, and inside that function you can draw things.
+
+So if you have, say, a Flex layout, which is a way of putting things on either a row or in a column, below each other or besides each other, you can say "flex", and then it takes a variable number of child functions, so to speak, and each of those functions is a widget. So if you want a button and a label, you take this Flex object, call this layout method, and the first function you give it, you draw the button, and the next function you give it, you draw the label. And the Flex object will make sure that the two widgets are placed either besides each other, or on top of each other, depending on the direction.
+
+**Mat Ryer:** So are they given bounds then at runtime? Does the layout component say "Okay, I want you to draw, and here are your bounds. This is the space you have to draw in"? Or does the layout component somehow apply that information before it then gets drawn? Do you see what I mean?
+
+**Elias Naur:** Yes. Indeed, there's a layout protocol. It's implicit, in the sense that you can't guarantee that the widgets follow this protocol. But what happens is that there's a global context, which I usually call the GTX. It's more like the Go standard library's context object... But it's the context for geoprograms. And that context at all times contains the current constraints, which is sort of what you were saying with the bounds. The constraints simply say "In the X direction you must be at least this number of pixels wide, but your maximum is this number of pixels", and the same thing in the Y direction, in the height.
+
+So you get constraints sort of as an input in this context, and it's up to the widgets to fill out a dimensions, a concrete size, width and height, that they chose to be layout-wise. So they can draw outside these bounds if they want to, but this is what will be used for layout operations. You can sort of imagine, this is a recursive process, so the Flex will give every child element some constraints, and will use that to place the next one, and so on.
+
+**Mat Ryer:** That's really cool. It's nice to know that, because of course, everyone's gonna probably -- if you're building an app, you're gonna need those kinds of layouts, to be able to say "Spread these elements out evenly" or "This is the one that can expand to fill the space, these other two are fixed", those kinds of things. They are very useful, and I saw in some of the source code examples it is really quite easy. You just sort of create like a tree of elements, isn't it? It works all the way down. Really interesting.
+
+\[31:53\] So what sort of use cases is this for? When you look at it on the GioUI.org, it does look like an application framework for building frontend apps. And by the way, on that website there is an image of an example view that you've renedered... And I was blown away, because there's a little Run button next to it, which I almost missed; and clicking that sort of ran the WebAssembly version of it, didn't it? Which is crazy. So it supports WebAssembly.
+
+**Elias Naur:** Yes. But not very well. Well, it supports WebAssembly in theory, but in practical terms the Go implementation of WebAssembly - which is an entirely different discussion - is very, very bad... Because one thing, WebAssembly is not that great either, because it lacks support for something such as threads, for example; you only have access to one thread... And it's interface -- for example for Gio I need the access to WebGL, which is the browser way of doing hardware accelerated graphics, and you essentially have to call a Reflect-like interface from Go, which then calls into JavaScript, which then invokes the browser's underlying WebGL implementations... And as you can imagine, that is very inefficient.
+
+And on top of that, the WebAssembly bytecode (the virtual machine of WebAssembly) is a stack-based machine, which fits very poorly to the current Go compiler, and so on. So the code is not very efficient. So it works on WebAssembly, and it's correct, in the sense that it's the same code that runs in the browser, that runs as a native application, but it doesn't run that fast.
+
+**Mat Ryer:** Yeah, you probably wouldn't use it with Gio. The reason I brought it up was just really -- I'm quite impressed with how portable it is, because of course it runs on Linux, it runs on macOS, iOS and tvOS as well, and Windows and Android, and things. So yeah, having that portability I think is another quite interesting thing, because of course, Go is cross-platform. But lots of the different choices that you have when you want to build lower-level things, you tend to have to be specific to a particular architecture sometimes. So how does it do that? How do you keep that abstraction? Is it just that you're dealing with the common low-level things, like drawing and receiving events?
+
+**Elias Naur:** I think the primary reason that it's so portable is that I made the same choices as Flutter did. I think they made the choices for the same reason - they wanted something that was maximally portable. And to do that, you simply have to, as you would in any other case, reduce your dependencies as much as possible.
+
+So Gio has almost the minimal dependencies you can have, which is simply somewhere to draw, and somehow to receive input events. Everything else is done in Go code. Actually, it's so portable -- I released a unikernel version of Gio. I have a port that runs without an operating system, and directly in QEMU; that's how portable it is. Because it really only needs somehow to draw the graphics, somehow to receive the input events, and that's about it.
+
+**Jon Calhoun:** So when I'm looking at this example, like the kitchen example on Gio's website, I think what's kind of interesting to me is that it's easy when you start running it to sort of just think "Oh, this is just HTML elements, or something." You click, you can add text, you can type, all that stuff works... But I'm assuming that this is actually either like a Canvas, or something like that, in the DOM -- is that what you're doing? ...and just writing to it.
+
+**Elias Naur:** Yes. It's a Canvas with a WebGL context attached to it.
+
+**Jon Calhoun:** Okay.
+
+**Elias Naur:** And that's another reason why even though we got Go to run very efficiently with WebAssembly, you may not choose it anyway to use in the browser, because you'll kind of have the same problem as you have with Flash, or with the Java applets of the old. It's much better integrated in the browser, but it's still not so well integrated that you can select this element and inspect it from the inspector... It doesn't integrate 100% with the rest of the page if you have that.
+
+**Mat Ryer:** \[36:01\] You could just use HTML and CSS, I suppose...
+
+**Elias Naur:** If you have something that you need to be very slick, then yeah, you should use something that is native to the browser. But of course, if you have a project which is primarily an app or a desktop application, and you need a quick and dirty way to draw that in the browser or activate that in the browser, then that's a perfectly viable way to do it, because the code is literally the same. So other than dealing with a different size, and so on, it's the same code.
+
+**Mat Ryer:** So like visualizations - it would be quite a cool use case for it, because it would be great if you could render some data visualization in Go. You might just render it into an image and serve it that way, I suppose, but... Yeah, anyways, it's a cool thing to play with. And I think one of the things Gio has is this playability. As soon as you see it, you feel like "I could build something with this." I can tell people have that sense about it.
+
+**Elias Naur:** And it's also very easy for me to do live examples, like you have examples on the Go Docs page, or the package go.dev. Just like you have with the kitchen example, essentially... I wrote an article comparing the immediate mode to the retained mode way of designing your user interface, and I used live examples. I used the DOM example, of course, which is native to the browser, but I could very easily integrate small snippets of Gio code and actually run it, so I can demonstrate.
+
+I hope and expect that that will be very useful for doing the future tutorials, and articles, and so on, in the same way as the Go Doc examples, which are also viable in the sense that you can change them and play with the examples, and run them from the browser. If you can run these examples from the browser, then I think the WebAssembly port is worth it just for that.
+
+**Mat Ryer:** Yeah, I agree. Absolutely. And it is great, if you go to that website - we'll put it in the show notes - there are definitely some things to play with.
+
+**Break:** \[38:04\]
+
+**Mat Ryer:** When you look at the website you do get a sense that it's for applications, but what about games? What about more fun things that you could potentially use it to build? Is that viable?
+
+**Elias Naur:** Yeah, sure. An issue you have with games is that you typically want a very low-level access to your graphics card. If you have anything else than the most basic games, then you need this axis. And Gio, for portability reasons, only more or less exposes the operations that are necessary to do 2D vector-based user interfaces well. So what I've done to cover this use case - because I really want to cover those niche use cases well, because one thing that happens when you introduce something like this to the world is that all the Go programmers say "Well, that seems like a good idea." But if you ask a so-called frontend developer, he will say "Why should I care about Go at all?" So I have to learn a different language just to use your framework.
+
+\[39:50\] So I'm trying to reach those niches where you have enthusiastic Go programmers anyway, or you have an existing bunch of code anyway, and I think that games is a very good example... Also because I used to do computer games in a previous life, about 15 years ago.
+
+So the problem is, of course, that you need the low-level access. And the great thing is that that is also solved by Gio being very portable. So what you can do is you can extract Gio and only use the widgets and the layouts and all the tools that are in the Gio Go code, the portable Go code. And then you can attach another renderer to it. So there's somewhere in Gio that translates these operations, say "circle there, rectangle there, background there, text there" - translates that to GPU commands. And you can take those commands and intercept them and merge them with whatever you have.
+
+So if you're writing a game, you typically use some kind of game engine, whether you're writing it yourself or it's some third-party thing, and you need all the drawing to go through that. By splitting up Gio from the actual operating system or native part, it's very easy to integrate with games. So yes, it's very much a thing that I'd like to see done with Gio. But the point is just that Gio in itself does not make writing games that much easier; it makes writing the user interfaces much easier, of course, I hope... But it doesn't really help you with the details of, say, writing a 3D game, or 2D game, or so on. So even though it's using the GPU, it doesn't give you the tools to efficiently and portable do an interesting game with.
+
+That is not quite true... So the guy (Egon) who did the logo for Gio, has done some interesting animations with Gio, which was done quite inefficiently, because Gio doesn't give him that much to work with... But he still managed to get something that renders quite quickly on his machine. So it is, I suppose, possible to do a game, if it's vector-based, and so on.
+
+**Jon Calhoun:** What is your ideal application that you'd like to see built with Gio? If something would just pop up and just blow you away and just excite you... Was there anything in mind whenever you were building it, that you'd really love to see built with it?
+
+**Elias Naur:** What I wanted to do, and I started to do, and I presented that very early version of a chat application which is completely decentralized, because I'm interested in decentralizing all the things, so to speak... So I did something called Scatter; it even has its own domain name, as Mat would say. I bought Scatter.im. The idea is - very shortly - that you use email as a message transporter, instead of have a centralized service... But something like that would be very interesting to do. And interesting, user-facing, popular app for phones... Because users on phones are the most picky.
+
+You don't see many new desktop applications done these days... And if you do, it's an administration tool, or some editing tool. It's a workhorse, it's a tool of some kind. But if you do something on the phone, it's meant to be for the so-to-speak general public. It needs to be polished, it needs to have all the corner cases covered, and so on. So that would excite me quite a bit. I've actually been hired to do an Android app, so I hope that I will come out at some point and blow you all away.
+
+**Mat Ryer:** Are you going to use Gio?
+
+**Elias Naur:** It's going to use, Gio, yes.
+
+**Mat Ryer:** Oh, that is cool. Well, you've mentioned the logo... I think projects should have logos, as a general rule; open source projects like this one. The Gio logo is very good, by the way. Compliments to -- who is it who did it?
+
+**Elias Naur:** Egon Elbre (if I've pronounced his name correctly). He does a lot of these Go illustrations, and logos, and so on. At some point -- I think I mentioned on Twitter some months ago, and said "Well, it has a name, it has a website, but it doesn't have a logo. If anyone has ideas...", something like that. And I think half an hour passed and he came up and said "I'll do one if you like." And we did a bit of back-and-forth on the Slack channel, and then he came up with the logo... And I completely agree, I love that logo. It perfectly embodies what I think Gio is. It's simple, it's vector-based...
+
+**Mat Ryer:** Immediate...
+
+**Elias Naur:** And immediate, yes.
+
+**Mat Ryer:** As soon as you look at it, it's there.
+
+**Elias Naur:** \[laughs\]
+
+**Mat Ryer:** \[44:03\] It is very good. But as a general rule as well, it's a nice thing -- for anybody that's getting into open source that doesn't feel like their coding is up to scratch yet, it is a kind of great way to get involved in other projects, is by doing logos or other kind of artwork bits and pieces like that. And I think it brings the projects to life. I don't have any data on it, nor have I done any research whatsoever, but 80% of open source projects with a logo probably are gonna do better, aren't they?
+
+**Johnny Boursiquot:** \[laughs\]
+
+**Mat Ryer:** And that's science.
+
+**Johnny Boursiquot:** And you've just introduced an artificial barrier for every open source developer out there. Way to go, Mat. \[laughter\]
+
+**Mat Ryer:** Wh-- what...
+
+**Jon Calhoun:** I mean, he did kind of make a tool that makes it a little bit easier to make gophers, that you can generate a logo...
+
+**Mat Ryer:** Yeah, yeah. I mean, to be fair, Ashley McNamara really deserves the credit for that project. My involvement was putting PNGs over the top of each other.
+
+**Jon Calhoun:** Okay, you did a little bit of work, and then Ashley did all the heavy-lifting, and then you got to say "I've got this really big number of combinations, and Ashley only drew this many illustrations."
+
+**Mat Ryer:** Exactly. I made the billions of possible gophers possible.
+
+**Johnny Boursiquot:** Yeah. \[laughs\] Elias, I'm interested in hearing what you define as success for Gio. Obviously, we've talked about use cases, what you like to see people create, what your ambitions are around being able to create with Gio itself... I'm curious, what does success look like for you with this project at this point?
+
+**Elias Naur:** I would say two things. The first thing is Gio 1.0, which is where I hope that we can say "Okay, the API is stable", in the same sense as the Go compatibility promise... So we can release a module, let's say 1.0, and then we're not going to change the API. If we are, we're going to release a version 2, and so on. So all the promises that will make people in turn say "Okay, I think this is a cool project, but I don't want to deal with the API changes all the time", and those kinds of people can start using it for real. So I would say 1.0 is a very important milestone... Also because to reach that you don't just need to cover all the necessary features to reach that 1.0; what is even more important - you need to have enough users so that you can be certain (or relatively certain) that you've covered all the things that would come up when you're doing an application.
+
+I can't say now that all the things that I've done up to now will be frozen, because I don't know whether a use case will appear that disqualifies that API that I've done. So when 1.0 is reached, I know that there will be enough programs using this and being hopefully happy about it, and so we can say "Okay, practical use has shown us this API is pretty good."
+
+The second thing is probably the funding of the project. That it's well funded, so that the buzz factor hopefully increases - or is it decreases? I don't remember. And also so it's viable, in the sense that the people working on it actually do it because they can be supported while doing it. And that's actually been done quite well up until now, in the sense that there's a handful of people that signed up for my sponsorship page of GitHub, and especially Emanuel from [Orijtech](https://orijtech.com). He did a very generous sponsorship, even though he's not really using Gio for anything.
+
+So sponsorship is very important... But also, what I'm going to work on is this Android app, which will both hopefully validate Gio as a practical thing, but also kind of say "Okay, it's not just open source in my free time", so to speak... Because it's full-time for me. I see this as a business for me. So yeah, viable in the sense of funding, and 1.0.
+
+**Mat Ryer:** That's brilliant. Well, it's that time again... It's time for our Unpopular Opinions.
+
+**Jingle:** \[48:11\]
+
+**Mat Ryer:** So who wants to go first? Anyone got an unpopular opinion?
+
+**Johnny Boursiquot:** I volunteer Elias. \[laughter\]
+
+**Elias Naur:** Okay, I wasn't actually sure what kind of unpopular, and I asked Mat in this email asking me for an unpopular opinion, "How unpopular?", I asked him.
+
+**Mat Ryer:** \[laughs\] How unpopular do you want it?
+
+**Elias Naur:** Yeah, that's it. So I brought two of them. One is very on-topic, almost obvious, and the second one is slightly - perhaps even very - off-topic. And they are both at least controversial, and maybe even unpopular. So which one do you want?
+
+**Mat Ryer:** Controversial, please.
+
+**Elias Naur:** Both are controversial.
+
+**Mat Ryer:** Oh, well in that case, either one.
+
+**Elias Naur:** Okay, I'll start with the very obvious, which is of course that in my very arrogant view, I think that the retained design of other user interfaces has really slowed down and wasted many developer resources to deal with it. But more importantly, that's just completely guesswork, and not fine science, like you're doing, Mat.
+
+I think that one of the reasons that there's a difference between a frontend developer and a backend developer, so to speak, or that there is a distinction at all, or a very common distinction is that many developers will simply be fed up with the tools that are available for doing user interfaces.
+
+So you sort of say "Okay, this is very tedious. It must be because I'm not very good at it." And then they sort of say "I must be a backend engineer, because I'm not capable of dealing with all this complexity, and tediousness, and why doesn't this work, and why do we have to use all these libraries?" and so on. So at least a controversial, and perhaps an unpopular opinion.
+
+**Mat Ryer:** It's a good one. Do you think if we hadn't ever had this retained thing, we'd be like on Mars by now? Do you feel like it's really held humanity back that much?
+
+**Elias Naur:** Perhaps the Moon then, again. A few more times on the Moon. \[laughter\]
+
+**Mat Ryer:** Okay, cool. That's perfectly fine. I'm not sure that enough people know enough about it for that to be considered unpopular. Well, what was your other one?
+
+**Johnny Boursiquot:** The next one, yeah. I wanna hear that one.
+
+**Mat Ryer:** I'm excited about the unrelated...
+
+**Elias Naur:** The unrelated one is short and simple. It is that I think that everyone should own a little bit of Bitcoin, and especially during this crisis we're in. That's short and simple, and hopefully unpopular.
+
+**Mat Ryer:** \[laughs\]
+
+**Johnny Boursiquot:** Well, it's not unpopular to me. I already own a few, so... I support that unpopular opinion.
+
+**Jon Calhoun:** I feel like you should have told us this before we aired the episode, so we could just ride that wave... \[laughter\] No, I don't think we're that popular.
+
+**Mat Ryer:** So you think this is gonna cause a spike?
+
+**Jon Calhoun:** I don't think so. I wish, but I don't think so.
+
+**Mat Ryer:** Yeah.
+
+**Elias Naur:** I remember bringing it up at a table at some lunchtime at the GopherCon, and the people around the table just looked at me, and the only question was "Isn't it really bad for the climate?"
+
+**Mat Ryer:** Oh, really?
+
+**Elias Naur:** I think I got the impression that Bitcoin is not very well received or seen affectionately in this community... But of course, I could be wrong. So what are your unpopular opinions?
+
+**Johnny Boursiquot:** Quickly diverting... Mat?! \[laugher\]
+
+**Elias Naur:** Quickly diverting. I think that I'm hearing that it's not unpopular enough... I'd like to hear genuinely unpopular opinions.
+
+**Mat Ryer:** Johnny, have you given an unpopular opinion yet? Because the thing about Johnny Boursiquot is I think he's too nice to have an unpopular opinion. What do you think?
+
+**Jon Calhoun:** I thought that was gonna be your unpopular opinion somehow. \[unintelligible 00:51:56.00\]
+
+**Johnny Boursiquot:** Yeah, too nice... No, I come up with unpopular opinions once every three episodes... So catch me in the next one.
+
+**Mat Ryer:** Okay, yeah. That's great.
+
+**Elias Naur:** Diverting again? \[laughter\]
+
+**Mat Ryer:** \[52:10\] He doesn't run in immediate mode, so... \[laughter\]
+
+**Johnny Boursiquot:** Nice, nice... \[laughter\]
+
+**Elias Naur:** He does more of the retained type.
+
+**Jon Calhoun:** I think we could talk more on the -- the frontend developer one is always weird to me, because I feel like there used to be the stigma that frontend developers weren't real developers... And especially now, it's really frustrating in the sense that when you're working in JavaScript and React and all that stuff, you're doing real development. There's no doubt about that.
+
+Now, if you're just designing pages, HTML and CSS, and you're just making it look good, I could maybe see the argument of "Well, you're not really dealing with logic as much at this point. You're just dealing with aesthetics." And it may be there. But going sort of against your opinion, I think part of the reason some developers don't do frontend stuff is because you kind of have to have that -- I almost wanna say "design sense." I've always felt like you have to have that design sense of being able to make something look good. Because if you can throw all the HTML DOM elements on the page, it still doesn't look like you know what you're doing, unless you can make it look pretty.
+
+So for me, that was always one of my struggles, as I -- I mean, it's kind of like you said, the tools are kind of not amazing, in the sense that I'll be there after an entire day and be like "I've got one page looking good, that's it. Where did my whole day go?" Whereas you're doing something backend and you're like "Oh, I just got all these things working, and it's great..." But it's also frustrating in the sense that with backend sometimes it's hard to demonstrate what you're building. It's hard to show people that.
+
+**Johnny Boursiquot:** You just said something that I think is perhaps -- I've gotten hints of that, and because earlier in my career I was doing more frontend and design kind of work, so I also got a sense of that as well... You're conflating all the different roles. I think most of us would be considered by frontend folks to be backend folks; folks who work with stuff behind the scenes, stuff you can't see.
+
+So by thinking that "Okay, I can sit down as a developer and just come up with a beautiful frontend application that not only looks good, but the user experience is good, and the functionality behind the scenes is good, and the remote server that this thing talks to is good..." The proverbial full stack, as we call it. Do you think that one person can do all these things well? And I'm not saying those people don't exist, I'm just saying that that's like conflating different disciplines. So UI is different from UX, which is different from being able to program given a UI, after the UX has been done.
+
+So these three things alone are different disciplines. Sometimes you have folks who are talented enough that they have all these qualities, and can also do the coding behind the scenes, and do the backend work, but that's a lot of things. And in my experience, if you're doing all these things, you are gonna have a stronger area of focus. Very rarely do you do all these things extremely well, all the time. You are gonna have stronger areas. But I think there's this sort of false belief that one person can do all of these things. Perhaps startups look for the one person who wears all the hats, that way they don't have to hire all the disciplines... But who knows.
+
+That's the idea, that's what I'm thinking - we shouldn't conflate all these disciplines, because they are disciplines in and of themselves. They require study, they require understanding and knowledge and expertise.
+
+**Jon Calhoun:** I think it's especially challenging because, like you said, startups will -- like, if you start with one developer on a startup, they kind of have to do everything. They might not be great at it all, but they have to do it all. And then when you go to hire, usually the way that gets split up is there's one frontend guy - and by frontend I mean he's doing UI, UX, he's doing probably some coding still... He's doing all these different things, all put together, and they just sort of define it as a frontend developer.
+
+\[55:53\] Then the larger a company gets, I feel like you're correct, definitely, that these things get isolated and you specialize in what you're really good at... And that ends up having much better results, I'm sure. But it's just very hard, because a lot of these smaller companies are kind of -- it's almost like everybody's hiring using the same terms, but looking for different things, and that makes it really challenging.
+
+**Mat Ryer:** You're right, the size of the team I think is important for this... Because David and I - we're building something, and there's just two of us working on it, and we're doing the whole stack. There's pros and cons to that. One of the pros, of course, is we can think about a user experience thing and know all the way down to the database how that's gonna work. When the problem is small enough and you can do that, you can deliver quite a really good experience, paying attention to all the bits in that.
+
+As the team grows, realistically that becomes much more difficult... Although I have seen small teams of essentially full-stack developers, where they basically there's no roles; everyone's just responsible for everything. Some people are naturally better at one area than others, and that happens even within backend development, too. But yeah, there are pros and cons to it.
+
+I think the more you can do, the better, because like you said, you get that nice, joined-up experience. But at scale, it starts to get difficult. And then of course, you need then even more people to be able to glue the other bits together... So suddenly information architecture, which is another UI discipline that's separate to UX design and UI design and stuff. So yeah, I think it's a tricky one, and I think team size comes into play quite a lot.
+
+One question I didn't ask you though, actually, was this one - how does Go talk to the GPU?
+
+**Elias Naur:** How does Go talk to the GPU?
+
+**Mat Ryer:** Yeah.
+
+**Elias Naur:** It depends on the platform, actually. In WebAssembly you simply call this syscall/js package, so the JS package, which is kind of a Reflect package for JavaScript. So there's a WebGL command \[unintelligible 00:58:05.07\] exposed as JavaScript functions, and you can call them directly from the Go program that is built for WebAssembly. But for the other platforms you typically use cgo to interface with a DLL, the dynamic library that exposes an API that gives you access to the GPU. That is typically OpenGL on most platforms, but for Windows -- you can also use OpenGL on most platforms, but the best thing to use is Direct3D.
+
+So there's actually two backends - there's OpenGL, and there's Direct3D. But on Windows - it's actually very interesting, because the syscall package is capable of dynamically loading DLLs, and then invoke functions from those DLLs. So you can avoid cgo on that platform.
+
+**Mat Ryer:** That's crazy. But of course, Gio's done all this work for us, right? If we use Gio, we get to get all that for free.
+
+**Elias Naur:** Yeah, you don't have to deal with that.
+
+**Jon Calhoun:** If I build something with Gio, does that make building it and sending it off to different platforms a little bit more challenging, or how does that work?
+
+**Elias Naur:** For Windows, because it's not using cgo you can build a Windows Gio executable from everywhere you have Go. For Android, you can build it if you have the Android SDK installed... And the NDK, because - well, it's Android. For macOS and iOS--
+
+**Mat Ryer:** EyeOS.
+
+**Elias Naur:** EyeOS, yeah... \[laughter\]
+
+**Johnny Boursiquot:** Good Lord...
+
+**Elias Naur:** So macOs and iOS - of course, it's Apple, so you have to have an Apple machine more or less, and you have to use their toolchain to build the cgo that enables you to access the native parts that you need.
+
+**Mat Ryer:** Building for the Apple TV is quite a promising, enticing little thing, isn't it? Imagine building your own TV apps in Go. We may have to do a hack day on that. That could be really fun.
+
+Yeah, it's a great project... Thank you so much, and thanks for coming on the show, Elias. I hope you will come back and talk to us again.
+
+**Elias Naur:** Thank you very much.
+
+**Mat Ryer:** It was great to have you. Thank you. And for everyone else, we'll see you next time.
+
+**Break:** \[01:00:32.12\]
+
+**Johnny Boursiquot:** When I came to the U.S. and I kept hearing "Casper, Casper, Casper...", I'm like "What is this Casper thing everybody keeps talking about?" He's like "Yeah, this is a childhood thing... Casper, Casper!" And I finally saw Casper, the Friendly Ghost, the original -- I don't wanna say black and white, but it was an old, old either short, or film, or something. And I was like "Oh, okay. Casper, the Friendly Ghost."
+
+Then I started seeing Casper on mattresses, and things, I started seeing Casper everywhere. I'm like "What?! I don't understand. This confusing." \[laughs\]
+
+**Mat Ryer:** I don't get that. What is it? Why was everyone talk about Casper so much, Johnny, when you moved to the U.S.? Is Casper all the thing? "You've not seen Casper?! You've gotta see Casper!" Everyone just loves it.
+
+**Johnny Boursiquot:** I know. Yeah, everybody's talking about Casper. I'm like, "I've gotta see this Casper cat." It turns out it was a ghost.
+
+**Mat Ryer:** It's like a Game of Thrones of its time. \[laughter\] It's a ghost, but it's friendly. It's such a twist!

Introducing your team to Go_transcript.txt

@@ -0,0 +1,341 @@
+**Mat Ryer:** Hello, and welcome to Go Time. I'm Mat Ryer. Today we're talking about how to introduce Go to teams that don't currently have Go. When is this a good idea, and when is it not? What do we need to know before we embark on this? Helping me get to the bottom of it today - it's Jon Calhoun. Howdy, partner!
+
+**Jon Calhoun:** Hey, Mat. How are you?
+
+**Mat Ryer:** I'm good, mate. How are you?
+
+**Jon Calhoun:** I'm pretty good.
+
+**Mat Ryer:** You notice I did that voice, because I had thought that Calhoun Jon sounds like a bandit from a Wanted poster in a Western, doesn't it? Like, "WANTED! 25 Dollars."
+
+**Jon Calhoun:** Only $25? I must not be a very good criminal. You have to adjust for inflation. That's actually quite a lot of money these days. We're also joined by author of Learn Go With Tests. It's only Chris James. Hello, Chris.
+
+**Chris James:** Hello, Mat. Good to see you.
+
+**Mat Ryer:** Welcome to Go Time.
+
+**Chris James:** Thank you very much. I'm very excited to be here.
+
+**Mat Ryer:** Yes, I'm excited to have you, mate. An interesting fact about Chris - not only did he write Learn Go With Tests, which a lot of our listeners I think will be already familiar with, both of his names are also first names.
+
+**Chris James:** Yeah, no one actually pointed that out to me before, actually.
+
+**Mat Ryer:** Yeah.
+
+**Chris James:** Interesting.
+
+**Mat Ryer:** You're in good company. Rod Stewart... Jimmy Stewart also is another one, and John Stewart. I could only think of Stewarts.
+
+**Chris James:** Yeah...
+
+**Mat Ryer:** Which do you go by, Chris or James?
+
+**Chris James:** Chris. James would be a mistake.
+
+**Mat Ryer:** Okay. \[laughs\] Okay. We don't like to make mistakes.
+
+**Chris James:** No.
+
+**Mat Ryer:** Okay, so let's get started then. What's a good reason -- why, first of all, do people want to introduce Go into teams? What are some sensible reasons for this? Got any ideas?
+
+**Chris James:** \[03:45\] I mean, Go's pretty good, ain't it? The reason it's my language of choice is because any old developer can write some software on their laptop and declare they've made something... But making software that can be used in a broader context, with a team of multiple people, and software that will live not just for a few months, but for many years - that's a whole different skill. And I kind of feel like Go's main strengths are cases towards that kind of thing. Things like the backward-compatibility guarantee I think is a really big thing. In a previous life I had to upgrade a Scala 2.8 project to a Scala 2.10 project. And that took me weeks. You compare that with -- like, in the Go world, it should in theory just be a little, comfy change and you're done.
+
+So when you bring in things like the testing story, the excellent standard library, it does make a sensible choice, I would say, for a general purpose language.
+
+**Mat Ryer:** Okay, that's interesting. I like the quote; I wrote down what you said. You said "It's pretty good, ain't it?" And I think it is. But is that enough of a reason? The fact that I love Go - is that a good reason for the team then to start doing Go? And I'm thinking also, if you're a member of the team that's not yet responsible for tech decisions, and some people -- it depends on how they organize themselves. Sometimes you have tech leads, or even managers, or even people outside of the team are responsible in selecting the tech. So is it enough that it's just something you like?
+
+**Chris James:** No, probably not. What you have to bear in mind is that programming languages - they're not a goal in themselves; they're a means to an end. So if you're thinking about championing Go in your job, try and put on your boss' hat, as you mentioned. It's unlikely that a CTO is gonna get a bonus because some systems were written in Go. I don't think that will be in her contract. She's probably more measured on how successful projects are executed, and I think there are more specific things, like uptime, performance, costs, and that kind of stuff.
+
+I think if you're trying to bring Go into your company - yeah, you shouldn't take that point of view that I just went with; take the point of view of your boss, and think about how can you help them achieve their goals using Go. Because then you can relate Go and its strengths to those problems, rather than just thinking about all the cool things about Go.
+
+**Mat Ryer:** Yeah, that's interesting. And you're right, there are some aspects of Go which do make this a better candidate, actually, than other languages, I would say. One of the examples, the thing that always springs to mind for me is the minimalism in the design of Go. The fact that there aren't that many keywords, there aren't that many language features. It's quite a cut-down language, especially if you compare it to C\# and Swift, and other languages. So in a way, I think it does have an edge to be a good candidate for this.
+
+**Jon Calhoun:** I think one of the ways it really works in that sense is that - let's say you're a Rails shop, and you use Rails because you can whip up CRUD-type apps really quickly, and Rails, Django, those types of frameworks are very good for that. But if you have some sort of background job, some sort of API endpoint or something like that where performance becomes more of a problem, or you need concurrency or some other thing there, the fact that Go is something where anybody who's programmed in almost any language can just look at the code and generally know what it's doing is a big plus. You don't have to retrain your entire organization to be like "Okay, we're gonna learn this new Rust thing." And Rust is one of those languages for me where I look at it and I'm like "I actually have to learn this. I can't just look at some code and go make some changes." Go, on the other hand, while it might not be idiomatic or perfect Go, I can understand it and make some changes.
+
+So I think one of the better reasons that you might have for looking at the CTO's eyes is if you have a problem that you need some other language for and Go is a good fit, you could probably make some solid arguments on that side, where it's gonna solve this problem, it's gonna be easy for us to pick up; all of those pluses are there. But I do think that as developers we have that desire to learn new things, and I think that might be a mistake sometimes.
+
+We talked about this in the last episode, where people wanna learn Kubernetes and all these other things, so all of a sudden they go out and introduce them to the project, and then when they go leave somewhere, it's like "Who's maintaining this? Who knows how to do it?" and nobody on the team does. So I think you need to avoid that sort of reasoning.
+
+\[08:21\] But one of the upsides to Go, at least for me, is that even if somebody does that and leaves, it's not challenging enough that somebody can't pick it up.
+
+**Chris James:** Yeah, I think I'd like to add to that... I think its simplicity plays a part in the hiring aspects of the organization as well. As I mentioned before, I used to work in a Scala place and we found it very hard to hire people who were technically strong enough to write good Scala; because it's really easy to write bad Scala. Whereas with Go - obviously, it's not entirely foolproof; you can make mistakes, and things... But certainly, over the past few years, when I've been working with Go and working with less experienced people, I find they can pick it up relatively easily. If they have some familiarity with another general purpose language, they can usually pick it up.
+
+Usually, the biggest stumbling block is typically when you're starting off, you're starting with a dynamic language, like Ruby, or JavaScript, and then going to a statically-typed language is sometimes a bit of a challenge... But once you get over that, it's all curly brackets and variables and things really at the end of the day. So that makes it quite a safe choice for our risk-averse CTO.
+
+**Jon Calhoun:** Yeah. I did notice that Dan in the Slack had mentioned that introducing Go tooling works well, or Go for tooling. And I think that's something that I've seen; I don't know if you guys are the same... But I've talked to a couple organizations where if they need to build a command line utility or something like that, either internally or for customers, they've found that Go is really nice, because you can get it to compile for every language, and all of a sudden you have this tool that would have been much harder to build in another language.
+
+But more importantly, I think one of the key things here is that almost every success story I've heard has involved a project that needed solved, and they proposed -- well, it's small enough that they can propose "Let's try building a solution in Go." And I think that's one of the big keys if you're gonna try to introduce Go to your team - you can't just shove it in there at some point.
+
+If you're starting a brand new project, you might be able to make the argument of "Hey, let's do this." But if you're not doing that, which I think is most people - most people aren't rewriting their app or starting something from scratch; so you look for some small thing that you can introduce Go with. I think that's probably one of the biggest keys to having a successful introduction.
+
+**Mat Ryer:** Yeah, I think you hit the nail on the head when you said the word "small". I think that actually is a good thing to think about. If there is a small problem that you're gonna solve, especially one that -- you know, sometimes these tasks come up where you just have to process some files, or process some data, or something, get something ready... And maybe it's only something you're gonna run once, but it's your job to solve this problem - that's kind of a perfect candidate for the very early stages... Because it's something that only you are gonna be using, maybe, initially, but it's still an opportunity to solve that problem with Go, maybe even using some of the features of Go that you're excited about, like concurrency, and things.
+
+I did this a while back with filepath.Walk, and I was walking files, and then playing around with different concurrency on the processing of those files. I had to do a simple task. I'm not great in the Bash command line; I'm sure a Bash wizard would be able to just do it with a single line of ugly something... But it was nice for me to explore the concurrency side of Go by tackling that problem. So in a way, I got to explore it myself. At the same time, I was solving a real problem that the team needed solving, and then it was nice to just do a little show and tell of it. I said "Look, this is how I did it. I put the code in the repo somewhere..."
+
+\[12:08\] Somebody had a similar problem, and they were able to take the code and adapt it, and just solve that problem. And we ended up having like a file walker kind of little tool that we could reuse and just change for different purposes. So yeah, that thing about the tooling, when you've got some little tool that you need, even if it's not gonna be given to customers or even other team members, but solving a real little problem I think it's key.
+
+And if it's small enough as well, you don't mind throwing it away. If it's a failed test, if you do a little thing and it's not working, or it's too hard... Or if you leave and the team are left with it, if it's small enough, then it's not that big a job to rewrite it or replace it if you need to.
+
+**Chris James:** Yeah. I think it's important, particularly if it's unfamiliar with yourself or with people you're working with to try and keep the problem space small - I think a lot of people when they're trying to learn anything, sometimes they sort of take on too many things, and spin too many plates. I literally did this with Go. I remember reading about Go ages ago, and I thought "What I'll do is I'll learn Vim at the same time", and obviously, I just got nowhere. I couldn't do anything; I couldn't edit text, and I couldn't understand the language. I got into a complete mess. And I literally left Go for like six months while I tried this, and went "Well, I can't do this." And it was actually I couldn't do Vim, it wasn't I couldn't do Go.
+
+**Mat Ryer:** You had to get the code right the first time, as you were typing it. You can't make a mistake.
+
+**Chris James:** \[laughs\] Yeah. So I think definitely try and keep the problem space small; small little utilities. And I think it's also worth saying that like, I think what we're gonna say is that context is really important. Quite a lot of developers have a lot of eyes watching what particular work they're doing; we've got the JIRAs, and the tasks, and all this kind of stuff, and every morning you say what you're doing, and stuff... And if you're flippin' about it, you might get into trouble. But if you're saying "I'm going to solve this specific problem, which is a problem that's causing this pain" and then go \[whispering\] "in Go", then it's okay. You can just do that. You're adding value, but also proving that Go is a viable tool for your particular business.
+
+**Jon Calhoun:** I think one key that's also easy to forget is the isolation... If you can solve an isolated problem, that helps a lot, because it's not shoved in there -- Mat mentioned something that processes files; that's isolated enough that if you need to throw it away and write something else in another language, it's really not that big of a deal. But if you throw it into this big microservices -- even that's not as bad. But if you throw it in this big thing where people then have to learn how to use it, and everybody on your team has to learn how to use it, that becomes a little more challenging. So I definitely think looking for those isolated problems help on that "I'm gonna solve this real pain problem", you whisper it's in Go, and then you say "It's this isolated thing." So if it doesn't work, it's really not a big deal. Like, "We tried it. We threw it out." And at that point you should have a pretty good idea of how to solve it in any other language, because you've already written the code; you're just translating at that point.
+
+**Mat Ryer:** There's something about the fact that when you're doing something that you're really interested in, you tend to do a better job. I certainly do. But if I was using Go for the first time to solve something, I would be very motivated to get that to work... So I think that's something that tech leads and managers could take from this. You can enable that sort of situation, you can enable that environment where people are able to have a little autonomy and explore things, and try different things and do little experiments... Which may fail, and that should be okay. You can still do like a brown bag session on it and talk about why it didn't work, if you're doing a Rust one...
+
+\[15:56\] Or in your case, James, maybe if you're stuck in Vim and you can't get out, or you don't know how to go and change some text, that could be an interesting little lesson, I think. So even if there are things that don't work, they're still interesting, they're still worthwhile, and they're still worth sharing, I think.
+
+**Jon Calhoun:** Speaking of things that don't work, one of our listeners in the Go Slack had mentioned that one of the hard parts is finding talent. And I think you will possibly get that pushback from managers, where they're thinking about hiring new team members, and growing the team, and actually being able to manage this project... And I know that there are going to be real concerns about "We can't introduce ten different languages to our codebase." That's gonna be a challenge. We can't introduce a language where you can't hire people for that, and I think that might be one of the issues now with Go, at least sometimes - people think it's a new language, it's going to be hard to find people.
+
+Probably, if I had a manager doing that, my response would be "Let me do something small and let me see if I can teach the team how to use it. If it's a quick enough transition, then that means we can bring other good developers on and they can learn Go really quickly." But if it's something where it takes too long or you feel like it's gonna be problematic, then we can revisit that. Kind of like you said, you can scrap it if it doesn't work. But that's kind of a way to do an isolated test and prove that it's quick and easy to learn... Because Go in my opinion is one of those things that while you might not be able to hire people that have been using it for ten years, you can definitely hire people that can learn it in a couple of weeks and do well with it... And I think managers also forget the other side of it, where there's people actively looking for jobs that have Go, because they want to learn it. And I think they discredit how much of a hiring tool that can be...
+
+**Mat Ryer:** Yeah, I was gonna make that point. I get a lot of people asking me "How can I find Go jobs?" It's people that either have just learned it, or they want to learn it. Or usually they know it and they're using it for some of their personal projects, and they want to turn that professional. I get asked that a lot more than I get asked the opposite way around. Carl Mendez in the Gophers Slack mentioned that -- he says "A lot of recruiters still don't know what Go is, and it can put small orgs off." I think that's probably because they hear people telling them to Go a lot, in their job. I always tell recruiters "Go away." I'm constantly saying that. So no wonder... Maybe we need to spend a bit more time with those recruiters and help them.
+
+**Chris James:** I think this is an interesting topic, and my own experience was -- about four years ago I joined a company, and I knew full well they weren't doing Go, even though I wanted to do Go... But I knew the CTO, and I knew the kind of environment he was trying to build; it was very much kind of like a learning environment, sort of empowering teams to build software the way they want to... So I felt quite confident that if I can get in there and I can get to know people, I think I could convince enough people to make that happen. And thankfully, I did.
+
+I think the hiring thing's interesting... Depending where you are, there's just a small pool of people who describe themselves as Go developers. But as we said, I think if you're a decent software engineer, you'll be fine. Certainly at the place I'm talking about everyone was writing JavaScript, and it was all happy and stuff... But we built a community of people who wanted to learn Go. And it started small, and it grew bigger and bigger. And we did all sorts of activities, which I guess we can go into later... And eventually, we got to that critical mass where it was no longer seen as a risk to write systems in Go. It was like "Well, we've got enough people here. We've proven over the past few months we are a bunch of peope here all interested in Go. We had written some little command line tools, like you mentioned... And then at that point, it just felt like a natural thing to do, rather than big business case, with lots of stress behind it, and stuff. It just felt like this is an okay thing for us to do.
+
+I'd say people should try and find good environments. You won't necessarily get to write Go on day one, but if it's a good environment, you might get to it in a few months' time.
+
+**Mat Ryer:** \[20:05\] How many people was it for you in your case before you felt like you had that critical mass?
+
+**Chris James:** To add some context, this place had about 30-40 software developers, I'd say... And at first I just talked to people about Go, going "It's really cool!" I got two or three people who were like "Yeah, actually this is pretty cool. I'd like to start to learn this." I think it's really important to have some allies with you to help you push this forward. Particularly if you're in a big(ish) organization, you can't spearhead this yourself. You need help. So I had those 2-3 people that I thought were gonna really help me out, and we started just doing things like book clubs, and doing activities at lunchtime, and stuff... And then yeah, I guess we got to a point where we had, say, 10-15 people who are described as "Wouldn't be offended at writing some Go." At that point, it was an easy conversation with a CTO, and I was like "This isn't just me making some snowflake system, and then I'm gonna run away and you're gonna be stuffed. There's gonna be plenty of people who can support the software."
+
+**Break:** \[21:09\]
+
+**Jon Calhoun:** So when you talk about a good environment, I assume one of the things that's there is like - if you work in an organization that's in crunch time pretty much all the time (which really means it's not crunch time) or even if you are in a good environment where... How do I -- basically, you're in a good environment, but you need to get a release out, or something needs to happen, I think one of the things that's key is you have to look for the right timing. You mentioned timing, too; you have to find a good organization that's actually open to this sort of thing, and then you have to find a good time for it inside that organization.
+
+And I think that's something that other people forget, too - you didn't just jump in and say "Hey, let's go write some Go" and pushed for it. You waited until you had the critical mass, you probably waited until like "Okay, we don't have any major things that need to be shipped in the next month. We've got a little bit of wiggle room here." You know, you wait for all those things to sort of line up, and you have a team that's on board... Because I think even then -- if you love Go and you think it's perfect for the problem, if you have a team that just does not like Go, it's not gonna work there.
+
+**Chris James:** \[24:00\] To your point - let's pretend that I did it all perfectly and executed \[unintelligible 00:24:02.15\] but actually, funnily enough, I think in my first few months there, we were doing an inception... And for those who are unfamiliar with an inception - it's a thing where you get everyone in a room and you have a meeting for two weeks... Which sounds horrendous, but actually, it can be quite fun and engaging...
+
+**Mat Ryer:** To go into each other's dreams?
+
+**Chris James:** It feels like it at times... But normally it's just some dull meeting room with some whiteboards, and things... But you know, you discuss the products, and you think how you're gonna build it, and stuff. I actually thought to myself - it's not unreasonable for me to say "This is a new project, so we could write a new system in Go, right?" And I remember saying it out loud, and I remember getting shot down so hard by a project manager... He said "Look, the developers spend time spiking out programming languages, and we've decided we're gonna write everything in JavaScript, because that lets us be efficient, and share code etc."
+
+And I bit my tongue, but I definitely remember being really salty about him, and thinking to myself "Who is this guy to tell me what's right and stuff... I need to add a bit more empathy. This guy - and again, think about incentives, and stuff; his job, how he's judged is in how well projects are executed. And here's this dev who's barely been there like two months, \[unintelligible 00:25:15.18\] all this risk. Because we can talk about how simple and safe Go is, but it's still a risk to bring in a new language; it doesn't matter how simple it is, it is risk.
+
+So the lesson I took from that is even though I feel like I'm right, you should definitely try and read the room, and think about "Is it really worth raising this at this point?" Because all it really did was it just sort of soured the rest of the meeting a little bit... Or at least it did for me.
+
+**Jon Calhoun:** One of the things that's sometimes hard to grasp as a developer, especially when you're building your own personal projects, you're like "Okay, it doesn't really make much of a difference if I jump from Go to JavaScript." There's not that big of a risk factor there. There's some risk, but it's not as much.
+
+But in a bigger organization, like you mentioned - and this isn't even like a huge organization, but if it's all in JavaScript, you can reuse things... So one example is some people build tools to generate the starting framework for a microservice, or something... And it generates a bunch of code for them, and if you jump to a new language, all of a sudden they lose all of that. Somebody has to somehow find a way to replicate it. And then there's the "How do we make the Go version act exactly the same as we expect the others to act?" Because I did see one project -- I think it was one trying to go from Rails to Go, and they tried to rewrite the entire API... And the issue they ran into was that Rails did a bunch of subtle things that they didn't realize with requests coming in, to manipulate data or do something else... And basically, long-term they learned that trying to get Go to act exactly the same way that Rails did was actually challenging, because there's all these things they didn't know that were happening, that were happening, that they had to account for, and they didn't wanna break anything.
+
+And I understand that if you have a working ecosystem at a company, it's hard to wanna break that... And that's why I think sometimes finding those isolated problems, again, is just really valuable early on, because you can make the argument for "Okay, it solved these problems really well, and even if we can't use it across the entire org, we can at least use it to solve a specific subset of problems.
+
+**Mat Ryer:** Yeah. In that case, it's probably Ruby having some magic, and over-delivering and under-promising... Where Go doesn't do that. Go tends to stick to the spec; it sticks to the letter of things, generally... Or at least there's not a lot of magic that's in there. That sometimes is something for sure to be aware of. But of course, that's a strength, as well as a weakness. The fact that Go is very expressive and easier to reason about, and there's no magic stuff happening...
+
+And I think the kind of architecture as well probably matters, right? If you've got microservices architecture, one of the advantages of microservices really is that you can have different languages. Maybe they're better-suited for certain tasks... Each microservice could be a different language.
+
+\[28:06\] Now, there's a cost obviously to that from maintenance, but you could probably make a good case in some situations for this component, this small component would be great in Go, and it's quite low risk; if it doesn't work, we've still got the other component. That's probably not the same in monoliths, I guess...
+
+**Chris James:** Yeah, I actually made the exact argument when we finally made the push... Because the thing about microservices is they're all hype, and it's an interesting architectural approach, but it's not without its costs. There's a lot of cost to having a distributed system. In general, distributed systems are just hard, right? It's much harder than calling a function if you have to do it over the network, right? There's other things to consider. So definitely when I was talking this through the CTO, I was like "Look, we've put all this investment in a microservices architecture, and one of the definite selling points in my mind of that architecture is it gives you that flexibility to just write a service in a new language, or in a different language, or the best language for the job."
+
+In our case, we had a service that we wrote in JavaScript - because we wrote everything in JavaScript - but it was on a very CPU-bound task, and it just ended up being incredibly slow... And it was becoming a real problem. We just couldn't actually deliver a good experience, because this thing was just taking forever to process these messages, and things. And I think it was a really nice example of just being opportunistic about it, and just going "Look, this thing is CPU-bound and it's really slow. We've spent months getting people involved with Go, and things...", so it became quite an easy sell; it wasn't like we had a poor-performing system \[unintelligible 00:29:46.23\] and just threw our hands up. We spent a couple of weeks trying to make it work, but we didn't really get it where we wanted it to go, so to speak... So I just went "Let's just try and write this in Go."
+
+Coming back to the scope of doing things - because we had written the software, even though it \[unintelligible 00:30:04.13\] we understood what it had to do. We had a good understanding of the ins and outs and what it needed to accomplish... So therefore that kind of reduced the amount of stuff we needed to do. We had to reproduce it in Go, but it just reduced the amount of risk, because it was like "Well, we know what we need to build." We already had all the black box tests around it that we could just reuse. We just plug in a Hugo system and cross our fingers... And thankfully, it was like an amazing success story for us. We went from ten Node instances with half a gig or RAM each to three Go instances with 128 megs, and it was almost ten times faster... Which was awesome. It was probably one of the coolest moments in my career, being able to go "Look what we did. We've invested this time in learning this new language, and it's actually had a business outcome", coming back to our CTO friend. We managed to complete that circle of value, I guess.
+
+**Mat Ryer:** I love that. You've mentioned a few things in that which I think are really worth pulling out... Knowing the ins and outs of a thing - so if you're gonna rewrite something that already exists, often figuring out the ins and outs is the hard bit. Implementation tends to be quite easy once you know what you're building... So yeah, that's another good point. If there is something that exists already, and you feel like Go can do a better job at it... And there will be lots of examples in Node; Go compiles right down for different architectures, it does do quite a good job at certain tasks, which other languages won't be able to compete with... So yeah, I think that's a great one. If you can do a better job in Go, that's gonna be a much easier story.
+
+**Chris James:** \[31:58\] Yeah. And it was nice, because we were basically stuck. We just didn't really know what to do. "We can't just keep throwing more instances at it and hoping for the best." It was just gonna run up a ridiculous bill... \[unintelligible 00:32:07.27\] Two to three weeks. I think we can do it in two to three weeks. And we did. It wasn't the smoothest ride; there were some libraries that were in Node that just weren't in Go, and we had to write that kind of stuff... But still, when I look back at it, it couldn't really have gone better, to be honest. That was really nice.
+
+**Mat Ryer:** I love hearing that. By the way, if you've just put two fingers up in the air and come up with two weeks, don't then deliver it in two weeks; it makes the rest of us look bad who can't estimate...
+
+**Chris James:** Yeah, I'll get kicked out of the union.
+
+**Mat Ryer:** Yeah. Although we don't know when.
+
+**Jon Calhoun:** To add to that point - Mat, you were talking about knowing the ins and outs is challenging... Recently, I had somebody reach out to me, and essentially they were trying to rewrite a library; I think it was in Python. It was a Python library that would take SQL, and I think was trying to spit out CSVs, or something like that... I don't remember exactly what it was doing, but essentially their confusion was that they were like "Go was supposed to be faster than Python." They wrote this tool that does this, and it wasn't faster, and they were like "I don't understand why it's not faster." And I do think that there are going to be times where either you don't understand the ins and the outs, as you described, or you don't understand the bottlenecks. Because in this case, there's a hundred different ways that it could have been bottlenecked that weren't the language. It could have been the database you're working with, it could have been writing to a file, it could have been a bunch of other stuff; who knows.
+
+So if you do have one of those cases where you're trying to bring Go to your team and it doesn't work, it doesn't actually give you those positive results you're looking for, how would you guys handle that, I guess? Would you give up on Go, would you try to tell them "Well, this wasn't the right one, but it was worth a shot"?
+
+**Mat Ryer:** Great question.
+
+**Chris James:** Yeah, it's a great question. I think in general with software development it's a real skill to try and lift up risk as soon as you can. That's the big thing about iterative development - you try and figure out what are the riskiest things, and try to tackle them first, so you can get an understanding.
+
+But I also think there's always a risk of introducing Go to your team that you're not being honest. We all love Go, and it's great, and we don't have to list all the reasons why, but it's not the magic silver bullet. It has its downsides and things, and it's always important to be quite realistic about this kind of stuff... And I think if you're going into an endeavor where you don't truly know if it's gonna be more performant, you need to lay that out upfront, and just go -- I mean, this is what I did with my thing. I said "I'm pretty sure that this will be way more performant in Go, but please don't fire me if it isn't." It wasn't my exact wording, but... You know, it's about expectation management, I think.
+
+**Jon Calhoun:** I think that makes sense. My brother was working on a project once, and he was doing something with WebSockets; it was really simple, whatever he was doing. It was like a simple live chat server that they didn't need to persist records; it was just kind of like "If you got it, you got it. If you didn't, great." And I think he was using Fyne... Is that the library that's in Ruby, or JavaScript, or something? So he was using some library to do the WebSocket stuff, and he found that under heavy load, this project was basically coming to a crawl. The WebSocket server was the issue.
+
+He had looked at some things to scale it up and to fix it, and at one point I'd basically looked at it and I'm like "This thing has way more than what you need. What you're actually looking for is super-simple. Let's see if we can write something in Go." And I did what you said, Chris - I told him upfront "I don't actually know if this is gonna be faster, but I have a feeling it will be, because we're gonna strip out everything we don't need, we're gonna keep it super-simple, and Go should be good at this. It should be good at handling many different requests on the same server, and doing all of that stuff."
+
+The end result happened to be very positive on that one. The server stopped crashing, and everything else that was going wrong stopped happening... And it worked really well. Luckily, this was a project that literally I think the Go code was 150 lines of code, or something... Something super-simple.
+
+\[36:05\] There was some stuff in front of it to deal with authentication or whatever stuff it needed to do, but the actual Go part was so small that even if it didn't work, we're like "Well, it's not enough code that it's a big deal if it doesn't work." And I think that was one of the nicest experiments I had with that, where we're testing it -- and this was before I knew Go that well even, so it was kind of a nice "See if it works and get some hands-on experience."
+
+But you're completely right, you have to be clear with everybody ahead of time that you don't know for sure this is going to solve the problem... But if it doesn't, I think you also learned something. You realize maybe "It's not CPU-bound because of the reasons we thought."
+
+**Mat Ryer:** Yeah. So I guess you should measure and really understand that first, as you're putting together the hypothesis of why you think Go is gonna do a better job... Because that is quite a good point.
+
+One thing that was mentioned in the Gophers Slack - and by the way, you're welcome to join us in Gophers Slack in the \#GoTimeFM channel; that's my marketing... Someone just did, so it works. \[laughter\] Someone mentioned this earlier, and this is something I've found - I worked with a team who had different languages, and over time they found that the bits that were written in Go kind of just kept always working. And that wasn't true for other situations. I think it was wrapped up in dependencies changing, and other things -- not just about the language, but other things going on in the team, too. But generally, the feeling was that the Go code was the stable code. This was the code that you could start to rely on. And in that case, it was actually because of testing. It was because testing is a first-class concern in Go. So you are encouraged that. And I don't need to tell you, Chris, because you wrote Learn Go With Tests... But yeah, having tests, caring about tests and having them for your code does improve the quality and stability of that code. So maybe there's something in that, do you think - the fact that testing is a part of the toolchain, part of the community, part of the ecosystem and the spirit of Go helps create more robust things?
+
+**Chris James:** Yeah, I think -- as we said, we need to be honest and realistic; you can definitely write automated tests in other programming languages, to be clear. But I think you touched on an interesting point where you talked about the community aspect of it, and the conventions.
+
+It feels like because tests is such a first-class citizen - it's right in front of you; you don't have to think about what particular testing library you have to use, or mocking frameworks, and all this business... It's right there in front of you. Because it's accessible, you do it. And I've found a lot of developers who have done some automated testing and things, but they were still quite new to it... But by doing testing in Go, where it's all very strict back, and there's not a lot of fuss and nonsense, you can just start learning the fundamentals a lot easier, and I think that really helps to build that more robust software. And I support benchmarking also fits into that as well. It maybe felt silly at times, but sometimes I was just writing benchmarks with test just because I could. It was so easy, I was like "I'll just do it for fun." Because why not? It's basically the same as writing tests, which I think is such a smart move by the language; you don't have to learn two different things, really. it's kind of the same thing.
+
+I think the cultural thing is a really interesting thing across all programming languages. They all have their cultures, and things. Ruby was also very up on the testing thing as well... Yeah, it's an interesting point to me.
+
+**Mat Ryer:** \[39:55\] Yeah. And the fuzzing proposal - a design proposal - is the same; they're folding it into existing -- the way that tests already work today. I look forward to Learn Go With Fuzzing. That book's gonna be chaos.
+
+**Chris James:** \[laughs\]
+
+**Mat Ryer:** But you're right, testing is -- a lot of decisions are made for you with testing. So that takes the cognitive load away, and then that does make it an easier path, doesn't it? Same for go fmt, the fact that the code is all formatted the same. There's probably a lot of other things where decisions are taken away, that actually helps when you want to introduce Go, because there isn't a debate about every little thing. It's like "Well, these things are just sorted out. This is how you do it in Go." And then you can get down to the important bits, which is hopefully solving the real problem you were gonna solve.
+
+**Chris James:** Yeah. I think when I was introducing it, a few of the senior developers definitely gave some grumbles when there was no collectiony things, like mapping of a collection, or filtering, or anything like that... And I kind of just said "Man, just get over it." The thing is, there's just one way to write a loop; we don't have to bikeshed about forEach, or map, or filter, or fold left, or fold right... And all this kind of stuff that in other programming languages feels powerful, feels alluring, and it feels very intellectual as well, at time; "Oh, I'm gonna think about this, and think about this..." You know, it's liberating that with Go it's just like "Oh, you need to iterate over something? Well, you know how to do that, because you learned that on day one", and that's it.
+
+**Mat Ryer:** That's right. And then you don't need to learn and remember what all the different variations do, like many, once, find, select... There's loads of them. And by the way, they will almost certainly come to Go if and when generics comes... But you're right, there's something quite nice about the fact that every time you want to iterate over something in Go, it's basically the same.
+
+**Jon Calhoun:** I think there's a lot of benefit, in the sense that not only is that stuff not there, but I think people don't abuse things to get the results they want, not realizing the ramifications of it. An example is people might do something where they're iterating over a list multiple times, and making their code really inefficient, because they just don't realize that's what those things are doing... Because on their screen they're writing one line; they're like writing "Find this thing". They don't realize it has to go through the whole list to find it. They don't really think about any of what's actually happening. Whereas with Go, you have to explicitly tell it what to do. So you're very clearly saying "You're running a for loop here. There's no hidden for loops; you're running a for loop." And that probably helps people write better code, or clearer code, at least for junior developers. I think senior developers probably understand what they're doing, but junior developers - I think it's where you can run into some of those issues.
+
+I did have one question for you guys, I guess... Earlier when you were talking about testing, and that being part of like the Go ecosystem, and everybody really pushes for good tests, one of the things that stuck out to me at least was that when I wrote code in dynamic languages like JavaScript or Ruby, it felt like it's not that people didn't want to write tests, it's almost that they gave up on them after the tests didn't do what they wanted them to do, or they didn't catch the bugs... And one of the examples I can give is you get nil passed in somewhere, and then all of a sudden everything just breaks, and all of a sudden you're writing a test for like "This shouldn't be nil", or something random like that.
+
+With something like Go, with the type checker and everything else being there because it's a compiled language, static types, it was almost like people complained about the extra work upfront, but the end result was that it allowed you to write tests that focused on the things that actually mattered, and the things that you wanted to test, versus -- I felt like in some of those dynamic languages, people gave up on tests because they got so sick of just writing a bunch of silly tests for all these weird cases... Really, they didn't wanna be testing those. That just felt like a waste of their time. And in reality, you wanna test bigger-picture type things sometimes. Or like if I run this function that's supposed to do X, Y and Z, I wanna see that it does X, Y and Z. I don't wanna go through and make sure that it iterated over these things in this correct order, or did something weird... You don't care about that.
+
+**Mat Ryer:** \[44:15\] Yeah, that's especially true in JavaScript, or dynamic languages, because I did find myself a lot of the time type-checking, making sure that it behaved when you passed incorrect things in. It's asking for a string, but I'm gonna pass it the entire window. Totally, you can do that in JavaScript. Just give it the window.
+
+So yeah, when I wanted really robust code, I did exactly that - I've found myself using testing so that I could make sure it would misbehave properly, or it would gracefully handle those weird things. So yeah, again, I think the minimalism and the simplicity in Go, strong types - its design helps with all that.
+
+**Chris James:** Yeah, I can relate it back to a story I was talking earlier, about how making the leap from dynamic to static for the less experienced developers was a bit of a challenge at first... And one thing I did was we had some kind of outage because of one of these silly things, where we pass through an object instead of a string, or something like that... I can't really recall the exact details. But you know, we had the post mortem and stuff, and we were "Alright, we'll write some tests to make sure this never happens again."
+
+So again, I'm writing these things, going like "Make sure you send through a string and all these things; these kinds of tests that feel silly, but actually you need them, otherwise you can run into trouble... And then I was able to at the following lunchtime club go "Here's that same problem in Go, and it doesn't exist. I'm running the compiler, it won't let me do it. I can't even run the tests. It will not let me do this." And I like to think that was a bit of a \[unintelligible 00:45:47.23\] moment for a few of them at that point as well, and they were like "Okay, I see the value here."
+
+It's just doing some of the thinking work for you, because it's too much. Programming is hard enough, without having to think about whether you can pass a string through a function or not.
+
+**Break:** \[46:07\]
+
+**Mat Ryer:** There's another thing... Sometimes it's tempting to really flex and use some of the language features to their fullest, and really show them off... But you can quickly end up with some complicated-looking code, and it'll probably go against your cause if you do that... One example is channels. There are some great things you can do with channels in Go. But actually, if what you really need is just simple concurrency, just consider goroutines and a wait group. So just learn the wait group, and goroutines, and how that works; you can do a lot, it can go a long way with just those two little bits and pieces.
+
+I think that probably applies as well... It's like, yes, we want to use all the cool features, but remember, one of the key value props of Go is that it's readable and it's maintainable... So that means it's worth taking the time to make sure that you write good, simple code, as simple as you can get it. What you really want when you show it to somebody - you want them to get it, you want it to click; "Yes, okay, I understand this." If, like with Rust, you're doing something and it's amazing what it's doing, but it's complicated, that's just a cognitive barrier. And no one likes other people being so much smarter than them. And you can make people feel silly just because of writing complex code, and that's worth avoiding, I'd say.
+
+Are there any other reasons why you shouldn't introduce Go? I think that's one, for the sake of it using those complex language features. I think another one worth avoiding is tech wars. There's almost certainly no point having the argument that Go is better than this language, and getting into that. What do you think?
+
+**Chris James:** Yeah, this is definitely something worth talking about. I think internet forums about programming are just the worst, right? You get these holy wars about Rust versus Go, and C\# versus Java, and whatever... And the thing is, you think it only exists on the internet, but it happens in workplaces too, and it can become incredibly toxic.
+
+I remember a long time ago basically getting so fed up with it I asked to get moved off of the project... And I looked back at it, thinking just how ridiculous it is that basically I couldn't stand a programming language war anymore, that I had to get off the project... But it's just what happened. And it's very easy to slip into it. I don't think anyone goes into the intention of being difficult, and things... It's just because we get passionate and we hold these things quite dearly, we can come across as being quite difficult and disrespectful to other people's decisions.
+
+So I think it's important to try and -- you should be able to speak positively about Go without disrespecting other people's opinions and decisions that have been made. In general, there's nothing worse than a new person coming in and saying how all the decisions that were made were wrong, and they're stupid, because they never understood the context behind those decisions... It's very grating.
+
+So you should be able to speak positively about Go, and that should be enough. And as we said, just be honest and realistic about it as well. It's not gonna turn water into wine. It's great, but be realistic about it. Don't insult people's intelligence about it.
+
+**Mat Ryer:** Hm. And if that doesn't work, you can always incept them and go into their dreams and convince them that way.
+
+**Chris James:** Indeed, yes.
+
+**Mat Ryer:** I know that's how you do it, James.
+
+**Jon Calhoun:** It is interesting that as developers we love to jump into projects and immediately think "I could do so much better if I rewrote this from scratch." I think every developer has thought that. And then the more you work on a project, the more you make changes, the more you realize that there's a lot more complexity there than you realized...
+
+**Mat Ryer:** Yeah.
+
+**Jon Calhoun:** \[52:01\] I think just being aware of that, and empathizing with the fact that not only did they have all that complexity to deal with, but they probably didn't know about it all upfront; they probably had to figure it out as they went, which makes things even more challenging.
+
+**Mat Ryer:** Yeah, especially if not everything is described in a test. Sometimes there's little changes that are made throughout the codebase that have a significant impact in some cases, and they aren't always covered explicitly with test code... So they're sort of hidden features, really. That's true.
+
+But also, rewriting -- I think I said this on last week's show... Hemingway said "There is no writing, but rewriting." I've paraphrased that now. Rewriting stuff does make it better. That's why microservices and small problems - that's why if you can rewrite something like that, you may have a good chance of making it better. But of course, you do have to be careful of what that thing is... And it's about those ins and outs, again.
+
+**Jon Calhoun:** I think the key is -- like, when Hemingway talks about rewriting, I doubt he means...
+
+**Mat Ryer:** Go code.
+
+**Jon Calhoun:** ..."Throw your entire book out and just start writing the whole book from scratch and it'll be better." He probably means "Go find a page and modify that page", something small and manageable. And I think the problem is people take quotes like that or take ideas like that and they think "Oh no, I'm just throwing the whole book out the window and we're just gonna start from scratch", and it's like, that's gonna be terrible. It's gonna be terrible for authors.
+
+**Mat Ryer:** So you think Hemingway would have been a microservices kind of guy.
+
+**Jon Calhoun:** I don't know if he would have been microservices, but you don't even need microservices to rewrite parts. If you have a monolith, you can literally be like "Okay, I'm gonna take this one thing that we're doing in our monolith and I'm gonna make it a small API that we communicate with. And you don't have to have microservices everywhere; it could just be like "This one thing's a small API", and you could go write it in another language. And it's gonna take a little bit of work to make that all work upfront, but it's not impossibly hard... And it's way safer than "Let's rewrite the whole thing."
+
+**Mat Ryer:** Yeah. So you should be considerate to what's really going on in your team, and the goals of the team. That is gonna be important; it can't just be "We're gonna just do this because we love it, or it's the best thing." You have to be careful and considerate of what's really going on, absolutely.
+
+**Jon Calhoun:** We mentioned it earlier, but if your team's in crunch time, that's a bad time to try to introduce a language. But even then, a rewrite or anything like that - you have to realize that a rewrite means you probably aren't shipping new features until the rewrite's done. And that could be six months easily, depending on the scope of the project. And that is never good for business, for the most part. If it fixed a bunch of issues, potentially it could be, but for the most part, you might have better luck spending those six months just doing bug fixes and hunting down issues and solving anything you can.
+
+I think that if you're going to push for Go, you need to find the right problems for it, and if you're trying to push it for the wrong problems, it's just not a good idea. Another example - Mat, you asked for reasons why not to... One example is if you have an entire team of expert Rust developers, you're probably not gonna get a lot of value out of jumping to Go. Rust is a harder language, in my opinion, to pick up and to get everything. If you have a team that's all senior developers that are doing great with it and they really love all the things that it provides, I would say that trying to introduce Go there is not a good idea.
+
+But if you have a team that's maybe writing some JavaScript, or some Ruby, or something like that, and having some of the stuff written with a compiler, and type safety, and maybe a little more performance, and a little bit of simplicity so that junior developers can jump in and pick this up is at all something you want, that's fine. But I think there probably are situations where your team just decided that you want senior developers who've been writing Rust for two years, and that's what you're looking for. And if somebody wants a job there, they need to learn Rust really well. I think that's not necessarily a terrible thing; it's not the most inclusive thing in the world necessarily, but I think every organization has to decide what's best for them, and if you're pushing Go to an organization where it's just completely against their core beliefs and values, it's just not gonna work.
+
+**Mat Ryer:** Well, we've reached that time again... It's time for Unpopular Opinions!
+
+**Jingle:** \[56:21\] to \[56:40\]
+
+**Mat Ryer:** So... Jon just finished his drink... \[laughter\] So do we have any unpopular opinions?
+
+**Jon Calhoun:** I need a new drink. Is that one unpopular...?
+
+**Chris James:** I've been trying to think about how many bridges I want to burn ever since I knew I was gonna be on the show... I was thinking "Where should I go with this?" I'm trying to keep it sensible, so... Here it goes - Go is just a general purpose language, and it's not gonna solve all your problems. A big theme of this conversation is how context is important, and your work environment is important... And if you want to add Go to your organization, you need to understand the problems, constraints, and it’s people. But if your environment is constant crunch, and all this business, even if you could wave a magic wand and make everyone write Go, and ignore all those problems and everyone's now writing Go, your job might be a bit more novel for a while, but you're not gonna enjoy it.
+
+And to make sure I stand by this unpopular opinion, I did literally go from a role of writing Go most of the time to writing Kotlin, which is obviously a sacrilegious thing to say on this podcast... But I did it because the environment and the opportunity - everything else around it seemed wonderful.
+
+So I think it's important to just have a bit of perspective around programming languages. They do have strengths and weaknesses and things, but they're not the ultimate source of success for general software engineering.
+
+**Mat Ryer:** Yeah, it's what you do with them.
+
+**Jon Calhoun:** I don't think that's gonna be an unpopular opinion.
+
+**Chris James:** Ah... Damn.
+
+**Jon Calhoun:** I think you need to burn more bridges.
+
+**Chris James:** Well, give me a minute...
+
+**Mat Ryer:** We do test these, by the way. The @GoTimeFM Twitter - we'll do a poll...
+
+**Chris James:** Oh, I saw. yes.
+
+**Mat Ryer:** So we'll find out if that really is popular or unpopular. If it's not unpopular, unfortunately you are contractually obliged to return and do another episode.
+
+**Jon Calhoun:** This is how we get guests.
+
+**Mat Ryer:** Do you wanna have another go?
+
+**Chris James:** Alright, one more. I think the Venn diagram of Java best practices and Go best practices is not a big circle, but it's not as different as people think it is. I think a lot of people like to talk about Java, and they're normally talking about bad Java, like an abstract FactoryBean, or whatever... Good Java developers don't think that's good either. You think about what good Java developers think about - it's all generally good software engineering stuff, like single responsibilities, separation of concerns, loose coupling, all that kind of stuff.
+
+**Mat Ryer:** Right.
+
+**Chris James:** So hopefully that will be unpopular...
+
+**Mat Ryer:** That's an interesting one. What stands out to me is the type hierarchies. You don't really do those same type hierarchies; you can't do them in Go.
+
+**Chris James:** Right, you can't.
+
+**Mat Ryer:** But when used properly, in languages that support it, they can be extremely good. The trouble is it gets abused, and we sort of then compare it to the abuse of it. But there is something to be said for the fact that you can't do it in Go. That kind of gives you a head start, or at least a better chance of writing cleaner, simpler code.
+
+**Jon Calhoun:** I think you're right that Go probably makes it -- that you have a better chance at it, writing the good, clean code... But I think any language, given enough time and enough environments, some group will find a way to write awful code, that people hate. If that's their version of that language, they're gonna be like "This was terrible." And that might happen with Go, I don't know... There might be people who just write Go exactly like some other language, and it just does not work well with the language, and people are like "Oh, it's a terrible language because of that."
+
+\[01:00:18.04\] I talked to somebody who claimed they basically saw code where they were using panic and rescue essentially like exceptions. They just decided "We're translating exceptions to this code, and that's what we have." And if you see that, you're gonna be like "This is terrible. Why didn't they just do raising exceptions? It would have been so much easier." And I can completely understand that; I just think any language, given enough time, people are gonna find a way to write what's effectively code that's bad, and that people don't like
+
+**Chris James:** I think what's interesting is - I think someone touched upon it - that there's definitely some anxiety around generics in Go. I think a lot of people seem to think that if and when generics come to Go, it will suddenly turn to Java. I think that's an understandable concern, but we were talking about culture around programming languages and things, and at least I'd like to think that the culture around Go will mean that they will be used tastefully, at least most of the time.
+
+When Java was first out, software developers were wearing suits, and UML was all the rage, and things; it was a completely different period of time and a different culture compared to what we have now. So I like to think it'll be okay, but I guess we will see.
+
+**Jon Calhoun:** I think it's gonna be like when developers first go to Go, they use channels, concurrency and that stuff way more than they probably should. At least that's been my experience with pretty much every developer who's tried Go out - they see these cool things and they wanna try them, and they use them too much. Then they realize that it's too much and they step back.
+
+I'm guessing generics are gonna be the same thing, where because it's cool and new, and they can do things they couldn't do before, you're gonna see a bunch of libraries that really shouldn't exist, that don't need to exist... But I think that at some point people are gonna step back and be like "Alright, it was fun doing that", but really for trying to get stuff done and get work done, let's focus on "Do we need this in generics? Do we need to actually do that?"
+
+**Mat Ryer:** Yeah, one argument for that is that the standard library should get a good set of useful generic things, in the standard library, so that we don't end up in a situation where all these different libraries that work slightly differently, and they're probably all gonna have the same names of the methods, and things... So I think they will do that; I think they will probably add -- at the same time that generics happens, there will be a set of generic data structures that we can coalesce around, maybe.
+
+**Jon Calhoun:** I think they'll also have to push back on that some. Java is an example of like every data structure you can imagine is probably in the standard library. I remember a friend of mine had an interview question at one point where they asked him to essentially build a least recently used cache with -- I think it was just basically a least recently used cache. But there's a linked HashMap in the Java standard library, that is effectively an in-memory least recently used cache... Because it used the link list to keep track of which one was used most frequently, and then the HashMap is how you access each thing directly when you need to access it... And I know when he got asked the interview question, he basically -- because they didn't actually phrase it as like you would know that's what it was... But he said "Do you mean you just want me to use this?" and basically wrote the code real quick on the whiteboard (it was like a whiteboard one). And the interviewer looked at it and basically was like "Yeah, if you know that's in the standard library without looking, you probably don't need to write this. You probably understand it well enough that it's not a big deal."
+
+But it was just interesting... If he'd been writing in any other language, that wouldn't have been a viable option; he would have had to actually write it out. And I'm wondering in Go if they're gonna have that issue, where people want every obscure data structure in the standard library, even though it might not necessarily need to be there.
+
+**Chris James:** Yeah, it's a lot of weight on the shoulders of the Go team... Because I was thinking about culture, and it's almost like the standard library informs some of the culture around the Go community, I feel like. A lot of people talk about if you wanna get an example of some best practices, there's a lot of good best practice within the standard library. You can just dig in there and have a look and see how it's done. So yeah... Good luck to them.
+
+**Mat Ryer:** Well, I'm afraid that's all the time we have today. Chris James, thank you so much for joining us.
+
+**Chris James:** It's a pleasure.
+
+**Mat Ryer:** Check out Learn Go With Tests. I recommend it. Jon Calhoun, you're coming with me to the local sheriff. I'm claiming that reward. See you next time!
+
+**Outro:** \[01:05:06.03\]
+
+**Chris James:** I can't believe how many times you called my James, Ryer.
+
+**Mat Ryer:** I didn't call you James, did I?
+
+**Chris James:** You called me James like three times. I was like "Are you kidding me?" \[laughter\]
+
+**Mat Ryer:** Oh, no... Did I really? That was because I made the joke at the beginning.
+
+**Chris James:** Yeah...
+
+**Mat Ryer:** Oh, you're joking. I'll tell you what - hang on; let me just record a few Chris'es for the edit. Chris. Chriiis...! Chris! Chris...? Chris. Yeah, that'll do. Chris?

It is Go Time!_transcript.txt

@@ -0,0 +1,27 @@
+[0.00 --> 5.64]  I'm Jared Santo, GoTimes producer and a loyal listener of the show.
+[5.64 --> 10.18]  This is the podcast for diverse discussions from around the Go community.
+[10.68 --> 14.06]  GoTimes panel hosts special guests like Kelsey Hightower.
+[14.80 --> 19.72]  And sometimes you can leverage a cloud provider and make margins on top.
+[19.80 --> 21.00]  That's just good business.
+[21.40 --> 25.08]  But when we're at the helm making the decision, we're like, yo, forget good business.
+[25.60 --> 29.86]  I'm about to deploy Kafka to process 25 messages a year.
+[30.90 --> 32.62]  It's nerd pride, right?
+[33.32 --> 35.90]  Picks the brains of the Go team at Google.
+[36.44 --> 40.54]  You don't get a good design by just grabbing features from other languages and gluing them together.
+[41.20 --> 46.02]  Instead, we try to build a coherent model for the language where all the pieces work in concert.
+[46.66 --> 49.62]  Shares their expertise from years in the industry.
+[50.20 --> 51.86]  Don't expect to get it right from the start.
+[52.12 --> 53.42]  You'll almost definitely get it wrong.
+[53.52 --> 55.52]  You'll almost definitely have to go back and change some things.
+[56.04 --> 59.14]  So yeah, I think it goes back to what Peter said at the start, which is just make your code,
+[59.14 --> 61.14]  write your code in a way that is easy to change.
+[61.80 --> 63.28]  And then just don't be afraid to change it.
+[63.56 --> 66.24]  And has an absolute riot along the way.
+[66.84 --> 70.60]  Yeah, you know that little small voice in your head that tells you not to say things?
+[71.22 --> 72.30]  What is that?
+[73.14 --> 74.06]  How do you get one?
+[75.74 --> 76.56]  You want one of those?
+[76.58 --> 77.74]  Is it like an in-app purchase?
+[78.62 --> 80.08]  It is go time.
+[80.46 --> 84.22]  Please select a recent episode, give it a listen, and subscribe today.
+[84.22 --> 85.88]  We'd love to have you with us.

On community and safety_transcript.txt

@@ -0,0 +1,373 @@
+**Johnny Boursiquot:** Hello, and welcome to this episode of Go Time. Once again, thank you very much for those of you who listen to the show regularly, and a special shout-out to our live listeners and those that engage with us on the Gophers Slack channel on a regular basis. If you also want to join the fun, do check out gotime.fm for all the details. Again, thank you so much. You all make this recording every week worthwhile, you make it more fun.
+
+I'm Johnny Boursiquot, I will be your co-host for today. Joining me also as co-host is Jon Calhoun. Say hello, Mr. Calhoun.
+
+**Jon Calhoun:** Hey, Johnny. How are you?
+
+**Johnny Boursiquot:** I am doing well, I am doing well. I'm feeling excited, because today we have a special guest, Denise Yu. Allow me to tell you briefly about Denise - she's an active member of the Go community, especially in the Toronto, Canada region. You might have come across her talks at GoCon; she's also spoken at DevOps Toronto, at various meetups, and she's kind of making a name for herself, in part because of the way she gives back to the community, something that as you all know is near and dear to my heart... And some of us have even participated in her drawing lessons that she teaches online, which is kind of fun. We'll definitely get into that one; I wanna know more about that, because I can't draw anything... So we're gonna get into that.
+
+She calls herself a developer and cat enthusiast. Welcome to the show, Denise.
+
+**Denise Yu:** Thanks so much for having me. I'm excited to be on. This is my first time doing a live stream -- or a live podcast recording rather, so... Definitely not nervous at all.
+
+**Johnny Boursiquot:** \[laughs\] You'll do just fine, you'll do just fine. So I've only briefly given folks an introduction as to who you are, but why I really wanted to have you on this show is because I think a lot of the things you do that are valuable to the community - I want more folks, more broadly to be aware of those things, and of you, because I think you're doing some awesome things, which benefits a lot of us, whether we realize it or not... And I kind of wanted to have you on the show, and -- especially, it seems to me the kind of role you do, even with your day job at GitHub, is all about taking care of people, taking care of community... So it lines up quite well with what we're gonna get into during the show today. But please, give us a little more about yourself. What don't we know about you that you want us to know?
+
+**Denise Yu:** \[04:00\] I guess we can start with what I currently do... I'm currently a software engineer at GitHub. I actually started about eight weeks ago, right when everywhere started going into lockdown. I am actually at my parents' place right now. My intention was to visit my parents for a couple of weeks, like one week, maybe two, and now we're starting week eight... So it's been an interesting experience, living at your parents' house at age 30, but it's good to stay put, especially now. And it's been a good chance to spend more time with my family, which is -- I think I may have realized I hadn't been home for quite a while, because of immigration reasons and just life reasons.
+
+So I work at GitHub. The team that I joined is called the Community and Safety team. This is a team that has a mission that I'm really excited about. I interviewed specifically for this team, and was like "No, I just want that one job. I don't want any other roles." I found out about this team last year, when I was at a conference called Write/Speak/Code, out in San Francisco. At that time Lexy Galantino, who's now one of my teammates was at the conference, speaking about her team... And basically talking about the ways that GitHub builds tools, basically combines policy solutions and engineering solutions to encourage people to become part of constructive and positive communities, and to discourage negative, unproductive behaviors that erode things like trust in the community, collaboration in communities, and that sort of thing.
+
+So I was very excited about the mission. I'd been doing open source software for a number of years. Very early on in my career I actually accidentally became the lead maintainer of an open source project... So I know what it's like to be on the other side of the table and be running an active project and feel overwhelmed by all the things that people are asking of you.
+
+So I was very excited about the role, super-excited to get the offer... I started eight weeks ago, and it's been fantastic so far. I'm having a lot of fun. This is my first time back in \[unintelligible 00:06:06.05\] in a number of years. I was working on infrastructure tools for the last few years with Pivotal, so getting back into Ruby on Rails stuff actually has been -- Rails has changed a lot in four years. The last time I wrote in Rails was over four years ago, and I was like "Oh... This thing that I struggled with four years ago is just a solved problem now", which is kind of nice.
+
+**Johnny Boursiquot:** Awesome, awesome. I'm interested in the journey that you've taken within the Go community. What brought you to Go, and what keeps you in the Go community?
+
+**Denise Yu:** Yeah, so I started writing Go at work last year, and I actually haven't written Go for that much time. When I worked at Pivotal, a bunch of teams had basically adapter code that was in Go... So I didn't get super-deep into Go my first two years there. Mostly my Go code was about taking YAML that's in this shape and then marshaling it into a data structure and changing it into a slightly different shape... Which is kind of like what app developers do; you get some input, change it around a little bit, and then output it to somewhere else.
+
+Last year I rolled on to a team that was working on an open source product called Concourse CI. Concourse is written 100% in Go. I think that was a pretty deliberate decision early on, because Concourse, first of all, needs to be highly performant. It's a CI/CD system. It's run on containers. A lot of the container APIs are written in Go, which is fantastic. Runc for example has a Go client... So I started learning a little more about Go, and trying to get more into the weeds of the language. And I definitely am still very far from being able to call myself a Go expert, but some of the things that I really like about Go is - believe it or not, I was actually thinking about this the other day... Learning Go conventions and the Go style of coding actually made me a better developer in other languages... And I feel like I go through this every time I learn a new language. Learning JavaScript made me better at Ruby, and now I feel like learning Go made me better at both JavaScript and Ruby...
+
+\[08:11\] And I'll tell you why - the reason why is because I love that Go is so opinionated. If you wanna write a conditional, there's only one way to do it. So if you read a conditional in someone else's code, it looks the same everywhere... And it's stuff like that, that low-level stuff. There's no reason to want to be creative with that, unless you're playing code golf, or something I guess... Which I guess you can write your own macros and things if you wanna do that. But I really like that when you're reading Go code, all you have to learn is the domain. You don't need to learn a whole different set of conventions. This is something that I've struggled with with reading Ruby and JavaScript code for many years... Because at pretty much every job I've had that's been in Ruby or JavaScript, we've always had a debate about what styleguide we wanna use... And this is always just such a timesink, because it doesn't matter. Ultimately, you should pick one and introduce some automation that enforces it. It's not a good use of developer energy.
+
+I have this idea, I think every team has a certain amount of -- almost like a friction budget. There's a healthy amount of disagreement that every team can go through, but you still feel good about what you've done at the end of the day. Some teams have a higher friction budget than other teams, depending on how gelled you are... But it's not a good use of that friction budget to argue about things like syntax and styling. You should spend that energy arguing about bigger things, like "Are we actually serving our users? Are we actually architecting our systems in the right way, or are we thinking about scale?" You know, the more interesting, open-ended questions... Not like "Oh, how many lines do you want your conditional to have?" That's not a good debate to have. So that's the first reason.
+
+The second reason I really like Go is because I actually think -- I've spent a lot of time thinking about the structure of functions, and I really love that Go, by and large, if you want to know what the happy path, the intended return value of a function is, you look at the bottom; the last line is your happy path. And many Go testing libraries -- or not the testing libraries themselves, but the way that people write tests kind of reinforce this standard.
+
+I think about, when I make a method call, the execution kind of ping-pongs through kind of like a pinball machine... You know, when you play pinball, the ball goes through the top and it goes down to all the flippers, and I visualize myself like flippering away the edge cases... So at the bottom, all that's left is the straight path. With that sort of mental image I started thinking more about how the Ruby code, and the JavaScript code, and the code in other languages that I've struggled to read the most has always been code that has a lot of cyclomatic complexity - a lot of conditionals, a lot of nested statements, a lot of visual misdirection. So sometimes the happy path will be nested inside three layers of ifs, and it's like "Well, how was I supposed to know that?"
+
+**Johnny Boursiquot:** Right, right.
+
+**Denise Yu:** But I think Go strongly discourages that, and encourages a kind of linear logic flow.
+
+**Jon Calhoun:** It's interesting that you mentioned the happy path, because as far as I can tell, there's nothing specific about Go (the language itself) that enforces that. It's just kind of the convention everybody's adopted, and I don't know if it's the fact that all the developers who were working on it were working on complicated enough problems that they were like "We need some sort of sanity check here" or what... But all these other languages - there's definitely no reason why they couldn't do it.
+
+I remember when I first started programming - I think it was Java, is what I saw it in... And there were people who made the argument that if you ever have an if statement, there should be an else statement. And in Go it's the exact opposite argument. If you can avoid that else statement, go ahead and avoid it. Just return or do whatever you can, and get everything back, so that it's not indented any further. So it's interesting to see those differences.
+
+It's also interesting to hear you talk about it this way, Denise, because we've had Dave Cheney on the podcast before; last time we had them, he basically had just published the Zen of Go website, which basically kind of fell out of a talk he gave -- I believe it was at GopherCon Israel, or something like that... But basically, he was making the case for not really treating idiomatic Go - what we consider idiomatic Go - as Gospel. And I could definitely understand where he's coming from.
+
+\[12:20\] At the same time, a part of me was like "Yes, absolutely, don't let the idioms in the community become a barrier, or some sort of gauge as to whether you're a good Go developer, or not so good, or whatever it is. Don't use it as a barrier." But I think in a lot of cases it also helps to instruct and guide. It provides this guard rail for folks to have an idea, "Okay, I could write Go like Java (I call that Gava), I could write Go like Ruby (I call that Guby)", you could write it to look and feel like other languages that you're perhaps coming from or that you're more comfortable with... But that would look a little weird as a Go project; that would look a little weird within the ecosystem of other Go projects. So when I read that code, if it didn't look like other Go projects, even though syntactically it could be compilable and there would be nothing wrong with it, but it would still feel wrong; perhaps that's too strong a word, but it would simply be not idiomatic.
+
+I'm wondering how you see the idioms of the community. Do they help/hinder, did they help you in your learning? Do you sometimes find yourself deviating? How do you treat what we generally refer to as idiomatic go?
+
+**Denise Yu:** Personally, I try to follow idioms as much as possible, but I also acknowledge that I've been programming for a couple years now, and for me learning a new language's idioms is not such a high barrier anymore, because I've done it for a bunch of other languages by now... The one thing I think makes onboarding a little easier is that Go ships with tools that in a lot of times enforces these idioms. For `go fmt` for example. I don't know if you consider spacing and indentation part of a language's idioms - maybe, maybe not. But if you just run Go Format, it's not even something you need to think about.
+
+**Johnny Boursiquot:** Right.
+
+**Denise Yu:** On the point of people writing Go like other-languages-Go - when I worked at Pivotal... And bless their hearts, when I started out, I didn't know what Go conventions were, and I just kind of copied the pattern that was in front of me... Which by the way, people always copy the pattern in front of them. So I think that fact doesn't get leveraged enough as a teaching tool... So we can do a better job at putting the right patterns in front of people early on in their learning journey with a new language or a new tool. I think that's a very powerful lever to pull on. But when I was at Pivotal, we very much wrote Go as if it was Ruby. We even used behavior-driven testing suites. We used the test runner in an assertion library called Ginkgo and Gomega, which I don't' know if you guys have come across that before...
+
+**Johnny Boursiquot:** Yeah, I'm familiar with them...
+
+**Denise Yu:** I've since learned, after talking to people who are outside of the Pivotal Go community that those two libraries are quite contentious out in the wild, because they ease the developer who knows Ruby and RSpec into learning Go... But you don't learn the Go way along the way; you're just sort of like insisting on following the Ruby way.
+
+One of the things that was a source of -- not tension, but just like a source of frustration for developers that I worked with at Pivotal was Ginkgo has a before hook and a just before hook... And the idea -- say you have a file that's for the whole package, or something... Maybe that package has methods that each need setup to be a little bit differently... So if you have your global before, you can nest contexts, and within each context you can run a just before, and then add some different flavor just for each context... Which - if you don't know about the existence of just before, and you go to the top of the file and you're like "Okay, that's weird... My before is running. What's this extra side effect that's swooping in right before this test executes?" And that would be the just before hook, which - I don't wanna talk about how many hours I've spent debugging that. \[laughter\]
+
+**Johnny Boursiquot:** \[16:25\] Yeah, that's the thing... It's like, there's a feature or an approach that these libraries bring, that if you don't know what the team's convention for using these things are, it could easily bite you.
+
+I'll admit, when I first saw those libraries, I was like -- I mean, the language I was using the most right before Go was Ruby. So when I saw these things, I was like "Hey, I can do RSpec style of development in Go. How cool is that?!" I saw it as really like an on-ramp into Go development. And the more I learned about Go and the standard library, what you can do without any additional stuff or any additional packages and things, the more I started to see the advantage of using the standard library and the things that it comes with, that you can do out of the box. I sort of stepped off the whole BDD train, but I started using assertion libraries.
+
+Testify has been and remains one of my favorites. Before I bring in Testify, I'll try things out with just plain, standard library, just with the testing package, and a lot of times it works just fine. I might add the cmp package from the Google folks, just to do comparisons and diffs and stuff for a test output... But really, between the standard library and that package, it offers the vast majority of what I need on most projects. Again, I'm not saying that if somebody wants to use those packages, will they make life easier for transitioning from a different language that has an assertion library; I'm not saying you shouldn't, but I always tell folks, "Hey, try the standard library first, and then see what you need." It's almost like "Don't start with all the things. Incrementally add things as you need them."
+
+Like you, I saw Go as -- it sort of made me a better engineer overall, because it forced me to take the simplest path, and then only really bringing in things that I needed it to, when I needed it to. Interesting how that works, but... I'm not fighting it.
+
+**Jon Calhoun:** I agree with you, Johnny - we need to try those things; it definitely makes sense. It's kind of like when you're getting in the pool, you don't have to just cannonball in the deep end; you can ease your way in, gradually... And if using those tools helps you, that's fine. But one of the side effects to that is, like Denise said, if you have a big company with a big project and they never really move off of those, or if they sort of stick with it, it's almost like they're using Go, but they're not really using it the way everybody else is. They're not necessarily getting the full benefits that could be there.
+
+Now, that doesn't necessarily mean that they're doing a terrible thing or anything, but it is potentially affecting their experience with it. So it's kind of a double-sided blade, where you've gotta choose when to maybe try something else or see if that make sense, and when it's not really a good learning tool.
+
+**Denise Yu:** Yeah. And the last time I worked at Pivotal on the Concourse team -- so Concourse is different from the other projects, because it is fully open source-facing, so anyone can send a PR to Concourse. So towards the end of my time there we actually were experimenting with not using Ginkgo and Gomega anymore, because that does represent friction for new contributors... Because Pivotal is probably one of a handful of companies that use those two tools aggressively. But if you've never seen that before and you're like "Oh, I wanna just submit this little bug fix", and then you go look at the 800-line Testify you're like "Oh, God... I've gotta add a test into this thing..." \[laughter\]
+
+It's much better to just choose actively the most popular path, which is to just use go tests, or something more lightweight.
+
+**Jon Calhoun:** You can see this in other languages, too. Even in Ruby, so many people use RSpec that while there are other great ways to test there, you almost have to use RSpec because everybody is used to it and they're familiar with it, and if you want your project to be open source, that's kind of the path of least resistance.
+
+**Denise Yu:** Yeah, exactly. That's a great way of putting it.
+
+**Break:** \[20:27\]
+
+**Johnny Boursiquot:** I wanna shift a little bit into -- actually, you spoke at GoCon... Was that two years ago, three years ago now?
+
+**Denise Yu:** Actually, it was about last year.
+
+**Johnny Boursiquot:** Was it? Wow...
+
+**Denise Yu:** Yeah.
+
+**Johnny Boursiquot:** Okay...
+
+**Denise Yu:** I know the last two months have felt like a year each.
+
+**Johnny Boursiquot:** Literally, ever since the lockdown, the days have started melding into each other for me... Now a week that goes by feels like a month. It's kind of crazy... But you spoke at GoCon, and the following year you joined the organizing team, right?
+
+**Denise Yu:** Yes. Yeah, this past year I was -- well, I guess I still am on the organizing team. Just because the conference didn't happen doesn't mean I'm not an organizer...
+
+**Johnny Boursiquot:** Right, yeah. So you joined the organizing team... And yeah, pretty much y'all had everything just set up and ready to go; you were reaching out to speakers, and putting the finishing touches on things, and then obviously the whole Covid-19 thing happened, and now you're basically having to cancel the conferences here... I'm sort of curious -- well, first of all, I'm hoping that you and the organizing team are gonna bring the conference back next year. When exactly - that's TBD, but hopefully that's something that's gonna come back.
+
+I'm wondering, now that you've been part of the organizing side of the conference, what have you learned from that experience? I'm asking purely from somebody who organizes events, and everybody has a little something they'll do different... So specifically within the context of the Go community, what have you learned about organizing this conference?
+
+**Denise Yu:** I learned a lot... I just didn't realize how many moving pieces there are to event production. Because I've also only done community one-off events, meetups and that sort of thing. Meetups and conferences are not the same thing. With a meetup you get one venue host, you maybe get a couple hundred dollars for pizza, and you just throw an event on Meetup and that's pretty much it. All the infrastructure is already there.
+
+But for a conference, you've gotta -- it took us two months probably to find a venue, because we were aiming for 350 people, and it turns out that's a very awkward size for an event, for venues. Venues are either small, sub 100, or they're over 500. There's not that much in the middle... And we also wanted to balance things like closest to public transit; would it be in a neighborhood that was relatively interesting? Could people go for a walk, go outside if they wanted to?
+
+We were trying not to do a hotel ballroom, because Toronto has a lot of interesting buildings, and I think we were hoping to find a venue that supported -- I don't know, something that wasn't like a giant international hotel chain. \[laughter\] Of course, we would need to get a hotel anyway, to put our speakers in, so we couldn't avoid giving money to hotels... \[laughs\]
+
+\[24:10\] So the venue that we ultimately found was the Isabel Bader Theater, which is part of the University of Toronto. I had actually learned about the theater because they do all sorts of events there, and my partner and I went there because one of our friends was performing her community orchestra there... And when we turned up for the concert, we were like "Wow, this is such a great venue. The seats are so comfy... Wouldn't it be great if we could have a conference here?"
+
+So as the conference time approached, we were looking at lots and lots of different venues. It's difficult also to find a venue that will give you flexibility over catering. We didn't need 100% flexibility; we didn't wanna bring in 100% of our own, or anything... We just wanted something that was at a reasonable cost.
+
+A lot of these venues are geared towards weddings. So the second you say "Large event" or "Conference", they're like "Let me tell you about our $99/head banquet package." And we're like "We don't really need $99/head. We want this to be affordable for everyone that's coming." So that's just the logistical side of it.
+
+In terms of -- what else...? This was a couple months ago, and I feel like my brain has kind of forgotten a lot of things in the last two months of just general pandemic chaos.
+
+**Johnny Boursiquot:** Did you work in a program as well? Like, try to figure out what kind of talk, and from whom, and all these things?
+
+**Denise Yu:** Yeah, so we kind of divided and conquered in terms of what each person wanted to work on... Because we had a team of six or seven people. My main area of responsibility was the program... Because I speak at a lot of conferences, I could deeply empathize with people, especially -- I feel like I'm always empathizing with the first-time submitter, or with the person who's kind of new to the community, who doesn't feel 100% confident in casting their net out there... So we went through several rounds of copywriting for the actual CFP language, and I spied a lot on other CFPs that were open at the time.
+
+Special shout-out to the DevOps Toronto team; I think they did a really good job of spelling out, like "This is the type of assistance we can give you if you're a new speaker. Here's who we want to hear about. Here are the things that you shouldn't be worried about. Don't worry if your talk is not technical enough; don't worry if you feel like it's too introductory. We anticipate a very diverse audience in terms of experience level, so there is room for every type of talk."
+
+One thing that we didn't fully align on and was kind of an ongoing conversation was who was the intended audience for this conference... Because Canada doesn't have another Go conference. This is the only Go conference in Canada. Certainly the only one in Ontario, certainly the only one in the Greater Toronto Area... So we kind of went back and forth on whether we wanted to prioritize local representation, because we wanted to use this as a chance to create some local voices, leaders in the Go community... But if we did that, we also wanted to make sure that whoever we were giving a platform to, we want a diverse speaker line-up. Even if it doesn't represent the current Go community -- because I know that in the last community survey I think only between 5% and 10% of Go developers self-identified as women, I think. It was a pretty low number, and our gender balance was much higher in terms of non-men to men; it was much, much higher than that. So I was really happy with that. Out of a line-up of 17 speakers, I think we had 5 or 6 non-men speakers...
+
+**Johnny Boursiquot:** Nice.
+
+**Denise Yu:** Yeah, it was really -- it was hard to find those people. We didn't do as much community outreach as I would have liked to. We just kind of ran out of time on that... But we did spend a long time going through the CFP.
+
+Oh, actually - I wrote an application for it... \[laughter\] I was out of work at the time, and I didn't like how PaperCall -- like, no offense if anyone from PapelCall is listening to this, but I didn't like PaperCall's built-in mechanism for sorting through submissions... Because my requirements were I want multiple reviewers, and PaperCall caps you at five people... But I was like "Well, we have seven people, and possibly more, if we decide to bring on guest reviewers." So I want an unlimited number of reviewers.
+
+\[28:22\] I want the author, the organization and identify details to be not shown by default, and I also want the ability to edit the content of the abstract, in case someone said "Oh, by the way, I'm blah-blah-blah, and I work for blah-blah-blah" just in the abstract itself. PaperCall won't do that for you; PaperCall just hides the name and the email something like that.
+
+So I actually built an app in Node.js, from the ground-up, that you export all your JSON from PaperCall and you just put it into the app, and it displays every single talk as it's own -- like a submission form. I totally \[unintelligible 00:28:55.11\] the login process. I put it on Heroku, and I was thinking -- I basically had a one-person hackathon for four days to crank this out... And the last night, I was like "Ohh, authentication... I really, really don't want to think about OAuth right now. I really don't wanna build a Twitter login, or something like that..." So what I did was I just put a map of keys and values into the Heroku app environment... \[laughter\] That was people's usernames and passwords.
+
+**Johnny Boursiquot:** \[laughs\] Okay, as an SRE, you've been officially slapped on the wrist. \[laughter\] Wow.
+
+**Denise Yu:** I know, so bad...
+
+**Johnny Boursiquot:** But hey, you shipped it... There's something to be said for that. And it worked. I imagine it worked for what you were looking for.
+
+**Denise Yu:** Yeah, exactly. And as soon as the review process was done, I spun down the app... I think. I should check on that, actually. \[laughter\]
+
+**Johnny Boursiquot:** So I can totally feel you with regards to having a diverse speaker pool. Folks just don't know how hard that process is, for a number of reasons. I've come across folks who are very talented, super-smart, that don't fit the typical demographic... And you say "Hey, come give this talk. You'll do great", but they have so much self-doubt, like "I don't think I can do this..." I'm like "No...! Believe me, you can!" It takes so much convincing... And obviously, I think our industry is to blame for that. For so long we weren't really paying attention to any of that... But now that we are, it's not something that's going to change overnight, and that's what I usually try to tell folks... Like, "Hey, this work that you're trying to do is hard."
+
+I know that it's hard because it's not going to change overnight. It's not like all of a sudden when we start paying attention to the problem you're gonna have all these great speakers that don't fit the typical demographic... Like, these people are just gonna come out of the woodworks - no. You're gonna have to do a lot of outreach, you're gonna have to find them, convince then, and really be more welcoming for these folks who step out and really take on these roles.
+
+So I definitely applaud you for doing that work, and I definitely know that it's not an easy task at all. I do wanna touch -- so your job as community and safety... Every time you say the title for that job, I'm like "Hm, what is that...?" So you talked about the policy aspects, and how that sort of works with the engineering policy aspects of things... I'm wondering, what is your day-to-day like? When you're worrying about community and safety, what does that mean exactly? What's your day-to-day like for that?
+
+**Denise Yu:** Sure. So my team is a very interesting team, because community and safety -- first of all, it's quite nebulous. Most people think "Oh, you're like \[unintelligible 00:31:48.06\] team. You're the support teams that handle user complaints, and abuse reports" and we're like "Well, no." GitHub actually has a separate team called User Policy for that mission.
+
+So community and safety exists to build tools to help maintainers grow sustainable and healthy communities. I think that's the shortest summary of it that I can think of. What that means in practice is there are a lot of different features scattered all throughout GitHub, where individuals can be either reactive or proactive.
+
+\[32:22\] I'll give an example of reactive... Let's say you're arguing with someone on a GitHub Issues thread, and they start becoming really abusive, and they start cussing you out, or just being unpleasant. So for a long time, all you could do was report that all the way up to GitHub admins... Which is a good strategy, and that's definitely a very valuable layer to have... But the problem there is site-wide admins don't have all the context; they would have to maybe go back through lots and lots of previous comments and try to figure out exactly what happened, what led to this moment, as part of investigating whether it is a high-priority incident to respond to.
+
+A feature that my team introduced is called tiered moderation. Most of this work was done before I got here, so I'm not taking any credit for the engineering work. This credit goes to the community and safety team before I got there. So tiered reporting means that - well, okay, if you don't report all the way up to site admins, who else can you report to? So a more logical person actually would be the maintainer team of an organization... Because if you've been working on an open source community for a number of years, you kind of know who the problem actors are, and you kind of know who has good intentions, but maybe struggles with written communication.
+
+So if you now give people the option to report to community maintainers, that's another layer of this sort of progressive escalation, I guess, that we give to community members. Community maintainers also might be in a better position to reach out to the person who's being aggressive and just find out "Hey, what's wrong?" Maybe someone is really stressed out because of this global pandemic that we're all living through. Maybe someone misinterpreted something. We sort of give people tools to assume best intent, but take action for the safety of the community. That's kind of reactive.
+
+Some proactive things we do are we will try to design the site in a way that encourages positive interaction. Now, maybe after Satellite - I'm not sure when this feature will launch, but if you go to open an issue on a repository, there will be a screen that asks you "Okay, is this a bug report? Is this a feature request? Is this just a question you have about using the app?" And if it's a question, it'll redirect you to somewhere else. That helps to make the issues fire hose more manageable for maintainers. I don't know if you've ever been on the receiving end of thousand of GitHub issues, but it's so hard to wade through, and a lot of people have come up with their own engineering solutions over the GitHub API to try to visualize issues in a way that's more manageable.
+
+So it's things like that... There's a lot of room for engineering. One of the things that I've been working on lately is improving abuse reports, and -- what else...? There's a new thing that I'm working on, but I don't think I can talk about it yet, because I would need to go through the release channels for that... So maybe I'll circle back to you in a couple of weeks about that.
+
+**Johnny Boursiquot:** Cool.
+
+**Jon Calhoun:** So would your team be the team that's in charge of releasing features like the templates they have for PRs for issues, and that sort of stuff? Is it that type of stuff, as well? Or is that outside the bounds of what you work on?
+
+**Denise Yu:** I feel like issue templates and PR templates are probably owned by the teams... Because there's a team that works on issues and a team that works on PRs.
+
+**Jon Calhoun:** I don't mean actually creating it, but the feature that enables that to work.
+
+**Denise Yu:** Oh... I'm not 100% sure. I can check on that though. There's a lot about what this team does and doesn't do that I don't know yet.
+
+**Johnny Boursiquot:** Yeah, that's part of the beauty of a brand new environment, brand new team, brand new job. You're learning on the job, right?
+
+**Denise Yu:** Oh, absolutely.
+
+**Johnny Boursiquot:** \[35:55\] I wanted to talk a little bit about the drawing lessons that you've been doing for the internet for the last few weeks. What brought that about? Tell us about that - what is that, what are you trying to achieve there? \[laughter\]
+
+**Denise Yu:** So for the last few years -- okay, so it's kind of a funny story why I started doing sketch notes in the first place... Back in 2018, maybe end of 2017, I worked at Pivotal in London. And if you were a member of the tech community in London around that time, most people know Pivotal as the office that will host anything... \[laughter\] We will host any meetup, we will buy you pizza, we will let you drink our beer... Just say the word, and if there's an employee who can be here physically to host it, then our space and our food and drinks are yours. A lot of my friends knew this fact, and they would bring their meetups \[unintelligible 00:36:47.01\] the person hosting them.
+
+So one day, my friend said "Hey, there's this sketchnoting community. Do you wanna host them?" and I said "I have no idea what that is, but that sounds great. Bring them over. We'll get pizza." So they came, and I actually ended up attending quite a few sessions. We hosted probably 3-4 sessions, and it was really cool.
+
+I always drew things when I was a teenager. My favorite thing to do to procrastinate doing homework was to draw anime characters when I was growing up... But I hadn't drawn for a very long time, and it was something that I had forgotten how much I enjoyed, until I started doing the sketchnoting meetups again... So a couple months after that, in early 2018, I did a product management rotation at Pivotal, and one of the things I had to do for my job was to push out lots and lots of communication... And I found that sometimes I would send an email to either all of R&D, or just a couple of other teams, and my emails would say "Breaking changes in our API. Be aware of this", and people would miss the email. It would get filtered away, and they would never see it, and then a few weeks later someone would come to me and say "Hey, my thing broke", and I was like "Yes, I told you this was going to happen. You just didn't read my email."
+
+So I started an experiment - I started sketchnoting my announcements instead of writing them in email, and engagement rate went through the roof. I was having VPs respond to my email, and I was like "I didn't even know this guy was on this mailing list, but that's cool..."
+
+So I found that that was a really powerful way to get people to look at things that you were saying... And I liked also the frequency of people coming and saying "My thing is broken." That went down a lot after I started doing this, which was great; that was my intention.
+
+**Johnny Boursiquot:** That was the goal.
+
+**Denise Yu:** Yeah. But then just for fun, I started trying to sketchnote people's conference talks. I would go to industry conferences, and also community conferences, and picked talks that I was kind of interested in, sat there and sketched. I started out just doing pen and paper. Eventually, I got an iPad and started doing it digitally... But I found that of all the interactions that I had at conferences -- you know, when you go to large conferences, you get kind of tired after a few days. Not every interaction is gonna be positive. If you're speaking, I find that people generally tend to be nicer than -- I don't know... But once in a while, if you're speaking, someone will ask a rude question, or a question that's not really a question... Things like that. You have mixed experiences there.
+
+But I've found that when I started producing art of other people's talks, people universally loved that. Nobody was ever mean about a piece of art that I created. The speaker was almost always so surprised, and so excited that someone had visualized their talk... So that kicked off a cycle where I was like "Huh. There's something here." If I can start off by sketching other people's talks and give them greater outreach... And actually, to this day, I still get a Twitter notification every few days, because a year and a half ago I went to the lead developer in New York City, and Tania Riley was there speaking about being blue, the idea of doing all the "non-technical" work that teams need to keep running... So I sketched that talk and published it, and every couple of days someone finds that and retweets it again. It just keeps coming back.
+
+\[40:15\] The talk itself is really great also, but having a visual accompaniment to that talk enables someone to sort of look at it and say "Oh, this looks really interesting. I now am going to go and invest 30 minutes in watching the original, full-length talk."
+
+So my intention is never to replace the information that's in the talk, it's more to give people a high-level overview of what happened in it, and help them make an informed decision about whether they want to invest more time digging deeper and going to watch the original.
+
+Inevitably, as I started producing more of this on my own, people started asking me "How do you do this? I wanna learn how to do this. Can you teach me?" I was like, "Um, I can try..."
+
+The first time I tried to teach it was at -- I think it was at Write/Speak/Code last year. The same conference where I met Lexy. So I ran a two-hour-long (I think one-and-a-half hour or two hours long) workshop where I broke down sketchnoting into a couple different basic skills. One of them is being able to draw humans and being able to give them emotions. Another is drawing shadow, like shading things 101. All of the lessons are geared towards people who don't self-identify as artists; people who say "I can't draw. I can't even hold a pencil."
+
+**Johnny Boursiquot:** That's me. \[laughter\]
+
+**Denise Yu:** And I think I've gotten a lot of encouragement from a few specific people... One of them is Marlena Compton, who runs AppearWorks; so Marlena and I, and also our other friend Kaitlin Gu are teaming up to -- we're running a whole conference about this in two weeks, called Let's Sketch Tech.
+
+**Johnny Boursiquot:** Yeah...
+
+**Denise Yu:** The other person that has really been such a vocal cheerleader is \[unintelligible 00:41:56.09\] from Microsoft. Basically, anytime I put anything online, Nithya's like "Instant retweet. Do more of this. You're awesome. I wanna see more." So I kind of wish that everyone could have a Nithya just in their corner, just cheering them on for whatever they wanna do...
+
+**Johnny Boursiquot:** Yeah, like they're always cheering you on... That's awesome.
+
+**Denise Yu:** Exactly, exactly.
+
+**Johnny Boursiquot:** I'm checking out the website right now... It's letssketchtech.com.
+
+**Denise Yu:** Yes.
+
+**Johnny Boursiquot:** You can register for different days... Yeah, the next one coming up on May 9th is "Telling stories with doodles." You can absolutely register for it. This is such a cool idea. I might have to jump in on this.
+
+**Denise Yu:** Yeah, please do.
+
+**Johnny Boursiquot:** Because I'm like -- I have a very clear understanding of the value that visuals play in communication. Some of my best talks, I had folks come up to me and say "Hey, that visual you had on this slide, or that diagram (or whatever it is), I could really appreciate that." And you can easily give them that sheet, or that image, or whatever it is, and it packs so much information, because you're visually representing -- they say an image is worth a thousand words for a reason, because you can communicate so much when you go beyond just text... And this is something I encourage folks who are putting talks together all the time... Like look, there are two very specific things you can do to make sure your talk, beyond the content itself -- your content can have value, but that's table stakes; there are two things you can do to make sure it's well-received. One, tell stories around it, and two, use good visuals. Don't fill your slide with bullets, and text... Nobody's gonna read that stuff. It'll just put everybody to sleep.
+
+If you can tell stories and if you can have visuals around things, illustrations... That's sort of the next level I wanna get to. I wanna have illustrations in my talk, I wanna draw some cats, like you're teaching people to do... \[laughter\] Because these things do make for a more engaging and a more fun experience. And proof you've got by starting to add these things, sprinkle these things in your announcements... And you saw the effect - people were reading them, they were engaged, and it solved the problem. So I think this is absolutely definitely worth it. So for all those listening, do check it out. It's letssketchtech.com. Yeah, I definitely will be checking this out myself, because I need some help, because my doodles are terrible. \[laughs\]
+
+**Denise Yu:** \[44:20\] The conference is also kid-friendly, so if any of your kids wanted to come...
+
+**Johnny Boursiquot:** Nice!
+
+**Denise Yu:** Jon's daughter might be too young for it... \[laughs\]
+
+**Jon Calhoun:** A little bit too young. So when you're drawing, do you also find that -- because you do ones about specific topics... So when you're doing that, do you find that it forces you to think about how to explain the topic from a different perspective? When you're writing up a blog post, you kind of have the world; you can go on as long as you want. And with a single-page sketch, I assume that that limits you and forces you to think more creatively about how to present this material.
+
+**Denise Yu:** Yeah, definitely. I've been really inspired by seeing Julia Evans' work over the years, and one thing that Julia has spoken about is the most difficult -- or I don't know, it's the most difficult for her, but she has called out that it's very challenging to distill a complex concept into something that'll fit a tiny space... Because when you're making these visuals, the point is not to capture everything. You wanna capture just enough, and then complement it with something visual. So that part is definitely tricky; that takes a lot of practice. I still sometimes err on the side of being too wordy.
+
+**Jon Calhoun:** I think that's hard to break, though... I've heard several people state that it's harder to write something that actually gets the point across in fewer words than it is in more words, for almost everything... And I imagine when you're drawing and doing stuff like that with it it's even harder, because you're trying to toss the visuals in there, you want it all to make sense, but you also wanna not just go on forever.
+
+**Denise Yu:** Yeah, definitely. One thing I do pretty often, especially if I'm making a drawing to force myself to learn about something new - I get people who know more about it to review my work. Or I just straight up partner up with someone who knows the subject matter really deeply.
+
+I have one drawing about the Raft consensus algorithm... Because - what else are you gonna do on a Saturday afternoon? "I'm just gonna draw the Raft consensus algorithm..." But I have partnered up with my friend Matt, who lives in London, and he is a distributed systems pro. He knows a lot about that. So I was like "Alright, I'll take care of the visuals. Let's pair on this, let's jump on a call. We'll do version one together, and then I'm gonna send you over the final version. Literally, you just tell me if I have captured anything inaccurately."
+
+**Johnny Boursiquot:** That is pretty cool. It also seems like you have a newsletter going on. Are you actively publishing content? Where do I sign up?
+
+**Denise Yu:** It's not really a newsletter, the way that some people use it to sort of like talk out loud... I am trying to only use that newsletter to blast out announcements about upcoming workshops and events that I'm taking part in, that are mostly free.
+
+**Johnny Boursiquot:** Cool.
+
+**Denise Yu:** I don't know, I'm kind of on the fence about whether I should send a blast to everyone to tell them to buy tickets for the conference... But I've been kind of lazy about it lately, because I've just been kind of low on headspace, with the pandemic and everything...
+
+**Johnny Boursiquot:** Right, right.
+
+**Denise Yu:** But my intention is to run more free workshops at a regular cadence. I initially started off saying "Look, I'm gonna do this twice a week", and then I was like "Oh god, that's so much work, to set up an event, and do the marketing, and everything..." So I just kind of stopped doing that. But I've recently started working with the Microsoft Reactor team, which has actually been fantastic. We did a workshop this past Saturday, and they were just so great. \[unintelligible 00:47:28.13\] They handled the event setup on Meetup, in like three different Reactors, and at one point we had like 105 people watching the stream, which was really wild. I had no idea that many people wanted to learn how to do tech sketchnoting... So that was awesome.
+
+So if I do future events, probably the more put-together ones will probably be in collaboration with the Reactor team... Which I'm pretty stoked about.
+
+**Johnny Boursiquot:** \[47:59\] That is pretty cool. It's definitely one of those things where you look at it and you're like "Yeah, this is kind of cool." And obviously, because I follow you on Twitter, I saw some of your retweets, and you were talking about the sketching stuff - I was like "This is kind of cool. Here's another person in the industry that I respect, and they have a hobby. I could use a hobby... I could use a hobby, too." \[laughs\]
+
+Yeah, this is awesome. Do keep doing it... I think especially in this season that we're in now, where everybody is sort of low on headspace, like you mentioned... Something like this, where it just kind of takes your mind off of things is absolutely valuable. Thank you for doing it, and I hope you will keep on doing it.
+
+Jon, anything else we wanna touch on before we ask Denise to drop her unpopular opinion? Denise, did you prepare an unpopular opinion?
+
+**Denise Yu:** I'll think about it right now...
+
+**Johnny Boursiquot:** Judging by the look on your face... \[laughs\] Judging by the surprised look...
+
+**Denise Yu:** I'll think about it right now.
+
+**Johnny Boursiquot:** Okay, good.
+
+**Denise Yu:** Okay, I have one.
+
+**Johnny Boursiquot:** Jon, do you have the music for Unpopular Opinions?
+
+**Jon Calhoun:** I do. Before I play it - Denise, it doesn't have to be tech-related specifically. It's kind of anything you want... So don't stress yourself out.
+
+**Denise Yu:** Okay.
+
+**Jon Calhoun:** Are you ready for the music, Johnny?
+
+**Johnny Boursiquot:** Hit me.
+
+**Jingle:** \[49:19\]
+
+**Denise Yu:** Okay, I thought of one.
+
+**Johnny Boursiquot:** Yeah, please.
+
+**Denise Yu:** My unpopular opinion is that buying and selling turnips in Animal Crossing is overrated.
+
+**Johnny Boursiquot:** Oooh. Okay...
+
+**Denise Yu:** Because I think Animal Crossing is an entire game about the journey, and grinding Bells is part of the journey. So if you sell a bunch of turnips for whatever price they peak it - I think that fast-tracks you past the parts of the game that I think are fun.
+
+**Johnny Boursiquot:** Okay, okay... That's good. First of all, I was gonna ask "What is Animal Crossing?", and you said it was a game, so thank you for that... \[laughter\]
+
+**Jon Calhoun:** Johnny and I don't play -- like, I don't play this game, but I've heard that it's a lot of farming...? Not like agricultural farming; farming as in like doing repetitive tasks a lot...
+
+**Denise Yu:** Yeah. There also is agricultural farming though... You could grow flowers and trees.
+
+**Jon Calhoun:** Well, a little bit of both then.
+
+**Denise Yu:** It's a game that came out on Nintendo Switch, and basically almost all of my friends are playing it. I only got it a few days ago, because I left my Switch in Toronto, and I'm in New Jersey now, so I have to actually order a new Switch to play with... \[laughter\]
+
+**Johnny Boursiquot:** Wow... It means that much to you!
+
+**Denise Yu:** Yeah, I know. I was desperate. I was like "Okay, I can't handle this FOMO of seeing everyone else's islands on Twitter anymore. I'm just gonna buy another one."
+
+**Johnny Boursiquot:** \[51:10\] Wow... \[laughter\] Awesome. Okay, well, for those who do play Animal Crossing - yeah, this will either anger them, or they'll be like "Yeah, yeah! I agree!" \[laughter\]
+
+**Jon Calhoun:** I think you could make the argument generally about a lot of other games though... I feel like people are impatient when it comes to games where there's some effort you have to put in to get someplace... They'll immediately take the shortcut of however they can get to the end, sort of, and then they miss that experience along the way.
+
+When I was younger, I played Diablo II a lot, and they had ways you could just mod your files and give yourself all the gear in the game... But then as soon as you did that, you were like "Well, I don't wanna play anymore, because it's not fun now that I have everything."
+
+**Denise Yu:** Yes, exactly.
+
+**Jon Calhoun:** And you realize very quickly that the journey is the enjoying part of the game, it's not actually the end result.
+
+**Denise Yu:** I agree with that.
+
+**Johnny Boursiquot:** I can agree with that, because I have played Diablo. I played the original, Diablo II, and everything really in the Diablo series from Blizzard... So I can definitely relate to that, yes.
+
+**Jon Calhoun:** Johnny.
+
+**Johnny Boursiquot:** Yes.
+
+**Jon Calhoun:** If you tell me that the mobile game is good, that's like an unpopular opinion that I'll hold you over for like two months.
+
+**Johnny Boursiquot:** Wait, there is a mobile version of that?
+
+**Jon Calhoun:** There is a mobile game.
+
+**Denise Yu:** A mobile Diablo?
+
+**Jon Calhoun:** Yeah, I'm pretty sure there is. Because at one point they were supposed to be announcing Diablo IV, and they announced Diablo the mobile game, and everybody got really, really mad.
+
+**Denise Yu:** Oh, yes, I think this rings a bell... I think I remember reading about this.
+
+**Jon Calhoun:** It was at Blizz - whatever their conference is - not that long ago; maybe a year ago, I don't know... But basically, they had this big thing like "Oh, there's the Diablo announcement" and everybody thought it was gonna be Diablo IV, and it was not. \[laughter\] They were like 'We have it for mobile" and everybody's like "What are you doing...?!" \[laughter\]
+
+**Johnny Boursiquot:** "What is this nonsense?" It wasn't idiomatic, you know? It wasn't something the community was familiar with...
+
+**Jon Calhoun:** And apparently, video game fans are very vocal about their unhappiness...
+
+**Johnny Boursiquot:** Yeah, I'm not gonna say anything. I might get in trouble. But yeah, they're a very vocal group. Hey, to each their own.
+
+Awesome. So this was fun, having you, Denise. Like I said before, I really wanted to introduce you to the broader Go community, because I think you're doing a lot of fun and cool stuff... And the role you're playing I think suits you quite well, in your gig... And yeah, we didn't even talk about the teaching stuff that you do; you help out with GoBridge as well, and all that stuff, and I'm sure folks will find out more about that when they look you up.
+
+I definitely wanna thank you for coming on the show. This has been fun, and I hope we can do it again.
+
+**Denise Yu:** Yeah, for sure. Thanks so much for having me.
+
+**Outro:** \[53:57\]
+
+**Denise Yu:** I've gotta get myself a podcasting setup like the two of you have...
+
+**Johnny Boursiquot:** \[laughs\]
+
+**Jon Calhoun:** Apparently, if you join the Go Time podcast and do it every week, they just send you a mic.
+
+**Johnny Boursiquot:** Heeey... Not a bad idea...
+
+**Jon Calhoun:** So it's really easy to do, you just have to show up every Tuesday.
+
+**Johnny Boursiquot:** \[laughs\] That's right. Every Tuesday at 3 o'clock Eastern Time.
+
+**Jon Calhoun:** It's actually not every Tuesday. If we get enough hosts, then we can sometimes take vacations.
+
+**Denise Yu:** Every Tuesday for the rest of your life?
+
+**Johnny Boursiquot:** \[laughs\] Well, you know, you can do like Mat; every now and then you take a break. Mat took a break today, so it's all good.

On the verge of new AI possibilities_transcript.txt

@@ -0,0 +1,439 @@
+**Mat Ryer:** Hello, and welcome to Go Time. I'm Mat Ryer. Today we're talking about the new possibilities in AI (artificial intelligence). We're gonna dig into that a little, learn about what it means, what it is... And since this is a Go podcast, we're gonna keep our eye towards Go on this journey.
+
+Joining me today - Jaana B. Dogan. Hello, Jaana. Welcome back!
+
+**Jaana Dogan:** Hello! Hi!
+
+**Mat Ryer:** How are you doing?
+
+**Jaana Dogan:** Good. What about you?
+
+**Mat Ryer:** Good, thanks. I'm excited, because we've got a couple of guests who know a considerable amount about AI, as you'll find out... It's Daniel Whitenack. Hello, Daniel.
+
+**Daniel Whitenack:** Hello! Great to be here!
+
+**Mat Ryer:** Welcome to the show. You are of course a host of our sister podcast, Practical AI, right?
+
+**Daniel Whitenack:** I am, yes. So... Shameless plug for Practical AI, if you're interested in those sorts of things. We release a show every week, so check it out.
+
+**Mat Ryer:** Yeah, well done. We're also joined by Miriah Peterson. Hello, Miriah.
+
+**Miriah Peterson:** Hi!
+
+**Mat Ryer:** Welcome to the show, thanks for joining us.
+
+**Miriah Peterson:** Anytime.
+
+**Mat Ryer:** Yeah, good. Well, we can just kick off if we're ready. I'm interested in just getting a very high-level idea for people that really haven't paid any attention to AI. Who wants to have a go at telling us a little bit about this? What is it?
+
+**Daniel Whitenack:** What do you think, Miriah?
+
+**Miriah Peterson:** I would say if they want to know, they can listen to Daniel's podcast.
+
+**Daniel Whitenack:** \[laughs\]
+
+**Miriah Peterson:** The way I like to explain it is what we think of AI is mostly just the typical problem-solving we do as software engineers; instead of having to think through it all ourselves, we're trying to get the computer to think through a little bit of it for us.
+
+**Daniel Whitenack:** I think that's a great starting point. When someone thinks about something an AI model does - maybe that's like recognize a cat in a picture... So if you think about the function that's serving is literally a function, like a software function. You give input data, and you get output data.
+
+We as software engineers can parameterize functions; we say "Oh, if I get this query string and it's this time of day, then do this thing" or something like that.
+
+\[04:05\] But an AI function or an AI model is essentially the same, in the sense that it's a function, it's just that the internals of that function are parameterized not through the developer's logic only, but through a process called training, which is basically like trial and error. So you give the computer a bunch of examples to learn from, it goes through some trial and error process called training to set these parameters of the function, but you end up with just a function written in code, and that's AI... Which might be disappointing for some people that think there's some type of robots always involved, or something... But it's related to robots.
+
+**Mat Ryer:** That was a disappointment for me initially, I must admit. It's interesting you talk about these functions then. So what does it look like inside it? Because obviously, if we write our own functions in Go code, we can see inside that function and we know every step that it's gonna go through. Is that the same for machine learning models?
+
+**Daniel Whitenack:** Well, Mat, how would you write a function that recognizes a cat in an image? Let's take an example...
+
+**Mat Ryer:** "If cat, then return true."
+
+**Daniel Whitenack:** So it could be an if statement, right? It could be a series of if statements that is like "If I see a bunch of red in the image in this area, then it's a cat. "That's a perfectly fine model that we can parameterize, but it's not gonna be super-useful. I don't know that much about cats, but I don't think they're all red. Yeah, so - series of if-then statements like that actually could be considered a type of machine learning or AI called a decision tree. The difference with the AI model would be that the parameters of those if statements, or the thresholds involved, would be (again) set through a separate process called training. We wouldn't go in and set them ourselves.
+
+There's kind of a numerable of these structures within a function that can be parameterized. One that people have probably heard a lot about is neural networks, which is kind of just like a bunch of sub-functions in a function that are all tied together in various ways, and could have up to like a billion parameters or so.
+
+**Mat Ryer:** Wow.
+
+**Jaana Dogan:** You mentioned a couple of times about machine learning... I wonder what is the difference between AI and machine learning nowadays. Because when I was a college student, we had a machine learning class, and they were trying to explain to us, you know, historically, this area started with AI, but then it became more about machine learning and pattern recognition... And they were trying to almost distance themselves from AI or that terminology, but now everything seems to be under the AI umbrella again.
+
+**Daniel Whitenack:** That's a great question. I'm gonna let Miriah take this one, because I wanna avoid it as much as possible.
+
+**Miriah Peterson:** Yeah, it's a question that keeps coming up a lot. The easiest way to explain it is AI is a general term for any process that involves some kind of a learned or patterned behavior. Statistical, repeatable processes end up being part of it, and just data science; machine learning ends up being part of it; the deep learning aspect of neural networks are part of it... And then to not confuse it, there's this generalized artificial intelligence, which is the idea that a computer can think like a human.
+
+So AI, I think, is more of just the marketable name; any kind of learned or statistical process. And machine learning is what we determine to be the actual computational training or method behind that sellable AI.
+
+**Jaana Dogan:** That clarifies a lot of things for me, thanks.
+
+**Daniel Whitenack:** Yeah, that's excellent. That's a tricky question, because if you ask it at some AI conference, you'll get as many answers. It's kind of one of those, and various people are opinionated about it in various ways.
+
+**Mat Ryer:** \[08:02\] So in this problem then of trying to find a cat, you mentioned you could use if statements and things to go and check the pixels, but of course, you don't know where the cat is gonna be in the image, you don't know (like you said) what kind of cat it is, so you wouldn't be able to write code in the traditional sense... So it makes sense then that we're gonna use example data or training data to come up with whatever those internals have to be. Are there any tricks to that? How many examples do we need, and what sort of challenges are around actually that data?
+
+**Daniel Whitenack:** Yeah, I kind of like to think about the training process essentially as almost like iterative testing, with table tests of some function. If you're thinking about the trial and error process, then you parameterize these bits of the function... And if you're wondering "Well, how good is my parameterization? How good did I pick my numbers?", then what you wanna do is you wanna try some examples and see how many you get right.
+
+The difference with a traditional software engineering function is that you always expect to get all the examples right. For your API endpoints you have a bunch of examples in a table; you want to get 100% of those right, and fail if you miss one.
+
+In the case of the machine learning or AI model, you're gonna have a bunch of example images. Some of them are gonna have cats, some of them are not gonna have cats. You would never expect to get all of them right, but you want to get as many as possible. So what you do is you choose some random parameters to start with, and then you run your examples through and see how many you got right. Maybe you got like 20% right, or something.
+
+Then you tweak your parameters a little bit and try again, and maybe you got 25% right, so you're kind of going in the right direction with your parameters... And you kind of just do this iteratively over and over, until you get the best parameter set that you can find. That's how the training process works.
+
+Now, there's various mathematics that help in that, in terms of not just randomly choosing parameters, but moving them in the right direction... But it's essentially that trial and error. Now, it depends in terms of the training data and how much you need, it depends on how complicated your model is.
+
+If you have just one if statement, then it's gonna be fairly quick to parameterize that, and you might not need that many examples. But if you have over a billion parameters, like some of these larger models that we see now, you're not going to find all of those parameters with 100 examples. You need very, very many examples, which is why with the scale of model complexity that we've seen over recent years, we've seen a similar sort of boom in how much training data is needed.
+
+At the same time, we've seen various tricks that allow you to adapt or fine-tune models, and not always start from scratch with your training process... Which has been one of the reasons why things are moving so quickly - there's this kind of idea of piggy-backing off of other's work. Google might have trained already on 200 terabytes of data, and you're just fine-tuning to a particular problem, so you don't need as much.
+
+**Jaana Dogan:** It's really interesting, one definition of machine learning back in the day -- I had a friend who decided to define it in a way that if you can't write the function, you just basically brute-force it, which really resembles what you're describing. If it's an easy function, you also have an easier option to maybe write it down yourself, if it's just one if, or whatever... But as soon as it's becoming more complicated \[unintelligible 00:11:53.16\] so the training phase is also getting more complicated.
+
+**Daniel Whitenack:** \[12:02\] Yeah, I totally agree with that. One of the things - and Miriah did a great job at describing the umbrella term of AI, but one of the main shifts we've seen in recent years is a shift towards neural networks, which if you think about what's happening there, if you have something like logistic regression, which is based on a simple formula, or different regressions, or maybe time series models that have to do with something with seasonality, and things that have some connection to reality (the model is based on some expert knowledge of how reality behaves), well that requires expert input into how you form this function.
+
+Whereas in recent times, the really interesting thing has happened where with these larger neural networks there's enough complexity in the neural network, there's enough parameters that you could essentially model any sort of relationship between your input and output, if you had enough data. So now we no longer have to rely on expert input as much in defining the function, we just kind of have a big function and a lot of data, and that allows us to do really interesting things. I'm not a linguist, but I can train a machine translation model, and know nothing about the two languages involved. That's pretty extraordinary.
+
+**Jaana Dogan:** Is this why they say they don't understand what's going on in that black box anymore? Because it's all magic.
+
+**Daniel Whitenack:** Yeah. I mean, at the end of the day, it's all code... But it isn't very interpretable code, let's say.
+
+**Mat Ryer:** You mentioned the recent burst in AI... What's actually driven that? Why are we not seeing and hearing so much about AI? I ask this with the context that -- I know some of the machine learning research was actually done back in the '50s... So it's kind of curious. Why now? Why are we seeing it all now?
+
+**Miriah Peterson:** I think now we just have more data than ever, more everything. They tend to - I mean, I'm not a historian, but I feel like somebody with historical knowledge tends to trace this AI boom back to the creation of the internet. All of a sudden you have internet, and now everybody's on it, and now there's a whole bunch of data. When you have a whole ton of data you can't go through, you try to figure out a way to process all of that data... And we've discovered that these neural networks can process a whole ton of data and can figure out patterns, and make leverage those patterns for our use, or for some kind of output.
+
+**Daniel Whitenack:** Yeah, I think it's kind of multi-faceted... I love that word whenever I get to utilize it. I think one thing is the availability of data, the other thing is the availability of compute... Right?
+
+**Miriah Peterson:** Yes, that's the other thing.
+
+**Daniel Whitenack:** Yeah, these models that are processing a lot of data need to process it quickly, and go through many millions of iterations, potentially... And so that requires things like GPUs, most often.
+
+Along with that... So it's data, it's compute, but then lastly it's kind of the realization that so many of these problems that we've been working on for so long, if you think about like edge detection in images - there were ways to do that before neural networks, or segmentation in images, or machine translation, or sentiment analysis... All of these things - there were methods for doing it before, but people have started to realize that all of these problems, this logic that we're interested in can be reformulated in the terms of a generalized machine learning problem, where you basically have some number of inputs in, a really complicated parameterized thing in the middle, and some number of inputs out. And most problems can be reformulated in that way without making it specialized to any domain. So now people have just kind of gone crazy with "Well, neural networks can do anything, if anything can be reformulated in this way."
+
+**Mat Ryer:** \[16:06\] Yeah, that's a really interesting thing. John on the Slack channel -- by the way, we're on GopherSlack, for anyone that wants to join the live recordings of Go Time... And you can ask questions, like this one from John; he's talking about the training data and the training process, and he asked "How does the model not just only learn what it's seen in the example data? How is it that it can see new data that it's never seen before? ...a new photograph of a cat, and it knows it's a cat still."
+
+**Daniel Whitenack:** That's a great question, that indicates a real problem in these methods. If you had a billion-parameter function, and a thousand pictures of cats and non-cats, then what's going to happen? Well, you're just going to be able to find the cats and the non-cats in those pictures really accurately. Almost at 100%. But you're not gonna be able to generalize, like Jon mentioned. This is a problem called over-fitting, and this problem has to do with your model being very complicated, but there not being enough variability in your data to generalize that model. And there's various ways to deal with this in the training process, including splitting out your data and actually optimizing around data that the model hasn't seen yet, or stopping early if you're over-fitting to the data that you're already seeing.
+
+There's even recent examples, like with OpenAI's work in robotics, where they intentionally put in some sort of randomness into the robotic simulation training data to actually make the model a little bit more robust. You might have seen these really strange pictures of robotic hands manipulating Rubik's cubes, and then like a stuffed giraffe comes in and hits the hand - that's their "stuffed giraffe" perturbation on the experiment... And they're able to handle that situation with the robot because they introduced this sort of randomness into the training process.
+
+**Mat Ryer:** It really highlights the fact that you want your training data to be as close to your real-life data as possible. I've done some work in machine learning as well in the past, and what would be very common would be that, say, customers want to detect or teach a model that it could learn to recognize logos, or something... And the training data was either a transparent PNG, or white background, very clear logo. And in the use cases, they're looking for this logo in live video streams, sports feeds, that kind of thing... And it was a kind of difficult challenge there. So would you say that it's important that your training data looks as much like your real-world data as possible?
+
+**Miriah Peterson:** Well, you want your data to be parameterized or directed -- I don't know, I can't find the right word for that... But you want your data to be appropriate for the problem you're trying to solve. We're really liking this cat problem... So if you're trying to get really good at figuring out if something's a cat or not a cat, then having data that has thousands of different kinds of cats, and each of them is a cat, and then having data that's not a cat, like having a whole bunch of sharks, would help it determine between the problem of shark and cat. But there's a lot of really interesting techniques that people have discovered to help with that problem, just by making different kinds of tweaks to your data.
+
+Say you have a whole bunch of different zoo animals and you have discovered that lemurs look really similar to cats, but you wanna still just figure out what a cat is - well, that's when you start manipulating your data. You can start adding random noise to it, and you can start adjusting size or pixel density... And then your model has to learn to pick out features that are not just maybe placement, or not maybe just ear size, but other things... And then it doesn't necessarily look like an actual cat anymore, because you've added different noises, but it helps to figure out the problem of only finding out what the cat is. So it really depends on what you're trying to look for and solve.
+
+**Daniel Whitenack:** \[20:25\] The example that I like to think of with this is if you train the models used in a self-driving car - let's say in Sweden, or somewhere - it's probably going to be pretty good in snow, and maybe certain types of environments. But if you say "Oh, now we've solved the self-driving car problem", and we ship this thing over to Australia - well, the first kangaroo that runs across the road, you're gonna crash your car and there's gonna be some catastrophic event. Your target environment definitely influences how you construct that data to train your models.
+
+**Jaana Dogan:** It's really interesting, because people have been saying that training is revealing some of the biases in the data that they have never seen before...
+
+**Daniel Whitenack:** Yeah, definitely.
+
+**Jaana Dogan:** ...which makes me also maybe ask more about "How do you evaluate the results?" We talked about testing, about the testing table, but you know, the actual evaluation is more complicated than that. What goes into evaluation?
+
+**Daniel Whitenack:** Yeah, that's a great question, and the answer is kind of disappointing, in that it sort of depends on the problem that you're solving. There's a lot of metrics that are used though that are geared towards certain problems. Most of the time, what you do is you say "Oh, I'm doing a machine translation problem" or "I'm doing an object recognition problem" or "I'm doing a time series forecasting problem" - what are the metrics that have been used to evaluate these? With object recognition you might look at something like accuracy, or precision, or recall. All of these have to do with false positives, and true positives, false negatives and true negatives, and how you balance those. In a fraud detection case you might really want to get all of the true positives, even if you get some false positives mixed in there. But that might not be what you want in another case, right?
+
+So your metric is very problem-dependent, and in machine translation you use this metric called "bleu", which has been developed specifically for that problem. Now, that's kind of separate from the bias issues that you talked about. So in addition to -- like, you could be very accurate, but still have bias in your training data, which will create a biased model. This has been shown with models that model recidivism for offenders coming out of jail... And the model will bias against black males, or something like that, because of the way that they've set up the data, which is obviously not something that we want to have happening...
+
+There's a lot of people working on this problem. There's great tools from IBM and others that actually allow you to evaluate bias in various categories in your training data, and also integrate that into modifications in your training process such that you necessarily aren't creating these biases, even if it's at the cost of accuracy or whatever metric you're interested in.
+
+**Break:** \[23:42\]
+
+**Mat Ryer:** I think that gives us quite an interesting foundation of AI. Maybe we could shift now and talk a little bit about Go's role in this. When we built Machine Box, we had Python for a lot of the internals of these boxes, and we used Go around that to do different things. That decision really came out because of the maths libraries and some of the other libraries that had already been done... How far are we now in that? What are the options for Go programmers if they wanna do their own machine learning?
+
+**Daniel Whitenack:** Miriah, what's your opinion? I know you have strong opinions here, so...
+
+**Miriah Peterson:** I do have strong opinions, unfortunately. I actually get asked at work all the time if we're gonna start doing our machine learning in Go, because I talk about Go so much... I think there's a lot here. There's so much at play, but I think as far how far we are - it depends on your problem. If you can use Go and get a good result, good accuracy or good whatever, then you can deploy your model, fine. But what I always tell people is -- I guess it's just support.
+
+I feel like there's a lot of community support that's lacking in maintaining the packages and the libraries that we have... So if you do run into an issue, you might not get the response back. There's a huge support for the Python stuff, and it's just -- you know, we've gotta get people in the community excited about using the Go libraries that are there, because they're awesome, they're strong; they just need more support. We don't want issues breaking production systems.
+
+**Mat Ryer:** So why is it so good in Python and not in Go then? Is it just because of the history?
+
+**Miriah Peterson:** There are tons -- PyData, NumPy, and there's a couple other organizations that support the open sourcing of these mathematical libraries and these data science libraries, and they're paying for people to work on them full-time. When you do that, there's just a natural stability that people feel secure about when they're using it. That's not quite there for Go, and I'm not gonna ask the Go team to take over our mathematical libraries and the data science stuff. I don't think that's their goal. I think we as a community have a ton of people using Go, and I think we have a need to start putting machine learning in our systems that we already have written in Go... So we just have to start using what's there and supporting the great tools that we have. We have a ton of great tools and libraries, they just need more support and more use.
+
+**Jaana Dogan:** I always had this impression that there's nothing much out there, so maybe it could be more of a knowledge share problem. Is there a way good way to understand what is the current state of things? Is there a requisitory that just captures what is out there?
+
+**Daniel Whitenack:** \[27:59\] There are a few... One of the great resources that people don't know about - there's a data science channel in GopherSlack, which if you're looking for anything, that's probably the quickest answer that you're gonna get, because people are very active there.
+
+**Miriah Peterson:** It's one of the most active channels I'm on.
+
+**Daniel Whitenack:** Yeah. So if I just go over there now - it looks like there's about 1,500 people in that channel, and it's fairly active. There are a few different repositories online. I know I created one at one point under the GopherData org on GitHub... Although I think it probably needs updating, and I would welcome some PRs there. And there's a few other ones floating around.
+
+We'll call out -- you know, getting involved in the Data Science channel will help, but also, there's a lot of great starting points for experimentation, but also contribution, like Miriah said. The Go Num family of libraries has a ton of things that are just like numerically-related, whether that's matrix manipulation, or regressions, or statistical tools - all of those things are there, which that is an incredible set of tools that people should know about. That's a great starting point.
+
+Other than that - yeah, it depends on what you're looking for. For the neural networks and AI side of things people might take a look at Gorgonia, and also \[unintelligible 00:29:18.29\], which are two projects that are very active. They're side-projects for people, so like Miriah said, contributions are always welcome... But they've come so far in terms of supporting things like deep learning, supporting things like CUDA integrations for GPUs, and that sort of thing. So yeah, I would recommend getting involved in the data science channel and those other things.
+
+The other thing I'll mention is, you know, as Mat said, Python has its strengths, but Go also has its strengths, so oftentimes where Python people struggle is not in the model training. They might be able to train a model really quickly and manipulate their data very quickly...
+
+**Mat Ryer:** But they can't pass a string in in a safe way, can they? \[laughter\]
+
+**Daniel Whitenack:** Exactly. Now when the software engineers come to them and they say "Oh, we wanna integrate this in our API layer", then all of a sudden there's a complete breakdown. So you have this really interesting thing where it's really easy to onboard into Python and do something quickly, it's really hard to productionize that stuff. It's getting a little bit better, but generally there's a lot of blockers there... Whereas if you can wrap things in your Go API and integrate things in that way, there's natural strengths there, there's a lot of great advantages there, and a great community around that... So there is a balance there.
+
+**Jaana Dogan:** Do you see any rewrites to Go? People might be experimenting with Python and then maybe if it's possible to achieve the same thing with Go - are they rewriting it when they are productionizing it?
+
+**Daniel Whitenack:** I definitely have seen a lot of people do training in Python, and then inference in Go. That seems to be a fairly common pattern. Inference - what I mean is you've trained the model and now you're utilizing it to make predictions... And if you think about what you're doing with that, it's the same thing you would do with any handler in an API, or another application; you're just sending data in, processing it with a thing, and taking data out... And that can be integrated in all the standard ways with how Go operates.
+
+**Mat Ryer:** Does a lack of generics in Go create a problem when it comes to implementing these maths libraries and things? Because I heard someone say that they tried to write something and they ended up using a lot of CodeGen to get the different types they needed in Go. Is that a problem you've come across?
+
+**Daniel Whitenack:** Personally, I haven't found it to be the main issue that I'm dealing with. Other people that are developing the libraries themselves - maybe they have other opinions, and I've seen also people mention that as well... So I don't know, Miriah, if you have any thoughts there...
+
+**Miriah Peterson:** \[32:04\] I feel pretty similarly. I think the generics question is more just people coming from a Pythonic way of doing it, and the way that it's worked before, and it just doesn't translate one-for-one back into Go. So there might be some workaround, but I don't think it's necessarily a hard block in the way.
+
+I think the bigger issue is there's just different thought processes, and we have to think without generics for the time being, or sometimes you've just gotta hack it.
+
+**Mat Ryer:** Does concurrency help at all, or is that training process -- you're bound anyway by other things? Does it matter which order you do the training in, and things?
+
+**Daniel Whitenack:** Generally, if you think in terms of Map and Reduce operations, training is one giant, big Reduce operation where you really need to load a lot of data into memory... At least in batches. So there's batching that can be done. And you can separate out the work between workers using some schemes, but it doesn't always result in benefits. It also depends on communication and other things... So generally, training is not really the place. But again, with inference and prediction, if you're doing batch inference, like if you want to now tag 100 million images as either cats or not, then obviously there's gonna be a huge benefit to doing that operation in parallel in some way. That's my take anyway.
+
+**Miriah Peterson:** I'm not gonna comment too much. I think concurrency ends up being a bigger benefit in the actual package development, and I haven't done a whole lot of actual implementation on that side, so I can't really comment effectively.
+
+**Mat Ryer:** Fair enough.
+
+**Jaana Dogan:** Is there any other language features that makes Go a better language for this type of job?
+
+**Daniel Whitenack:** That's a great question. I find that Python, in the sense that -- like, people say it's very readable, but in the sort of Bill Kennedy way of describing what readability is, would be very not readable. When you're doing this sort of operation and something goes wrong, with Python the stack traces, like -- I've been running some experiments recently, and some of these errors in the stack traces I'm getting out are just so hard to deal with, because I have no idea what's going on under the hood, or at least where to trace things back to...
+
+So I feel like Go's readability - not in the sense of things being concise, but in the sense of things being clear and logical - I think that is an extreme benefit to this sort of process, where the things that you're wiring up, like these functions, could be very complicated, these models... There could be a lot of integrations that are important, like with CUDA, or other things... So I think that that clarity definitely helps.
+
+Obviously, the culture around testing and robustness and integrity is something that I think is a great benefit in Go, which is obviously the place where people hit blockers in terms of Python.
+
+On Practical AI we interviewed a couple of different people around this topic, but Joel Grus, who works at the Allen AI Institute - they have a package called AllenNLP, which is written in Python, and they really have to do a lot of work to enforce type labels, and utilize CI/CD to do that type-checking on all of their code to actually make sure that their code is robust and operating... So it takes a lot of extra work to build that stuff into Python, where for us it just comes for free.
+
+**Mat Ryer:** \[35:51\] Yeah, sometimes -- especially when people are new to Go from other dynamic languages, that's a bit frustrating that it's so strict... But of course, that pays dividends very quickly.
+
+**Daniel Whitenack:** Yeah. I just pulled up -- there's this other package called SpaCy, which is a Python package, and their underlying library that they use for machine learning is called Thinc. And they just made a big new release of it and they branded it as "A refreshing functional take on deep learning", and one of their first bullet points of why it's so awesome is type checking. "Develop faster and catch bugs sooner with sophisticated type checking." And from the Python perspective, it's like "Oh, we can do that!" People are starting to realize that we're moving past this phase of "Let's train models as quick as we can", and into a phase of "Let's build products and integrate AI into products." And when you start thinking like that, then you'll have to consider a lot more than just training things quickly.
+
+**Mat Ryer:** So is this technology -- obviously, it works at a big scale... I mean, whenever you see those little boxes that say "I'm not a robot" and you have to click, sometimes it then says "Okay, just to make sure you're not a robot, just tell me where all the street lamps are in this image, or where the cars are" or "Which of these images contain cars?" And this is basically us all training Google's AI brain, right? So at a big scale, when you've got lots of data and all these different users, obviously you can make some real use of this technology... But what about small companies? What about people that don't have much data? Should they still pay attention?
+
+**Miriah Peterson:** Always. I'm gonna throw my company under the bus a little bit... I've been fighting this a lot at work. I'm a huge AI \[unintelligible 00:37:42.07\] we fight to find a lot of problems where we can use it. We're not super-big; we've got about 13,000 customers, about 60 engineers, and we have one model in production and it does really great sentiment analysis on text replies. If you text your dentist and say "Yeah, \*thumbs up\* I'm gonna be at that appointment", we know it's a confirmation. It does it great.
+
+**Mat Ryer:** What if you say "Thumbs up\* I can't make it"?
+
+**Miriah Peterson:** Well, usually we get people cussing us out, and that's how we know we've texted somebody that's not a customer.
+
+**Mat Ryer:** Oh...
+
+**Miriah Peterson:** But see, it does that kind of a thing, and we did start at that very early; it was one of the first things we integrated, because it was a problem we noticed. We noticed that the text reply "C for confirm and infer no" was not a good user fit. The more dynamic relationship was a great user fit, and all we needed was our text. That's a really small dataset we were able to use. I think we used in open source one, and then we were able to actually create a game for our employees to add more data. So we didn't start with a whole bunch of data, but we gradually got more and it fit really well. And it was just a really small use case. So I think there's always a problem, and a small company can benefit from using machine learning to really just make their products better.
+
+**Mat Ryer:** Great. I mean, next time try Y and N for yes and no, instead of C and N. Just an idea.
+
+**Miriah Peterson:** C for Confirm.
+
+**Mat Ryer:** Confirm, or No...
+
+**Miriah Peterson:** N for No...
+
+**Mat Ryer:** \[laughs\] Yeah, but classically, Y and N, but yeah... I see what you mean. Still, that natural conversation is better anyway though.
+
+**Miriah Peterson:** Yeah.
+
+**Daniel Whitenack:** Yeah, and I was gonna say, too... A part of what I mentioned around the rapid growth in AI is due to pre-trained models and transfer learning. These are two ideas that really benefit people that don't have a lot of data. So pre-trained models basically mean someone else has already trained a model for recognizing cats, which they definitely have, many times, on the internet. So you actually don't even have to go through that training process, you can just pull down a model from the internet and run it in your own code, just to do the predictions.
+
+\[39:50\] There's also tons of APIs where you can integrate this sort of functionality that's AI-driven via API. So whether that's sentiment analysis, like Miriah is talking about, machine translation, object recognition - all of these sorts of functionalities are already available out there... And then transfer learning is basically the idea that someone's already trained a model to do almost what you want; you just have to tweak it a little bit.
+
+This is something that I use all the time, because I work on a lot of technology for lower resource languages, natural languages, like languages that people speak... And the languages that we work with, typically we might have 30,000 samples of parallel data between English and this language. If you look at large-scale machine translation models, they might have nine million examples. So we have almost no data.
+
+What we do is we really leverage things like "Oh, well there's a lot of data for English, right?" and someone's already trained a model for English. Well, if we wanna train a model for Singlish, which is a dialect of English spoken in Singapore, then we can start with the English model and then add in our small amount of data to fine-tune the model for the situation we want. So you no longer have to start from scratch, you're kind of standing on the shoulders of giants, in that sense.
+
+**Miriah Peterson:** I've run into this - just to build off of a lot of what Daniel said... I've had people ask me a lot of times where to get started with AI, and I usually say find an API that works. Google's got great APIs, Microsoft's got great tools... And while you're using that API, start growing your data, so that you can find what you need -- I mean, so that you have good data, and once you get enough, you can start doing transfer learning and really fine-tune your use case. But don't push the problem off. Start with an API and then get ready to make it better.
+
+**Break:** \[41:57\]
+
+**Jaana Dogan:** We talked a lot about fine-tuning and tweaking things... Is this like you are tweaking the model primarily, or would you rather try to have a layer on top of that model to maybe eliminate some cases, or gather some more information about the incoming data, maybe pass things to different models, and so on? I've seen different approaches when it comes to composing different solutions... What's your opinion on this?
+
+**Daniel Whitenack:** It really depends on if you want to do the same task, but for different data, or if you want to do a slightly different task. If you want to recognize -- I don't know, what's close to a cat? Miriah said lemurs are close; I'm not totally sure if that's true...
+
+**Miriah Peterson:** Squirrels, fuzzy-faced animals, some dogs...
+
+**Daniel Whitenack:** \[44:02\] Yeah, you could start -- like, if you wanted to adapt for raccoons, you're doing the same task, you're doing object recognition... So really, what you wanna do is tweak the parameters of the model; you don't necessarily want to change the model structure... Whereas a lot of times in natural language processing or in other areas there's open language models, which will tokenize your data and create a learned representation of language, but they won't complete the task that you want... Maybe that's sentiment analysis. In that case, you might have to add layers onto the model for the specific task.
+
+When people have heard of these things - maybe they've heard of BERT and ELMo, and Transformers, and GPT-2... These are all really large-scale language models; they really are meant to be fine-tuned to these different tasks, and meant to be generalized across them.
+
+**Mat Ryer:** A friend of mine, David Hernandez and I made a face recognition technology once... And the way we did it, essentially, was using transfer learning. We took a model that knew about loads of people, specific people, and then stripped away some of the layers so that it didn't quite know all the people, but it knew faces. And then just by giving it new examples - and not very many new examples - you could then specialize it just for your specific cases. So the net result was you could just train a small group of faces and it would be able to determine who those people were.
+
+So yeah, I think that approach, transfer learning, using other APIs, using other things - I agree with that. I think we should all be hacking a bit more on AI and seeing what it can do in our use cases... Because you might be really surprised; hopefully very pleasantly surprised.
+
+**Daniel Whitenack:** Yeah, one project worth mentioning that's very hackable and easy to get into is GoCV. GoCV has the ability to pull in TensorFlow models for things like object recognition, and do things like face recognition, or finding a certain person in an image, and those sorts of things... And really, you can do that sort of thing just by pulling in the models, and they have great examples of that. You really don't have to have anything special. They've done a great job of setting that up, if you're interested in something that's easy to onboard into.
+
+**Mat Ryer:** And Natalie Pistunovich did a good talk about using TensorFlow with Go, and essentially using the client libraries there in that way. That is a real option for people. You don't have to be down in the weeds of a machine learning model and all that complexity... To be making use of the technology.
+
+**Daniel Whitenack:** Yeah. People have a vision of an AI person as like a really professor-looking person at a chalkboard, scratching away math problems, or something...
+
+**Mat Ryer:** But you don't have a chalkboard, do you?
+
+**Daniel Whitenack:** I don't. I would prefer a chalkboard over a whiteboard, but that's a whole another subject...
+
+**Mat Ryer:** Yeah, that's for next week.
+
+**Daniel Whitenack:** Maybe I have to give some unpopular opinion in the show, right?
+
+**Mat Ryer:** Yeah, that could be it.
+
+**Daniel Whitenack:** I'm giving mine early - chalkboard over whiteboard. Anyway, I feel often as a practitioner more like a cook than a professor, where really I'm just taking someone else's recipe, I'm adding my ingredients in, which is my data, and then I get out something that I can use in my code. I pull a great model from Google, or OpenAI, or someone, I combine it with my code according to their recipe of how it should be formatted, and then I get out a model that works for me and I integrate it into code. It's much more like that than it is other crazy things that people might have in their mind.
+
+**Mat Ryer:** Yeah. I think that's a good lesson. I think it's always worth thinking like that, frankly... Because too often we want - for good reason - to be able to do all the bits ourselves, and we don't need to; and sometimes it's better not to.
+
+\[48:05\] Barnaby in the Slack channel was asking about object recognition. When we were talking earlier about detecting cats, and things... With computer vision, how is AI different to that?
+
+**Daniel Whitenack:** Yeah, I think with computer vision -- if you think about how this was used... Let's take an example of manufacturing and automation, which was something that people think a lot when they think about AI... So cameras have been used in manufacturing for a long time to detect edges, and insert things into slots, and do all sorts of things... But in the sort of computer vision way of thinking it's almost like -- one of our guests on Practical AI described it this way, "It's like you're finding your way around in the dark. You find this edge, and then you move two centimeters this way, and do this operation with your robot, and then move two centimeters down." It really has no idea what the thing is, it just knows there's an edge here, and "I'm moving this far."
+
+Whereas with object recognition or more modern methods, really you are saying "This is the type A slot where I put in part X. And this is the type B slot, where I put in part Y." So it's really recognizing things in the surroundings, and performing operations based on that.. Not just working off of shapes and edges and that sort of thing.
+
+**Mat Ryer:** Yeah. And I would also add to that - mash them up, if that helps. It helped us in the past. We used a combination of computer vision and machine learning, and the results sometimes were really staggering actually, so... Yeah, that's great.
+
+Well, I think, Daniel, you jumped the shark a little, but it's time for our regular slot... It's time for Unpopular Opinions.
+
+**Jingle:** \[50:00\] to \[50:18\]
+
+**Mat Ryer:** So, let's see... Daniel, are you gonna stick with yours, or have you got another one? Chalkboards over blackboards, you prefer...
+
+**Daniel Whitenack:** Well, on my chalkboard I would probably write something to the effect of "AI is the same as machine learning", which is my unpopular opinion. \[laughter\] Probably I just lost a lot of Practical AI subscribers and followers on all sorts of platforms, but that's my unpopular opinion.
+
+**Mat Ryer:** Fair enough. Do you agree with that one, Miriah?
+
+**Miriah Peterson:** That machine learning is AI?
+
+**Mat Ryer:** Yeah.
+
+**Miriah Peterson:** I use the words interchangeably all the time. That doesn't mean that they always have the same definition; that just means they're synonyms. Synonymous words don't have the same definition.
+
+**Mat Ryer:** Yeah, good point.
+
+**Daniel Whitenack:** \[laughs\] Burn.
+
+**Miriah Peterson:** \[laughs\]
+
+**Mat Ryer:** That's the best "Burn" though, ain't it? I like those complex "burns" that you have to look up later, to find out why you're insulted. Do you have an unpopular opinion of your own, Miriah?
+
+**Miriah Peterson:** I have an opinion, I don't know if it's popular or unpopular...
+
+**Mat Ryer:** Let's have it and we'll find out.
+
+**Miriah Peterson:** But I think it's generally speaking unknown... I do a lot of community work, and as part of that I go off to speak at bootcamps, and I tell them how amazing Go is, and how awesome meetups are, and that they should go to all of my meetups and my conferences... But my sales pitch for it is that, by going the meetups, they're contributing to open source the same as a GitHub PR. Your community is open source, just like GitHub... So I think that everybody that goes to a meetup is contributing to open source, and that's my unpopular opinion.
+
+**Mat Ryer:** Oh, that's a great one. That might not be that unpopular.
+
+**Miriah Peterson:** \[51:55\] I know, but it's not verbalized. It's not formalized, so it is unpopular if it's not formalized.
+
+**Mat Ryer:** Okay. That's the second one you got me on... \[laughter\] That's water-tight, legally, so I can't argue with that one, Miriah. Jaana, have you got an unpopular opinion these days?
+
+**Jaana Dogan:** On this topic?
+
+**Mat Ryer:** Anything.
+
+**Jaana Dogan:** I'm just too under-qualified for this show.
+
+**Mat Ryer:** Oh, well... No, not at all.
+
+**Miriah Peterson:** Me too, it's okay. \[laughs\] I just talk at people a lot, and everybody thinks I know what I'm doing. \[laughter\]
+
+**Mat Ryer:** You didn't mention your company earlier when you...
+
+**Miriah Peterson:** Weave?
+
+**Mat Ryer:** That's it, yeah. Well, you can't point to your hat, because it's a podcast. They can only hear us.
+
+**Miriah Peterson:** I'm gonna point to my hat every time I say Weave.
+
+**Mat Ryer:** You can point to it, but one saying is they won't know that you've done that. Also, that hat looks like it was knitted...
+
+**Miriah Peterson:** Oh, it was definitely knitted by a machine.
+
+**Mat Ryer:** Yeah. AI...
+
+**Miriah Peterson:** Definitely. Robotics.
+
+**Mat Ryer:** Yeah. But not woven; that was the thing. So that's disappointing, but... You know, no problem.
+
+**Daniel Whitenack:** What about you, Mat? What's your -- you've never had an unpopular opinion, have you?
+
+**Mat Ryer:** No, they're always popular. I can't help it.
+
+**Miriah Peterson:** Never mind, I'm not gonna be that mean today.
+
+**Mat Ryer:** No, you can do it. Go on, you can say it. It'll get cut out if it's mean to me.
+
+**Miriah Peterson:** I was just gonna say that you're English, so all of your opinions are unpopular in America.
+
+**Mat Ryer:** \[laughs\] Okay, yeah. That's definitely gonna get cut out. \[laughter\]
+
+**Miriah Peterson:** Okay, I told you it was kind of mean...
+
+**Mat Ryer:** Not really... Not really. I'm just catching up on Slack to see if we've got any other bits and pieces...
+
+**Daniel Whitenack:** So people were mentioning that the infra around machine learning and AI, written in Go... That's maybe a good topic to mention here at the end; there's a lot of great infrastructure written for machine learning and AI, and it's written in Go, because great infrastructure is written in Go. Things like Pachyderm, Seldon, Dgraph... I don't know, I'm probably missing some others. Those come to mind immediately. Obviously, a lot of Python people even use Docker, and other things...
+
+So I would say if you're wanting to run machine learning/AI pipelines at scale, Go is definitely your friend, even if you don't know you're using it.
+
+**Miriah Peterson:** Yes!
+
+**Mat Ryer:** Great.
+
+**Miriah Peterson:** Anybody that's using Kubernetes is using Go for machine learning, so...
+
+**Mat Ryer:** Well, we had the CTO of Cloudflare on, and he told us that essentially anybody using the internet is in some way using Go somewhere, so... It is kind of cool.
+
+**Daniel Whitenack:** You're probably using Python as well, right? If that's the logic...
+
+**Mat Ryer:** Yeah, maybe.
+
+**Jaana Dogan:** \[laughs\]
+
+**Mat Ryer:** Yeah. And definitely JavaScript.
+
+**Daniel Whitenack:** Oh, yeah.
+
+**Mat Ryer:** And PHP, Jon said... Okay, that is all the time we have for today, so thank you so much to our special guests, Daniel, Miriah and Jaana, and we'll see you next week on Go Time!
+
+**Break:** \[55:13\]
+
+**Mat Ryer:** ...keep it short and sweet. Then this is where the music plays... But this is now the after-party, so... Now you can--
+
+**Daniel Whitenack:** Nice.
+
+**Jaana Dogan:** But we're still live, um...
+
+**Mat Ryer:** Miriah, now you can say whatever you like.
+
+**Jaana Dogan:** We're still live.
+
+**Mat Ryer:** Yeah, that's true. We're still live.
+
+**Daniel Whitenack:** \[laughs\]
+
+**Miriah Peterson:** But I can say whatever I want?
+
+**Mat Ryer:** Now you don't have to hold back... I was joking, because you weren't holding back.
+
+**Miriah Peterson:** Well, then I should tell everybody to go to my conference.
+
+**Mat Ryer:** Yeah, actually you should, anyway. Which is it?
+
+**Miriah Peterson:** GoWest.
+
+**Mat Ryer:** GoWest. And which direction, how do you get there?
+
+**Miriah Peterson:** Well, generally speaking, if you fly West, since the world is circular, you'll get there eventually.
+
+**Mat Ryer:** So you might as well call it GoEast.
+
+**Miriah Peterson:** Well, if you're in California, East would be the quicker way to travel.
+
+**Mat Ryer:** Yeah, but I hope people won't. I hope they will stick to the guidelines/the code of conduct set by the conference - you have to go West to get there.
+
+**Miriah Peterson:** Go West, yeah.
+
+**Mat Ryer:** I think that's absolutely fine. And when is it, and where?
+
+**Miriah Peterson:** So May 8th is the conference day. We have workshops May 7th. The idea is we wanna do a regional conference that highlights a lot of the regional talent and still brings in a lot of bigger names to the area. There's a huge Go community here in Utah, I know there's one in Phoenix, Arizona, and in Denver... So we're trying to just take all of those communities and give them a huge, awesome, amazing conference to celebrate it.
+
+**Mat Ryer:** Great. And is the conference in Utah?
+
+**Miriah Peterson:** It is. I live there, in Sandy, Utah. It's a nice 20-minute drive South of Salt Lake. A quick little stint over from the airport.
+
+**Mat Ryer:** Yeah. And it's a lovely state, Utah, isn't it?
+
+**Miriah Peterson:** Yeah. And in May it'll be just gorgeous. It's green, you'll still have snow in the mountains, but it's warm.
+
+**Mat Ryer:** Perfect.
+
+**Jaana Dogan:** What is the website?
+
+**Miriah Peterson:** GoWestConf.com.
+
+**Jaana Dogan:** Okay.
+
+**Mat Ryer:** Great. And if the editors are listening - which they're not - they'll put that into the show.
+
+**Miriah Peterson:** You know, I would be so happy... \[laughter\] I would jump. \[playing Van Halen - Jump\]

Organizing for the community_transcript.txt

@@ -0,0 +1,619 @@
+**Mat Ryer:** Hello, and welcome to Go Time. I'm Mat Ryer. Today we're talking about organizing in the community. You know all those meetups and conferences you go to, and there's food there, and great content, and everything's just worked out.. Well, it turns out that that's quite difficult to achieve, and we're gonna dig into that a little bit today, and learn a bit more about that, I think.
+
+Joining me on today's show, we have Ronna Steinberg and Natalie Pistunovich. Hello!
+
+**Natalie Pistunovich:** Hi!
+
+**Mat Ryer:** Welcome to the show.
+
+**Ronna Steinberg:** Hi, Mat.
+
+**Natalie Pistunovich:** Thanks for having us.
+
+**Mat Ryer:** We also have Johnny Boursiquot. It's only Johnny Boursiquot. Hello, Johnny.
+
+**Johnny Boursiquot:** Hello, Mat.
+
+**Mat Ryer:** How's it going?
+
+**Johnny Boursiquot:** It goes. You know, I've been indoors for a little while, so it's good to have some human interaction that basically aren't my family. So hello, everybody. Happy to see you! \[laughs\]
+
+**Mat Ryer:** No pressure... There's loads of pressure now. Great! We're Johnny's friends for the week. We're the ones he's gonna get to hang out with. So I actually asked our guests for some mini-bios, and they're not so mini. I didn't realize how much you do for the community, I must admit.
+
+Natalie, three meetups and three conferences you organize. That's phenomenal. The one that I know about, the GopherCon Europe, is a phenomenal conference. It's excellent.
+
+**Natalie Pistunovich:** Thank you.
+
+**Mat Ryer:** Yeah, thank you so much for that. And Ronna, organizer of the Women Who Go Berlin and the diversity scholarships for GoBridge.
+
+**Ronna Steinberg:** Correct. I think I'm dragged into every adventure that Natalie sends me on \[unintelligible 00:03:14.18\]
+
+**Mat Ryer:** Are you happy about that, Ronna? Is that a problem, or is that okay?
+
+**Ronna Steinberg:** No, I think it's great. It's working so far, so no complaints here
+
+**Natalie Pistunovich:** The thing Mat missed from my bio is that I'm a developer advocate at Aerospike, but more importantly, I'm a big fan of Ronna. So it's great to have this adventure--
+
+**Ronna Steinberg:** Is that on your bio?
+
+**Natalie Pistunovich:** This is in my bio... For this show. \[laughter\]
+
+**Ronna Steinberg:** I missed that part. I didn't read that part, I'm so sorry. I think it's a mutual thing; I think a lot of people know about this, so... \[laughter\]
+
+**Mat Ryer:** Well, that's lovely. Actually, so that our listeners get to know you, for those that haven't met you, maybe we could start off with a working-from-home tip from each of you. That would be a great new regular slot of the show... Working from home tips. Ronna, do you wanna go first?
+
+**Ronna Steinberg:** \[04:15\] I think Natalie should actually go first. I mean, she does have more experience with this. I have more experience I think on managing this part.
+
+**Mat Ryer:** Okay. Natalie, would you like to go first?
+
+**Natalie Pistunovich:** Absolutely. I've been working more remote than not for a while now, and the thing that helps me more than anything to start the day is if I'm going somewhere. So I have a good breakfast, I enjoy my coffee, brush my teeth, wash my face, and change into the loungewear, which is also comfy clothes, but it's not the pajamas.
+
+**Mat Ryer:** I see. So you make it sort of that thing of making a distinction about when you're in work mode versus in home mode.
+
+**Natalie Pistunovich:** Yes.
+
+**Mat Ryer:** Yeah, that's a good one.
+
+**Ronna Steinberg:** Okay. So for me, it's more about the quality of delivering pretty much anything. A lot of us are very used to the agile way of doing things, delivering things very rapidly... I think everybody -- because communication is harder right now, it creates a massive amount of ping-pong. So whether it's defining a task, whether it's actually delivering a bug fix - whatever it is, it needs to be in very high quality.
+
+The same goes for code reviews. If you write a message to somebody \[unintelligible 00:05:35.25\] try to be mindful of other people in this process...Because it's addictive, right? Delivering things very quickly is addictive; right now I think everybody needs to slow down a little bit.
+
+**Mat Ryer:** That sounds great, and I think you're right about this idea of asynchronous communication. When you're in an office with somebody, the conversations can be very different. They could be very short little sentences, lots of nice pleasantries amongst it... But yes, if it's asynchronous, it is worth spending a bit more time on that clarity, and things... Especially if we're gonna have different schedules as well. I think that's a great one, Ronna.
+
+**Ronna Steinberg:** Thank you.
+
+**Mat Ryer:** So how did you get into organizing then? Natalie, what did your involvement in this kind of come from?
+
+**Natalie Pistunovich:** Yeah, that's a good question. I've been trying to think about that, and probably it goes back to since forever. I've been on the students' union in university, and throughout life, and then when I moved to Berlin, I went to some conference that I found, which was the Berlin Dev Fest, organized by the local Google Developer Group communities... And I just spoke to somebody randomly, and he said "Oh, you're doing Go? You should help me organize the Go meetup." And that was \[unintelligible 00:06:56.21\] who was organizing the Go meetup for the first few years in Berlin.
+
+Then I came to the very first meetup, and he said "Oh, by the way, I'm moving to London. Good luck." And that's how I became the organizer of the Go User Group. \[laughter\] And then from there it came to other meetups, and the idea for GopherCon started with really admiring GopherCon in the U.S, but not having a European version of it, and just saying "Alright, let's do this."
+
+**Mat Ryer:** Yeah. So when you moved to a new city then - is this quite a good way to meet new people as well, and to kind of jump into existing communities?
+
+**Natalie Pistunovich:** Definitely.
+
+**Mat Ryer:** Yeah. And GopherCon EU, for those that don't know, actually moves around, doesn't it?
+
+**Natalie Pistunovich:** GopherCon Europe does move around... Not only in EU countries, but in all European countries.
+
+**Mat Ryer:** Yeah.
+
+**Natalie Pistunovich:** We love you, U.K. people. \[laughter\] Even though there is a GopherCon U.K.
+
+**Mat Ryer:** Right. Well, there was one in Tenerife, which was -- nobody believed me when I said I have to go there for a conference, by the way... \[laughter\] No one believes that.
+
+**Natalie Pistunovich:** There was one in Iceland, in Reykjavik.
+
+**Mat Ryer:** \[08:09\] Yeah, the Iceland one was also excellent. And this year, what's happening with--
+
+**Natalie Pistunovich:** Berlin. Well, this year, if you'd asked me a few months ago, I would have told you Berlin, and this is where it would end, but now there is a pandemic so... Hopefully Berlin. We'll see how things develop.
+
+**Mat Ryer:** Yeah. And what about you, Ronna, then? How did you get into organizing as part of the community like this?
+
+**Ronna Steinberg:** Well, it's actually a strange story, really. I lived in Berlin, I was a gopher in Berlin for two years, and then I moved to the Netherlands. And accidentally, I signed up for a meetup in Berlin, for a Go meetup in Berlin... And I canceled probably two minutes later, but somebody was stalking the list of people... \[laughs\] And that somebody was not Natalie, it was Vanessa Ortiz, who was the organizer of Women Who Go Berlin back then. She was looking for women who were doing Go to come and mentor.
+
+She immediately checked and found my LinkedIn, and like "Oh, you know Go... You've been doing it for a while... Can you come help?" But I wasn't even in the city. Then I decided to move back to Berlin, and I was basically handed down the chapter, an already made chapter. It took about two minutes for me to decide when she asked me if I wanted to take over, to decide to do it... And I really thought it was a bad idea, but I did. I'm very happy that I did it. And yeah, the rest is history... As well as my friendship with Natalie.
+
+By the way, Natalie gave a talk about new people and about organizing meetups and about her journey at GopherCon in Denver. I advise anyone who is interested in insights to watch it. It's very good.
+
+**Mat Ryer:** Okay, cool. That's probably available on YouTube now, isn't it?
+
+**Ronna Steinberg:** yes.
+
+**Mat Ryer:** So Natalie, maybe you could give us a clue - how could people find that? What was that called?
+
+**Natalie Pistunovich:** Closing Keynote 2018.
+
+**Mat Ryer:** Cool. That'll do. Just trying to give everyone the search terms... \[laughs\]
+
+**Ronna Steinberg:** I think it was something -- The Value of New, or...?
+
+**Natalie Pistunovich:** The importance of beginners.
+
+**Ronna Steinberg:** The importance of beginners, okay.
+
+**Mat Ryer:** Yes, I remember. Very good. So yeah, we do recommend that. We'll put that in our show notes as well.
+
+So some of the conferences are kind of commercial, others are more sort of just community, and they balance the costs and things... What is it that drives you in particular to keep going with it? Because I've spoken at some conferences, and done a few of the little bits and pieces, but not much... And even just that small exposure, I realize how difficult it is. So why do you keep doing it? What is in it for you?
+
+**Natalie Pistunovich:** The chance to do some anarchy. \[laughter\] So I never had a conference shirt in the size XS until it was my conference.
+
+**Mat Ryer:** Really?
+
+**Natalie Pistunovich:** And then it was ranging from XS to 5XL. So I like onesies, for babies, which is really cute. But also, other types of things, like -- having spent some time living in Kenya and talking to people and learning how hard and disconnected the regular everyday in certain areas from what I'm used to from Europe got me really realizing how hard it is to be part of this if you're not living in a place like the U.S. or Europe.
+
+\[12:09\] And just doing a call for paper usually will not get you to reach such people... So I get a chance to use the network that I built there to every year to fly in speakers, or attendees. From Africa, from Latin America, from Asia... It is nice to be able to do this.
+
+**Mat Ryer:** Yeah, because otherwise there probably isn't a conference or even a meetup near them, is there?
+
+**Natalie Pistunovich:** There is.
+
+**Mat Ryer:** They do have meetups and conferences?
+
+**Natalie Pistunovich:** There are meetups. There's a meetup in Nairobi. This year GopherCon Africa was supposed to kick off in Nigeria. See how that goes as well, with all the global plans changing... But yeah, you see how big the disconnection is... And being able to help a little bit to bridge that gap is a huge reason for me to do this.
+
+**Mat Ryer:** Yeah, that's great. That is great.
+
+**Ronna Steinberg:** There is a very exciting community right now in Nairobi, and I suspect Lagos as well, but... An incredible, incredible engagement over there with Go. How many people do they have right now, Natalie? I think you probably know.
+
+**Natalie Pistunovich:** How many people sign up to the meetup in Nairobi?
+
+**Ronna Steinberg:** I think so...
+
+**Natalie Pistunovich:** Lower hundreds...?
+
+**Ronna Steinberg:** No, I don't think so. I think it's higher. I'll find it. You do you, and I'll find this information.
+
+**Mat Ryer:** I did a remote meetup in Lagos. They do have a -- it's a great community there.
+
+**Natalie Pistunovich:** In Nigeria for sure it's bigger.
+
+**Mat Ryer:** Yeah. And if there's anyone out there that wants to speak at a conference or a meetup and it's their first time, doing a remote one actually has quite a lot of benefits. You can kind of be in a space that you feel safe in, and everything's more familiar and comfortable; you can still deliver a good talk, and hopefully as long as all the internets work as they're supposed to, it was a great experience. So I do recommend that.
+
+But you can't really beat actually going and physically being a place, and that's the advantage for conferences. That's the advantage to fly them over.
+
+**Natalie Pistunovich:** Yeah, yeah.
+
+**Mat Ryer:** Did you find that thing, Ronna?
+
+**Ronna Steinberg:** Yeah, I have an update. In the Nairobi Gophers group there are 870 - this a round number - members. Woo-hoo to them!
+
+**Mat Ryer:** Yeah, 870...
+
+**Ronna Steinberg:** It's impressive. Very impressive.
+
+**Natalie Pistunovich:** Yeah.
+
+**Mat Ryer:** And there will also be lots of people hopefully listening where they sort of -- they don't feel like there is representation, or maybe there isn't a meetup in their area... What can they do? I mean, it doesn't sound easy to just start a meetup, is it?
+
+**Natalie Pistunovich:** It depends a lot on all kinds of things. If you have a place you can go to and ask "Please give me space", then it's definitely easier to roll, because you can host a remote talk, like you said; you can find maybe a speaker from the community, or give a talk yourself, or just host a fish bowl session, a QA, watch some other talk together.
+
+But finding a space is not always obvious. If your company is able to host this, this is great. If you know some co-working space that will give this, this is also great. But this is never taken for granted.
+
+**Ronna Steinberg:** Well, I guess it really depends on the city, right? If there are other active communities in your area, you can approach them and see how they're doing it. That's what I did. It was very helpful for me. I did not pretend at all that I could do it on my own from the start.
+
+**Johnny Boursiquot:** Before we move on, I am curious as to what some of the challenges beyond venue - and I'm not sure if food tends to be a part of the meetups, when they're organized in those environments, as they tend to be... You know, with U.S. or Europe-based meetups. So beyond venue and food, which are common, I would imagine, across all meetup organization - but beyond these, what is the most challenging aspects for, say, a meetup in Lagos or a meetup in Nairobi? Beyond that, what are some of the other concerns that you have to contend with, that you necessarily don't have to in the States, or in the U.K, or something like that?
+
+**Natalie Pistunovich:** \[16:26\] I think that's a great question. From the little bit that I got to see, Go is not yet as popular, so you can find many people individually working on this... But the network is still being built. For comparison, the Berlin User Group I think has almost 1,000 people, and many of them work in companies that are using Go.
+
+And even in the Berlin meetup it's sometimes hard to find people to give a talk, because they think their project is boring, because they think "Oh, it's just what I'm doing at work." So if this is a language you're learning for fun, and you're coming to listen and not necessarily to give a  talk, then having enough speakers is harder. But this is harder with any community that is beginning, and I think that Go is just not yet as popular there.
+
+Another thing is what Mat mentioned, doing the remote talks. This is great, that the internet worked all the way through. This is also something that's not obvious, and you probably need to pre-record this as a back-up, or just count on everybody's patience if there's some kind of a hiccup with the internet.
+
+**Mat Ryer:** Yeah. The London Gophers, actually - they put all of their talks online as well. They're just on YouTube. That's nice for anyone obviously that can't make it, but it is also nice to get a sense of what these meetups are like... But you don't really see some of the main advantages, which you get from actually physically meeting in person... Which we perhaps shouldn't dwell on at the moment, but eventually we're gonna be back to normal, we're gonna be back at meetups, and stuff.
+
+So what are some of the advantages for people attending conferences and meetups, do you think?
+
+**Ronna Steinberg:** The way we do things at Women Who Go Berlin is actually -- what's happening right now globally is a drawback for us, because the way that we do this... Because when I started out there were very few women that we could sign that were professional in Go, and this was a Women Who Go chapter... And I wanted to train people as fast as I could, so I came up with a format where I go to companies that want to host Women Who Go Berlin, and we would come up with a challenge that has something to do with what they do.
+
+Then our members would do a real-life, industry-level challenge; something that takes 2-3 hours. By the way, you mentioned at the beginning the waveform going up when you clicked your fingers... We drew a waveform at a Soundcloud meetup; that was pretty cool.
+
+**Mat Ryer:** Wow.
+
+**Ronna Steinberg:** Yeah. There is a bunch of them. We do them once a month. Of course, right now it's a bit of a problem... What it does - it actually leverages the fact that everybody is in the same room and then I can group them up to make sure that people will finish the challenge, which is ideally what should happen. We do cool things, and it's a lot of fun. So yeah, this is a very big setback for us, so we're trying to figure out what to do right now.
+
+**Mat Ryer:** Yeah. I mean, moving that all online is for sure gonna be a challenge. So I didn't realize, that's not just a meetup that you're describing, Ronna; that is where you actually get together and work, and build something.
+
+**Ronna Steinberg:** It's workshops, yeah.
+
+**Mat Ryer:** That's really cool, and that's a whole different ball game as well. So much to actually get your hands on and get stuck into. I find hands-on experience is a great way to learn for a lot of people, especially me... So yeah, learning by actually building is a nice thing that others ought to probably think about as well.
+
+**Ronna Steinberg:** \[20:12\] Yeah, it comes from my own experience. I mean, if there's one thing that I learned in my career, it's that I learned best when I was paid to learn something... \[laughs\] And I was actually doing it on the job. So I'm giving people the opportunity to learn on the job. They have access to mentors from the hosting company, the hosting company gets to maybe try and recruit people off of that, so they can you know.. a bit of a market going on, I'm not going to deny it. Hopefully, we'll get a lot of people recruited, and a lot of people hired, and we'll increase the community.
+
+**Mat Ryer:** Great.
+
+**Natalie Pistunovich:** Yeah, I think you touched a really important point, which is exactly this place to seek for opportunities... Because yes, it's possible to chat with everybody in an online event, but it's not the same as talking to people from the hosting company, or talking to people from other companies who are looking to hire... Many people find their next jobs in such events, or in meetups or in conferences.
+
+When your pool of opportunities of companies that you can go to is smaller, because there's not so many big companies, or there's not so many companies that are using Go... To answer your question, Johnny - I think this is another big difference between meetups that I noticed in Europe versus Africa, just from being present in those two mainly.
+
+**Johnny Boursiquot:** I'm curious if you've encountered perhaps any cultural barriers as well in those environments... Or is the age group young enough and progressive enough to not be hindered by certain things that hold folks back? And I'm gonna be blunt here and basically say that in certain parts of the world, Africa included, which is kind of what we've been talking about here, there are men who cannot stand having a woman go up on stage and try to teach them something. That's a very real thing.
+
+In some parts of the world we might think that's backwards, but there are parts of the world that are still very anti-women being in any sort of leadership (or otherwise) positions. Even if it's for a 25-minute talk. That is something that can be perceived as challenging authority in some ways.
+
+There are these kinds of things that still exist, that I'm wondering if they play a role, at least from what you've seen... If any of these things are playing a role in maybe the frequency or how well those events go, how welcome and inclusive they end up being... Do women end up coming back to these events? Do they feel safe enough to be part of these communities? As we all know, there's toxic environments everywhere, so what kind of challenges are they facing in those environments, that might be different from ours.
+
+**Natalie Pistunovich:** It's a really great question, and my answer is probably really biased, because I've only participated in the events that I was invited to, obviously... So I did not get a change even to be part of such events. But probably the most -- so Kenya felt to me very open-minded in that sense, and probably the place that had the most people with traditional outfits, if that's any indicator, was actually Mauritius, an island in Africa.
+
+\[23:49\] Even there, men that had a very traditional outfit, with this costume, and the head cover still were there, and listening to my talk, asking questions... But I'm sure that this is because they read the agenda and they saw that there's gonna be a woman speaker, and if they would not want that, they would not show up there in the first place.
+
+So definitely that exists, definitely that's a problem. This is not something that I personally experienced, because usually it's announced who the speakers are. But it's really important that you're raising awareness to this, so thank you very much.
+
+**Mat Ryer:** Yeah, I was gonna say - there's also places that aren't hospitable to women... Hacker News is also there. \[laughter\]
+
+**Natalie Pistunovich:** Well, you can hide better there. \[laughter\]
+
+**Johnny Boursiquot:** Yeah, you can lurk. You can have a male(ish) codename/username or something.
+
+**Natalie Pistunovich:** Yeah.
+
+**Mat Ryer:** Actually, I spoke at a conference in Lviv in Ukraine, and I don't know really, but just superficially, the mix was about 50/50 men and women, which did look strange at a tech conference, I must admit. And I actually made the point that, in a way, I reckon if you can get over that - because we do have that problem in tech; we do have this problem, and if we can get away from it, we're gonna just amplify all our potential. That's the thing. I feel like we hold ourselves back by these kind of old-fashioned attitudes. And I made the point to those young - they were quite young - students that if you can solve this problem (or maybe it's just not a problem at all), it gives you an advantage in the world, which I do believe.
+
+**Ronna Steinberg:** I want to say something about a different perspective. I think there's a lot of misconception around this topic. There is a big misconception around this. There is something to be said for hard work, and this idea that hard work is going to get you anywhere in life. The people who believe in these things, that if you tried hard enough, you will get there, and therefore this just-world bias (that's what it's called), and therefore if a certain population is not getting there, they're just not willing to put in the work etc. There is a lot of misconception around that, because it's a beautiful idea, and the people who believe that are usually very good people.
+
+The reality is that hard work is not going to get you everywhere. For some people maybe it will, maybe it won't... And we're seeing right now how much security there is in that, right? So I think we probably shouldn't mark certain people with certain ideas as the enemy, or target them, because they're not bad people; they just don't necessarily understand what is happening... Especially in this industry, because it's very easy to overlook certain problems that we have. It's a big manifesto, I know.
+
+**Mat Ryer:** Yeah, it is.
+
+**Johnny Boursiquot:** For me personally -- personally, I think I've been fortunate enough to have been part of the industry for a long time, and I've seen that thing you're talking about, where folks are like "Well, you got to where you are because you worked hard." Well, yes, but I've also seen people who work just as hard or harder than me, and because maybe they didn't know the right people at the right time, or they didn't have access to a certain person or a certain company, or whatever it is...
+
+This thing about the work alone gets you to where you need to be - I think deep down everybody knows that's not true. It's never been -- that's table stakes, everybody has to work hard... And those who are privileged enough that don't have to work as hard, and they just have the connections, basically can just skip a few steps... And again, more power to them, whatever that ends up being for them.
+
+\[27:52\] But I think if everybody is working hard -- what we're trying to do... I'm gonna speak for Natalie and for Ronna for a little bit here... I think what we're all trying to do, because I think we're all part of this organization of community events group, I think what we try to do is basically create opportunities for people to learn and engage. We're broadening the surface area for connections to happen, for learning to happen. We're creating opportunities.
+
+We can't open some doors for people, because -- I don't know about you, but I don't have that much pull. I can't place somebody at a company and say "Hey. Here, have a job." I can't give you a job; I can create opportunities for you to come, and to meet, and to learn, and to participate, and to engage, and to give some of yourself and some of your time... And the more you give, the more you receive.
+
+So I think we all create opportunities for these things to happen. We broaden the surface area. That's something that just wasn't being done at all just ten years ago. Events weren't focused on creating opportunities for people; they were focused on show-casing certain individuals' ability to talk, and to sell, and to pitch. If you've noticed, meetups - we didn't call them meetups back then, but... Meetings of this type were -- I remember when I first started going to meetings; they were so high-level... They were not geared towards beginners. They were not geared towards being inclusive... And I'm not just talking about gender, or race, or anything like that; it's almost like the guild of wizards... You show up and they're talking about these very high-level concepts.
+
+I have nothing against that, but the focus wasn't on building community per se; it was like "Hey, we all know magical incantations for things. Let's just get together and talk." It's kind of like an old boys' club, where you get your scotch and your cigar and you're kind of complimenting each other, patting each other on the back kind of thing. That was really what it felt like. The focus wasn't on creating opportunities for learning. I know I'm kind of generalizing a little bit here, and please forgive me for that, but... That's just been my personal experience.
+
+So to some degree I've seen that in our tech careers, and have decided rather than complain about it, to do something about it. I think that's why in part we're all in this, and doing this... But I'm 100% behind what you said, Ronna. It's not just about working hard. That's a given. In my mind that's a given. It's about basically finding the opportunities, and in some cases helping to create them yourself, and hopefully that comes back around and it works out for you as well... But I think the more options you try to create for others, the more you're gonna find yourself. And it's not just hard work. Never been.
+
+**Ronna Steinberg:** I'd like to add something to that, if that's okay...
+
+**Johnny Boursiquot:** Please.
+
+**Ronna Steinberg:** ...because Natalie and I, we talk about this a lot. So to anyone who might be listening to this, we've put a lot of effort in this scholarship, we've put a lot of effort into bringing people from really all over the world to certain conferences where they can maybe find opportunities, which is really what we want to increase.
+
+But then, there is always this concern that people who do come, are not necessarily right now - and that's also okay - at a place where they can lever it, where they can just make up their minds, jump on an opportunity, and "Wherever it takes me, I'm just gonna go." And it's okay, but, you know, if I could speak to anyone who applies to the scholarship, what I would like to tell them is just be aware that the people that you're going to meet, at least some of them can actually -- it may not be you, Johnny, or me particularly, but there could be somebody out there that is right now recruiting... Like, create the space, and when opportunity knocks and you're ready, you might actually take it.
+
+I think it's very important, because we do put a lot of work into this, and we're always worried that -- you know, on the one hand we don't wanna pressure anybody. But on the other hand, we don't want to see opportunities go to waste.
+
+**Break:** \[32:12\]
+
+**Mat Ryer:** Could you tell us a bit more about the scholarship then, please?
+
+**Ronna Steinberg:** There's a lot to say about this scholarship... \[laughter\] I wrote a document once for GoBridge on how...
+
+**Mat Ryer:** \[unintelligible 00:34:04.28\]
+
+**Ronna Steinberg:** \[laughs\] ...on how I do this. I started with having no idea what I was doing. Natalie basically pinged me one day -- it's always on Facebook Messenger; don't ask me why, but all opportunities are always showing up on Facebook Messenger... The place where you least expect it to, and you don't take seriously, ever.
+
+**Mat Ryer:** Yeah... I haven't had any opportunities on Facebook Messenger.
+
+**Ronna Steinberg:** \[laughs\]
+
+**Mat Ryer:** I tell a lie, I did have the opportunity to have some hair with one of the filters, so I enjoyed that, but... Not quite what you're talking about, is it?
+
+**Ronna Steinberg:** Important too though..
+
+**Natalie Pistunovich:** Well, for context, for a long time on the Facebook Messenger is where Ronna and me would chat about life. We just used that as our regular communication channel.
+
+**Ronna Steinberg:** It's embarrassing, really. \[laughs\]
+
+**Natalie Pistunovich:** We have since then upgraded and we have a part of that on WhatsApp. \[laughter\]
+
+**Ronna Steinberg:** And Slack, and then there's all the gophers Slacks... All of them. So there's the Gophers, the Women Who Go one, there's the Women Who Go Berlin one... So all of them.
+
+**Natalie Pistunovich:** The GDE one...
+
+**Ronna Steinberg:** I can't find the messages. So whenever I want to search for something, Natalie sends me \[unintelligible 00:35:24.00\]
+
+**Mat Ryer:** Yeah, I know. That's pain.
+
+**Ronna Steinberg:** But yeah, so she asked me to do this and I didn't have any idea how something like that could be done. And I immediately then said yes, obviously. Because, you know... \[laughter\] That's just how I do things. So then I found out, "Okay, I do need to create a form for applications."
+
+Then it turned out that it's quite a headache, trying to figure out what to say in that form. Because if people are applying -- for instance, how do you ask somebody what under-represented group they belong to? Because when we asked them what under-represented groups they identify with, people who are not necessarily a native English speaker just said "All of them. I identify with every single fight, from every under-represented group out there." So I think there's a lot of that going on...
+
+\[36:32\] And then how to get people to trust you over a form, with personal information that is quite sensitive, and... A lot of things that happen in there. Then there is forming a committee to figure out who's going to go. It is very hard to tell people that they are going to measure other people, and it's a very difficult task, I think. I'm glad actually that I don't rank people, and I don't decide who goes, because that makes my life a bit easier. And then take all of that data...
+
+And then there is the logistic of hotel, flights and all of that stuff. Some of these people are flying for the very first time in their lives, and they're going to be doing it on their own.
+
+**Mat Ryer:** Wow... Wow. That changes a lot, doesn't it? That changes a lot, if you think about that.
+
+**Ronna Steinberg:** It is scary for me. Once in a while I find myself like "Okay, do they know what to do if they miss a connection? Do I know what to do if I miss a connection?" \[laughs\] I find myself struggling with these questions... And there's a massive leap of faith happening on the week of, and then hopefully the magic happens and it works out. It's not easy.
+
+**Mat Ryer:** So you must be able to build a lot of trust though, because people wouldn't do that if you hadn't already done that, right? If you hadn't been able to build that foundation of trust, probably you wouldn't get people being able to come.
+
+**Ronna Steinberg:** I think people want it enough, so that they can take... And yeah, we email a lot; I get to know pretty much everyone pretty well, who is going \[unintelligible 00:38:15.10\] It's not easy. For instance something that has occurred to me when we were in the U.S. - and this was actually not a GoBridge scholarship, I think... There was a woman - she did not have travel insurance, and we're in the U.S, so what happens if tomorrow she needs an X-ray.
+
+So I think these scholarships really need to take things very seriously. We are fortunate to be doing this in Europe; if we're talking about GopherCon Europe, it's easier, because things are not as expensive. We can definitely figure it out. But otherwise, I would probably have decided to fund everyone's travel insurance.
+
+**Natalie Pistunovich:** What I would like to add is some things that -- probably one of the biggest challenges is actually to find people to fill out the scholarship.
+
+**Mat Ryer:** Really?
+
+**Natalie Pistunovich:** Because you made a form, you're tweeting this on social media, and you're saying "Hey friends, please retweet this." You're tweeting this from the conference social media, you're tweeting this from GoBridge. And then you have the same reach, of the same people who know and are familiar with those conferences. But how do you reach those people in the different meetups we talked about, that are not part of this smaller community of Go developers in Europe, Go developers in the U.S.? Starting to reach those for me feels like half of the work.
+
+What I did in the first year was go to Meetup.com, go to WomenTechmakers.com - all those platforms list different types of meetups - and just start searching for keywords. I personally texted on Twitter and on Meetup and on Facebook to those different groups, "Hi, Women Who Tech in Lesotho. Please share this with your attendees." Being able to reach people outside of your immediate or even first or second degree of connections is incredibly hard.
+
+**Mat Ryer:** \[40:23\] So you find existing communities and go there. That's one way.
+
+**Natalie Pistunovich:** Yes, this is one thing you do. Another thing is that in those events, in those meetups, whenever I get a chance to give a workshop or a talk somewhere that is not Europe and not the U.S, I'm making sure to stay in touch with as many people as possible, so that later I'll be able to tell them "Hey, this is the time for the scholarship. Please share this with anybody you think can be relevant."
+
+This part, of finding enough people who are actually outside of your reach, to fill out this form, is a lot of work.
+
+**Mat Ryer:** Well, I can imagine actually all of it sounds like a lot of work.
+
+**Natalie Pistunovich:** Yes... \[laughs\]
+
+**Mat Ryer:** So if we have anyone listening, who is this scholarship open for? Where can they get more information about it? Because you never know - there might be someone listening who just don't know about this.
+
+**Natalie Pistunovich:** We share this on all the Twitters we can - the one of the conference, the one of GoBridge, on our personal ones...
+
+**Johnny Boursiquot:** The one thing I'll add to that is reaching people outside of your bubble or your sphere of influence, whether it be second, third degree - if you can manage somehow to do that, I think one of the things that I personally try to do is to actually meet people where they are. Not just online, but actually physically. The reason why I enjoy being close to Baltimore is that there's a thirst there for technology and for learning, and a lot of people there look like me. So when I organize a workshop, I do my darndest with the help of other community members, which I rely on quite a bit to help get the work out there and to bring more of those folks into the workshop.
+
+You kind of have to be willing to go into those communities, because those people are not gonna come to you per se. First of all, they may not even have heard of your event or your conference or your meetup or your workshop, whatever it is.
+
+**Natalie Pistunovich:** Exactly.
+
+**Johnny Boursiquot:** So the job of reaching them is one. Second - you have to go to them, because they've been overlooked for decades; your one little event is not gonna do much to all of a sudden make them feel comfortable about coming out and being exposed and vulnerable out in this community that you and I may find to be quite navigable, because we've been in it for a long time. But for them, it's extremely intimidating.
+
+And here's the other thing, too - the importance of representation, the importance of somebody seeing Natalie giving a closing keynote at a GopherCon...
+
+**Ronna Steinberg:** Or you.
+
+**Natalie Pistunovich:** Or you, yeah.
+
+**Johnny Boursiquot:** Indeed... For those who don't know, I did do it last year. But yeah, to me that closing keynote talk you gave was inspiring, because we know in the Go community specifically that a lot of the people that are coming into the community are brand new. There are now more beginners in the Go community than there are experienced folks. That's the truth. And that number is only gonna continue to grow. So we have to somehow make it okay -- we have to say "Hey, this is a community of learners, of beginners."
+
+Maybe you're learning Go for the first time, maybe you're coming from a different language, or a different community; it doesn't matter. In some sense, we are all beginners. Because even the experienced folks within the community, we are learning from the beginners as well. It's not like we reached a pinnacle and we stopped learning. We're still learning, too. We're all learning together.
+
+\[43:59\] So the importance of actually making it okay to say "Hey, you know what - we know it's intimidating from the outside, but let us help you be part of this community", I think it's huge. So yeah, kudos to you, Natalie; that was a very enjoyable talk. It kind of gave me courage to do mine the following year. But yeah, I think these things are important.
+
+**Natalie Pistunovich:** Well, thank you for the great talk.
+
+**Mat Ryer:** Yeah, they're both good. \[laughter\]
+
+**Johnny Boursiquot:** Thanks, Mat.
+
+**Natalie Pistunovich:** They're alright...
+
+**Mat Ryer:** Johnny had a quote which was great, and it said "Community gives back." It's this idea that don't just expect the community to just give stuff to you; it gives back. And the best way to be part of that community is to sort of get stuck in. I think that point resonated quite well. It definitely makes sense.
+
+When you're organizing something, it's all about that organization and all the effort that goes into making the plan, or is there an element of improvisation that has to be there, because when it comes to it, you don't know what's gonna happen?
+
+**Natalie Pistunovich:** \[laughs\] Of course, there's a huge place for that. There's a saying that what's not flexible breaks, and you can make all the plans you want, but if you insist on 100% sticking to them and then something unexpected or different from the plan happens, if you are not open to improvise, this will not work.
+
+**Mat Ryer:** I'd love to hear any stories that either of you have for things like that. I love it when things go wrong, but then the last minute you save the day. That's the kind of story we want. What we don't want is "Something went terribly wrong and it ended in tragedy."
+
+**Natalie Pistunovich:** And "Zis was dee plan", yeah.
+
+**Mat Ryer:** Yeah. \[laughter\] If it was the plan all along, that's an entirely different -- that's sort of evil organizing in the community. We'll do a separate episode on that.
+
+**Natalie Pistunovich:** Well, this is the reason I was proposing the name Chaos Engineering for this episode, because making a conference is a lot like that. You do your best to have a good -- the equivalent of uptime would be "Things functioning as planned." You always have to be prepared for when it's not going well, because preparing for when it does go well is the easy part.
+
+**Mat Ryer:** It's kind of like writing Go code. We deliberately handle errors all the time, don't we? We sort of expect things to go wrong. It kind of works.
+
+**Natalie Pistunovich:** Yeah, pretty much. Error-driven development, right?
+
+**Mat Ryer:** Yeah, why not. What about you, Ronna? What's the worst thing that's happened, but it's been okay in the end?
+
+**Ronna Steinberg:** The worst thing that happened... There's been a few. I never remember them. They never make an impression... \[laughs\]
+
+**Mat Ryer:** Oh, that's good.
+
+**Ronna Steinberg:** But yeah, I think if you have a mature enough community, things can break and nothing will happen. There was this one incident where I twisted my ankle on the way to give a talk for Natalie's \[unintelligible 00:47:03.00\] GDG Golang, and they ended up having a fish bowl instead, that ended up being amazing; an actual experience for diversity, I guess. It was a victory there. Do you wanna tell the story?
+
+**Natalie Pistunovich:** You summarized it pretty well.
+
+**Ronna Steinberg:** I wasn't there, so I actually don't necessarily know what happened exactly, because I wasn't there.
+
+**Mat Ryer:** You weren't -- yeah. You can't tell the story because of your ankle. \[laughter\] This is great. It's like your ankle broke on the way to telling the story. \[laughter\] So Ronna, what did they do instead? \[laughter\] I missed it -- it sounded like you said there's a fish bowl.
+
+**Ronna Steinberg:** I was told a lot of stories about what happened, but I actually cannot give a first-hand account for this.
+
+**Natalie Pistunovich:** \[47:52\] So we started the fish bowl session with Alan, who was the first speaker, and myself. We proposed the topic of becoming a GDE. Then there were some questions around that. Then the next topic was -- I cannot remember, but I'm pretty sure the topic that impressed Ronna the most was when one of the Women Who Go members who came to this meetup, she started asking the crowd on being a beginner in Go. And she got a lot of great tips, and was practically handling the crowd on her own for about 15 minutes; she learned a lot, taught the group a lot...
+
+**Mat Ryer:** That's great.
+
+**Natalie Pistunovich:** Ronna, was this what you were focusing on?
+
+**Ronna Steinberg:** Yeah, I guess. I think it's like the story of the year, as far as I'm concerned. By the way, she just got a job as a full-fledged gopher, so... Congrats to her.
+
+**Natalie Pistunovich:** Alright!
+
+**Mat Ryer:** Oh, there you go.
+
+**Ronna Steinberg:** Yeah, good work.
+
+**Natalie Pistunovich:** \[unintelligible 00:48:50.03\]
+
+**Mat Ryer:** She wasn't responsible for you breaking your ankle though, right? \[laughter\]
+
+**Ronna Steinberg:** The twisting. I didn't actually break it.
+
+**Mat Ryer:** Twisting, sorry.
+
+**Ronna Steinberg:** I probably dramatized it just a bit too much, I guess. My ankle really didn't fit into my shoe, and I really couldn't walk... But the next day I could, and I felt very silly.
+
+**Mat Ryer:** Oh, no.
+
+**Natalie Pistunovich:** This entire meetup was one huge adventure, because it was the first meetup of this year, 2020... And usually we have the last meetup of the previous year, so in December, in the beginning of December, and then the first meetup of the year in January we have in the end of the year. So this is the one time where we have more than a month apart, but almost two months apart.
+
+And it's the holiday season, so everybody is out of office, and longer vacations, and so on... And our hosting company was not answering the emails. They were not answering on Slack and they were not answering on Twitter. And it's been a few days that they were unresponsive, and we started pinging them in a slightly more intense manner, and then what ended up happening is that the day before the event I went to their office in person, and said "Hello, we'll have a meetup tomorrow", and they said "Oh, actually --"
+
+**Mat Ryer:** You tried Slack, you tried Twitter, you tried email, you tried all the ways that you can get through to Ronna.
+
+**Natalie Pistunovich:** I also called them by the number on Google Maps.
+
+**Mat Ryer:** Final resort.
+
+**Natalie Pistunovich:** No answers, yeah. And then I showed up in person, and they said that since they committed - usually companies commit a few months ahead; we have this kind of a queue - they decided not to use Go, so they would not be interested in hosting us anymore...
+
+**Mat Ryer:** Oh, nice.
+
+**Natalie Pistunovich:** Which is normally fine, but when it happens one day before the meetup, this is less cool. \[laughter\]
+
+**Johnny Boursiquot:** And they could have said that, right? \[laughs\] That would have been so much easier.
+
+**Natalie Pistunovich:** Yeah, yeah. And then I started tweeting "So, anybody feel spontaneous and would like to host this? Because we have two great talks, and people have babysitters and whatnot, ready to come to listen to talks..." And then after a few hours they decided "Well, we made a promise, we would like to keep the commitment" and so the meetup went through as planned. This was a perfect way to start the year. It's like a gun in the first act, to what's gonna come to this year in general.
+
+**Mat Ryer:** Yeah. And I like to think that that company then allowed the meetup to happen, and then saw it and thought "You know what - we're not gonna use Rust. We're gonna go back to Go."
+
+**Natalie Pistunovich:** Oh, this would be amazing. I should ask them, you're right.
+
+**Mat Ryer:** If we make this into a Netflix special, that's what has to happen. \[laughter\] You've just gotta give people what they want.
+
+**Break:** \[51:40\]
+
+**Mat Ryer:** So as attendees, as people that are gonna take part in this community but aren't involved in organizing, what things can we do to make your lives easier? That's always an interesting thing to think about, and often the answers surprise me, so it might be nice for people to hear. Is there anything in particular that we should or shouldn't be doing, that we do, that we probably just don't even know we're doing it?
+
+**Natalie Pistunovich:** I would be curious to hear after answering this what is the most surprising answer you heard.
+
+**Mat Ryer:** Okay.
+
+**Natalie Pistunovich:** So the two things that I would love for attendees to do is 1) assume everything is on good faith. This includes things like a change in plans, or not doing things as you hoped it will be, or misunderstanding something, and so on... And mainly being the first person to follow the instructions. For example, when you say "Break is over, go back to class", leading the group to go back into the room.
+
+There's not enough words to stress how helpful it is to stick to the agenda and to make things smooth. To clap whenever it's the time, to laugh if somebody said a joke and it's funny enough for you to understand that this is a joke. To tweet, to share whatever you learned on social media - if it's Twitter, if it's writing about this on your blog, and so on... All those small gestures really help a lot.
+
+**Mat Ryer:** Brilliant. My problem is I sometimes try and do a joke, but I miss out that bit of letting people know that it's a joke; that's the bit that I miss out. Otherwise, I feel like they would be laughing, but... Yeah, they just don't know it's a joke. They just think it's a sentence.
+
+**Natalie Pistunovich:** \[laughs\]
+
+**Johnny Boursiquot:** So if you have to tell them it's a joke... It's not.
+
+**Mat Ryer:** Right. I don't tell them. It's just awkward silence. \[laughter\]
+
+**Natalie Pistunovich:** Ronna, what do you think?
+
+**Mat Ryer:** Yeah, what about you, Ronna?
+
+**Ronna Steinberg:** Keep your RSVPs up to date I guess would be the first thing that comes to mind... Because we don't want to waste food. I'm not gonna say that we're wasting beer, because this is Berlin and we never waste a beer, but... At least the food. We don't wanna waste food.
+
+**Mat Ryer:** And also, if it's sold out, it frees up a slot for someone else to go, doesn't it?
+
+**Natalie Pistunovich:** Yeah.
+
+**Mat Ryer:** So if you can't make it, go back into whatever tool you said you were gonna make it with, and let them know you're not gonna make it. That is important, especially -- I mean, the London meetup, the London Gophers is packed, it's sold out every month (or was). So that's especially important if you're at capacity.
+
+**Natalie Pistunovich:** Yeah.
+
+**Johnny Boursiquot:** So one story - I know you like stories, Mat, so I'm gonna give you a story...
+
+**Mat Ryer:** Gather around, everyone.
+
+**Natalie Pistunovich:** \[laughs\]
+
+**Mat Ryer:** Could you do it really close to the mic, Johnny, so it sounds like you're all close?
+
+**Johnny Boursiquot:** And I'm gonna use my deep voice for you... \[laughter\]
+
+**Mat Ryer:** Good. People will love that. I'll get letters.
+
+**Johnny Boursiquot:** \[laughs\] So in the early days of community organizing I learned a lesson that I've never forgotten... And to this day, it is one of the core principles that I hold to when I'm organizing meetups and things like that.
+
+I had organized an event, and kind of dealing with the same logistical issues that Natalie was just talking about... Basically, I had planned for an event to happen, in a particular venue, on a particular date, I ordered food and all that stuff, and sort of doing all of that logistic stuff... And then it turned out that last-minute the location we were gonna host the event - they had an event happening on that day; it was very last-minute, and we had to scramble. Virtually the same story, Natalie, you just talked about... You're sort of scrambling last-minute.
+
+**Natalie Pistunovich:** \[laughs\]
+
+**Johnny Boursiquot:** \[56:03\] And then thankfully, I had -- I think it was 2-3 days. It wasn't the day before, like Natalie's, but close enough to be like "Oh, man...!" Maybe it was my ego, I was just kind of bruised a little bit, and I'm like "I don't like this." So I decided to cancel the event.
+
+Later on, after I'd organized a make-up or follow-on event or whatnot, later on I had somebody come during that event and tell me "Hey... You know, the last time you had organized this event, I had gotten a babysitter, I had made arrangements..." They were from an under-represented group, so they were already struggling financially to be able to set time aside from work. They probably worked multiple jobs and they had to take time off, and they had to find a babysitter, go out of pocket...
+
+There's a certain side of this tech industry that we all live in that we just don't see... Because we're all making good money, and we're all technologists, coders, developers, programmers, whatever it is... We don't see those trying to get in; we don't see the same struggles. Because if we did struggle like that, we're long past that, so we've forgotten.
+
+So for those of you who have forgotten, remember this - the folks that are trying to get into this community, the folks that are trying to be just like you, and you are the person they wanna be when they grow up... Folks that are trying to be just like you, they have to overcome a lot more struggles than showing up to an event. They have so much going on in their lives. In order for them to actually make that event, they have to make some sacrifices that you may not ever hear about as an event organizer, or even as an attendee sitting right next to them.
+
+So that lesson, basically -- I took that to heart. I was like "You know what - I cost somebody time, money, anxiety..." I created an unwelcoming, non-inclusive, non-caring situation, and I'm just glad they were able to come back and try again the second time. But after I learned that lesson, I'm like "You know what - if I'm going to put together an event, I'd better make sure that I maybe have a backup, or have some sort of alternative..." I try to go out of my way, at least try ahead of time -- there's some things you can't control as an event organizer, and Natalie and Ronna will tell you that; you just can't control for all the variables, and you don't have to. You just have to factor in - there are some people that are gonna make certain sacrifices to be part of something that you're putting together.
+
+And you as an attendee - there are some things you can do. If you RSVP for an event, especially one that's been sold out, and you no longer want to come, just un-RSVP. Create a spot for somebody else who perhaps has made some sacrifice so that they could be there, for them to be able to make it. And if you're going to a conference and they have a scholarship program like Ronna organizes, give money to that. These are some very real, very tangible ways you can actually help.
+
+Community organizing, conference organizing, meetup running - that's not for everybody. We're not asking everybody to do that. But as a fellow human being, there are some things you can do to actually show support and be supportive of those who are doing that work.
+
+I just wanted to put that out there, because I think sometimes we all forget that our day-to-day, the way we go about things in tech - there's an entire other side of that that we don't see or we don't get to experience, because we're so far removed from it. But it is there.
+
+**Mat Ryer:** That's a great thing. Thank you. Well, it's that time again... It's time for our regular slot, Unpopular Opinion!
+
+**Jingle:** \[01:00:02.06\] to \[01:00:20.07\]
+
+**Mat Ryer:** So who's got an unpopular opinion they would like to share? It can be tech-related, but doesn't have to be. It can be about organizing, but it doesn't have to be...
+
+**Natalie Pistunovich:** Ronna, do you wanna go?
+
+**Ronna Steinberg:** I can go. Unless you want to go.
+
+**Natalie Pistunovich:** Please do.
+
+**Mat Ryer:** We've got ourselves a polite-off.
+
+**Johnny Boursiquot:** \[unintelligible 01:00:27.10\] So polite, yeah.
+
+**Natalie Pistunovich:** Livelock! \[laughter\]
+
+**Ronna Steinberg:** You don't have to have two cases to generalize behavior to an interface, to take an interface or to use an interface, as far as I see it. This is coming from my experience before Go, and I think it works very well with Go as well.
+
+Interfaces are a good way to implicitly document what kind of behavior you anticipate from the type that you're taking in, and what kind you don't. It creates a small subset of things that the type should be able to support.
+
+The first thing that you even see when you look at the function signature is the types. If you have that documentation, it can be a little bit more loose. The behavior of the concrete type can change, but your function is decoupled for generations to come, and any maintainer will understand what they're supposed to be doing with this code, and what this code is supposed to be able to do, and especially what it cannot or shouldn't be doing, or you didn't anticipate that it should be doing.
+
+There is another point there, because in Go, when using an interface, we are causing an allocation to the heap, so there's something to say about that. It's a performance enhancement, though... It's not necessarily something that you should be considering. But if your program does require that kind of zero allocations consideration, you might want to forget about everything that I just said... But I think that we should treat this as a limitation in Go, and not as something that is wanted and is a good experience.
+
+**Mat Ryer:** Yeah, interesting. I hear sometimes - and I think I've said also sometimes - talk about this idea of early abstraction... But what you've said really does make sense. If you're using an interface to tell a story, or -- I like what you said about you can just use a subset of what you need. So if there is a concrete type that you're thinking of when you build your thing, but you're only gonna use a couple of the methods, having an interface with just those couple of methods in is a great storytelling, and communicates very clearly the intent of what you're gonna use that thing for. And of course, the fewer methods, the easier interfaces are to implement as well.
+
+That's a great one, actually, Ronna. Thank you very much.
+
+**Ronna Steinberg:** Sure thing.
+
+**Mat Ryer:** Yeah. That's going in the Hall of Fame, that one.
+
+**Ronna Steinberg:** Really?! Oh, my god.
+
+**Natalie Pistunovich:** Wooh! \[clapping\]
+
+**Mat Ryer:** We don't have a Hall of Fame, but--
+
+**Ronna Steinberg:** I am bowing right now. You can't see me, but I am bowing right now... \[laughter\]
+
+**Johnny Boursiquot:** See, Natalie might have ruined it. She was clapping, so they're gonna think they have to cut that out. \[laughter\]
+
+**Natalie Pistunovich:** Keep, keep!
+
+**Ronna Steinberg:** You're not cutting that out...!
+
+**Natalie Pistunovich:** This is the sign for keep, right? \[laughter\]
+
+**Ronna Steinberg:** You are not cutting that out! That remains forever. Okay...
+
+**Mat Ryer:** That remains... And what about you, Natalie? Do you have one?
+
+**Natalie Pistunovich:** Yes. I missed the note that the unpopular opinion should be Go-related, because this is a Go-related podcast...
+
+**Mat Ryer:** It doesn't have to be Go-related.
+
+**Johnny Boursiquot:** It doesn't have to be, no.
+
+**Natalie Pistunovich:** Oh, yay! Alright.
+
+**Mat Ryer:** Yeah. Anything.
+
+**Natalie Pistunovich:** \[01:03:52.14\] Then... It has a longer preparation. So diversity needs to be improved on many different fronts. Some fronts have improved, some fronts have improved less... It's hard to say there is any front or any type of diversity that is fixed, and the problem is no longer there. And I'm putting a lot of work into gender diversity, Ronna is putting a lot of work into general diversity... But for me, gender diversity is not the main goal. I don't have a main goal when I'm working on diversity.
+
+One goal that is very close to my heart is bringing people from different countries, like we spoke today. And I prepared some numbers. GopherCon Europe speakers who are born, raised, and live in an African country in 2018 was 10%. It was supposed to be higher, but too many people had their visa either rejected, or just never responded to.
+
+So they applied all the documents, we prepared everything, we helped the speaker gather all the documents they need, and they just did not hear back. We ended up with having 10%, so one out of the ten speakers is actually coming from an African country. And this still was a way bigger effort than reaching 50/50 speakers of men and not men. That was GopherCon 2018.
+
+In 2019 it was slightly better. Two out of the twelve came from an African country. In 2020, the current line-up is 4 out of the 28, which is still pretty much the same. Cloud 9 it's also one of eight, and BSides, which is a small local conference about security that I'm doing, is one of six. I have yet to break from the limit of 20% of the speakers coming from an African country, which is a personal mission of mine... And it's really hard.
+
+Now, this was all the preparation. Here goes the unpopular opinion... A conference that has a line-up of speakers who are diverse, but still all of them are coming either from Europe or from the U.S, is not the most diverse thing. What I am looking to evaluate the conference whether they are really diverse on my criteria - I'm looking at which countries the speakers come from. If it's not an easy country, where the visa is not an issue, for example, or giving such an opportunity is a big deal versus a huge deal, this is what makes the difference for me. It's not a good recommendation to give to conference organizers to focus on this, because it's basically recommending people to do a lot of extra work.
+
+Working on somebody's visa requires your legal preparation with probably five different types of documents. On top of that - this is just from Europe. I'm sure the U.S. is harder, at least in these times.
+
+So on top of those types of legal documents, you have to prove all kinds of financial support, you have to prepare things like flights, even though they don't even have a visa yet; they already need to present tickets. Accommodation, which is maybe slightly easier... Plenty of documents to show that those people will go back, so a rental contract, health insurance, connections, like if you're married or you have parents, a job, last salaries... So basically you're spending many hours with each person you want to bring, whether that's a speaker or an attendee, just making the legal parts of it. Still, I find this incredibly important as an aspect of diversity, and I think this has not been discussed enough.
+
+**Ronna Steinberg:** \[01:07:59.02\] I'd like to add to that one of the issues that I was struggling with. I had also my own struggles with bringing people to GopherCon Europe as attendees with the diversity scholarship... Something else that has been nagging at me for the past three years is how intimidated they are by authority. Depending on where they are - some places more than others - they treat authority very differently than what I would, or a lot of people would.
+
+Also, in a lot of countries these services are completely privatized. So they don't actually meet a German trying to enter Germany, they will meet some private company, and we don't actually know anything. Germany might on its own, for instance, say "Great, so somebody's going to come in. They have hotel, they're going to pay here, there's going to be some business done locally - that's great for our economy", but I don't know necessarily what is going on through the heads of a private company, and at the end of the day how people are going to be treated when they show up.
+
+I have no idea how to guarantee, for instance, how this meeting is going to go, and there is a lot of fear around this... And yeah, definitely not discussed enough. Thank you, Natalie, for bringing this up.
+
+**Mat Ryer:** I wonder if we could make a video of seeing the end-to-end process, just to show that it's at least possible. Because I understand that reluctance to -- especially if you've never been on a plane before, just imagine that. And you're gonna just fly somewhere, and someone's definitely gonna meet you at the airport? I just can't imagine those things happening, probably.
+
+**Ronna Steinberg:** They are very adventurous, I'll say that. We're definitely getting \[unintelligible 01:09:52.22\] I love every single person so far... And also the people that we couldn't bring, really. It's definitely a privilege.
+
+**Mat Ryer:** You know, all the diversity that the Go community has makes it a better community, so thank you to everyone on this and everyone listening that does help with that, does contribute. I think we should try and be the most diverse tech community there is, why not.
+
+**Johnny Boursiquot:** So I don't have a lot of unpopular opinions too often... \[laughter\]
+
+**Mat Ryer:** It's a good opening. Strong start. But...?
+
+**Johnny Boursiquot:** \[laughs\] And maybe it's not even an unpopular one, but here's something I've been thinking about a lot lately. I'm not sure if it's because we're all home and can't really go anywhere, shouldn't really go anywhere, and just more time to think, but... Diversity efforts shouldn't only be the burden of those carrying it. If you recognize the benefits of a diverse community or workplace, then get off your tush and help make it a reality. If you've been wishing that your local tech community was more diverse, stop agonizing. Organize.
+
+You can play a role in this. You can't leave it up to others, least of all those that are trying to step out from that shadow, from that cloud. It's everybody's responsibility, not just those who have to deal with it. So there's room there.
+
+**Mat Ryer:** That's a fair point. It sounds like a targeted attack on me, but... \[laughter\] As a general point, I think it's absolutely right. It's a good point. That's why talking about this stuff is important, because this is how we find out about these things. Because this stuff is not obvious to people that aren't in these groups. So yeah, good/fair point. Community gives back, again; it's what you were saying in your keynote, Johnny.
+
+[Johnny's keynote](https://www.youtube.com/watch?v=_Pc0bzz4-gM) is also still available on the internet. You can find it with your favorite search engine by typing in "johnny" or something... I don't know the exact keywords, but I assume "johnny boursiquot keynote". Mind you, spelling Boursiquot is also a bit of a challenge for people, isn't it? Maybe...
+
+**Johnny Boursiquot:** It can be. But \[unintelligible 01:12:42.14\] will help you.
+
+**Mat Ryer:** My watch has found it. \[laughter\] Well, that's all the time we have today. Thank you so much Natalie, Ronna. It was great. Thank you again for all you do, and we'll see you all next week.

Play with Go_transcript.txt

@@ -0,0 +1,349 @@
+**Jon Calhoun:** Alright, welcome everybody! In this episode of Go Time we're going to be talking about Play with Go. This is a new project that was created by Marcos and Paul, our guests... So I guess I'll just start with introducing them. First we have Paul Jolly. Paul, do you wanna say hi?
+
+**Paul Jolly:** Hi!
+
+**Jon Calhoun:** And then we have Marcos Nils. Marcos, do you wanna say hi?
+
+**Marcos Nils:** Hola! Hi, everyone. Marcos here.
+
+**Jon Calhoun:** We also have our panelists - Carmen is with us. Carmen, I believe you were involved with this, correct?
+
+**Carmen Andoh:** Yeah. Hi. I was.
+
+**Jon Calhoun:** Okay. So it'll good to get your perspective on this. And then lastly, we have our newest panelist, Kris Brandow. Kris, do you wanna say hi to everybody?
+
+**Kris Brandow:** Hello, everyone.
+
+**Jon Calhoun:** Okay, so I guess to get started - Marcos and Paul, do you wanna tell us a little bit about yourselves and what this project is you've been working on?
+
+**Marcos Nils:** Yeah. You were saying, Jon, that Carmen was involved. Actually, I believe Carmen is the reason why in the very beginning this was possible. The short story is that I met Carmen and Paul at the Go meetup in London. I was happening to be living in London not so long ago. I'm in Argentina right now, by the way... And speaking about Go, of course - it was funny, because it was a session, if I recall correctly, about testing.
+
+\[04:01\] And the presenters were discussing about different testing tools. After the presentation I reached out to Carmen and I told her "Hey, I actually wrote a Go testing library." And I don't know if someone recalls -- it's called Goblin. It was one of the first BDD libraries that was written for Go. I'm coming from the Node.js world, so what I tried to do is basically replicate Mocha into Go.
+
+Anyways, we started drifting conversations, and then I talked to her and I showed her Play with Docker, which is where Play with Go takes inspiration from. She introduced me to Paul, and I'll let you, Paul, tell the rest of the story.
+
+**Paul Jolly:** Yup. So I'm Paul, and as Marcos said, we met at London Gophers. In the last few years I've been a co-organizer of London Gophers. And Carmen, I think you were actually speaking at that particular London Gophers. We were hosted at Cloudflare, from memory... This was all in the pre-Covid days, when I think we had about 250 people there.
+
+In addition to organizing London Gophers in my spare time, I also maintain a number of open source projects, and co-organizers of the Golang Tools group as well. But as Marcos was saying, he and I have been working on Play with Go for the last six months. It's a series of hands-on interactive browser-based guides that introduce the tools required to work with the Go programming language. So it's really like a zero cost way of getting up and running, beyond the Tour of Go and beyond the Go Playground, to actually get programming with Go.
+
+**Jon Calhoun:** Okay. So I guess my next question would be what exactly is Play with Go? Why would somebody wanna check it out, and what problems does it solve, I suppose?
+
+**Paul Jolly:** Yeah, sure. So the simplest way to describe it is to think of -- the Tour of Go is very familiar to people as a way of starting to learn the Go programming language... But once you want to go beyond the language and actually wrap your hands around something and create something, you need to understand the Go tooling. And that's where you need to get something set up locally, generally speaking. You need to install Go, have your computer set up with an editor etc.
+
+So Play with Go is a very low-cost way, that doesn't require you to have anything installed locally. As I said, the guides are browser-based. So when you open a Play with Go guide, on the left-hand side you've got the guide, which reads as a guide, or a tutorial as usual. On the right-hand side of the page is an actual interactive terminal that is hosted on Google Cloud. So this is effectively real compute resources that are live in the browser for you. So as you're working your way through the guide, you've got command blocks in this left-hand side, which is the tutorial part, and you can actually click on those command blocks and they will execute in the right-hand side, the terminal, on that, in the Google Cloud compute resources.
+
+For example, you could have a command block at the left-hand side that simply just ran *go version* and that would run the command on the right-hand side and show you the output from it. So that's the interactive nature of it.
+
+You can also create files as you're going along. For example, let's just imagine you're creating a guide that explains how to create tests in Go. You would obviously create yourself a package, and then within that package you would have perhaps a regular package file and then a test file, as well... And you would create these two files by having code blocks in the left-hand side of the guide, and then with the final command block you would click on the command block and that would run go test, and you'd see all this happening live within your browser.
+
+So this is a bit of an experiment really to see whether this kind of interactive type of documentation works... Because the Tour of Go and Go Playground are incredibly popular as a way of getting to know the language, but then at that point there's kind of nothing; you're left to your own devices. And this is a way that, as Carmen said in her closing keynote at GopherCon, of trying to lower the barriers to entry, if you like, for the next one million Gophers, i.e. not place any requirements on them to have anything installed locally, and just try things out in the browser, in a nice, fun, interactive way.
+
+**Jon Calhoun:** \[08:14\] So if I understand this correctly, that would mean that this is -- like, if you wanted to show somebody exactly how running go test works, you could do it there. But then you could also jump into other tools like go fmt, and things like that, and actually show them how it formats code, and different things like that.
+
+**Paul Jolly:** Absolutely, yeah. And it's not just limited to tools that run just on that machine. As you can see, if you actually go to play-with-go.dev, there are a number of module-based guides on there as well, that show the full lifecycle of a module, from just creating your first Go package, to then creating and publishing a module, to then actually getting that module through the Go proxy. So it is really a beginning-to-end story when it comes to modules, as well as tools themselves.
+
+**Jon Calhoun:** So with something like this, do you imagine it being used -- I guess where I'm seeing this is another use case is the fact that when modules were first introduced, it was sort of hard to wrap your head around it unless you actually grabbed the code that had it, and installed it, and then got to play with it a little bit. Do you see this as something that would also help with that, where they're doing proposals and they can actually show documentation and see how it would work?
+
+**Paul Jolly:** Yeah, absolutely. That's one of the things that we tried to do with the launch of Play with Go as well. Jay Conrod, who works on the Go team, actually put together a guide that demonstrates a new feature of modules that's gonna be present in Go 1.16, which is the ability to retract a module version.
+
+Let's say for example you accidentally publish version one of your module when you didn't mean to; you intended to stay on version zero, i.e. unstable. Module retraction allows you to say "No, hang on... That was an accident. I didn't mean to publish that." So not only can you look at the documentation that Jay has already written for module retraction, but you can actually go and see how does that workflow actually pan out when you're doing it in practice. So there's a full example there of you publishing a module, accidentally publishing a v1 etc.
+
+So it does give you a good way of demonstrating to other people "Okay, what does this actually feel like when you're using it? Is the UI/UX around this tooling correct?" A form of early feedback, if you will.
+
+**Jon Calhoun:** Okay. Marcos, you mentioned this was based on Play with Docker. Is that a similar type of project, and what specifically about this makes it related to Go? Or is there anything specific that makes this more Go-specific than Docker?
+
+**Marcos Nils:** Yes, good question. Play with Docker is actually a project that I started with a friend that is called Jonathan; we need, of course, to give him attribution, because this wouldn't have been possible without his efforts. We started it in 2018. It's funny, because it was actually part of a hackathon project that we did in a weekend, I believe, because we were basically developers.
+
+The reason why we did that is because we saw, I believe, someone that you know in the Go community, which is Jérôme Petazzoni, an ex Docker employee. He was actually trying to teach Docker to people. And the way that he did that in the past was provision a dedicated VM for each participant of his workshops, and then he needed to provide five VMs per student; because you needed to make a cluster, and so on.
+
+And the UX around that experience wasn't the best one. Basically, he literally gave a piece of paper to each student that he had to print beforehand, and then each student would have to have an SSH connection to the terminals, and then the whole thing became very difficult to manage.
+
+So we were looking at this workshop with Jonathan and then we said "Hey, this could be easily done in the browser, where you could have all your sessions in a single place. You can jump between terminals and you can run Docker there, you can expose services, you can access all services with a host name, super-easy..." And then in -- I believe it was one night that we sat and started doing some coding, and then we got the PoC running. Of course, Play with Docker is an open source project and it's written in Go...
+
+\[12:21\] After that, as I said before, when I met Paul, I told him "Hey, this could be easily translated to anything that's pretty much terminal-based." Like either Go CLI, it could be any programming language... We even did a Play with Kubernetes. It's also something that is based on Play with Docker. We did Play with Nomad, and now Play with Go. We're trying to bring new experiences for just people that need to learn through this type of experience.
+
+**Jon Calhoun:** So you said that Play with Docker is open source. Is Play with Go also open source?
+
+**Marcos Nils:** Yes, it is. Everything is open source. If you go to the Play with Go GitHub organization, you're gonna find several repos there. You have the Play with Go, which is the main repo that basically brings all the components together. There are different components that I believe Paul can explain better... And then there are some other repos out there that are basically like the website, the tool that generates the guide and makes sure that the guide is consistent...
+
+And this is a very important point, Jon, because you were mentioning about, for example, leaning how to use modules, right? It is funny, because if you go to the Go blog right now and you see the first blog entry about using modules, you're gonna find that there were some examples using Russ'es repos in GitHub, or his basically vanity URLs... And you're gonna find that those do not work right now, because the URLs changed, or maybe the repos are in the wrong version right now...
+
+The benefit of Play with Go is because everything is automated and tested, we run the complete tests on each iteration; every time we do a commit, we add a new thing, we change a guide, we update a guide. Paul did an amazing job on rerunning that stuff constantly, so we can always guarantee that the tutorial is gonna be updated and working.
+
+**Paul Jolly:** Yes. And Jon, back to your point of it being open source - yeah, it is an open source projects. One of the questions we've had from folks who've actually tried it out so far is why do need me to log in when I'm accessing play-with-go.dev? It's a great question, and it's kind of related to the fact that this is an open source project. We're not logging in in order to steal email addresses, but as I said in the introduction, there's a real container that is running per guide that you launch within play-with-go.dev. So there are real compute resources, and hence money that we are spending whenever anybody ever opens a guide.
+
+So yes, we are an open source project, and we're funded by sponsors entirely. But the focus, as Marcos was saying, is that this really is a project that is by the community, for the community, to coin a bit of a phrase there. Whilst it does look like this is an official Go project, this is an open source project, and the intention is that anybody should feel free to contribute in various different ways, whether that be just feedback on the site itself, contributing to Play with Go, Play with Docker, or any of the open source projects that are a part of it... Or if anybody wants to contribute content, as well. There really is sort of an open door as far as how we can make this content platform better. As Marcos said, there's ways in which we've tried to lower the barrier to contributing by making sure that if anybody does contribute any content or any fixes, we can know instantly whether there are any problems with any of the existing guides. It should be a fairly resilient system.
+
+**Jon Calhoun:** So you mentioned that it's a funded project... Before we move on, do you wanna take a minute to just mention some of the -- do you remember who the sponsors are? I don't wanna put you on the spot...
+
+**Paul Jolly:** Well, at this point we should actually say a big thank you to Carmen, because Marcos said that it was Carmen who helped originate this project. And it's been through the support of Carmen and her team at Google that this has actually happened in the first place.
+
+\[16:07\] I think along the lines of Carmen and her team sponsoring Staticcheck and TinyGo; that's where the support of Play with Go comes -- it's sort of that same kind of sponsorship. So Carmen, a huge thank you to you, because I think it's this kind of model of supporting open source projects is one that obviously Play with Go has originated from... But I think it's a fantastic model as well, because it then helps those projects to support other open source projects as well... And I should just call out a couple that we're actually using. We're using Staticcheck, but we're also using Gitea as part of Play with Go...
+
+So this model of larger organizations sponsoring open source projects like this is, I think, a very healthy one for the ecosystem, specifically the Go ecosystem here... Because it encourages those open source projects to flourish, and as we know, quite a lot of projects, including Go itself, have benefitted from a lot of open source projects in the Go ecosystem itself. Let's just take Staticcheck as an example - it's benefitting massively at scale from projects like Staticcheck.
+
+So in different ways, these open source -- so Play with Go is a very different sort of project to Staticcheck for example, but it is looking to grow the Go ecosystem and the number of gophers out there. So supporting open source projects like this and the people who contribute to it. So this is something I think we want to look at down the line, as well as if there are people who become just absolute top contributors of content, or reviewing of content, or fixes to the site itself... Those people should actually be rewarded in some way for doing that, because that's the whole point of this sort of ecosystem. But I've been talking a lot on that; I think Carmen is the expert to actually speak to that point a bit more.
+
+**Jon Calhoun:** So Carmen, it seems like the Go team is actually putting money towards making sure that educational material and teaching people Go is a big focus. Is that actually the case, and do you wanna talk about what's going on there?
+
+**Carmen Andoh:** Sure. So one of the things that we have a problem with in terms of strategy is we have a finite number of engineers on the Go team... And I always am hoping to tap into them to write blog posts about modules, or update documentation... I know Effective Go needs to be rewritten with modules in mind... There's just a lot with regards to educating.
+
+So we launched learn.go.dev, which was kind of a beta site curation that takes in and just helps organize some of the information... But at this point it's pretty static, and it could be akin to the Wiki on GitHub for Go. So we realized pretty soon on, either Google has to hire a ton, dozens and dozens of content creators and tech writers to scale this for the next 10 million (or however many million), or we need to think about an open source strategy for this.
+
+The other thing - when I researched the learning landscape of Go, I recognized that the current material that is existing that is produced by the Go team is static, and it's text-heavy. We do have the Playground, and we do have the Go Tour, but that's the extent in terms of being able to provide modalities of material.
+
+So we rely largely on third-party people to either do training, so Ardan Labs or Gopher Guides, or we have people who are enterprising enough to create Gophercises, such as yourself, Jon.
+
+\[19:48\] So the missing link, I feel, is that as both Paul and Marcos have alluded to, there was a big gap in what I call the interactivity and the ability to make documentation not stale, and have it come alive, so that people can try out what's being meant by it... And it also is a change in -- so we have a change in people for Go coming to the language; lots and lots of people. But we also have people that are a different kind of persona, which is mostly enterprise. And when you dig deeper into the persona of an enterprise or a person who uses Go professionally, they're not similar to the early adopter.
+
+The early adopter is curious, and they probably want to try out all the new languages, whereas the professional programmer using it at work - they just want to get the job done. They wanna see what it looks like, and they're very practical and pragmatic, and they don't wanna wax eloquent about the philosophy a bit, or the context a bit... They just wanna get it done.
+
+So we've heard that we can call it as meat and potatoes, or beans and rice, but really to the heart of it... And I think that play-with-go.dev is that gap that fills for everything that's happening with Go right now.
+
+So a long-winded way, Jon, of saying "Yes, we're looking at ways, with limited funds, to be able to scale out content that would be useful to a lot of people, but also get the ecosystem involved, and empower them..." Because some of the most useful materials that I've found is when people wanna share and teach. Everyone wants to feel useful. So I think this is a convergence of both strategy and product that is trying to meet all of those things at once... And I have a lot of hope and confidence that we'll achieve that.
+
+**Paul Jolly:** I think I'll just briefly add to what Carmen said... What we have there with play-with-go.dev at the moment is the first cut. It's really important to not see that as the finished article, because it's a starting point in terms of content, but it's also a starting point in terms of the UI, the UX... And the point of making it open source is that people will have really good feedback, and we have had some excellent feedback from a technical writer on the Go team, as well as Bill Kennedy, as to how we can make the UI/UX better, but also how we can write the guides, this style of interactive guide in a different way, as well.
+
+So I strongly encourage people who have had a play around with it - and we've already had some feedback as to how we can tweak this, and "Okay, what is the target audience for this guide here? If it is this type of audience, then actually you're gonna need to cut down the introduction and not make quite so many assumptions about the technical proficiency of somebody who's starting here." So yeah, there's so many gaps to be filled still, but I think we at least have a solid starting point here for, as Carmen was saying, this different style of learning.
+
+**Break:** \[22:58\]
+
+**Jon Calhoun:** So given that this is a starting point, I assume that means obviously you want to expand it, and make it better, and improve... So in my mind I guess there's a couple of ways people could help. Are you looking for additional sponsors? Is that one way that organizations can help?
+
+**Paul Jolly:** Yeah, I think we would definitely welcome that, because I think there's a number of ways in which these guides could actually be improved, just sort of building on this starting point here of this interactive guide. As Carmen was saying, there's a large enterprise contingent effectively that's going to start jumping on the Go ship fairly soon. They have already started jumping on... And one of the excellent bits of feedback we've had from Bill Kennedy has been that a fairly significant percentage of those people are gonna be Windows-based.
+
+So quite honestly, if you have a -- as you'll see if you go to play-with-go.dev, these are Linux-based guides at the moment... And actually, to make it a welcoming place for someone who's coming from a Windows background. It would be really nice to actually have a Windows container that is running, where you have either a command prompt, or a WSL 2... And guess what - it'd be fantastic if instead of just having a plain terminal, you had VS Code integrated there as well, with absolutely zero setup costs required on these. So then you're just opening a guide, and much like Marcos said, in the early Play with Docker days you had no requirements on the user to install anything on their machine and they've got everything up and running for them. Here it's the same type of idea. This should be possible as well when it comes to learning Go, and when you actually need to come to use tooling. So you should just have the editor in the browser with you, as well as the terminal.
+
+Marcos, you should just chip in here, but that's one area where further sponsorship would enable -- that's a fairly significant build, to get those two features supported... But that to my mind would be where sponsorship would be well-placed, I think.
+
+**Marcos Nils:** Yeah, I completely agree with you, Paul. Right now, as you said, cloud resources and the ability to expand to different platforms - I think it would be very appealing to us to maybe host Play with Go in a place like Packet where we could run ARM64, especially since now everyone is gonna have M1 which you can run on the ARM as well. And of course, Microsoft Windows would be amazing to try it out.
+
+**Paul Jolly:** I think sponsorship should also be considered to be going towards content creators as well, back to Carmen's point... This really needs to be viewed as an exercise as well in trying to grow an ecosystem for creating content. And we've got the agreement from members of the Go team as well, that for those guides which are core fundamentals, that they will help to review those... And I'm sure we're gonna have an army of people who are actually gonna try out these guides and give feedback where the guide doesn't read well, or whatever.
+
+The hope is that we can use this platform as a start on the content creation side of things as well... So I think it would be important to reflect that sponsorship should really go towards ongoing costs, in some way, of people's time in that respect, as well.
+
+**Jon Calhoun:** \[28:02\] So you'd mentioned that sponsoring and helping from an enterprise perspective would help expand what's there, and also help with contribution... For individuals, I assume either giving feedback or contributing new guides is a good way to -- is that one of the ways you expect or hope people will be able to contribute individually?
+
+**Marcos Nils:** Yup. I think we cover pretty much all the aspects of the platform regarding contributions. It can be either from the content side of things, as I said, it can be from the UI side of things as well... Take into account that neither Paul, nor myself are UI experts, not at all. Very far from that, actually. So the UI needs a lot of rework, and some polishing, as we said before...
+
+Another very big aspect that we are already working on is translations. Right now, as you might have probably seen, pretty much everything is in English. We are, of course, looking forward to expand it. The platform is already based on a multi-language architecture, basically, so it's very easy to extend the current guides in a different language. So we really foresee a near future where we could have tens of different languages due to the easiness to basically extend it. I don't know if you have any other things in mind, Paul...
+
+**Paul Jolly:** No, no. I think that's a great point, and I can't believe we didn't already talk about that... As well as extending it to multi-platform. The idea is that it should be possible to write a guide that you write the script for it once, essentially for a platform, and then you could say "Okay, if this script is running on Windows, then instead of running this command here, you run that command there", which of course, there are gonna be platform specifics...
+
+And as Marcos said as well, when you're then presenting that to the user, you want it to be in their local language. So the vision is sort of that when you land on a guide, you say "Hey, I'm on this platform, and this is my chosen language here." And if there exists a translation for it, for that guide, and then \[unintelligible 00:30:04.19\] present that to the user... And we also show it to them on whatever platform they're interested in as well.
+
+So again, back to Carmen's point from earlier - that's really being as welcoming as possible. If we can provide people something in a language that's familiar to them, for their chosen platform, it's such a low-friction way to get started. They don't have to install anything; they have VS Code in the browser. It's just a "Okay, fantastic. I'm up and running. It feels like I'm making progress already."
+
+**Jon Calhoun:** That makes sense. So what is that contribution process like? How would somebody get started if they wanted to either provide a translation, or to write a whole new guide... What would your advice be?
+
+**Paul Jolly:** Yeah, this is a good question. So we're just in the process of finishing up a guide that explains how to write a guide at the moment. That's awfully meta, but... That will give somebody a starting point. Literally, how you check things out, how you create a directory for the guide itself, what structure the guide takes. It's just split into two basic parts, which is a markdown file and then a script, which takes the form of a QFile.
+
+So the guide process itself -- the guide that explains how to create a guide will just walk you through that entire process. There's people who will be on-hand to answer any questions you might have on Slack as well... So kind of expecting that initially, that guide contribution process will have a few rough edges. We're gonna work through those with -- Jon, we're working with you as well on the guide around Go testing... We're gonna be working with Bill Kennedy and a number of others as well. So that's hopefully gonna smooth off those rough edges for the guide contribution process.
+
+Language translations are not something that we support just yet... But again, that's where with a bit more sponsorship, dare I say, towards the buildout required for that, we can easily get language translations in there as well, as the multi-platform support.
+
+**Jon Calhoun:** \[32:00\] So you mentioned briefly that it's a markdown file, and your scripts, which is a QFile... And I had a chance to go with you and look at a lot of those in more detail... So for anybody who's wondering, it's kind of like you write a markdown file, inside you that you put a special command that essentially says "This is the code or the script that I want you to run at this specific step." So when you're writing in other languages, you just keep those script commands at the same spots, and you just replace the actual text with whatever language you want. So that's what the process looks like.
+
+And Marcos, earlier you mentioned that these documents were testable... Meaning that if I write a guide, you can actually verify that the guide still works, whereas some of the guides on the Go blog right now aren't always working. Do you wanna talk a little bit, one of you, about how that works and how that current setup you have enables that?
+
+**Marcos Nils:** Yeah, I guess Paul would be the right person to explain in detail.
+
+**Paul Jolly:** Sure, okay. Yeah, Jon, it's exactly as you said. We have a tool that's called Preguide that was written -- actually, a bit of a back-story to Preguide. When I first was talking to Carmen and Marcos about this project, the idea was that I was initially going to start by writing a few guides, and then we sort of stepped back and said "Okay, what does this look like? What do we need to change?" etc.
+
+But the guide writing process to start with was pretty frustrating... Because as soon as I'd got halfway through the first guide, I had no way of checking whether the steps that I'd written still actually worked. And so I fairly quickly thought to myself "It's a problem for me now. This is just not scalable at all when it comes to multiple people contributing, potentially in separate languages." And as we agreed at the beginning of the project, the community aspect to this is going to be very important. So having a process by which people could contribute was fairly important. So that's where a tool called Preguide was really born.
+
+It's a simple tool that, as you said, takes the markdown part of a guide, and the Q script, and effectively smashes the two together, producing a markdown output. And that separation of the markdown, which is the prose, if you like, from the guide, from the script part, then allows for multiple translations, but it also allows you to add a bit more structure to the specification of the script as well.
+
+We talked about the ability to have multi-platform... If you've actually defined the steps in your guide in a more structured way - and we're using Q for that, as I said - you can then much more easily say that "Actually, you know what? At step number five here, if we're running on Windows, instead you need to run these commands here." Or step number six...
+
+So that allows you much more easily to have not any platform-specific variations in your script, but even language translations as well. So the canonical Hello World program - you might actually choose to provide a localized version of that, which would require you to have a localized version of the file that you're running as well.
+
+So splitting apart the prose and the script was a key part of that, and the structure that Q gives us within the script specification allows us, as I said, to have those specializations. Preguide is then just responsible for pooling all of that together, running a whole lot of sanity checks, and so it effectively acts like a compiler as you're writing the guide. You can just have this very tight edit and compile iteration loop. And whenever you change anything within the script itself, as Marcos was saying earlier on, it actually just reruns the entire script to validate that all the steps work as you expect, and that the output is as you expect as well.
+
+And Jon, as you were saying as well, then using this sort of text template style, templating within the markdown file, Preguide takes the output from the commands that have run in that automated process, the commands that have run in Preguide, and then generates for you a markdown file. And that's the markdown file that you ultimately see rendered on the Play with Go website, that is the combination of the markdown template and the script output as well.
+
+**Jon Calhoun:** \[36:05\] That's really neat. For anybody listening who's kind of curious, what this also enables is let's say Go releases 1.17, or Go 2, or whatever; they could actually take all the guides, run them and see if they work in Go 2, and they would actually know immediately which ones work for it and which ones don't. So if you've even gone on the internet and read a tutorial and been like "Okay, this is exactly what I want", and then you realize that nothing works because it's an old version of whatever you're using, this will hopefully prevent that type of issue, and really make it easy for them to maintain... And even if you need to update a guide, you don't have to get rid of the old one. You can almost have a guide for the new version and a guide for the old version, and have that backwards compatibility.
+
+**Paul Jolly:** Yeah, that's exactly it. So just to build on that slightly... This is something else that we have - effectively the building blocks for you, but would just require a tiny bit more build, is this concept of multiple scenarios. Multiple scenarios is sort of like a variation of the multi-platform support. But Jon, just as you were saying, you would have multiple scenarios for like Go 1.15, Go 1.16... Because there are subtle differences between versions, where the command go is concerned; the output is slightly different, and if you're not able to show the user a guide in the version of Go that they're using, it's quite easy to get spooked sometimes by something that is not quite working as you expect... And with Go 1.16 in particular, there are some subtle changes to the way that modules work that will actually require that new ones to be explained, particularly if you're coming from Go 1.15.
+
+So yes, not only do we have all of these, as Marcos was saying earlier on - we have all of these guides being checked in CI every time there's a change made to any of them. It's trivial to say "Okay, it's time for us to release all these guides as Go 1.16. What changes are actually required in them?" and then we'd be able to validate them and then publish those with confidence that every single one of the guides works.
+
+**Kris Brandow:** I'm also kind of wondering if this would be a helpful tool for library maintainers; it could be like an integration... Because I know one of the things I always struggle with is when I find a new package that I wanna use, and it's just like "Oh, alright, how do I get started with this thing? How do I actually use it?" It'd be really helpful if there was just like an easy guide that the maintainer could just update, that has like "Okay, here are the steps you need to do. Here's exactly what you need to do to use this package, or this library."
+
+**Paul Jolly:** Yeah, Chris, that's exactly right.
+
+**Marcos Nils:** Not only would we just mention about the fact about writing guides, but we realized that this also allows to close the gap between the language maintainers or contributors and users... Because imagine that you develop a new feature, or you want to get immediate feedback about something that you're basically coding, and instead of writing a comment in GitHub, and the fact that you have to tell people to clone or fetch the latest changes from a branch and build the thing locally, you could just for example leverage Play with Go, like "We're looking for a sponsorship", and basically have a way to show that to your community super-easy and super-fast, and telling your users "Hey, can you please go to this, try it out and give me feedback to see if that's the expected UX functionality etc?" And maybe eventually you could even start playing some sort of sandbox and then share that session to someone else, so they can reproduce what you're also doing there. Similar to the Go Tour, but imagine it for CLI tools as well.
+
+I believe that the possibilities here are endless. It's just a matter of putting the focus right now on teaching Go, the community, and see what are the immediate quick wins we can do for people to get engaged on the project.
+
+**Carmen Andoh:** Marcos mentions a really important point for the future, which is - we talk about ecosystems, and this tool can be that force multiplier for library authors, for companies that are trying to ignite their own open source communities based on Go.
+
+\[40:11\] Again, when I think about all the things that his can enable, and the new future of open source - I know that sounds very pie in the sky, but I think that the old open source which is static and just text-driven is gonna be no more in a couple of years... And this is the harbinger. We're gonna see readmes that are interactive. We're gonna see issues where you can click and test there, instead of having to (like Marcos said) get your own environment set up.
+
+This can be a way to bridge the gap between beginners, for whom -- not only do they have to learn the things in order to be able to contribute, but they also have to overcome the biggest gap of all, which is the setup, which was mine for years. I think I lost one or two years about seven or eight years ago, because I was too shy to ask about the setup; I thought it was just me, and then people were like "No, it's this janky thing that you've gotta do, and you've gotta pull in this thing, and we forgot to document it etc." If this is just done for you, then you can get right to it and then learn along the way.
+
+**Paul Jolly:** Kris, I've actually got a really good example to pick up on the point... Gio is a project that I think has actually been talked about on Go Time before as well. Elias, who's the creator of Gio -- I sort of gave a quick demo of Play with Go to him, and he said "Hm... Hang on a second. Is there some way that we can not only have these interactive guides where you're running these programs on this remote session - can I actually render something back to the frontend as well? Because Gio being very much a visual thing, obviously, it's gonna be pretty useless to run a Gio program on a headless remote container. Can we render the results back to the frontend?" And again, that's absolutely something we can do, because Gio is cross-platform and can render via WebAssembly. That's the plan, is to try and give that a try as well.
+
+So Gio would be able to use something like Play with Go to not only, as you said, Kris, explain "Okay, this is how you would get started with your first Gio application", but actually render the results to you as well. So then, if you can imagine this sort of interactive environment where you've got VS Code there as well, the user hasn't actually had to install anything at all and they're already running a graphical Go application within the browser. So that's a pretty powerful offering, I think, exactly as Carmen said, because it just can't settle that friction for people who are getting started... And I don't think back when I started programming what sort of friction it was, particularly when it came to Linux, or anything like that. It's just a very different story these days, and I think it's a good one for newcomers to the language.
+
+**Kris Brandow:** I'm also thinking of the Playground, which I think is a natural extension of this at some point... Because I know there's lots of times where I have some application or some little thing I wanna try out, and my modules directory is filled with all these x.kris.whatever... of just like "Oh, I wanna play with this little thing", but I can't go to the Playground, so I can't import other packages into the playground... And this just feels kind of like a precursor to us actually having a playground where we can import third-party packages, and maybe even have something that's a little bit like a gist, where it's just like "Oh, here's my files, here's everything. I want you to go run them, and I can share them with other people and collaborate."
+
+**Paul Jolly:** \[43:39\] Yeah... To the credit of Go Playground, it has actually got some limited support for third-party imports at the moment. It's limited somewhat by how long it takes to then download, fetch and install those; I think it's a fairly conservative limit there. But I think you're right, Kris. This was kind of the thinking with getting this full lifecycle of module publishing going on, because so many guides that I started looking at - and I've been looking at the modules in Vgo since the early days, which is almost three years ago now, unbelievably... So many of them stopped at saying "Alright, and now at this point you Git push", and then there's quite a lot of hand-waving going on, saying "Oh, you Git push and then you can do a go get, and you tag, and you push that tag", and you're stuck; if you can't follow something at that point, and/or it's not working, then you are stuck. Whereas with the Play with Go guides that we've got - not only have you got that remote container, but it actually automatically creates remote repositories for you as you need in order to complete that guide. So you have the full experience of actually publishing modules and getting them via the proxy, exactly as the end user, albeit they'll be publishing to a different module path, exactly as the end user should be experiencing it as well.
+
+**Break:** \[45:03\]
+
+**Jon Calhoun:** Also, a positive side to this is that right now teaching somebody how to create their first package and then import it and use it is -- you pretty much have to install Go to do that... And the Playground doesn't allow you to do that. So this type of thing should work really well for that, because - even if it's really like a package with one function and you just wanna export that function and show them, there's not a good way to do it. A guide like this really enables that type of education.
+
+**Carmen Andoh:** One of the early things that turned into -- when I was trying to improve learn.go.dev, we said "Wouldn't it be wonderful if we had added to a Tour of Go a tour of modules?" And this is what that became. Because we recognized that there are an existing 1-2 million existing Go users who already have a firm grasp on the language, but need the next step. And this is exactly what that is.
+
+**Jon Calhoun:** Yeah, I think that'll help a lot, because I know -- like, I've just actually helped somebody today... They had some code and they were importing a package, and they just didn't quite understand - like, if you're not using modules, that the GOPATH actually matters. They had something that worked, and then they moved the code somewhere else, and it wasn't working, and they're like "Why is this all breaking?" And just having to explain that to them and show them "This actually matters, where your code is... Or you need to use these modules...", which is then, all of a sudden, introducing a whole new thing that you don't necessarily want to introduce to somebody right away when you're going to confuse them with it. So having options like this would really help with that.
+
+**Paul Jolly:** One other thing as well, as Carmen mentioned the wiki earlier on... There's actually a really good wiki for Go modules, but it's a bit of a double-edged sword, because it's so easy to edit the content. As we've also kind of discussed, it's so easy for that content to Go out of date. It's impossible to check that it's still valid... And all of these sorts of things make for a fairly daunting experience; if someone says "Oh, just go and take a look at the Go modules wiki for an explanation of how to do that." It's just a -- well, good luck if you're coming out the other side, if that's actually gonna work as intended.
+
+\[48:05\] And one of the things that -- I had an example of this last week, actually. Somebody asked a question in the modules channel on Slack... We were just debating a point - if you have a module, which is major version, greater than or equal to 2, what strategies do you have for actually publishing those multiple major versions? And it took me literally ten minutes to pull together a very simple guide that demonstrated the different options available to you.
+
+I then published that guide. It's not actually part of the index on play-with-go.dev now, but... I know that guide works, and I can just put some prose around it, a bit of explanation as to what I've done in each step... But it's a really crisp way of sharing with people -- Kris, this is the point you were making earlier on as well, of "Okay, if you wanted to structure your module with a branch-based strategy or a subdirectory for your major version modules, this is how you do it." And not only is it the explanation of it, but you know it works.
+
+So Jon, as you were saying earlier on, as we go through the new Go versions that come up, we just know this is gonna continue to work, this guide, because we can check it as part of CI every single time.
+
+**Carmen Andoh:** And I also think -- what's been weighing on my mind a lot lately is the software supply chain security; this new software reuse problem, which I believe is a collective hurdle that we have to get over and figure out... And I believe that the modules toolchain is along one of the better language ecosystems that has figured that out. But now, with play-with-go.dev, we can even be able to do some early testing and sharing, and you can do some scrapbook or sandboxing of pulling in these dependencies and kind of checking them, without having to worry about -- you know, yes, we have other ways of sandboxing; we sandbox on VMs, we sandbox in the cloud, we sandbox on our own machine, but this is just yet another way to do that. Again, when you iterate on all the different use cases of this, this is one of those great ones.
+
+**Marcos Nils:** I have one more observation... I believe this week we \[unintelligible 00:50:13.02\] about it with Paul. After looking at the analytics metrics of Play with Go, we'll realized that 50% of the audience is coming from mobile devices. Unfortunately, our awesome UI skills that we mastered with Paul in these past weeks - now we can look for a job and say "You are an engineer, Paul. Don't worry." Surprisingly, the site works pretty well in landscape mode on a cell phone, on any regular cell phone.
+
+So now, going back to what Carmen was saying regarding that the tutorials and the instructions should be interactive, or are going to be interactive in the near future, if you actually for example see a new release of either a CLI tool, or a Go feature, or whatever, if you can give a lot of people access to that information super-easy in their phone devices while they are commuting to work, or while they're on their sofa, without a computer in their hands, that enables a whole set of learning paths that are just unbelievable. And the numbers talk by themselves. 50% of Play with Go users are coming from mobile devices.
+
+**Carmen Andoh:** That's incredible.
+
+**Marcos Nils:** Yeah.
+
+**Carmen Andoh:** I'm really curious, because I remember making a joke about merging a PR via my phone on the GitHub mobile browser... And I've just thought, you know, I'm coding, and I'm reviewing, and I'm using my little thumbs to check the code, and I just wonder "Will someone write a tutorial or a guide only on a mobile device?" That is truly like wow to me in terms of the future.
+
+**Kris Brandow:** Yeah. I'm really excited for the -- hopefully, if we can get the barrier to entry low for writing guides as well... So I feel like there's so many things that I've searched around the internet to figure out... Semantic import versions is definitely one of them, where I'm like "How does this work again?" If someone could write one guide, that you could just go to and find... Because yeah, when you go to the wiki, I grep around, or I use Find in Page, and I'm like "I think I've found the information that I want", and I just really wind up asking Bryan Mills like "Hey, how does this thing work?" and he just tells me. But most people can't do that... So just having some real good guides around that would be fantastic.
+
+**Carmen Andoh:** \[52:25\] Yeah. And that's how we can scale someone like Bryan. For those unfamiliar, Bryan is on the Go team, working on the go command in the open source team... And I wish I could scale 100 of Bryan - to write guides, to do content... But he's gotta make the go command better, and I think that's why I really believe that if we continue to do it this way, we can totally scale that content and have that.
+
+**Paul Jolly:** Yeah. Kris, I'll just build on what Carmen said there by saying that getting that barrier to entry low for content creation is key. As I said, there's gonna be a few rough edges as we get started there, and feedback is very welcomed there... But I think an important part in having this building block now, as Marcos was saying earlier on - the ability to rerun a guide and just check and validate that it actually works. Because if you've got that sort of edit/compile cycle with writing guides just as you have code, you as the developer can just very quickly - as I did with that example I cited from last week - in ten minutes put together something that is the barebones demonstration of what is going on in a situation... And then actually you can have other people contributing to the process as well by putting a nice introduction in place, actually writing the guide around it, the prose itself.
+
+So I think there are ways in which to make that barrier to entry nice and low by making sure the tooling is good, and fast etc. But also, if we have a good process for review and editing as well that is very light touch... Imagine a world where we didn't actually have this automatic checking of guides; it would be pretty hairy as a reviewer, because you're basically saying "Okay, I've now got to go through and check it myself", and then if there's another change made, you've gotta go through and check it again... That's just not scalable at all.
+
+So the barrier to entry is not only the tooling thing, all the content creation itself, but it's the review process as well. So that's where, picking up on Carmen's point - yes, we need more Bryans and Jays, absolutely; but if we can make their job of review of these guides much simpler, then that's how we can scale things as well.
+
+And I'm saying this very much in the context of the go command, but Kris, back to your earlier point - could this be used for other tools, other libraries etc? Absolutely. So the tools and technology that exists for Play with Go is open source, and it should be able to be used for other people as well.
+
+**Jon Calhoun:** So if people have additional questions, or they wanna learn more, where do you recommend they go at this point?
+
+**Paul Jolly:** The best place to start is play-with-go.dev. That is the main site. And @playwithgo is the Twitter handle. There's also a Play With Go channel on Slack, as well as the GitHub project that Marcos mentioned earlier on. Happy to answer any questions that people have, take any feedback... It's a good starting point, but there's plenty still to be done, not only with Play with Go, but just effectively the experiment that we're trying here, this interactive form of documentation. As Marcos was saying earlier on, there's many, many ways that people can contribute, and a feedback is a massive form of contribution.
+
+**Marcos Nils:** One last thing that I would like to highlight - it's also a very nice, interesting project to start basically learning about different aspects of technology. Play with Go has a UI, it has Qlang, which is a very trendy technology - or language; I don't know how to describe it, but it's very interesting to learn. It has Go, of course, it has infrastructure, it has containers, because all the infra is running on containers; it has Docker... It has security concerns, because as you may imagine, running a container where you can basically compile and write code, and basically play funky with bandwidth, and some things - it's a challenge, as well.
+
+So if you are into DevOps, coding, UI, content creation, UX, whatever you feel like you can contribute, please talk to us, so we can help you out in whatever you need.
+
+**Paul Jolly:** \[56:23\] And we're gonna be starting to mark issues in the GitHub issue tracker as "Good first time issues", "Help wanted", sort of categorize things a bit better. And Marcos also had the suggestion of pulling together a GitHub project board, so people can see "Okay, where are we now? What things are planned?"
+
+But kind of drawing back to your earlier question about sponsorship - sponsorship will actually help this project to live on in terms of its infrastructure cost, but also where it goes next.
+
+**Jon Calhoun:** We're kind of running out of time, and I forgot to prep you both for this, so I apologize... But it's time for our Unpopular Opinions segment.
+
+**Jingle:** \[57:00\] to \[57:16\]
+
+**Jon Calhoun:** So if you've never heard Go Time, basically, if you have one and you want to share, we ask you to share an unpopular opinion. It can be related to tech, or anything at all; I'm trying to think of some past examples... One example is that the subway (I think) was the fastest form of transit in New York... Maybe they said buses; I don't remember which one they said. But we've had different ones in the past.
+
+Others have been more tech-related, so people have said like -- I'm trying to think... Does anybody remember some off the top of their head? I'm drawing a blank.
+
+**Paul Jolly:** Didn't you have someone who was speaking from GitHub say that Git is too hard to learn as a command... Wasn't that the most recent one, I think?
+
+**Jon Calhoun:** I think we had that one. We've also had -- we've had somebody say that dogs are bad in the workplace... Different things like that. So if you have any unpopular opinion you'd like to share, we can talk about it a little bit. Usually, what happens is after the episode has aired, Jerod will actually take the unpopular opinion, cut it out and put it on Twitter with a poll, to see if it's actually unpopular or not.
+
+I will warn you that, because most of the audience is gophers, sometimes things that would be unpopular in the whole ecosystem of developers is popular inside of the Go ecosystem. But do either of you have something you'd like to share?
+
+**Marcos Nils:** Going back to Paul's comment around Git - if only you had a good platform, Paul, to learn interactive things...
+
+**Paul Jolly:** Well, that's just it, right? \[laughter\]
+
+**Carmen Andoh:** \[unintelligible 00:58:33.06\]
+
+**Paul Jolly:** If only someone wanted to write some content for the Git command... \[unintelligible 00:58:39.00\]
+
+**Marcos Nils:** We need to build something. We need to build something right away. \[laughter\]
+
+**Jon Calhoun:** I feel like those guides are gonna have to be like -- you're gonna have to have 20 hours of guides, because there's all sorts of weird things to do in Git that nobody remembers, because you do it once a year.
+
+**Paul Jolly:** Oh, exactly.
+
+**Marcos Nils:** Now it's called Git Restore, right? They killed me with that one. It's not Git checkout anymore. I have one, which is -- it's interesting, because I'm actually on my vacation from code from work this week... And we just decided with my girlfriend that we wanted to bike around the city... And we realized that even though the city has all these new bike lanes, basically the problems regarding how the city operates are the same ones. People not respecting rules, either drivers, other cyclists, the people that walk on the street as well... So the unpopular opinion is that it doesn't matter the different ways or means that you can implement to try to fix an issue; if you don't basically change the core, it will not happen.
+
+**Jon Calhoun:** Okay.
+
+**Carmen Andoh:** Totally not unpopular. This is gonna get like a 90% agreement rate on Twitter, I bet you, Marcos.
+
+**Kris Brandow:** Yeah...
+
+**Paul Jolly:** Okay, I've got one that is Go-related... Is that allowed?
+
+**Jon Calhoun:** That's completely fine.
+
+**Paul Jolly:** I don't know whether it's -- we'll see whether this is unpopular, or controversial, or both... But Go modules will be the last dependency management system for Go.
+
+**Marcos Nils:** Interesting.
+
+**Jon Calhoun:** I don't know how to respond to that one.
+
+**Carmen Andoh:** \[01:00:10.24\] Kind of like the way that npm is for Node? Hm... \[laughter\]
+
+**Paul Jolly:** Well, I guess there's a number of -- understandably, coming from a background of their different dependency management systems, it feels like for me there is a real net benefit to what modules brings. Yes, there are bits that people find frustrating, but I think that the focus that it places on the module author to respect non-breaking changes and/or bump major versions - it kind of places the onus on the author, in much the same way that it places the onus on the author of Go code to write code that is easy for the reader to understand. It's exactly that right shift, where you wouldn't just leave the reader of some code to just make a whole load of mistakes as they're trying to understand your code. You make it as clear and easy to understand as possible.
+
+In exactly the same way, I think it's sort of encouraging - helping to encourage, at least - module authors to be more responsible. Not that they're being irresponsible, but you perhaps get the idea.
+
+And there's obviously tooling improvements along the way that are desperately needed, and I think we're sort of in the Go way just taking our time to work out what they are... But it does feel like modules is very much a net positive to things. So the unpopular/controversial opinion is that it will be the last dependency management system. There will be no others.
+
+**Marcos Nils:** For Go, right?
+
+**Paul Jolly:** For Go, yeah. Well, other languages are welcome to copy... And guess what, there is another language that is going to copy Go's module-based Q is gonna follow the module and package-based approach. It's influenced extremely heavily by the way that Go and its tooling works; the modules will follow a very similar pattern.
+
+**Jon Calhoun:** It would probably be more interesting if Peter Bourgon (I think that's how you pronounce his last name) was here, because I don't know if he would agree with you on that one... I don't think he likes semantic versioning I think that's the bigger underlying issue. And if you get rid of semantic versioning, I don't know if Go modules would stick around or not.
+
+I do agree that sticking with something that's built into the Go tooling is enough of a net gain that even if you don't like 100% of it, you'll still use it... It's kind of like go fmt; even if you don't love that format output, there's enough net benefit from everybody using the same thing that it's worth doing.
+
+**Carmen Andoh:** But ten years later go fmt now becomes ossified into the way that modern programming languages add as a feature. We do not wanna fight over tabs and spaces... And also the other benefits that it provides.
+
+**Jon Calhoun:** I still fight with tabs. \[laughter\] The only time I end up doing that is whenever I'm copy-pasting code to put in -- I write an email newsletter; so when I'm doing that, it's easier to turn to spaces, so it's not real wide on the screen... And I know there's some CSS to change that, but it's just sometimes easier to have spaces.
+
+**Carmen Andoh:** But it's not Go... Right? You're in a free-form--
+
+**Jon Calhoun:** I write Go, I format it, and then I put it into my editor and I replace all tabs with two spaces, and then I put it in my email... \[laughter\] I'll just say "Oh, that's not me. That's the email processor."
+
+**Paul Jolly:** I think there are people who definitely don't like the way some of the more opinionated parts of modules... I think the thing that's been interesting, as I said, over the almost three years since Vgo was first announced is that it's really hard, it's a massive space to try and summarize what your objections to something might be, because there's so much context that -- I'll speak personally here... If I consider my opinion on it, it is limited to my experience of modules and/or packaging and/or version dependency management... And it's really difficult to communicate to a very wide group of people why you think in the widest possible context modules is or isn't the right solution.
+
+\[01:04:13.26\] So it's a tough space to win in an argument, I think... And I think that's where having a really good evidence-based approach to things is useful. And that's where having an experiment that was Vgo in the start, that people could try out, that they could give feedback on, and the Go versions that have happened since Vgo, which has been an iteration on the modules experiment, have been fantastic... Because the Go team have been incredibly involved in the tools working group, but also in the modules channel on Slack, getting feedback from people, understanding "Okay, what are the rough edges here?" and trying to think about solutions in that wider context of "Okay, you've got a specific problem here, but actually this aspect of modules is trying to address this wider problem here."
+
+So it's a very difficult balancing act, and one which if you look at a very specific use case, I don't think you're ever gonna say "Yes, modules got it absolutely right." But it's that wider context, which I find very hard to appreciate... Because guess what - I'm not at Google, I'm not at some of these larger organizations. So that's effectively the tough job that Russ, Bryan, Jay and everybody who's responsible for modules has.
+
+But I think there's been some fantastic stuff that's come out of modules; if you look at the pkg site; these are real wins, to my mind... And they don't demand lock-in into the system either; that's another nice feature of it - you're not locked into it. And I think the tooling is getting there. This is something the Tools Working Group, which over the last 2-3 years has been talking about this, is kind of trying to help move along a bit. It's getting there, but there is plenty of room for improvement there, too. And kind of to the point of Play with Go is that if the tooling improves, there's now (guess what) a nice way of showcasing it to people, as well.
+
+**Kris Brandow:** Yeah, I kind of feel like whether you fall on the side of liking modules or not liking modules has a lot to do with whether you're like a library maintainer or just a consumer... So I think historically, modules from a library maintainer's perspective have been really, really difficult. I think it is getting better now. I think the tooling and the knowledge about how to do things is improving... But even at work we have this problem with semantic import versions, and we try to switch something over to modules, and it wasn't using modules before, but it was already on v2, and then when we switched it to modules it was importing the old, old, old code... It was all sorts of broken, and it took us a while to figure that all out.
+
+**Paul Jolly:** \[01:06:46.29\] I really wish that we had Play-with-go.dev two years ago, where we could spin up guides that demonstrated to people "Hey, you're in this situation..." This was actually one of the examples that Bill Kennedy gave as feedback... He said "You know, you can do these as guides, but you could also present them to people as problem-solving situations as well." So you launch a guide, which is in fact a problem, and you explain the background (the problem) that says "Okay, you've got a module that's in this state here. Now upgrade it to a new major version." And how do you actually do that. So you can present challenges to people in this type of environments as well, as a way of enforcing learning.
+
+So yeah, Kris... The frustrations that people have - it would be great to just say "Okay, let's just write a guide from that" or "Let's write an explanation of how you get yourself out of a hole." Because guess what - that's what we should be doing, is providing examples, documentation to people to help them when they find themselves in a bit of a corner, that the tooling has landed them in, fundamentally.
+
+**Jon Calhoun:** That type of tool would help a lot with misunderstandings, too... Because I remember when Vgo first launched, people misunderstood how it worked pretty drastically, in some ways... And all it took was actually grabbing the tools and running it locally to verify that what you thought was a bad idea and it wasn't actually how it worked. But the number of people who didn't do that seemed astronomically high... Especially on Hacker News and sites like that where they might not have even been Go developers, and they're just like "This is a terrible idea." You know, people like to read the headlines and comment. So it could definitely help in that type of situation.
+
+**Paul Jolly:** Yeah, for sure. And that was, I think, part of the thinking behind Jay's module retraction guide; it just gives people a zero cost way of having a click through and saying "Oh, okay, this is how module retraction works. Fantastic. That's gonna solve my use case." Done.
+
+**Jon Calhoun:** Okay, so I think that about wraps it up. Is there anything else that you guys wanted to talk about, mention, to bring up, that I did not get to?
+
+**Paul Jolly:** Nothing. That just covers it from--
+
+**Marcos Nils:** Yeah, I think that's it from my side as well.
+
+**Paul Jolly:** We can say a big thank you to Carmen again.
+
+**Jon Calhoun:** Well, Paul, Marcos - thank you for joining us and explaining everything. Carmen, thank you as well. It's been a while since we've had you. Kris, it's good to have you on your first episode as a host. This is your third or fourth episode in total though...?
+
+**Kris Brandow:** Um, third.
+
+**Jon Calhoun:** Okay. It almost doesn't count at that point then, I'm sorry. \[laughter\]
+
+**Carmen Andoh:** You're old hat by then.
+
+**Jon Calhoun:** Once you've been on the show twice, it's just like "Yeah, he's just here all the time."

Pow! Pow! Power tools!_transcript.txt

@@ -0,0 +1,431 @@
+**Johnny Boursiquot:** Hello, and welcome to this episode of Go Time. I am Johnny Boursiquot. Along with Mr. Jon Calhoun, we are joined by a special guest, Thorsten Ball. If you've been in the Go community for a while, his name should sound familiar. He's the author of two very educational books on writing interpreters and compilers in Go. How are you, Mr. Ball?
+
+**Thorsten Ball:** Hello! I'm fine, how are you?
+
+**Johnny Boursiquot:** I'm doing well, I'm doing much better now because we're gonna be talking about tools, power tools. Before we get into that, Mr. Calhoun, how are you?
+
+**Jon Calhoun:** I'm doing well. Excited to get into this one, because there's a lot of interesting things here.
+
+**Johnny Boursiquot:** Oh yes, indeed. So, one thing we will not do - or I'm hoping we won't do, in talking about tools... and to be specific, we're talking about not just tools built in Go, although that's certainly part of it, and not just tools unique or specialized for Go development, but really all the things that make up your developer workflow. It could be editors, it could be containerization that you use locally, it could be how you do live reloading and testing, it could be your shell... All the tools that we as developers use to build and ship software.
+
+But going back to my original point - we will try very hard not to start any wars over things like editors and whatnot, because everybody has their favorites... And it's the same thing for every tool on this list here. Maybe you have something that you really like, and you think it's the best thing since sliced bread, and that's okay, too. We're here to talk about a wide range of things, and then hopefully tie it all back to our daily Go development.
+
+So let us start -- well, where to start...? This is such a broad field... We can start with editors, but I really want us to keep it -- not go too deep in the weeds a little bit. Personally, I use a combination of VS Code and Vim, depending on what I'm doing, the environment I'm in... Either one of these tools works for me. Thorsten, what do you think?
+
+**Thorsten Ball:** I would say the number one tool that I would need on any computer that I work with would be a terminal, a shell. I often think about this question, because it's an interview question where I work, at Sourcegraph, so I get to ask it a bunch of times and discuss it a lot... And I think my answer would be a Unix environment, a shell, where you can boot up a bunch of tools.
+
+\[04:08\] I use Vim, so I start up Vim a bunch of times, I put it in a background with Ctrl+Z, I run ripgrep, everything is in a tmux session... I spawn a ton of different shells, close them, spawn new ones, close them...
+
+When I tried to switch to VS Code, for example, this was the thing that I noticed the most - it's not a specific syntax highlighting or theme or whatever, it is the ability to quickly run stuff in the shell and close the shell. So that's my number one must-have tool.
+
+**Jon Calhoun:** I think it's interesting how everybody is very unique there. I see people who use VS Code and use the terminal built in... I can't do that. I'm just like, "I'm sorry." I want that terminal to go away and leave me alone, because it's just not what I'm using. Because I just have global hotkeys that bring up the terminal, and that's how I get back and forth between the two. But I use VS Code more than probably anything else.
+
+**Thorsten Ball:** You have a global hotkey and it brings up the iTerm...
+
+**Jon Calhoun:** Yeah, it brings up iTerm.
+
+**Thorsten Ball:** ...thing at the top?
+
+**Jon Calhoun:** Yup.
+
+**Thorsten Ball:** And then what do you type in?
+
+**Jon Calhoun:** Whatever I need to do, depending.
+
+**Thorsten Ball:** But how do you get the output, for example, back into your editor, or something?
+
+**Jon Calhoun:** What output do you want me to have back in the editor I guess? It kind of depends. Part of this is probably depending on what you're working on, too. If I'm working on a web server or something, there are a lot of ways... You can actually plug in the two, so that you can actually hold in Cmd and click on a line number, like a source file with a line number, and it'll open that up in VS Code. There are ways you can connect those things that they'll often work.
+
+**Thorsten Ball:** Yeah.
+
+**Jon Calhoun:** So you can actually jump back and forth and do things like that. The other option is just to sort of know shortcuts... So if you know it's this file and this line number, go back to VS Code and do whatever you need to do to get there.
+
+But I do agree with you fully that a Unix environment is something that is really hard to not have in some capacity... Because this is a question I get asked by all of my friends that aren't developers - they always ask "Why don't you just run a Windows machine? Why do you need this Mac/Linux/or whatever it is?" And it's so hard to explain until you go try to do some sort of development in Windows, and when you wanna do those command-line tools or anything like that it's like "I have to learn an entire new workflow", compared to what I'm used to and what everybody else seems to be doing.
+
+**Thorsten Ball:** It's hard to explain, right?
+
+**Johnny Boursiquot:** To be fair, the Windows development story from the lens of folks who are used to using a macOS, or something, or even a pure Linux environment - the story there in the Windows environment, I hear... I'm not a Windows user; I have nothing against Windows, I'm just not a Windows user... But I hear that story has gotten much better. I hear folks talk how happy they are with WSL, the Windows subsystem (I believe it's called) for Linux...
+
+**Thorsten Ball:** Yeah.
+
+**Johnny Boursiquot:** ...and it provides -- you can't quite tell that you're not in a Linux environment. From what I can tell, it's pretty much very similar, if not close to being the same thing. So I hear that story has sort of evolved a little bit. Every now and then I'll sort of peek at "Hey, what are the folks on the Windows playing around with these days?" But to me, the issue I've always had with Windows environments for tooling has really partly been driven by the development community itself.
+
+A long time ago, when I first started doing coding and whatnot, all the tools that I needed to use were being built for Linux or Mac machines. So that sort of drove me towards adopting these tools, because that's where all the best tooling - at least for doing command-line kind of work, which I really enjoy doing - was being built, at least first. Over time, they were being ported, and... You know, with the Windows environment it always felt like we had to use some sort of a graphical user interface to use these tools.
+
+\[07:56\] The moment I got a taste of the CLI, I was like "Okay, I wanna do everything in the command line." And Windows back then was just not -- and I'm talking like Windows '98, Windows 2000... To this day, Windows 2000 still remains my personal favorite, because that was the last one I used before I jumped to using a Mac... Actually, I went through Ubuntu first... But yeah, it feels like - and I'm sure some folks that are hearing this later are gonna corroborate that, and some folks in the channel with sort of jump in as well... The Windows story for tooling and development has gotten much better over the years.
+
+**Thorsten Ball:** Yeah.
+
+**Jon Calhoun:** When you look at things like Docker too that are coming out, they've all sort of helped push that... Because these days, anytime somebody asks -- even on a macOS, there are easy ways to install things like Postgres, but if you are familiar with Docker, there are much easier ways to just spin something up that's running Docker than installing it.
+
+So knowing that that'll work on any operating system if you understand Docker is really helfpul, and I think that also really helps support that argument you're making where all these operating systems are kind of getting close to parity.
+
+**Thorsten Ball:** I wonder whether Go itself has had some impact there... Cross-compilation is really easy to do, and I released a new CLI today, or a new version of our CLI today, and there's a Windows build included... And I'd never thought about it. It's just done automatically.
+
+If I compare this, for example, to Ruby, which I worked in for a long, long time, this was never really supported, and since it's not a compiled language, you always have to bring the interpreter. You basically need to wait for the interpreter to build in all of these features. It's a slightly different thing, and I wonder whether cross-compilation in recent years has maybe allowed more people to use dev tools on Windows. Just a thought...
+
+**Jon Calhoun:** I definitely think that's true... I remember the first time I tried to learn Rails I didn't know the Unix environment that well; I was still relatively young... And I tried it on a Windows machine and basically gave up after a while. I was like "This is not working on this machine. I need to get a Linux installed before I can even mess with this." And I gave it up for at least a couple months until I came back to it, and I actually had a Linux machine to do it at that point. So I know for some things having that ability to compile on different platforms - that definitely helps.
+
+**Thorsten Ball:** It's funny, right? ...we said we don't start with the editors flame wars, but now we ended up talking about operating systems... \[laughter\]
+
+**Jon Calhoun:** It's not a flame war though, it's more of a... It's nice to see they're all coming around and you can do everything.
+
+**Thorsten Ball:** It's one of the two hot topics of the internet in 2005, or something... Either editors or operating systems.
+
+**Johnny Boursiquot:** Or operating systems... Well, you did touch on the containerization stuff, and we were talking about Docker - I was thinking in my head, I remember there was a period when I used Vagrant a ton. Vagrant was, in my mind, the original tool that allowed me to have separate environments, and back when I worked at the agency I had different environments for different customers. I could boot up an environment and have everything - that particular customer - the tools, the products and services that we were working on for them, I could separate these things and easily spin up an environment to work in that particular project, or that particular project... So for me, before there was Docker, there was Vagrant, and I still look back fondly on that project...
+
+But yeah, these days you can really -- with containerization you can kind of... Any sort of environment you want, you can kind of put together, combine them, and remix them however you need. It's a brand new world we've been living in. Well, how old is Docker now - 6, 7 years old? I can't remember. But yeah...
+
+**Thorsten Ball:** Yeah, seven.
+
+**Jon Calhoun:** Some of us are slow to adopt fully... \[laughter\]
+
+**Johnny Boursiquot:** We're the laggards on the hype curve?
+
+**Jon Calhoun:** \[11:54\] I've lagged a little bit with some of it. I mean, mostly because a lot of my projects are just me, so I don't experience the same problems that teams experience... But I have been on teams where -- I worked on one Rails project where one of the libraries we required got pulled from a package manager, or wherever it was... I think it was on Brew, or something, but it got pulled.
+
+So every new person who came to the team would go to get set up and it wouldn't work, because that wouldn't be there, and the newer version would break something... When you experience that, you're like "Okay, I understand why this exists, and I definitely want it", but by the time Docker came around, I was no longer on that project so it didn't matter as much. Sometimes you don't learn and you're like, "Meh, I'll wait till I need it", and then when you need it, you're smacking your head off the wall.
+
+**Thorsten Ball:** Yeah. I don't wanna go off on a rant here, but what you said, as in "I never experienced a problem, so I didn't get it, what it's for" - I think this happens so many times that people just don't add the context of the thing that they're describing.
+
+Five years ago, people were saying "Oh, Docker is the cure for everything. You've gotta use containers." This was one side of the debate, and the other side was "I don't get it. What's so hard about installing this and that, images or something?" and what they're missing is Person A is working for 15 different clients, and in 15 different environments... Person B has a small, highly-skilled team of developers that build their own AWS images, that they upload every day, and they have a build process for it, so it's guaranteed to have the same stuff on every machine. And all of this is lost when they talk about the same thing, and they just talk past each other...
+
+That happens all the time in programming, where people say "This is the best thing of all time." It's a person working for a 5,000-developer company, talking to another person working in a little agency with 3 developers. They go "You should use this", the other person goes to their boss and says "Somebody on the internet said we should use this."
+
+**Johnny Boursiquot:** \[laughs\]
+
+**Thorsten Ball:** If you could just put them in the same room, where they would have to carry a nametag with them, like "I work for company X, with so many employees, and we do this and that", that would clear up so much stuff.
+
+**Johnny Boursiquot:** You need context. When you wanna bring something in, you don't wanna just say "Oh, so-and-so uses this thing, so it must mean it's good." The fact that some company or somebody here or on the internet uses a particular thing, that doesn't mean it's necessarily good for you. They had their reasons for choosing it.
+
+There's this meme out there of folks using Kubernetes around blogs, or something... It's like having a giant 18-wheeler, just carrying a small little package in the back, on a truck bed. So it's like, again, using the right tools for the job, which is - I think we've touched on this in previous shows as well... There's so much of -- you see other community members doing something, or maybe somebody you respect/admire, you see them using a particular tool, or a particular technology, and in the back of your mind you're thinking "Well, I should be more objective with my selection criteria... But so-and-so is using it", so you start using it too because of that. There's this peer pressure that we don't wanna acknowledge; that's always in the back of our minds, and sometimes it acts on our decision-making without us even realizing it.
+
+**Thorsten Ball:** Yeah. I openly admit that I still like using the terminal, because I thought it was cool when I was 14 years old. That's probably the main reason... \[laughter\]
+
+**Jon Calhoun:** I think that's also why -- you get people who like Vim, or anything like that, and they stick with it once they've learned it, because it's productive for them... Even Johnny -- like, if I'm using VS Code, it's not that I think it's better than Vim, it's that for me particularly it's more productive, because I don't have to learn something new.
+
+Speaking of that stuff, and things where people jump around, let's talk about some tools that you guys used for databases, and that sort of thing. I say "speaking of that", because you see people jump from one database to the next hot one... Whenever MongoDB came out, I remember every tutorial and everything on the internet was MongoDB... You know, the NoSQL phase, and all of those phases. I'm curious, what are you guys using? When you're building a new project or doing something like that, what sort of tools do you find helpful on that front?
+
+**Thorsten Ball:** \[16:15\] Postgres as the database. I don't know how it's pronounced, actually. I got confused a couple days ago. Somebody wrote somewhere it's PostgreSQL, or something...
+
+**Jon Calhoun:** I think it's usually referred to as Postgres, but it is like--
+
+**Thorsten Ball:** Yeah, Postgres.
+
+**Jon Calhoun:** SQL is in the end of the name, I believe.
+
+**Thorsten Ball:** Yeah. The one with the elephant...
+
+**Johnny Boursiquot:** \[laughs\]
+
+**Thorsten Ball:** Yeah, that's the database of choice for me, for at least the last 6-8 years... And on Mac I use Postgres.app, which is this neat Mac app in the status bar at the top, where I can spin up a bunch of different Postgres versions, like 9.6, and 11, 12, and whatever... And yeah, control them. And then on the command line I have the psql, the command line client for Postgres... Which can be configured, if people don't know this; read in the dotfile. You can set a different prompt, you can turn on autocompletion, you can turn on different pagers, which change the way query output is printed... That's pretty neat, and you can customize it. That's basically it for database needs on my side.
+
+**Johnny Boursiquot:** So we're not at the Unpopular Opinion segment yet, but I'm not sure if what I'm about to say is unpopular or not, but unless I have a requirement to have some sort of a distributed system where I need my data highly available, and I need the different instances of my services or backends to be accessing the same data, I'm okay with shipping a single-purpose -- basically, one program, and shipping a flat-file, or maybe using BoltDB, or Badger, or something like that to have my application use just a local file to serialize data and read data back out of.
+
+You'd be amazed how many -- at least of your own personal, but also some production applications that don't need a distributed data store. You'd be amazed how far these things can take you. You don't always need a giant server to handle your storage, you don't always need that. Sometimes I think we just default to that out of habit, but you don't always need that... And it can actually simplify your deployment story if you just say "You know what - this thing needs to just read some file locally, and it'll serialize information back down, ease it and that's it. You move on.
+
+**Thorsten Ball:** That's really interesting... Do you have one process usually that reads in a file when it boots up, and reads it in memory?
+
+**Johnny Boursiquot:** Yeah, during start, typically, when I'm initializing my applications, I do flag-based and environment-based configurations. So I read in those flags, I read in the environment variables, and this is where the location of the file, and basically during initialization I just read in things, and if I need to deserialize whatever it is, I do it then. Usually, for the lifecycle of the application, basically there's just one thing in charge of writing back to storage.
+
+Then if the application were to shut down, it'll handle signals gracefully; it'll just clean up after itself, make sure the file gets written to and closed properly...
+
+**Thorsten Ball:** Can you give an example of which kind of data you write into the file? Is it a line-based record?
+
+**Jon Calhoun:** I can give one example... When I first launched Gophercises, which was just like a free coding course type thing, my user system was very basic. You basically signed up, I would email you a URL that would log you in, and that URL would basically stay the same forever for each person... But each person had a unique URL to log in. And that entire system was built with BoltDB backing it.
+
+\[20:07\] So while I did have to write users, it was such a rare occurrence that it didn't matter that it was BoltDB, because 99% of the time I was just reading data... And reading when a user logged in, or reading if they were trying to access different course materials, things like that. So it worked really well... And I wasn't hitting some millions of users scale; I think the highest that setup ever went to was like 25k-30k users, but it was never concurrently. It was spread out. And because of the way the app worked, people refresh a page once every ten minutes when they're done with the video. It's not like they're hitting multiple pages... So it just worked really well for that, and it made deployments and everything like that really easy. Even backing up the system was as simple as copying a BoltDB file and being like "Okay, I've got a back-up of my database now."
+
+**Johnny Boursiquot:** One of the little pet projects that I'm working on right now is writing a port scanner, and basically just being able to ship a binary and say "Hey, you write the results of your port scans to this BoltDB database." Then at some point I send that to an S3 bucket, I just package it up as an object, store that, and I get the guarantee that S3 is not gonna go down on me... At least that's more reliable than anything I can build.
+
+And even if I have a highly concurrent program, I sort of have a fan out/fan in approach and all my data gets written safely in. I don't know, it just works...
+
+**Thorsten Ball:** Would you say you don't need querying then, or indexes, or something? It's just a data dump, in a sense?
+
+**Johnny Boursiquot:** Well, you can have more sophisticated stuff. I can think of Bleve, for example; it's a goal-based indexer. So if you need searching capabilities, you can absolutely do that, and Bleve itself happens to support multiple storage back-ends... And BoltDB happens to be one of them. I'm sure there's a ton more. So chances are, if your use case, again, is not one where you require a highly-available distributed system, then storing on your local file system is fine, it's okay.
+
+**Thorsten Ball:** I mean, I would say this breaks down as soon as you have multiple processes, even on the same machine, because then you would have concurrent writes, or reads to the same file, you need to share them between different processes, so...
+
+**Jon Calhoun:** I mean, even BoltDB - you can't have two open connections to it at once. It doesn't let you do that. So you have to know from the get-go that you're either designing a system around that, so one process loads up everything, or you have to consider the fact that if you get to that point and you need to change what you're doing... But I do agree with Johnny that there are a lot of cases where people jump to these -- I mean, it's the same when you see people jumping to the NoSQL solutions when they really didn't need it. Consider your possibilities and what you expect that application to do, and if that works fine for your use case, then go for it.
+
+**Thorsten Ball:** I have to say, I'm super-fascinated. This is to me like somebody saying "Yeah, my car doesn't have an engine. I just have a little hole at the bottom, I stick my feet out and just walk..." \[laughter\]
+
+**Jon Calhoun:** I mean, I can tell you - I've actually seen things in production with SQLite powering them...
+
+**Johnny Boursiquot:** Yup.
+
+**Thorsten Ball:** Yeah, I mean... Hacker News is file based as far as I know... So I think it comes down to what exactly do you do with the data. Hacker News, for example - you have posts, comments, whatever; it's a pretty simple hierarchy. You can see how you can map that onto a file system. Every post is a folder, every comment is a file, whatever. Then you can use the file system... And the file systems nowadays are incredibly optimized and fast. We have SSDs... As soon as you can query the data you want without having to use an index, which is one of the big benefits of a database, then I guess you're fine. But as soon as you need different queries, and you need to group data, and you want it to be performant without loading it into memory - I guess then you need a database.
+
+**Johnny Boursiquot:** \[24:07\] Yeah. If your needs extend to wanting to actually slice and dice a data, then I'd say "Hey, get the data out into some format you can actually work with." If you need to do SQL queries on that thing, dump it out; have some sort of transformer that takes the serialized format that's in the file and converts it into records in a database. Or if you wanted to use some NoSQL, whatever the latest and greatest thing is, that everybody is jumping onto, if you wanna do that, then write a transformer for that, too.
+
+So you can transform the data into whatever format is optimal for your use case, for what you're trying to get done.
+
+**Thorsten Ball:** Which brings me to another favorite tool of mine, jq.
+
+**Johnny Boursiquot:** Yup.
+
+**Thorsten Ball:** You could basically write a database with this, right? You could dump your data in a JSON file, and jq is this little command-line utility that allows you to query data in a JSON file, or in a JSON input stream, whatever... And you can map over every entry, say "Give me this field, give me that field, create a string out of these two fields", whatever. I use this ten times every day.
+
+**Jon Calhoun:** I assume you're working with a lot of APIs that are giving you JSON data then...
+
+**Thorsten Ball:** Yeah.
+
+**Jon Calhoun:** Okay.
+
+**Thorsten Ball:** Or just simple -- for example, at Sourcegraph our configuration system is JSON, so you have an editor where you can put in JSON... So there is a lot of JSON involved, and I actually have -- it's really trivial, but I have a mapping in Vim that pipes the JSON that I just selected to jq... So it formats it, and then you can pipe out and query it with jq, too.
+
+**Johnny Boursiquot:** Cool.
+
+**Thorsten Ball:** So I get a lot of mileage out of having jq installed
+
+**Jon Calhoun:** See, that type of stuff is interesting to me, because I've found that it really depends on what your current project is. Jq is one of those tools that I've used in the past, and when I used it, I used it a good bit... But then when I wasn't using it, I found that I would have a new system and wouldn't install it for months, because I didn't realize I needed it... And then you go to use it and be like "Why isn't this installed right now?" It would just be a confusing thing.
+
+Some others that stuck out to me like that were things like Postman, and Paw, and a couple other tools that are... I think ngrok was one that does -- so Postman and Paw, for anybody unfamiliar (we'll try to put them in the show notes) are tools that are used to make API queries into... They're basically really good tools for testing and asserting different things with APIs. Then ngrok was a tool that I've used that does -- it's like a tunnel... Is that what it--
+
+**Johnny Boursiquot:** Tunneling, yeah.
+
+**Jon Calhoun:** Basically, it allows you to expose some local host process to the web... And one of the processes I used that a lot for when I did web development stuff was anytime you'd need webhooks and you wanna test them locally, you want a tool like that where you could spin up your local environment, expose it to the web, get a webhook and actually use it... Because otherwise it was a nightmare to say "Does my code work?" And the last thing you wanna do is "I'll ship it to production. It'll work..."
+
+**Johnny Boursiquot:** \[laughs\]
+
+**Thorsten Ball:** Ngrok is one of these tools you recommended to somebody if they're working with webhooks, for example, and they always say "This is amazing. I just boot it up, it works. I get a little URL... This is my web app now, exposed to the internet securely etc. How could I live without this?" Then you don't have to work with webhooks anymore, and suddenly you never use it again. It's one of these tools, as soon as you need it, it fits the use case perfectly.
+
+**Break:** \[27:43\]
+
+**Jon Calhoun:** Another one that I've noticed along those lines... So there's the new GitHub CLI tool that came out. And when it came out, it reminded me of the hub tool which I found -- again, one of those ones where like when I was creating a lot of new repos and doing certain things like that with the team, I found myself using that tool more often. Then later - I don't know why - I just stopped using it as much. I don't know if it was that I wasn't interacting with a large team as much, so I didn't really have as much of like an issue process, and all the other stuff... I could just use Git and figure out things that way. I'm curious if there's any others along that line.
+
+When you guys are working with source control, what tools do you find useful? Source control, and I guess maybe project management type stuff with your team - what tools are you finding useful in that front?
+
+**Johnny Boursiquot:** I have this -- I don't even know how to describe it. Every time I use -- I've used Hub, definitely, in the past... And everytime I use one of those tools, which is almost really if you think about them, they're just like some sort of abstraction, or some sort of a wrapper around tools you already have locally. I always feel like somehow I'm cheating. I always feel like "You should know the actual wizardry and flags and things that the git command lets you do", I always feel like I somehow am not giving the appropriate level of attention that I should when I use these tools... And a lot of times I've stopped using these tools, because I'm like "You should know how to do this with the actual tool, and not get a wrapper, or some sort of thing around it", or something. It's a weird thing, but I do experience that sort of feeling.
+
+**Thorsten Ball:** I use Hub a lot, or at least -- the new GitHub CLI has the same feature... I use it to list issues, basically. To talk to the GitHub API. I don't use it to create PRs or switch branches, or whatever... I do all my Git locally, with Git itself, on the command line. I have a bunch of helpers configured to have nice log output, a few aliases for faster git checkout, create feature branch, whatever, all of that stuff... And that's basically it.
+
+I never used one of the graphical git clients. I'm weirdly proud of it, even though I know I probably should have used one of them when doing an interactive rebase, or whatever... Or you know, you've ran into some weird conflict when merging one branch into the other and rebasing it en-masse, or something... But yeah, I do like it on the command line, and I've found that - you know, I'm speaking from the experience of sitting next to a colleague who used Git Tower (I think it's called), the graphical client for Git...
+
+He often had problems when he tried to rebase or he pushed to a branch or something, and stuff broke... And it was a combination of Git Tower in the back, automatically fetching new things, but him assuming "Oh, it's safe to rebase, because I haven't fetched the new stuff...", and then doing a manual rebase on the command line, but the graphical client getting in between him and the Git history and breaking stuff... And I'm not saying I'm right in what I'm doing.
+
+**Johnny Boursiquot:** \[laughs\]
+
+**Thorsten Ball:** \[32:19\] There's some (I think) advantage to knowing the abstraction layers, in a sense. You know, I haven't used it, so I can't really speak from experience, but I can imagine that if you do stuff in a graphical Git UI client that abstracts a lot of stuff away, it is as easy to shoot yourself into the foot as with Git on the command line... Because you don't actually know what's happening in the background, and for some use cases you might need to know.
+
+**Jon Calhoun:** I view it as similar to people who use a framework or an ORM when they don't know SQL, for example. It's one of those cases where it's great to get you started and get you familiar and get you some sort of framework to learn from, but you shouldn't stop at that point. You should look at other ways to use it, and maybe sort of get slightly familiar with the command line, so that you really understand what it's doing behind the scenes... And then I think from that point on you can sort of branch out and actually be a little bit more sophisticated with what you're doing.
+
+**Johnny Boursiquot:** When you said that, that immediately raised a flag in my mind... I'm having flashbacks to the days where I used to use WYSIWYG tools to build websites. If you've ever used HoTMetaL PRO, or DreamWeaver, or all these things...
+
+**Thorsten Ball:** FrontPage...
+
+**Johnny Boursiquot:** FrontPage... Oh my gosh, FrontPage... These tools - presented this layer between you and the actual code and the actual markup and scripts that got written... And at some point I was like, "Okay, if my objective is to get something done as quickly as possible, the graphical user interface is going to help me there." 99% of the time it will do the right thing, but in that 1% where it fails, I'm gonna have no idea how to actually fix it, and I'm more likely to just start over at some point and try and get it right, than I am to get under the hood and figure it out. But that is with the objective of getting something done as quickly as possible.
+
+Whatever it is that I'm trying to get done is not an area where I desire to develop mastery or expertise in. If that area is something I really want to understand and master, then I will go through the pain of actually getting under the hood and doing the reading - reading the docs, reading the manuals - basically getting to the esoteric knowledge that allows me to become a master in that thing. But again, we have 24 hours in a day, and I don't know about you, but there's only so much I can spend time mastering, so I have to be deliberate about which thing am I willing to spend a good deal of time mastering... And these days, that field has narrowed for me quite a bit.
+
+So if it means using a graphical user interface to get something done that I don't care to master or has no real impact on my livelihood as a professional, then I don't mind using the GUI. I'm fine clicking around in an interface somewhere and getting things done, so I can get back to the things that I actually care about. That's the lens that I bring to it, I guess.
+
+**Thorsten Ball:** Yeah.
+
+**Jon Calhoun:** Yeah, that definitely makes sense. I guess when I view the GUI stuff -- I've never used it for an extended period, but it is something where I've suggested people... Until you understand what Git is useful for, it's sometimes nice to look at it in this graphical view and just sort of get a feel for "Okay, what is a PR? What are these changes?" and then from there you can go back and start looking at the command line and looking at how things are different.
+
+For example, I don't know how you guys typically do your PRs, but one of the things that I've been really adamant about, I guess, is that I like one commit to sort of be one complete change, like one idea... So where some people do a PR that's actually composed of ten commits, I will actually in my own personal history squash that all together into one big commit. And there are some problems with that, depending on what review tools you use, and things like that... But it is very helpful in some situations.
+
+\[36:14\] If you've ever been on a team where you're trying to figure out where was this bug introduced, or when was this feature implemented or something, sometimes having that "every commit is one complete thing" is really helpful. I think that's something that'd be hard to explain to somebody the differences and the nuances, unless they have a broad overview of how Git works... And sometimes the graphical stuff can be useful to get that across. But then, like you said, if you really wanna get that mastery for it, you're unlikely to get that in a graphical user interface.
+
+**Thorsten Ball:** First of all, I have to add that I was really anal about my Git history. Perfectly written commit messages; I did fantastic commit messages. I put so much love into them. And now at work we do squash and merge of pull requests. So they're all gone, like tears in the rain...
+
+**Johnny Boursiquot:** \[laughs\]
+
+**Thorsten Ball:** ...so now I just don't care anymore. It's like "Fix? I don't care..." They're gonna just put stuff up, it's gonna get squashed anyway.
+
+**Johnny Boursiquot:** "Fixed stuff." \[laughter\]
+
+**Thorsten Ball:** Yeah, "stuff", "fix", "let's try again", "wtf"... But I wanted to say... I think there's another dimension to it besides mastery, and that is these tools - we can say graphical tools, but what I mean is IDEs or integrated environments; you know, FrontPage is what triggered this thought... These tools are great as long as you stay in them. But as soon as you have to break out of them, you're lost. Because they get their value from being integrated. And just an example that I thought of, because you mentioned DreamWeaver, I used FrontPage back then...
+
+So you would code up your web page in FrontPage, clicking all of the stuff together, moving it around, layouting it, and then you search on... What was it called back then - Altavista, Google? You search for "DHTML mouse cursors that follow your mouse", or something. And then the tutorial says "In your HTML add this to your head and add this to your body." And then in FrontPage you try to figure out "How do I get to the HTML?" and you get to the HTML and then it turns out "No, the head looks totally different here than it looks in the tutorial. It's not even close."
+
+I think this is the same when you talk about IDEs, for example, where you have a nice interface to configure your build, or your test, or whatever... Like, which flags etc. As soon as you want to run another tool in this process, you're lost, because now you have to go around the tool and try to inject your own thing into it... Whereas if you started with combining your tools, you might have an easier time. But that's just a really generic thought on... I don't know, maybe it's about using tools from the bottom up and combining them, or just starting from the top down and just replacing what you don't need.
+
+**Jon Calhoun:** I think some of it comes from figuring out which of those tools are the ones you need to replace too, though. Because like Johnny said, we don't have enough time to use everything. And even when it comes to tooling, one of the things I'm very guilty of, and I'm curious if you too are as well, is I will install all these different things and be like "Oh, this is gonna help me be so productive." I'll set up keyboard shortcuts and all these things, and a month later I'll have forgotten 95% of them. And that 5% that I've retained is like "Okay, cool. That's really helpful." But I'm still just like "Why did I spend all that time setting all these other things up, when I just don't use them often enough for that to stick in my head?" Whatever problem I thought it was gonna solve wasn't a problem that was big enough to justify a tool.
+
+**Johnny Boursiquot:** \[40:03\] I have a guilty conscience with some of the tools that I use... Because I've found that sometimes I use the hunt for new tools to do things perhaps I already know how to do one way, albeit "the long way", or something... I'll go on a hunt for tools that helps speed things up. And when I sit back and I reflect, I'm like "Why did I spend two hours of my time? Two hours that I know I could have spent somewhere else, on something more productive. Why did I spend this time experimenting with these other things, with these other tools that would shave maybe 5 seconds on something/whatever I'm doing now. I've figured out that I use that as a form of procrastination to not do the actual work.
+
+**Thorsten Ball:** That sounds bad, but I would also add that... It's fun.
+
+**Johnny Boursiquot:** \[laughs\] It is.
+
+**Thorsten Ball:** If you can shave off a couple of seconds... And if you enjoy it, that is good, because it gets you to do more work in the end. Or do more of the things that you wanna do... As in, you have a guitar - hang it in the background; I also play guitar. And if you talk to people who play guitar for a long time, they start talking about equipment, and it's the same thing. They go out and they chase this dream of "This next amplifier is gonna give me that sweet sound, and then I'm gonna finally sound like \*whoever\*. That next pickup on the guitar is gonna give me that" etc.
+
+One thing I recognize this - as long as it gives you fun or joy while you practice or play or do your work, it's good, because it gets you to do the thing more often.
+
+**Jon Calhoun:** I've also found when you're looking at those tools, it's so hard to decide which ones you're actually going to use, and which ones are going to be the 95% you don't ever touch. There have been a couple tools where I've set them up or wanted to use them and thought "Eh, I might use this. We'll see", and that's the only one I end up using out of everything I installed that day or that week, or whatever the timeframe was... And I'm just like "How was that the one that I'm using all the time?" and you wouldn't expect it to do much.
+
+One silly example is I went at one point and set up a script just to make sure when I run go test that it has colors. And I think what I did was anything with a file and line number is yellow, anything that's like a fail is red, and anything that's a pass is green. And it's just a silly little Bash script, nothing crazy; when I did it, I'm like "Oh, this will just be nice to have. No big deal." And now I find that having those colors makes it much easier to read tests, to the point that when I run go test without it, I'm like "Oh man, I've gotta change whatever's running the test to run it with colors."
+
+It's silly, but at the same time that's just something I've trained my brain to look very quickly and find "Okay, where's the actual line number in the test?"
+
+**Thorsten Ball:** I think tools should come with -- or you should be able to install them with an expiration date.
+
+**Johnny Boursiquot:** \[laughs\]
+
+**Thorsten Ball:** So let's install this, and if I don't use it in the next 30 days at least five times, you remove it. That would be cool.
+
+**Jon Calhoun:** I honestly feel like project management tools need that for anything you put in the task list.
+
+**Johnny Boursiquot:** \[laughs\]
+
+**Jon Calhoun:** I've been in so many teams where you put everything in the backlog, and then there's things that are like six years old in there, and you're like "We're not gonna do this. If we haven't done it in six years, it's not happening."
+
+**Johnny Boursiquot:** Right. Just put a TTL.
+
+**Thorsten Ball:** Some person will say "But we might get to do this; let's just keep it in there; let's not delete it."
+
+**Johnny Boursiquot:** Someday. Someday.
+
+**Thorsten Ball:** Yeah.
+
+**Johnny Boursiquot:** Every task and every JIRA instance out there in the world, or Trello boards or whatever the cool kids are using these days - everything needs a TTL. After a while just eject it.
+
+**Jon Calhoun:** Even if it pops you a little reminder that said "This is about to expire. Is it something you actually need to do?" Something to make you consciously say like "Yes, we're gonna do it" or "No, we're really not gonna do that. It's fine."
+
+**Thorsten Ball:** It should be the other way around. It should delete it and say "I deleted this. If you really wanna bring it back, send me an email."
+
+**Johnny Boursiquot:** \[laughs\]
+
+**Thorsten Ball:** Actually, I think Trello... They started -- if you had a Trello card in a Trello board and you didn't move it for a couple of days, it started to show cracks, or something. It aged--
+
+**Johnny Boursiquot:** \[44:20\] Oh, some sort of visual cue?
+
+**Thorsten Ball:** Yeah.
+
+**Johnny Boursiquot:** Yeah, "Maybe you don't really care about this thing you think you care about."
+
+**Thorsten Ball:** Yeah.
+
+**Jon Calhoun:** See, that sort of tooling is hard to figure out, because I think for some people the tool -- it's not about actually having the task in your task management system, it's about getting it out of your brain and somewhere else... And the only time you can get it out of your brain is to write it down or put it down somewhere... Because I know I'm guilty of this. I will write things down that I know the chances of me doing this in the next two years are virtually zero; but if I don't write it down, I'm gonna think about it randomly for the next two weeks, and I don't wanna do that.
+
+So it's almost like you need some way of handling that, whether it's that aging process, or something... But it's hard to find tools that take that into account, I think.
+
+**Break:** \[45:13\]
+
+**Johnny Boursiquot:** That's actually a good segue into the non-developer-related tools that we use. We touched a little bit on the project management stuff, but that's gonna be part of lives, right? So if I sit down to do work for eight hours for my employer, a lot of things are happening. It's not like I'm sitting there from 9 o'clock to five o'clock without moving, without doing things, without having the world trying to get my attention for something... There's a bill that needs to get paid, I need to pick up my kids from the bus stop... All these things are always going on, and like you say, John, you need to be able to offload things out of your mind.
+
+I'll be in the middle of writing a lot of code, and all of a sudden I remember something that I need to do in the house somewhere. I'm like "Oh, I forgot about that thing", and I need to quickly be able to make a note about it somewhere and get back to what I'm doing. That's how our brains work. it doesn't give you a choice as to when things pop up. Heck, books have been written about how we try to facilitate our brain to get into flow... But we have no control over that, so we have to be able to adjust and roll with the punches, so to speak, that our brain keeps throwing our way, to sort of still manage to get things done.
+
+\[47:54\] So along those lines, I use a test management tool called Things. It has a nice keyboard shortcut, it quickly brings up a little HUD display, I quickly type something in, and I hit enter, and it's not on my brain... I think. I hope.
+
+So things like that, like task management - that's something I rely on quite a bit. And I use my calendar, believe it or not, to actually track my time. What I'm working on when, where does my time go... At the end of the day I go through a process where I look at my day and I'm like "Hey, what did you get done today?"
+
+If I don't have things on my calendar, I account for my time. That's my biggest pet peeve, not knowing where my time went... So I use the calendar to track that stuff. At the end of the week, I'm like "Okay, you had a productive week" or "You didn't. Do better."
+
+Things like that - I find that the world around us affects our ability to be productive and get things done as we develop code... But these things too require tooling, I think.
+
+**Thorsten Ball:** Now that you've mentioned it, I'm constantly surprised at how effective it is to just dump stuff out of your head into a to-do list, a notes file etc. Every time, it's the same. You go "I don't know how to manage all of this. It's so much, I cannot keep track of this." You get really anxious and you get stressed out. Then I go "Okay, I need to write down what I actually need to do." Then you end up with ten bullet points in a markdown file somewhere, as I do... And it's so much better. Then you look at the list and you go "Okay, it's not that much. Seven out of ten of these are not high-priority. I can do these next week, whatever. Out of the three of them, this one is number one. Let's just work on this."
+
+I'm super-surprised how effective this -- isn't this from Getting Things Done, or something? ...where one of the big rules is just "Dump the stuff before you actually try to do it, or try to tackle it. Just dump it somewhere and then try to tackle it when you have the time."
+
+**Johnny Boursiquot:** Yeah, get it out of your head,
+
+**Jon Calhoun:** I will say I have learned that dumping them in the wrong spot is a terrible thing to do. For a while, what I would do is I would just email myself. If I found an article I wanted to read or anything, I just emailed it to myself... Because my phone and everything had an email client, so I'm like "Alright, this is an easy way to get it to myself."
+
+Then I found later that cleaning my inbox or prioritizing things was just a nightmare, because I had all these things in there that I'm like "Well, I'm not ready to archive this, because I haven't read it, but I really don't have time to read it today..." It was just like this massive mess, where I was just like "Alright, I need to get all of this out of my inbox and be able to clean it up." That plus having a snooze feature in Gmail has been something that have really helped me clear my head, in that sense.
+
+So if I have something where I'm like "I wanna deal with this email in a week", snoozing it for a week is really useful. Or if you have a flight, and you're like "I want this information, but I don't wanna think about it until then", that type of feature has been really useful.
+
+My wife has always yelled at me because I would sit in the kitchen and be like "Here are the things I need to do..." and she'd be like "Why are you telling me?" and I'd realized I was offloading to her... And she was like "This does not help me de-stress." \[laughter\] I'm like, "Yeah, I need to not do it with her..."
+
+**Thorsten Ball:** Yeah. Did you actually find that if you lose things, that it's a problem?
+
+**Jon Calhoun:** Like an email?
+
+**Thorsten Ball:** I'm saying this because I have a to-do list in Todoist, which is the tool I use... I call it my backlog. It is basically all of the dreams, it's a list of things that I would do if I was the person that I try to be, basically. A couple months ago I was honest with myself and said "I'm never gonna do this" and I just deleted the list.
+
+It turns out I do this a bunch of times, I delete my to-read, all of the books I wanna read, all of the music I wanna listen, podcasts I wanna listen to, all of this... And I delete it every half year, or something, and it turns out I'm not missing anything. If it's really important, either I do it right away or it stays somewhere in my conscious for the next week or so. Otherwise it's just not important. Or if it becomes important, it will pop up again.
+
+\[52:12\] I don't know if it's anxiety or whatever it is that you feel stressed out, or I feel stressed out, as in "Oh no, I cannot delete my to-read list", or whatever it is. Like, 50 books I've been collecting as "I want to read these for three years." It turns out no, I'm not gonna read them, and it's okay if I just delete this list.
+
+**Jon Calhoun:** I think generally speaking you're right, getting rid of the list is completely fine. One quick example where it was not that was I started doing an algorithms course and I knew I was gonna be doing this, but I had to finish a couple other projects. So I'm like "It's like six months out." So when I'd come across something where I'm like "Oh, this is a really good inspiration" or something that I thought I'd wanna read it around that time, then I needed to have a list for that specific thing. But usually, I was like "I know exactly where this gets categorized." So it just came down to like "I need to get it to something that's for when that time comes." So it wouldn't be like a backlog task list, it would be like "This course, reading list", so I'd actually know how to classify it.
+
+I agree with you, if it's just like a generic backlog, chances are you don't need it. Where I ran into issues is that I was just emailing myself all of these links, and some of them needed to be put somewhere and some of them were just things I was just never gonna get to reading... So it was kind of hard, because I just had too much going on at once.
+
+So what I've started doing more recently is I just have a notion board is a list of just reading links. When I see things, I'll throw them in there, and every so often I'll go through and throw them into the correct areas where they might relate to... And anything else, I'm like "If I haven't read it to this point and I don't wanna read it right now and it doesn't go anywhere that I need it in the future, it gets deleted."
+
+**Thorsten Ball:** Yeah.
+
+**Jon Calhoun:** But I have to be more diligent about it in that sense, because like you said, you can get a big list of things that you don't really need, but you convince yourself you need.
+
+**Thorsten Ball:** Yeah. I guess the difference is that one is archiving stuff for later retrieval, because you know you're gonna use it... So I do all my note-taking in markdown files that are in a Dropbox folder. Let's talk the next two hours about my note-taking system... \[laughter\] But that is the short version.
+
+Stuff I put in there, a lot of the stuff, I know I'm not gonna look at on a regular basis, but I know it's there. I can search it, I can grep for it, whatever... For retrieval purposes. But I think there's a lot of tools that can solve this problem, except that they're marketed as to-do lists, or task managers, or check-off lists of whatever.
+
+If you search in any app store for lists, you will find to-do lists. Things you have to check off, things you have to do. It's not "Maybe someday want to read books", it's always "To read." Maybe this is where this bad feeling comes from, as in -- you know, I curated this list of interesting things, but it says "To read", so now I feel bad about throwing it away, even though they're not interesting to me anymore. I don't know, maybe we need a different class of tools for this.
+
+**Jon Calhoun:** I will say that you brought up markdown, and I think that's something that every developer should learn, only because it's such a useful tool for note-taking, and for -- I use it for so many things; it's kind of insane to me... Between note-taking, or just writing up readmes for a project, or anything along those lines... To the point where I don't even like using a preview for markdown, I just prefer looking at markdown with syntax highlighting, because that's just what I'm used to.
+
+**Thorsten Ball:** I see.
+
+**Jon Calhoun:** So every time I see these tools that are like "Let me give you a life", I'm like "I've written so much markdown at this point that that does not help me." But I definitely think any developer - that's something they should get familiar with, because I don't see markdown going away any time soon, and I suspect their life would be at least more proficient when they're writing docs, if they know how markdown works.
+
+**Thorsten Ball:** \[56:04\] Yeah. The problem is once you have to start using a product that doesn't support markdown. Then you get mad... Like Confluence, for example. I don't wanna pick out a single tool, but... I do all my writing in markdown, locally, in a file somewhere; for example, Google Docs. There's no easy way to transfer markdown to Google docs, so you have to do this whole dance of previewing markdown in a tool that lets you copy the preview with the styling, so you can then paste it in Google Docs, including the formatting and all of that... And that is really, really bad.
+
+**Jon Calhoun:** Yeah. It's like you said, when you leave the one tool you're leaving to go somewhere else, it's problematic. And that's not even one that you would expect to have that same issue. But even the tools that do support it... Like, Dropbox Paper has a way you can export into markdown, and then you can basically paste markdown in there, and it pretty much always works. But sometimes the way they export just isn't quite -- I don't know, it's just weird sometimes. I feel like it doesn't always match exactly what I'd expect it to be, so it's still tricky... But I still encourage people to learn markdown, because I'm like, "This is something I use a lot."
+
+**Johnny Boursiquot:** Just don't go build your own markdown rendering engine, or something... I think we have enough of those, right? Unless you wanna do it for learning purposes, or something.
+
+**Thorsten Ball:** Yeah, build a static site generator instead. \[laughter\]
+
+**Johnny Boursiquot:** Yeah, isn't it time for a new Go router, or something? An HTTP router? I think we're due for another one. \[laughter \]
+
+**Thorsten Ball:** Yeah... Or a logging framework.
+
+**Jon Calhoun:** Is that your unpopular opinion, Johnny, that we don't need any more routers?
+
+**Johnny Boursiquot:** Yeah, I'll stick with that one... Speaking of Unpopular Opinions.
+
+**Jingle**: \[58:06\]
+
+**Johnny Boursiquot:** Thorsten, do you have one?
+
+**Thorsten Ball:** Oh, boy... I thought I had a lot, but I'm not sure right now, now that you're putting me on the spot. I think -- this might be unpopular, but I guess in this circle it's not... I wrote a blog post a couple of weeks ago and I've been thinking about this ever since... I think that some of the tools we use when programming might have an effect on us and how we program that is not always positive.
+
+For example, you use a code prettifier, like Prettier, or GoFormat, or whatever... It kind of makes you ignore all of the formatting, which in GoFormat's case I think is fine... But if you do really adventurous JavaScript, for example, with React, JSX, whatever, you start to ignore what is easy to write and easy to read code just because the formatter makes sure that it is properly formatted... As in, you can just type stuff into your VS Code that is probably configured, and hit Save, and it just formats all of the stuff. The linter works, the compiler works, all of this works... And I think in that case the tools might be optimized too much for writing, but our tools for reading code and consuming code haven't necessarily kept up with that yet. I don't know...
+
+**Jon Calhoun:** In some ways, I would say that I agree with it, but in others I disagree.
+
+**Thorsten Ball:** Same here. \[laughs\]
+
+**Johnny Boursiquot:** That's why it's unpopular. \[laughter\]
+
+**Jon Calhoun:** Whatever I say I agree, like -- if you look at anybody who's just getting into programming, they very often don't pay attention to the details when it comes to formatting. They'll have really weird indentations everywhere... Unless they're in a language that enforces it. But a lot of times if they're not, they'll just have -- I've seen all sorts of weird codes just indented weirdly, it doesn't make any sense to my brain, and I'm just like "I don't know what you were doing here. Were you just randomly hitting Tab sometimes, or what was going on?"
+
+\[01:00:12.28\] I think part of it is just because when you're just getting started, you don't realize how important that stuff is... So sort of learning to manually do it and getting an appreciation for the benefit of it is useful. And if you have a tool, like you said, that just formats for you, they don't really think about "Why do I need to do this? What's the point of it?" because it just happens. On the other hand, I don't know... I don't know if I could live without those tools, so...
+
+**Thorsten Ball:** I'm not saying we should live without those tools, I'm just saying they might be a feedback loop that is not visible, where you kind of shape your code based on how easy the tools work with this kind of code. For example, I worked in Ruby, as I said, for a long time, and I can honestly admit that I wrote some code that was a feature that was meant to be tested, and probably meant to be thrown away if it doesn't play out... And since Ruby is a dynamic language and it doesn't have all of these tools that for example Go has, it is really hard to find all of the instances of an identifier, of a method, or a variable, or whatever.
+
+So you would write this code with such a verbose name, just so you can easily grep it later on... You know where I'm getting at, right? I wrote code specifically because of the limitations of the tools that I was working with... And I think the other way also has an effect, where the tools give us a lot of power when writing code, but we lose the power when we read the code... And I'm just saying, it's -- look, I know that's really philosophical and it's probably not relevant to your day-to-day programming, so feel free to ignore this. I'm just saying this might be a thing to consider.
+
+**Jon Calhoun:** Yeah... I can definitely say I've seen the opposite, where if I'm using a language where it's common to -- like, Go does this; if you have a struct and you have some fields in the types, it always puts an equal amount of spaces between them, so that all the types line up. I think it was Ruby, when you're doing maps, or whatever the heck they were, the hashes - it was common for people to sort of organize them that way, with an equal number of spaces... But if you didn't have a formatter that did it, you kind of avoided doing it, because you were like "I don't wanna go manually fix this or tweak this..."
+
+So I've seen it the opposite way for sure, and I could see people leaning towards one certain style because -- that's not a good example; that probably makes it more readable, but I could see people leaning towards a certain style because they know their editor will auto-space things. So I could see it being the opposite, where while this other style might be easier, maybe the formatter doesn't quite like it, or it formats it weirdly, and all of a sudden it's not as useful as a style.
+
+**Thorsten Ball:** \[01:03:03.03\] Yeah. That's exactly my point. People often say "Oh, typing is not a bottleneck. Programming pure thought stuff. This is all just abstract ideas, and I've just gotta write it out", as in writing it out is just a manual last step. You could basically dictate it to somebody, or something. But then, you're too lazy to add some spaces to format the thing. It then comes down to this, like "Okay, I added this new line, the code works... Yeah, I should format it, but I'm actually too lazy to add five spaces on five lines", or something.
+
+So the act of writing code is still really real, in a sense, and it's still really bound by the tools we have available, and also limited by the tools we have.
+
+**Jon Calhoun:** I can definitely say, going the opposite way, one of the ways that I think Go has helped with that is when you see slices and the last one is required to have a comma afterwards - that type of formatting in my opinion encourages the correct type of behavior, because then you're encouraged to space things out the way you'd kind of want them to be, because you don't have to worry about "Oh, well later I'm gonna have to add that comma there", or something.
+
+**Thorsten Ball:** Yeah.
+
+**Jon Calhoun:** So it definitely goes both ways. At the very least, I would agree that formatters can strongly influence whether or not you write code that is readable, depending on how that formatter is set up.
+
+**Thorsten Ball:** Yeah, yeah.
+
+**Jon Calhoun:** So if you have a good formatter, I think it encourages better styling and more readable code, but a bad formatter I could see doing the opposite.
+
+**Thorsten Ball:** Agree.
+
+**Johnny Boursiquot:** Well, on that note, I think we've touched on a number of different tools that some of us use as developers, and others for it to help us manage our thought processes, I think. Actually, I find that at this point, at this stage in my career - maybe I'm just getting up there, or something... But at this stage in my career I find that managing what I work on, so knowing what to work on is becoming increasingly more important than the nuances of a particular tool. It's always getting the most bang for your buck, and that really comes in sort of managing your time... And if you put effort into one tool or one task, you know how much you're gonna get back for that investment.
+
+So yeah, it's definitely -- thinking along those lines, I think it sort of brings a different lens, a different level of clarity to the tools that we use in our day-to-day.
+
+This was fun, Thorsten. Thank you very much for coming on the call and sharing your opinions, some popular, some not... \[laughs\]
+
+**Thorsten Ball:** Thanks for having me, I enjoyed it.
+
+**Johnny Boursiquot:** Alright. And thank you for joining me, Jon. With that, we thank you for listening, and we'll catch you back in the next Go Time.

Quack like a wha-_transcript.txt

@@ -0,0 +1,681 @@
+**Mat Ryer:** Hello and welcome to Go Time. I'm Mat Ryer. Today we're talking about abstractions and interfaces, and we're obviously gonna deep-dive on Go interfaces, and look at some patterns and things there. Joining me today, it's Mark Bates. Hello, Mark.
+
+**Mark Bates:** Hello, Matthew. How are you doing today?
+
+**Mat Ryer:** Good, thank you. And yourself?
+
+**Mark Bates:** Not bad. It says here on my show notes that I'm supposed to mention BitBar and complement you accordingly.
+
+**Mat Ryer:** Wow, that's very kind of you to say, Mark.
+
+**Mark Bates:** So this is my mention of BitBar...
+
+**Mat Ryer:** Yeah, thank you very much.
+
+**Mark Bates:** ...and I'm complementing you accordingly.
+
+**Mat Ryer:** You've surprised me there, yeah... But I'll talk more about that later, because --
+
+**Mark Bates:** Oh, is that not the sponsorship portion of the show? Is that not where I --
+
+**Mat Ryer:** No, \[unintelligible 00:02:19.09\]
+
+**Mark Bates:** Oh, sorry. Sorry.
+
+**Mat Ryer:** That's Fastly. \[laughter\] We're also joined by Johnny Boursiquot. Hello, Johnny.
+
+**Johnny Boursiquot:** Hello, Matthew. Wait, your first name is Matthew? I didn't know that.
+
+**Mat Ryer:** Yeah, yeah.
+
+**Johnny Boursiquot:** Really?
+
+**Mat Ryer:** Yeah, Mat's the shorter version of it, but it's just one t.
+
+**Mark Bates:** Whow, whow - Mat is short for Matthew?
+
+**Mat Ryer:** Yeah...
+
+**Mark Bates:** I've just been...
+
+**Mat Ryer:** Mind-blown.
+
+**Mark Bates:** Totally. I did not see that coming.
+
+**Mat Ryer:** And John is short for Jonathan. And speaking of which, Jonathan is Jon Calhoun. Hello, Jon.
+
+**Jon Calhoun:** Hey, Mat.
+
+**Mat Ryer:** How's it going?
+
+**Jon Calhoun:** Good.
+
+**Mat Ryer:** Good. We're gonna talk about interfaces and abstractions today, and I thought, since you've done a lot of training material and stuff, it might be cool if you could kick us off, and tell us what is an interface, and what are they for.
+
+**Jon Calhoun:** Yeah. I mean, at its very core, they're just a way of defining behavior that you want. When we talk about code, a lot of times you look at structs and you'll see very concrete things that say what a user is, or all these different things... But whenever you're actually writing code, a lot of times you don't care specifically about the type that you're getting, you don't care if it's a user or if it's an admin, or if it's something else... You just care about some specific behavior that it might have. In Go, this is typically represented with methods of some sort.
+
+**Mat Ryer:** Yeah, so an interface is a type that just lists out methods, and then any of the type that happens to have those same methods can be used wherever that interface is requested.
+
+**Jon Calhoun:** Yup.
+
+**Mark Bates:** \[03:51\] The example I always use when I'm doing training is like an entertainer interface. So if I'm starting a club, some sort of an entertainment venue, if I use a concrete type, if I say "I want to use this concrete type; the concrete type is Beatle. Anybody who's a Beatle can play at my club." Well, there's only two people in the whole world who can play.
+
+Admittedly, if I got one of those two, I could easily pack the house. The other one would be tending bar. But that's concrete behavior. I can only fill it two nights of the year, possibly. If I accept an interface, if I say "Anybody who's an entertainer, anybody who can play something, whether it be a guitar or a flute, or can read poetry, or an improv troop", they can all entertain; they all have this play method on them, just like a Beatle would. Now I could have Paul McCartney come play, I can have the flutist come play, I can have that dance group come and perform, because they all implement that. They're not concrete anymore. To me, that's always a clear analogy, but... Maybe not.
+
+**Jon Calhoun:** See, I like that, because it's a good way of showing that you can also do interfaces that are like a long-running process. Anything that can play and that might block for a half hour... You know, everybody sits down and listens to an entertainer play. Or you can have behaviors like if you're dealing with packages in the post office, all you really care about is "Give me the dimensions." You don't typically care what's specifically in the box. You might have something like "Is this hazardous?", a couple things like that. But once you've checked those things off -- and those are sort of more behaviors that just give you some quick data back and they don't necessarily block, but interfaces can cover everything on that broad spectrum of "A server that can start up any type of server" or it could be "Just give me some information."
+
+**Mat Ryer:** Yeah, and I love in Go that you don't have to explicitly say that you're implementing an interface. In a lot of languages, when you create your type you actually list out all the interfaces that you're going to implement, and then the IDE usually helps you enforce that and make sure that you put all the right methods in, so that you satisfy the list. It doesn't work like that in Go. In Go -- it's called structural typing, so it's kind of like duck typing, but because it happens at compile time, it's called structural typing, apparently.
+
+But the duck typing idea is if it looks like a duck and it sounds like a duck, it's a duck. And it's kind of like saying "Yeah, so here is the interface with a few methods, and even if you didn't know about this interface, you can still implement it. Or you can write interfaces to things that already exist, or that other people have written." That turns out to be really quite powerful, as well.
+
+**Mark Bates:** Yeah, the implicit over explicit is really where it shines in terms of the interfaces... And I know a lot of new people coming to go - I've seen from class - really struggle with that bit, understanding that just because they've written a method, they are now implicitly implementing an interface... And they get hung up on "Well, how do I know that I'm implementing that interface?" Like, "Well, it's not important until you need to use it as an interface." That's the beauty of it. And you say "Well, how do I know?" You just look at the docs; what is this thing taking? It's taking a writer. "What's a writer?" A writer is anything that implements the write function that takes a slice of bytes and returns an int and an error. That's the beauty of it. You just kind of do it, you don't have to worry about tying into all these other things. It also means - and we can also talk on this later - that you can break a lot of dependencies, too. You can keep dependencies out of the mix by using interfaces in ways too, as well, which is quite nice.
+
+**Johnny Boursiquot:** Along those lines, my favorite use of interfaces is to leverage its ability to provide that sort of independent means of decoupling packages, the dependency between packages. For example, I do a lot of work with the AWS SDKs, and for example when writing a lot of data to (say) DynamoDB, I don't necessarily have to bring in the AWS SDK in the DynamoDB interface or implementation anywhere near my code. I can simply create an interface that I expect my code to use, and basically have that interface be local to my code, not even export it to the rest of the application at all... Have that be local to my code, and maybe in my main package when I'm initializing my application, I can then basically initialize a value that represents a client to my DynamoDB server, and then pass that in.
+
+\[08:16\] And as long as it satisfies the interface I've defined for my code locally, everything is good. My code didn't have to know anything about the fact that it's even a DynamoDB implementation at all. It can be anything that actually implements that interface. So that allows you to create that separation, that decoupling, because of that implicit satisfaction of those methods. Then it really allows you to keep your code separate and not have to depend on any sort of actionalities at all. That's my favorite part of using interfaces.
+
+**Jon Calhoun:** I'm glad you brought that up, because like Mark was saying, a lot of people got hung up on this fact that "How do I know if I'm implementing an interface?" and I think it's a weird paradigm to get used to. You kind of lift that responsibility off your shoulders, and it's the person who's using the type that actually has to care about "Is this going to be implementing an interface?" and then they define the interface.
+
+Like you were saying, Johnny, your code that needs something, that interacts with a database of some sort - it defines the interface and it doesn't even necessarily have to export it to the rest of the code... And it's weird to get used to that when you come from another language like Java or something, where you're explicitly saying "Here are all the interfaces I'm implementing", and that's very different from the way it is in Go. In Go you just write your code and then if somebody wants it to be an interface, it's their job to define the interface and make sure that it's the right one.
+
+**Mark Bates:** Yeah, a lot of -- especially, again, new developers, don't realize that you can create non-exported interfaces inside of a function or a method, to check right there. You don't have to export them, you don't have to have tons of interfaces. You can say "I'm looking for one very specific thing", create an interface in-line right there... And it's amazing. It's so wonderful that you can do stuff like that. Because you can even then turn around and create your own default implementation of that interface using functions and types to have a back-up in case the thing you're looking for doesn't exist, or is nil, or whatever. It's such a wonderful way of working and asking for and getting more and more enhanced functionality.
+
+**Johnny Boursiquot:** Along those lines, I've seen way too many times where as I'm writing my code, if I happen to -- I used to create public interfaces all the tie, and then I realized "Okay, first of all, I don't need to." And I came to a point where I'm like, every time I create a public interface, I'm kind of implicitly saying to whoever is gonna use this package, this code, that "Hey, you can actually depends on this, because I've exported it", thereby making it hard for me to actually change that later on if I wanted to.
+
+Like every other type, if you don't need to export something, don't. By keeping it local and private to the package, I'm basically saying "Hey, this is what you should expect to send in, or you can actually read the code, the implementation and see what interface you're expected to satisfy", and that enforces that separation. It removes the temptation to have my interface be in your code.
+
+**Mat Ryer:** It's interesting you bring that up, Johnny, because Eric Fouga on the Slack channel in \#GoTimeFm was actually talking just about that thing. He says that he likes the idea of providing the interface with the implementation, because you get this sort of explicit storytelling, I guess... And he's apparently challenged this before and people have said "You don't need to do it" or "It's not necessary", or something. He asks for a more concrete reason why it is bad to ship the interface and the struct, say, if you've got a package. What are the pros and cons.
+
+**Mark Bates:** I don't think it's wrong, and I don't think that's what Johnny was saying. I think what he was saying, like most code, is start with the least amount exported, and export what you need as you go. And I can tell you from very much so first-hand experience that I'm feeling a lot of pain around a lot of this - exposing too much of your API too early, and exporting too much of it does cause problems. It causes a lot of problems down the line in terms of migrating things, dependencies, things get stuck, and it becomes difficult to work with.
+
+\[12:19\] If you start by exposing nothing and then expose the things you need as you go, that's really very useful. So yes, there are very much so reasons you should expose interfaces. I don't think anybody would ever say don't. The standard library is littered with them; they are very useful. I think what Johnny is saying and what most people are advocating is "Don't expose the ones that people don't need to know about", the ones that are just useful for you, inside of your package. Only expose the ones that people need to fulfill to work with your package. Maybe I'm misunderstanding the question, but that's the way I was viewing it.
+
+**Mat Ryer:** Yeah, I think that's right. And I think I have a special place in my heart as well for single-method interfaces, for a kind of similar reason like about the whole minimalist mindset of keeping everything as tiny as possible. And doing that even down to the interface level, there's some surprising things that can happen, which only work with single-method interfaces. One example is just being able to use a function type, like the handle func is the great example of that. For anyone that hasn't seen that code, go and look up the handler func and Handler types in the HTTP package. It's not very much code, but it's very cool how there's a function type which happens to match the signature of the ServeHTTP method in the handler interface... And it too implements the ServeHTTP method and then just calls itself. So it's this kind of weird inception. It's the weirdest little thing that I think you encounter in Go often.
+
+**Mark Bates:** It is. It's a beautiful \[unintelligible 00:13:55.20\] I know that wasn't an intended thing; it was just a fall-out from the way the type system is designed. For those of you that don't know, in Go you can declare your own types. We do "type foo struct" and that's declaring a new type based off of struct, or based off of interface. We can do it off of ints, we can do it off of slices, we can create new types off of anything, including functions. And when you do that, then you can put methods on that new type, and that method can implement in this case an HTTP handler and then just call itself. I use it all the time; it's a wonderful little thing, especially for those single-method interfaces. Depending on what the other methods are, you can easily mock those out, too.
+
+**Mat Ryer:** Yes.
+
+**Mark Bates:** "Yes..." Thank you, Mat. \[laughter\] That was deep insight into what I just said there, I appreciate that.
+
+**Mat Ryer:** I was contemplating challenging it, but I was just gonna let it go... But actually, it only works with a single-method interface, that trick of doing the function thing... Because there's only one function it can call. Unless it's like a Close(). Sometimes you get like a no-op Close() and then you can implement those, actually, and it just doesn't do anything.
+
+**Mark Bates:** Yeah. But on the testing side it's incredibly useful. That's where I use it all the time, it's to implement testing versions of these interfaces.
+
+**Mat Ryer:** Yes, that's another use. If you do have some kind of concrete dependency, like you're gonna send an email, and you're using a package from SendGrid, let's say that they didn't export an interface, so you only have a struct to work with. If you wanna stub that out and test the code that you're writing to make sure it uses that SendGrid API int he way you expect, if that's indeed the kind of test you wanna do, then that can be quite tricky if you forget that you can write the interface after. You can write an interface that just essentially describes the same methods that you're gonna call in the original SendGrid API. You use that type instead in your real code, and then you've got an opportunity to build your own stub version of that, that you can use for testing.
+
+\[16:07\] Sometimes you can't avoid the situation of having to test those types of dependencies if you wanna unit-test something. And for those cases, that's incredibly useful. So it's really worth remembering that you can write your own interface about something else. It doesn't always have to be the other way around.
+
+**Jon Calhoun:** Another one that's come up with some of that weird stuff is any type that chains (there's method chaining) can be really hard to use an interface for, so you almost have to wrap the whole thing in something else that returns interfaces, and sort of define your interface there. It can get frustrating at times, but it's just kind of the way it is.
+
+**Mark Bates:** Yeah, method chaining is a real drag in that respect.
+
+**Mat Ryer:** Yeah, it's not very Go, actually.
+
+**Mark Bates:** No, it's not.
+
+**Mat Ryer:** And what we're talking about is these fluent APIs where every method call returns the object itself, so that you can add --
+
+**Mark Bates:** Or a clone, or a modified, or a new version of it.
+
+**Mat Ryer:** Yeah, right. The same type, yes...
+
+**Mark Bates:** Yeah.
+
+**Mat Ryer:** ...depending on what it's doing, yeah. I get it. And in some languages, they really work well. But in Go - Go is very strict about types, and in this situation it's very difficult for you to not replace wholesale some of these concepts regardless.
+
+**Mark Bates:** It's funny, I ran into, I think, the very first time I've ever really wanted generics in Go the other day. It was all about interfaces. The problem I had was I have two identical interfaces, and all they had was one method on them that returns a string. that's it, just a plain method, called name, returns a string.
+
+**Mat Ryer:** Both the same --
+
+**Mark Bates:** Both called plugin, they both have a method called name, they both return a string, they're identical, just in different packages. But because they're in different packages, they are now different types, and you cannot use one as the other in, say, a return. Even though they implement the exact same interface, they're not the same type, so therefore they don't work. That was the first time where it's like "Well, the compiler could tell that. That information is there. They are identical, so they do implement each other. They are interchangeable interfaces, so therefore their types really shouldn't matter." That's a case where generics would have solved that problem.
+
+**Mat Ryer:** Yeah, so this was Russ Cox when they did that alias. Do you remember that type "alias"?
+
+**Mark Bates:** Oh, yes. It's still there, you know... It's here \[unintelligible 00:18:38.19\]
+
+**Jon Calhoun:** I mean, I've used it before, I just -- I feel like you're something you're not supposed to be using when you use that alias. That's the hard part with it. I've used it occasionally to experiment with some stuff, and it just feels like I'm doing something naughty, that I'm not supposed to be doing... And I'm like "I probably don't wanna advertise this code now..."
+
+**Mark Bates:** Yeah, it doesn't quite feel right when I use it, too. I'm with you.
+
+**Mat Ryer:** Well, it was a fix, I think... And yeah, it didn't quite do its thing. Now, what's interesting - we're hearing breaking news from the Slack channel. Marwan is actually saying that in 1.14 there could be some changes. "I don't think that changes"-- okay, I'm reading it live as we speak, for some reason... It's like proper live journalism, isn't it? ...no, it's just me reading out Slack. I'm distracted by Slack, even now. \[laughter\]
+
+**Johnny Boursiquot:** The overlapping interfaces, yeah.
+
+**Mark Bates:** That's the overlapping in one struct. I don't think that's the same.
+
+**Mat Ryer:** Yeah. Because you can just do that with structs, can't you? If you've got two structs that have exactly the same fields, you can just cast one to the other and it's a very cheap operation. Is that right? I think that's right.
+
+**Mark Bates:** No, you can cast the type that's based on the other type.
+
+**Mat Ryer:** What do you mean "based on"?
+
+**Jon Calhoun:** I'm pretty sure what Mat's saying is right, I just -- it's one of those things that every time you happen to do it, you're like "Let me go ahead and write this real quick and make sure it works." \[laughter\]
+
+**Mark Bates:** \[20:03\] Yeah, make sure this works...
+
+**Mat Ryer:** Exactly.
+
+**Mark Bates:** Well, it's like, if you have a type MyInt based on int, you can cast it back and forth between MyInt and int, so I guess you can do that with a struct, too.
+
+**Mat Ryer:** Yes. Exactly. With structs if there's same fields and same structure, essentially, you can do the same. Just the name of it, and then brackets, and then pass the other type in.
+
+**Mark Bates:** Yeah. That's not at all weird Go code... \[laughter\]
+
+**Mat Ryer:** Exactly. And the fact that -- I mean, it just feels so brittle. But I guess if one of the structures changes, you get then a compile error. It's a compile-time error, because the types are no longer compatible. So maybe it's quite reliable, really... But it's surprising to see, because it looks like you're calling a method, actually, and that is quite strange.
+
+**Johnny Boursiquot:** The 1.14 feature that's coming with regards to the overlapping interfaces is that now if you actually have two interfaces that have the same method, before 1.14 you couldn't do that. Now, as long as they match obviously, you can do that. And obviously, your implementation can only have one -- say you have an open method, or something; your implementation can only have one anyway, so... Basically, the fact that the embedder and the embedded have the same thing kind of makes it mute. So now you're allowed to do that. The compiler won't let you, so that's the new thing.
+
+**Mat Ryer:** Oh... I didn't know you weren't allowed to do that before, actually... Funnily enough. Yeah.
+
+**Johnny Boursiquot:** Yeah, I think I never tried to do that. \[laughter\]
+
+**Mat Ryer:** Yes, certainly I haven't. I didn't know you couldn't do that. I thought it'd be alright. \[laughter\]
+
+**Johnny Boursiquot:** Well, now it is.
+
+**Mark Bates:** It doesn't solve the problem I had, but yes, it is useful that that fix is there.
+
+**Break:** \[21:36\]
+
+**Jon Calhoun:** I know this is really delayed, but earlier we were talking about single-method interfaces... I think the one thing that I wanna point out is that one of the aspects of them that I really like is just that it makes writing closures and turning them into an interface much easier. Because otherwise, using a closure for an interface would be a nightmare.
+
+**Mat Ryer:** Yeah. I do this a lot with handler func, again, actually. My handlers, which usually are methods on some server type - when called, they return a function. They return a handler func, essentially. In that case, the compiler will cast the type for you if it matches, actually, for that function case. So essentially, you get that little closure environment that you were talking about, Jon, where you can do some setup, you can prepare some resources if it's a web request, which it is in this case... And then in the body of the function you return - that's the real handler that gets called every time.
+
+So it's a tiny bit of indirection, but what you get from that is you can have per-handler dependencies just passed in as arguments, you can have the little setup code all in one place, near to where your actual handler is being done (the work of it). And similarly, you can have request and response types also in that space as well. And it keeps them all in one place, out of the way.
+
+So for some projects, that's quite a nice little neat package, a little neat way of designing these services. This is a good time for a commercial break. Please purchase... \[laughter\]
+
+**Mark Bates:** This dead air, brought to you by BitBar. \[laughter\] Bitbar. For all your stroking of Mat's ego needs.
+
+**Jon Calhoun:** We're gonna find out after this that BitBar has not been working with the latest update, or something...
+
+**Mat Ryer:** \[23:59\] No, it works. It doesn't really need many updates, frankly. It's kind of like done. For anyone that doesn't know, it's a little project which puts the output of any script or program into your Mac menu bar.
+
+**Mark Bates:** And the contents of your password manager into Mat's email. \[laughter\]
+
+**Mat Ryer:** No. \[unintelligible 00:24:21.15\] mistake on my part.
+
+**Mark Bates:** Open developer tools.
+
+**Mat Ryer:** Yeah... Actually, it's a nice example really if we're talking about abstractions, because the key point is the little tool doesn't really do anything; it just calls another program. And then the output of that program is what basically builds the menu bar and the menu that you get when you click it. So it's a kind of perfect example of an abstraction that really worked, because there's hundreds of plugins now for this, and they all do wildly different things, none of them that I could have imagined when I just made Bitbar for the one case that I had it for.
+
+So it's nice. That's the side of interfaces that enables other people. If you do provide an interface or a very simple way for people to integrate and extend what you're doing - if that's easy, then more people are gonna do it. And the point of having that there surely is for people to use it. It's enabling other people to also build on top of what you're doing. So even for its own sake it's great, but obviously in business it has massive value, as we've seen, as well.
+
+**Johnny Boursiquot:** So we've talked about all the benefits of using interfaces... Can you think of reasons when you should not use an interface?
+
+**Mat Ryer:** That's a really good one... I've definitely in the past overdone it. I've definitely done cases where I've overused interfaces when a simple struct turns out to be much simpler. I tend not to do that anymore, because I tend to start with the structs first and then let the interfaces find themselves, or reveal themselves over time.
+
+**Johnny Boursiquot:** So you don't design your code; you don't try to abstract too early by saying "Oh yeah, this thing's gonna receive an interface."
+
+**Mark Bates:** So Mat, can you go on record right now as saying you don't design your code?
+
+**Mat Ryer:** Well...
+
+**Mark Bates:** Come on, Johnny just asked you a question - do you or do you not, sir, design your code? \[laughter\]
+
+**Mat Ryer:** I feel like it designs me... I don't even know what that means. \[laughter\] Well, yes. I mean, obviously, you do. And interfaces are a great way to do that as well, especially if you're collaborating with people. You could say "Well, we know that our two things have to communicate, so let's agree on the interface between them and we can both build towards that. So in those contexts it's great. But yeah, of course, it is useful if you 're just sketching out concepts, actually.
+
+Sometimes in my notebook I'll actually write out Go interfaces to try and think about what these things are gonna be doing, and stuff. But yeah, I do tend to wait... If I'm doing a package, I want that to be the smallest possible footprint, so I am definitely in that camp of -- I wouldn't have an interface unless it was an extremely important part of this package, like the io.Reader, io.Writer, those kinds of types.
+
+**Jon Calhoun:** Building on what Mat said, I think one of the downsides to jumping straight to an interface is that it causes you to think "Oh, I'm gonna have three implementations of this", and starting to focus on breaking things into multiple versions, when sometimes that's just never the case. The classic example is typically your database. You're like "Well, what if we switch out for another database?" But in reality, most people never do that. So it's one of those -- it's not that you can't do some of that stuff to make it easier for you, but it doesn't make sense to bend over backwards to make this possible later, when in reality you're probably not gonna do it.
+
+**Mat Ryer:** \[27:56\] Yes. And often, whenever you think like that, the detail actually doesn't allow it anyway. Two different data stores often behave very differently. You wouldn't treat them the same. So it's more likely to encourage bigger changes anyway, isn't it? So I completely agree.
+
+**Mark Bates:** When I design interfaces upfront, they're almost never correct. It's because you're guessing; you're taking a wild stab at what you think the interface is, and especially if you go ahead and publish that. Now, I've been doing a lot of work with interfaces recently, and I can tell you that a lot of what I've been doing now is working with problems that I do understand and problems that I do know what these interfaces need to look like now, and how people are using them. But even then, I'm still saying "What's the simplest I can get away with?" and see how far I can push that before it starts breaking, and before I need a second method, or a concrete type, or something further down the line.
+
+**Mat Ryer:** Yeah. In the Buffalo project you're quite flexible. It's kind of like a framework, and it's flexible. It lets people plug different things in and out, doesn't it?
+
+**Mark Bates:** Wait a minute, I feel like you jumped about ten steps... \[laughs\]
+
+**Mat Ryer:** Oh, really?
+
+**Mark Bates:** Yeah, yeah...
+
+**Mat Ryer:** I mean generally though, Buffalo -- the reason why interfaces are important, and these kinds of concepts... Abstractions are important, and they're especially important in the Buffalo project, because of the nature of it. The fact that you can use different technologies.
+
+**Mark Bates:** Well, yeah. The reason I jumped ahead was I think Buffalo does a terrible job today of doing that.
+
+**Mat Ryer:** How does it work today?
+
+**Mark Bates:** Well, today we have a lot of hard, concrete types all over the place, lots of dependencies, we've got a plugin system that goes and searches your path for executable binaries named a certain thing, and asks them for information... It's very slow... Generally, as a whole, the Buffalo project was very much so -- like a lot of projects, I started it when I first came to Go, and I started writing Ruby for Go, basically. We all bring our baggage with us, right? So a lot of this has grown over time, with just me making choices that at the time seemed logical, or at the time were what I just knew how to do, because I didn't know Go well enough to make those choices.
+
+And then as projects grow, things evolve, and people come in, and changes are made, and new requirements are added on, or whatever. So today, what we have in Buffalo isn't as pluggable as I want it to be, and it doesn't achieve the goals I want it to in terms of saying "I don't wanna use [gorp](https://github.com/go-gorp/gorp) I wanna use [gorm](https://github.com/jinzhu/gorm). I wanna make this as seamless as possible. I don't wanna use gorm. I wanna use nothing. I wanna use ego templates, or raymond templates", or whatever templating you want; or whatever it is you wanna do, right now you can't do that in Buffalo.
+
+So I definitely have to go back to the drawing board, and we're currently rewriting it all now using a completely different system, but all interface-driven, using pretty much all of what we've just been talking about.
+
+**Johnny Boursiquot:** I have to ask though... If I'm going to use a framework, I want it to make some decisions for me. I want it to be opinionated. Personally, I think that's the reason why I use a framework, and not one of the reasons you use a framework, right?
+
+**Mark Bates:** I'm with you.
+
+**Johnny Boursiquot:** So if you're now telling me you're gonna provide this whole new pluggable system that can basically take any ORM tooling you want, it can use any UI interface you want, all the bits and pieces - if you make everything pluggable, then do you not create another problem? Now you have to document patterns. "Hey, you could use this set of things. This for ORM, this for template generation, this for that..." It's almost like you're pushing the decision to the user of the framework, as opposed to being opinionated about it.
+
+**Mark Bates:** \[32:07\] I'm absolutely doing that. I think a little bit cleaner than you might be imagining it... Buffalo today - right now you can say "Generate a new app" and you get this whole web stack, and it's got Node, and it's got Pop, and it's got Plush, and all that sort of stuff. That's that very opinionated that you were talking about. There's also a flag and you can generate a JSON one, which is slightly different. And that won't ever go away. We will still have those -- Rails calls them templates, but I'm not quite sure exactly... Kind of default presets, if "presets" is a good word, where you'd say "Give me the web preset", and Buffalo will ship with a few of them... And you're gonna get a Go file that has all those plugins in them, and you could just pull them out, or add your own, or whatever... Or you could come up with a different present that your company has of all these plugins, and just use that instead.
+
+So yeah, there's always gonna be opinions, and it's just like -- you know, Rails basically generates a Basecamp for you whenever you do Rails new... \[laughter\] Buffalo new will always generate the Basecamp for me, I would assume... Or something like that. But we need to make it easier for other people. Not everybody wants Pop, not everybody wants these things. And I know myself, I have hit points where I'm like "I need to do X, Y and Z" and I can't, because I don't have the hooks in the tooling, I don't have the hooks in the library itself...
+
+I mean, we talk about tooling in CLIs, and you start talking about how do you get versioning, and stuff like that... But that's getting way off this track.
+
+**Mat Ryer:** Do you think if you sit and design for much longer before you started Buffalo, that you would have come to these realizations just by exploring in your mind, or do you think the process was important?
+
+**Mark Bates:** Oh, god no. I think everybody else would probably agree with it - you can't design stuff like this in a vacuum. If you've never written a web framework, and managed a web framework, and all that goes along with something like Buffalo for example, or if you're writing Docker, or whatever tool/project it is you're talking about, you can't just start one of those in a vacuum, and say "I know how to solve this problem." Problems are always infinitely more complex than you know. Always. It doesn't matter the domain.
+
+So no, I could not have come up with a better design than I did when I first started writing Buffalo. What I can do is spend the last six months going on a kind of a connaissance,  a vision quest, if you will, for code, trying to figure out what that needs to be. What it needs to be to be truly idiomatic, and pluggable, and easy, and dependable and trusted. You can only do that sort of a thing with time and experience.
+
+**Mat Ryer:** Yeah absolutely. So in a way, this next API has emerged in some ways out of what you had before. But also, of course, it's not to say you shouldn't do any design. I mean, that's what you're doing now, when you're thinking about this - you're taking everything you know before and putting it into a new design... So of course there's value in that.
+
+**Mark Bates:** Yeah. We're currently rewriting the entire CLI project to a v2, using pure Go and interface-based plugins to really drive us. We're about 70% done, including some major pieces like generate some command, generate resource, and build, and test... And so far it's holding up beautifully. We've got some very small interfaces, not a ton of them. They're all standard libraries, there's no Buffalo types. Everything is a plugin; even the subcommands are plugins... And it's all managed with just a slice of plugins. It's ridiculously simple in its concepts, but really powerful. You could build really amazing things with just a few interfaces if you line them up correctly, and think about what it is you're doing. And you set yourself a space to work in.
+
+\[36:12\] For me, it's been understanding that everything is a plugin. If you take something like Buffalo generate, that generate command is just another plugin, and it implements the one interface you need to be a subcommand of Buffalo, which is a main function that takes a context, root, string for where you are, and the slice of arguments returns an error. That's it. Now it's a subcommand of Buffalo.
+
+That generate plugin issues three or four interfaces maybe, that say "Hey, if you implement these, you're gonna get these different lifestyle hooks when you run Buffalo generate", one of them being, say, a subcommand of Buffalo generate, like resource. And that's it.
+
+So you can write your own implementation to generate. If you speak those couple interfaces, you can write your own drop-in replacement for it, or any of the other things. So it's not about a lot of interfaces, it's about targeted interfaces, it's about defining the scope of where your interfaces are.
+
+**Mat Ryer:** Yeah, I like that idea, which I think everyone could actually use potentially. You don't have to be building Buffalo for it to apply... But that idea of having hooks into something. So if you do have some process that's kind of a closed box process, you may want some hooks into that. Having different interfaces for each hook, essentially - each method gets its own interface - and then they get to just implement the methods that they care about, you can of course check if a type implements an interface in Go very easily. And if you use the two-argument format, then you're not gonna panic when they don't implement that, so it's pretty safe. So you could use that pattern to allow other people then to hook into your own code, a bit like how you've done it for Buffalo.
+
+**Mark Bates:** Yeah, exactly. One of the examples I like to use is the Buffalo dev subcommand, which currently watches your Go files, compiles them, restarts your app every time you're working, which when you're working with a compiled language, it's great. So every time you go back to your browser, it's the fresh app again. And the same thing with Webpack. But the problem is you can't add your own build scripts. You can't say "I want something else that's watching my files and running my tests." You can't have something else that maybe is starting up a Docker service. There's no way of hooking into that build lifecycle.
+
+But you can easily add a couple plugins, and this is exactly what the develop plugin for Buffalo does now - or will do - in v2. It's like "Okay, we've got a before develop and after develop", so if you wanna set up some stuff, you need to launch Docker, write some files, run migrations before everything starts up, do that... There's a teardown you can hook into... And then there's a develop that \[unintelligible 00:39:00.05\] You can implement the developer interface and get spun off in a goroutine with everybody else to run your things you need to. And again, that's contexts, string, slice of strings, and error, and your context gives you all that cancelation. You can easily test async code if you're taking a context as your first argument.
+
+In this case, testing this plugin that runs all these things in a goroutine was super-easy. I just wrote another plugin that implemented that one function, and then I just canceled the context when it ran. That was all I needed to do. So they're easily testable, and you can hook in with so much ease. They're really powerful if you start thinking about interfaces in the right way. And yeah, you can do some pretty amazing stuff.
+
+**Mat Ryer:** You've reminded me of another one that's great, and Jon and I were talking about this the other week, as well... It's that idea of being able to wrap things with interfaces. A bit like how the middleware things work in the HTTP way. You have a function that takes in a handler and returns a handler, and then what you can essentially do is create a new handler that does extra things before and after passing the execution on to the other handler. So that thing of wrapping is actually quite useful.
+
+\[40:23\] One trick that you can use as well - if you've got a long-running io.Copy operation and you want to cancel that with context, you can create a kind of a reader with a context yourself... Which essentially wraps another reader and intercepts the read method (that's obviously the first one that gets called), checks to see if the context has been canceled by checking the err method; if that returns an error, the read method can return the error. If not, it passes it on to the inner reader. That's a way that you can actually get a cancelable io.Copy, which - it's really cool to think that just because of these basic interfaces you can add actually quite a lot of power just by thinking about it in the right way.
+
+**Jon Calhoun:** The reader is a really fun one to experiment with. I would definitely encourage anybody trying to wrap their head around this idea to spend some time with that. When I was messing around with the context -- Mat and I were talking about "Is it possible to cancel a reader?" and for whatever reason we hadn't read the whole thread on the GitHub issue, where somebody actually proposed just wrapping it like we said... But in the process of looking at it, I was like "Alright, let me go ahead and just throw this context in there and just check to see if it's canceled and just stop it." Well, one of the issues you run into is if you're doing really small files to test it, your one read will just read the entire file in one method call... So it's like, "Well, that doesn't work."
+
+But then you can quickly be like "Okay, can I make another reader that limits it to reading five bytes at a time?" Now you have an easy way of saying "I can chunk this and make it a little bit easier to see when it cancels", and I can actually have another one that's set that after it reads maybe eight bytes, it actually cancels the context. So you can do these things to sequentially exactly see what's happening and make sure your code is doing what you think it's doing. It's really weird at first, but it's also really cool seeing how much control you have over these things by just chaining these interfaces together.
+
+This all stems from a single-method interface, which is the crazy part. It's not like we went ahead and had some really complicated types; it was just a read method.
+
+**Mark Bates:** The single-method interfaces are really key for stuff like that... Because like we've been talking about, you can just create those types right there in your tests, and have them do whatever you need them to do. And they're just an interface. Whether it's read five bytes and cancel, whether it's capture the arguments and whatever that came into this function so that you can check them later and then cancel the context, or return some error you want it to return... You can just implement those types right there, implementations of them, using simple functions, or slices, or whatever you need.
+
+**Mat Ryer:** And it'll never get as complicated as abstract classes and big class hierarchies used to in C\#, because this technique really only works well with tiny little interfaces. So I think Go protects us a little bit there. There's another trick you reminded of when we talked about wrapping - if you're doing an HTTP response where you're writing to a file, and you're copying or you're writing to that file, if you want to see what's being written out, you can actually just replace the writer with an *io.MultiWriter*, and pass it an *os.STDOUT* as one of the writers. *os.STDOUT* is a file, so it actually implements *io.Writer*, and you pass in also the original writer. So it still carries on doing what it was doing before, but because of that Multiwriter, you also see it printed out into STDOUT.
+
+So again, not many keystrokes and suddenly you can peer inside your code without having to open up a debugger and things; those are difficult to use, especially when you're dealing with byte streams, and things.
+
+**Jon Calhoun:** \[44:03\] There's a lot of really cool ones like that in the io package. *io.TeeReader* is another one that does kind of like what Mat was saying, I believe, except whenever you're reading, you can actually pass in something that will write everything that it reads to that output. So you can actually have it write to STDOUT everything that it's reading from a file, so you can actually see "What am I actually reading from this HTTP request body, and what does it look like?" and you don't interfere with the rest of your code, you just wrap it real quick, test it, look at it and visually see "What am I getting?" and then you can remove it as soon as you're done.
+
+**Mark Bates:** Yeah, *io.MultiWriter* is awesome. I use that one all the time, just for that purpose, just for debugging what I'm expected to see, if I'm generating files, or whatever... It's like "Why am I not seeing that?"
+
+**Mat Ryer:** Yeah, it's important in some cases, isn't it? And sometimes you don't wanna interfere with what it's doing. You don't wanna invoke the Heisenberg principle; you wanna be able to observe it and for it not to change behavior.
+
+**Jon Calhoun:** I mean, nothing's worse than you're trying to debug, and in the process of interfering with it, you break it yourself, and you're like "It was never gonna work after I did that..."
+
+**Mark Bates:** \[laugh\] Yeah, right...
+
+**Mat Ryer:** Yeah. I saw a great example which involved putting a log line. The log line slowed the program down enough that the behavior changed... And it was obviously the thing you do when you're debugging something - you're gonna put some log statements in. Even that can interfere in some cases.
+
+**Mark Bates:** Yeah, I used to have weird ones in Ruby, where just the act of printing it would cause something in the function. Whatever it was I was trying to debug would get kicked off and it would actually produce different results when you printed it versus when you just executed it.
+
+**Jon Calhoun:** I think that's one that catches beginners off-guard. If they're dealing with a linked list, they'll iterate through it to actually print it out, and then they won't realize that their list is pointing to the end of the list, which is nothing... And then they'll be like "Why is it not working anymore?" \[laughter\] I've seen so many beginners get messed up by that. It's like, "No, you need to reset back to the front of your list", and if you don't have a pointer to that anymore, you're done. So printing out really screws you up.
+
+**Mark Bates:** Yeah.
+
+**Mat Ryer:** That's a great one. Well, in Ruby, of course, you could just do anything. There weren't any rules. \[laughter\] Someone probably took the to\_s method and just wrote their own, and did something crazy in there, and that's it.
+
+**Mark Bates:** Well, it usually was never even anything that mean or intentional. The to\_s was probably calling some other method that gave you a default value, and it was maybe calculating an evolved value, or something...
+
+**Mat Ryer:** Right, right. Or it maybe had some caching logic that malfunctioned.
+
+**Mark Bates:** Exactly, yeah. So it wasn't necessarily-- I mean...
+
+**Mat Ryer:** I'm not so sure it wasn't. I wasn't suggesting Ruby people go around casting spells on each other, or anything...
+
+**Mark Bates:** No, no, no... I certainly have never modified the plus sign on the American Ruby to do division to my co-workers, ever. \[laughter\]
+
+**Mat Ryer:** Yeah, because why wouldn't you want a language that lets you change what the plus symbol does?
+
+**Mark Bates:** Hey, you know what - it made debugging fun. It was an adventure every time.
+
+**Johnny Boursiquot:** I don't know if I want an adventure when I'm debugging. \[laughter\]
+
+**Jon Calhoun:** I mean, when I was young I did...
+
+**Mark Bates:** You didn't like grepping for source code that didn't exist? That wasn't a fun time for you, Johnny?
+
+**Johnny Boursiquot:** No, I don't miss method missing. \[laughter\]
+
+**Jon Calhoun:** You could implement every interface that way.
+
+**Mark Bates:** Yeah, but you couldn't find it, so it's hard to say it was missing. Method missing itself isn't defined anywhere. \[laughs\] I do miss Ruby sometimes. It was fun to do. You could do some really fun stuff with things like method missing, but...
+
+**Mat Ryer:** You could of course do some very appropriate use, as well... I did see some great examples.
+
+**Mark Bates:** Oh, absolutely. But honestly, you look at Rails, and one of the things that made Rails Rails was method missing. And a lot of Rails is based entirely off of method missing. All that magic that everybody loves in Rails is essentially using method missing. Sometimes well and sometimes not so well.
+
+**Mat Ryer:** \[48:11\] Yeah. So for anyone not familiar, basically if you call a method on an object, if you do that in Go, if you call a method and it's not there, that's a compile-time error. In Ruby, it'd just let you do that, but then it would just call like a catch-all inside, called "method missing". Then it gave you a kind of second change of seeing if you could do something with it. And a lot of the Rails things - you could write things like "find by name and age", and then that becomes a new method that you just invented...
+
+**Mark Bates:** Yeah, you could basically parse the name of the method and generate -- in that case it was generating queries for SQL... And also (in Ruby) if a type didn't exist, a module or a type, you could also capture that and define types on the fly. I had a library that distributed Ruby, and if you'd just ask for any type inside of a module, it would just create the module, it would create the type and connect it to a remote data source somewhere for DRb stuff... And it just did all that by capturing those error hooks where things don't exist in Ruby. \[laughter\] That's -- I know... Isn't it terrible?
+
+**Johnny Boursiquot:** Oh, my god...
+
+**Break:** \[49:36\]
+
+**Jon Calhoun:** So getting back to Go...
+
+**Mat Ryer:** Yes, please.
+
+**Jon Calhoun:** I feel like we've talked about interfaces a bit... How have we not talked about errors? I feel like that's something we should talk about.
+
+**Mat Ryer:** An error is probably the most important interface we have in Go, actually...
+
+**Johnny Boursiquot:** The best part of Go, you'd say, Matthew?
+
+**Mat Ryer:** The best interface in Go.
+
+**Mark Bates:** I thought BitBar was the best thing... \[laughter\]
+
+**Mat Ryer:** BitBar is good. There's no bones. No one ever said it's not. \[laughs\] Yes, Mark, you have said it's not... But you shouldn't phone me at 3 AM to tell me.
+
+**Mark Bates:** Well, when should I phone you to tell you?
+
+**Mat Ryer:** Leave a comment on Hacker News, like everyone else. \[laughter\] There's office hours. Please.
+
+**Mark Bates:** You answer phone's full by midnight. I have no choice... \[laughter\]
+
+**Mat Ryer:** Yes. Well, Jon...
+
+**Johnny Boursiquot:** Errors. \[laughter\]
+
+**Jon Calhoun:** So I guess it depends what we wanna talk about. The first obvious thing is, for anybody who's unaware, errors in Go are just an interface. It's an interface that just has the single error method, and it returns a string. And it's weird how powerful that ends up becoming, because it allows you to return nil, it allows you to just return any specific error type you want...
+
+\[51:55\] I find that really useful, because you'll see all this code where people get to return specific errors, and you can actually check them and see what they're doing. It's probably led to some bad patterns too, but it does let you do a lot more with the code that you otherwise could have. So I'd like to explore that more, but I don't really know where to start. Any suggestions?
+
+**Mark Bates:** \[laughs\]
+
+**Jon Calhoun:** I mean, there's a couple things I'd like to look at. The first one is, for you guys, if you're writing code, do you return specific error types, or do you just return an error that has a method and just tell them "Look for this method with an interface"?
+
+**Johnny Boursiquot:** My first pass is with the simple error types... And then if the program gets complicated enough where I care, where basically the call site needs to do different things depending on the kind of error it is, then I'll start using typed errors.
+
+**Mark Bates:** I'm assuming, Jon - real quick, just to clarify - you're not advocating that we don't return the error interface. You're just asking whether we use simple fmt.Errorf, errors.New or custom errors.
+
+**Jon Calhoun:** To give you examples, io has specific errors like end of file, and different things like that.
+
+**Mark Bates:** Sentinel errors, yeah.
+
+**Jon Calhoun:** So there are some like that... But then by using that, you then make anybody who's using your package have a dependency on your package, which a lot of the times when you're using interfaces, your goal is to get rid of that dependence. But then the other side of it is you could return an error that just looks like the error, but then they have to actually check "Does it implement this interface where maybe it has another method of some sort?" And then more recently, one of the things that makes that even more confusing is with all the wrapping of errors, when you start wrapping interfaces, you lose access to some of the embedded methods that are there... Which is something we didn't really get into, but it is a more challenging thing to tackle.
+
+**Mark Bates:** Oh, are we out of time? We've gotta go? We can't talk about this anymore? Oh, no...
+
+**Jon Calhoun:** We can always do another episode on the more advanced stuff...
+
+**Mark Bates:** \[laughs\] As Mark tries to skirt out of the issue...
+
+**Mat Ryer:** I've used the error interface... I use errors.New by default, for sure...
+
+**Mark Bates:** See, I use fmt.Errorf by default.
+
+**Mat Ryer:** Yeah. Well, I tend to use that -- there's an errors package. Dave Cheney, by the way, was the one that coined "sentinel errors". They're the special variable error types that you return. The context package does this. It has canceled and deadline exceeded to errors, that you can then see why the context has stopped. So yes, that's nice, but as Jon said, it becomes part of the API, doesn't it? It becomes a part of the public surface of it, so you then can't change that. You live with that. That's then a design decision.
+
+**Mark Bates:** Sentinel errors also offer a problem in that they can be changed at runtime.
+
+**Mat Ryer:** Oh, right.
+
+**Johnny Boursiquot:** Oh, that's fun.
+
+**Mat Ryer:** Because they're just variables.
+
+**Mark Bates:** They're package-level variables. So you can redeclare io.EOF at any time.
+
+**Johnny Boursiquot:** \[laughs\]
+
+**Mat Ryer:** Julie Qiu does a good talk on finding dependable dependencies, Mark, which I really recommend you watch again.
+
+**Mark Bates:** Okay. Oh, I've seen that talk, yeah.
+
+**Jon Calhoun:** Dave Cheney has a good write-up on making constant errors, but I don't think everybody does it. But it is possible.
+
+**Mat Ryer:** Yeah but it depends what you're gonna do with it, that's the thing. So it's nice to think "Oh, we'll build this system and all these errors will be strongly-typed, and everything will be brilliant", but what's the real use? I mean, you're gonna end up just sticking these errors in a log, or is there gonna be a notification at some point if it's mission-critical?
+
+**Mark Bates:** So what I've found myself doing is I find sometimes for myself internally sentinel errors can be very useful in a few different places. So if I need one of those -- so not even a sentinel error; let me take that back. I just often might need to return the same error in multiple places. File not found, or whatever the stupid error is; resource not found. So I might declare that as a non-exported variable error at the top, that I can just return, but it's not for anybody else to use. It's not a sentinel error, it's not exported...
+
+**Johnny Boursiquot:** It's documentation.
+
+**Mark Bates:** No, it's just more so I can say "return file not found err" as opposed to "fmt.errorf file not found." I can kind of declare the error once and return it. But I'm not telling you to check for it, I'm not making you aware of it. It's just so that I don't have to change --
+
+**Jon Calhoun:** \[56:07\] A shorthand.
+
+**Mark Bates:** Yeah, it's a shorthand. Exactly.
+
+**Mat Ryer:** Well, it does let you change it in one place...
+
+**Mark Bates:** Exactly.
+
+**Mat Ryer:** ...so your methods would just return the error interface. Externally, it just looks like a normal error.
+
+**Mark Bates:** Exactly, yeah.
+
+**Johnny Boursiquot:** Right.
+
+**Mat Ryer:** Of course, it is a normal error, because you either use errors new to make it, or it has somehow that error method on it.
+
+**Mark Bates:** Yeah. And I've been leaning towards the behavior-driven errors... Again, in the last few months, as I've been working more and more towards using interfaces a lot more. That makes more sense to me in terms of asking for information... But I don't return a ton of errors that are customerized like that anyway... But we do have that -- losing the embedded history thing becomes a problem.
+
+**Jon Calhoun:** Yeah. It's tricky, because one of the cases that I'll use errors with extra methods on them for is like if I'm building a web server, I sometimes like to differentiate between an error where I can actually expose some information to the end user, and an error where the end user just needs a generic "something went wrong" error. That's it.
+
+Because I've seen many applications that'll just expose the error every time, and I'm like "That's probably a bad idea." You shouldn't be just printing out strings when you don't really know what's in that string when it gets to the end user.
+
+And then the other area I've seen it useful is if you have users submitting forms, or they're doing something... On the back-end of your code you might have the same code handling an API, and handling forms... So it might want to return something that says "This field is wrong, or it's invalid" or whatever, and then on the front-end you kind of render that differently.
+
+If it's an HTML page, you're gonna render an input box with a red line around it. If you're dealing with JSON, you might have something that says "This is the field that's wrong", to try to help out the developer... So there are some errors that that's useful, but when you start wrapping them, it becomes a little bit trickier. And it's not impossible -- like, with wrapping it's not impossible, luckily, but that's the one case of interface embedding that doesn't cause you to lose it... And that's because of the wrapper type (is that what it is?) that has the unwrap method...
+
+**Johnny Boursiquot:** Wrap error...
+
+**Jon Calhoun:** Yeah. That's the only interface where the name of it is not what the method is... \[laughter\] It always throws you off. But because of that, you can actually write errors.as, or errors.is... I forget which one it is, but use one of those two...
+
+**Johnny Boursiquot:** As, is... Yeah.
+
+**Mark Bates:** Yeah. Were... \[laughs\]
+
+**Jon Calhoun:** You end up having to define a bunch of variables ahead of time, and it's kind of... It's not pretty-looking, but you can do it. So it not being pretty kind of makes you only do it when it's important. So there is one upside to that - you just don't throw it in there for everything. It has to be important enough for this code to look kind of ugly.
+
+**Mat Ryer:** Yeah.
+
+**Jon Calhoun:** But it is tricky sometimes...
+
+**Mat Ryer:** On that too, whenever you have APIs that return errors, or if you're gonna show them in the UI somewhere, I personally think that should be its own explicit mechanism in your code. I don't think we should use error for that. I think error in Go code means something's gone wrong; not that this field doesn't exist, or you don't have permission to access this resource. Those kinds of things should be, I think, done explicitly, because for these reasons it's too complicated, and you expect these different things to know too much about each other. But yeah, that was just sort of an extension on that... Otherwise, I completely agree.
+
+I have to say, we are approaching that special time where we launch our new radio slot...
+
+**Johnny Boursiquot:** BitBar? \[laughter\]
+
+**Mat Ryer:** It's time for your unpopular opinions.
+
+**Jingle:** \[59:42\]
+
+**Mat Ryer:** \[59:50\] So let's go. Actually, for the first time we have an unpopular opinion from our Slack channel. Dylan writes that interface names should be adjectives, rather than er verbs. SO he prefers "closable" to "closer". What do you think about that? Is that unpopular?
+
+**Johnny Boursiquot:** All I'll say is sometimes it is hard to twist a name into following that convention. I mean, I'm with Dylan on that one; you don't have to be dogmatic about it. Sometimes just for readability's sake it just makes more sense to go with what makes sense, right?
+
+**Mark Bates:** I use a combination of both, because some are "-ables" and some are "-ers". Some are more describing, and some are more doing. Some are more verbs and some are more adjectives, or adverbs, and I think that's fine. I don't think you have to be dogmatic about it.
+
+**Jon Calhoun:** Yeah. I think it's nice to have a general guideline to get everybody on the same page, but it's not -- it's kind of like the... It was mentioned in the Slack channel as well, the "accept interfaces return structs." It's not a rule, it's a guideline to get you moving in the right direction... But there's always exceptions to that.
+
+**Mat Ryer:** Yeah. \[unintelligible 01:01:02.25\] on Slack says that they use a prefix for their interface names. And I know that in C\# it was a tradition to use like iClosable, so that you know it's an interface. Does anyone use prefixes or suffixes or anything like that?
+
+**Mark Bates:** No...
+
+**Johnny Boursiquot:** If I see i in front of any interface, that developer and I are gonna have a little chat. \[laughter\]
+
+**Mat Ryer:** You refuse to implement it. "I'm not gonna implement that, ever. I'm not gonna implement that interface", which takes a lot of work.
+
+**Mark Bates:** Well, again, other languages do it, and it's idiomatic in other languages, so I think that's fine for those languages... In Go it's not idiomatic, so... You know, if a PR came across that had that for me, I would probably ask them to change it, just because it doesn't conform with idiomatic Go. Not for reasons I may or may not agree with, it's just that's kind of what it is.
+
+**Jon Calhoun:** All of these are interesting too, because -- so an example of a company going completely against style guides for a language is Google was pretty notorious for going against the Python style guide slightly internally. And even when the creator of Python started working at Google, he had to suddenly not use his own style guide, which would have been frustrating, I'm sure... But I think if you have an organization where your entire org is using the i prefix, then by all means, keep it consistent there; that's probably more valuable than being idiomatic. But if you're working on open source, then you need to conform to whatever the norm is there.
+
+**Johnny Boursiquot:** I'd say find out during the interview... \[laughter\] Because I would seriously have a problem with that.
+
+**Mark Bates:** "So Johnny, the interview is over... Do you have any questions for us?" "Do you use any prefixes when naming your interfaces?" "Well, yes, we do." "I'm out..."
+
+**Jon Calhoun:** "I just need to see some code. I need to see some legit production code with interfaces in it."
+
+**Mark Bates:** "I need a 10% bump, or I'm out." \[laughter\]
+
+**Mat Ryer:** I like that there's still a price though...
+
+**Mark Bates:** There's always a price.
+
+**Johnny Boursiquot:** Like, "I'm willing to overlook this, but you've gotta make it worth my while."
+
+**Jon Calhoun:** They'll have an intern just write plugins for everything you use that just removes it...
+
+**Mark Bates:** \[laughter\] Exactly.
+
+**Johnny Boursiquot:** It just hides it and puts it back in during commit...
+
+**Mark Bates:** Yeah, exactly. It rewrites on save. \[laughter\] Your own version of go fmt. Just puts an i in front of every interface.
+
+**Mat Ryer:** GoTroll. It could be a tool. We could make that.
+
+**Johnny Boursiquot:** \[laughs\] Nice.
+
+**Mark Bates:** Unpopular opinions, I guess...
+
+**Mat Ryer:** What about you, Bates? Have you got an unpopular opinion, mate? Have you got a popular one?
+
+**Mark Bates:** No, I don't have any popular ones... Everybody knows that. It was difficult choosing \*an\* unpopular...
+
+**Mat Ryer:** Yeah, that's the bit that you struggled with, wasn't it? The \*an\*.
+
+**Mark Bates:** That was the bit that I struggled with... And I think I'm gonna come up with "I don't like the way that the main package and the main function is designed."
+
+**Mat Ryer:** I see.
+
+**Johnny Boursiquot:** Explain, explain...
+
+**Mark Bates:** \[01:04:05.04\] Explain... Yeah, I think it promotes global scope, for example, and os.Args comes to mind. We were just talking about you can redefine io.EOF, and the problem with CLIs is if you're not immediately taking that os.Args and handing it off to something else, it's hard to write tests around; everything's kind of globally-scoped. Present working directory - well, technically it's global... Again, it makes it hard to test if you're talking about those things.
+
+So I feel like that, and a context. We have no context when we're in there. And admittedly, that was all after; context came out later. But if the main package was exportable, if we could call it, if the main function was exportable, and took a context, a current working directory, the arguments, and returned even a basic error... They'll let us do os.Exit() or whatever, but if we return an error, just do a default exit of some kind. I think that allows for better-tested CLIs, nicer-looking code that Go can give us that information at runtime. That's not difficult information to give us... And I think it promotes generally a better way of writing our CLIs.
+
+Right now I feel a lot of CLIs get written in the main function by accident, just because people are hacking away, trying to get something right, and then they've got a big, long main.go file that's not very well tested, or broken out, and other people can't make use of that CLI without compiling and shelling out.
+
+**Mat Ryer:** I completely agree with that, actually. I solved that problem though by -- I have a little run function, and that takes in the Args, and it takes in an io.Reader and the writer (if there's STDIN/STDOUT) and returns an error. And then I just have a standard little main... I do create a context in that main, which is canceled when Cmd+C is hit the first time. That cancels the context.
+
+**Mark Bates:** Right.
+
+**Mat Ryer:** Then the second Cmd+C exits the program. Because you don't wanna be annoying people if it's hanging for too long, or something.
+
+**Mark Bates:** Right.
+
+**Mat Ryer:** And then you can write test code and just call run and pass in the different slice of string for your arguments. So you could pass in a different writer; you could use a buffer, so you can read then what was written by your tool...
+
+**Mark Bates:** My biggest problem with that - which again, giving it to another function is a good thing... But my biggest problem there is as a third-party I still can't use your code programmatically from Go.
+
+**Mat Ryer:** Yeah, they're different. Packages and programs are fundamentally different.
+
+**Mark Bates:** Yeah. So the way I'm trying to solve it in my code now is my main is very simple - context, background, get the slice of Args, get the working directory, and then hand all of that off to an exported main function that takes those things in a package that I can then work with... And I basically don't have to look at the main.go file ever again, right? Now I'm just kind of off in Go land, and you can come along and import it, and you can pass it a context, a working directory and some Args and start using my CLI in your program. And it's really nice and clean, and kind of top-level... I don't know, I've been finding that as a pattern it's been working really well for me recently.
+
+**Mat Ryer:** \[01:07:22.26\] Do you shell out or do you call them directly?
+
+**Mark Bates:** What do you mean?
+
+**Mat Ryer:** Do you create a command exec and run an actual process? Is that how you run things? Or do you just call...
+
+**Jon Calhoun:** I think he just has a method; he calls a method -- or a function, on another package.
+
+**Mark Bates:** Yeah, exactly.
+
+**Jon Calhoun:** He might name the other package Mark, and it might have a main exported function, and he calls mark.main inside of his actual main that doesn't do much.
+
+**Mat Ryer:** I bet he does have a program called that.
+
+**Johnny Boursiquot:** I was thinking the same thing. \[laughter\]
+
+**Mark Bates:** I've been leaning towards a CLI package, and then having a type -- not even a top-level function, but a top-level type there, whatever it is, and that has the main function on it. Again, no scope; I don't want any global scope here. A zero value struct should be able to handle that CLI. And like I said, it's a pattern I've found has been working really well for me, because then I can kind of manipulate whatever I need to just with those three pieces.
+
+**Jon Calhoun:** It is nice though that Go makes it easy enough to do that. You've found a pattern that works for you, and you can sort of build around that. I get what you're saying, but I also feel like because it's so easy to just build around it, that it's kind of not that much of a limitation.
+
+**Mark Bates:** No, it's not necessarily a limitation. It's just an unpopular opinion, by the way... \[laughter\] I'm just saying, if they were to rethink it for v2, those would be my suggestions for --
+
+**Mat Ryer:** Well, on those bombshells of suggestions...
+
+**Mark Bates:** Wait, wait, where's Johnny? Isn't Johnny supposed to come up with one today?
+
+**Johnny Boursiquot:** No, man. That's next week.
+
+**Mark Bates:** Oh, that's not cool...
+
+**Johnny Boursiquot:** I need time to think about this.
+
+**Mark Bates:** Oh yeah, fine...
+
+**Mat Ryer:** He's too nice, he's too nice...
+
+**Mark Bates:** That's right, he's too nice. I remember now. You broke into a cold sweat when we said we might upset somebody... \[laughter\]
+
+**Mat Ryer:** Well, that just makes him a nice guy, doesn't it? \[laughter\] Mark's trolling him...
+
+**Jon Calhoun:** Why can't these two upset each other? Mark's nowhere near too nice...
+
+**Mark Bates:** Oh yeah, absolutely...
+
+**Mat Ryer:** And Mark trolls him for being nice. That's where we've got to... \[laughter\]
+
+**Mark Bates:** That's how evil I am.
+
+**Mat Ryer:** Well, welcome to the internet... And I'll say our goodbyes. We've reached the end of the show. Thank you very much, Mark, Johnny and Jon. Hopefully everyone's learned a little bit about interfaces and abstractions, and grappled with them, as you go into your future endeavors. We wish you all the best, and we'll see you next time.
+
+**Break:** \[01:10:02.11\]
+
+**Mat Ryer:** Recording!
+
+**Johnny Boursiquot:** I see bytes, and kilobytes...
+
+**Jon Calhoun:** Alright, that's good...
+
+**Mat Ryer:** That's great. I want that as my ringtone.
+
+**Johnny Boursiquot:** Kilobytes...
+
+**Mat Ryer:** "Incoming kilobytes..."
+
+**Jon Calhoun:** Alright. Are we cool with me going live?
+
+**Mark Bates:** Yup.
+
+**Mat Ryer:** Yeah, man.
+
+**Johnny Boursiquot:** Do it.
+
+**Jon Calhoun:** Alright, we're live. Two minutes early.
+
+**Johnny Boursiquot:** Well, we still have to wait like two minutes, yeah, to let people in...
+
+**Mat Ryer:** Okay, okay. Two minutes...
+
+**Mark Bates:** Let's all sit in uncomfortable silence then...
+
+**Johnny Boursiquot:** Well, I don't think it has to be silent.
+
+**Mat Ryer:** It can still be uncomfortable and we'd be talking... \[laughter\] We don't need to be--
+
+**Mark Bates:** Well, if we do that now, then what are we gonna do for the next 60 minutes? \[laughter\] Not talking, make things uncomfortable...
+
+**Mat Ryer:** Yes.
+
+**Johnny Boursiquot:** Wait... You all weren't kidding about that? \[laughter\]
+
+**Mat Ryer:** No, we just genuinely ran out of things to say.
+
+**Jon Calhoun:** That's it, folks... We're done with the episode.
+
+**Johnny Boursiquot:** \[laughs\] Come back next week.
+
+**Mark Bates:** Is that it? That was amazing. That hour flew by. Thank you for having me.
+
+**Mat Ryer:** It's the best hour I've ever spent with you, Mark...
+
+**Johnny Boursiquot:** \[laughs\]
+
+**Jon Calhoun:** In silence, huh?
+
+**Mark Bates:** Definitely. Absolutely.
+
+**Mat Ryer:** Yeah, we're not -- this isn't the show yet...
+
+**Mark Bates:** I hope not.
+
+**Mat Ryer:** No.

Reflection and meta programming_transcript.txt

@@ -0,0 +1,367 @@
+**Mat Ryer:** Hello, and welcome to Go Time. Today we're talking about reflection and what that means in Go. We'll talk about the Reflect package, what you can do with it, and some really interesting use cases for it, some examples around the standard library even, and then we'll get opinionated about it, no doubt.
+
+Joining me today, it's Jaana B. Dogan. Hello, Jaana.
+
+**Jaana Dogan:** Hello!
+
+**Mat Ryer:** Welcome back. How are you doing?
+
+**Jaana Dogan:** Good! How are you?
+
+**Mat Ryer:** Yeah, not bad. Thanks. Jaana, if you don't mind me saying, you don't sound that happy to be here... But don't worry, this will cheer you up. Jon Calhoun is also here. Hello, Jon!
+
+**Jon Calhoun:** Hey, Mat. No, she looks like she's deep in thought about something. Nobody else can see the video that we can, but she looks like she's deep in thought about something.
+
+**Mat Ryer:** Yes. Well, she's reflecting.
+
+**Jon Calhoun:** She's reflecting.
+
+**Jaana Dogan:** I'll reveal what I'm doing... Exactly.
+
+**Mat Ryer:** Okay, so maybe we should just start at the beginning at a high level. For anybody not familiar, what is reflection, and what does the reflect package give us?
+
+**Jon Calhoun:** At a high-level, it's kind of just a way of almost meta-programming, or interacting with code at runtime; that's the way I view it. I don't know what the official definition is, but everything I've ever seen that's using it is - while your code is running, you want to examine other pieces of code, or look at other things and find some information about them, or maybe try to modify different aspects of them... So it's these things where rather than doing it as a developer at coding time, you're doing it later, when the program is running.
+
+**Mat Ryer:** Yeah, so dynamic languages do that all the time, don't they? Ruby, and I think even JavaScript... Because in JavaScript you can take a string and add methods to it at runtime, and you can kind of do anything you want, it's a really flexible language. And one of the things about Go is it is a strongly-typed language, so it sort of deliberately doesn't do that, but of course, the Reflect package is an exception to that rule.
+
+**Jon Calhoun:** \[04:11\] Like you said, the dynamic languages - it's almost like it's not even thought of as something separate... Like it's just part of the language. It's just something people naturally do. And if you've ever coded in Ruby or one of those languages, it seems so natural, because you see everybody doing it. It's not something that would stick out in any codebase. But when you get into Go, not only is it specific, like you have to import this reflect package, but it's also very limited in what it can do, and I think that's intentional, and it doesn't sort of go with the story of what Go is trying to achieve.
+
+**Jaana Dogan:** Coming from a strongly-typed background, I was about to say that reflection is everything that your type system cannot provide as a first-class capability... But then I was looking at the Wikipedia page, and that's why I was so confused about the definition. I was in the thoughts, and Mat was thinking that I was being sad... It says "Reflection is the ability of a process to examine, introspect, and modify its own structure and behavior", so basically it's everything. And that actually makes sense, if you think about the meaning, what reflection means in daily language.
+
+**Mat Ryer:** Yeah, yeah.
+
+**Jaana Dogan:** So I think it's not even bounded to that little -- I mean, I try to over-specify in my mental model; it's just more of like everything. Everything about introspecting and modifying the structure and behavior.
+
+**Mat Ryer:** Yeah. I mean, when you do type assertions in Go, in a way that's a kind of version of this, isn't it? At runtime you're saying "Here's a type, and we don't know what this is, but I'm gonna assert it to a particular type, and if that's successful, I can then branch off and do something." So in a way, that's kind of reflecting, isn't it, a little bit? But that's still happening at compile time, isn't it?
+
+**Jon Calhoun:** Yeah, you can definitely put more checks there from compile time... But the actual checking - I'd assume some of that has to happen at runtime, because you don't actually know.
+
+**Mat Ryer:** Yeah, you're right.
+
+**Jaana Dogan:** Yeah, assertion is happening at runtime, so you can say it's an introspection piece, and it's actually reflection. But the type system has provided a very nice feature, for us to do it in a nice way, rather than depending on a reflection package, or something. So you can say that "Yeah, that's a reflection feature", but represented in the language with more synthetic sugar.
+
+**Mat Ryer:** Right, yeah. And it does help as well, it does some checks; you can't do invalid-type assertions, and things. The compiler will help you at some point. But you're right, of course, it has to be at runtime. That's kind of the point of it.
+
+**Jon Calhoun:** Yeah, it's definitely interesting in that sense. The type assertion one is one of the ones that everybody's probably seen; I'd say that the second most common one people have probably seen is struct tags. And while they might not be using them directly themselves to read them, I'd say that most people who write Go have at least seen struct tags and wondered "What is this thing?" So I think that might be another one to jump into with Go, because I definitely think that's the second most popular use case.
+
+**Mat Ryer:** Yeah. So anybody that's not familiar -- and you see this particularly if you're working with JSON data... You can actually put a string after a field name in a struct, and that string can be parsed at runtime, and of course, it can take meta-data out of it... And the JSON example allows you to specify the field name, so you can have a different field name to the one that you're using in the struct. And you can also optionally choose for it to not include that field at all. And you can also -- there's like a syntax with a string and then a comma, which is quite a strange part of Go, actually. It's quite unusual, isn't it? There's not much else like it. And you can also tell it to omit the field if it's empty. So if it's a default value, then it won't be included in the JSON object.
+
+\[08:17\] I remember when I first saw that... It really felt like almost a temporary thing, but it has proven to be very useful and pretty effective, in particular for that use case. But Jon, you wrote something that used struct tags as well, right? That form project.
+
+**Jon Calhoun:** Yeah.
+
+**Mat Ryer:** What was that?
+
+**Jon Calhoun:** So I've done a couple different things... Historically, I've used reflection in a lot of different projects. I come from a Rails background, and -- Rails is essentially a large exercise in using reflection. That's kind of how I view that whole framework. So it was nothing as crazy as that, because in Go I just don't think that makes as much sense. But I wanted to write some code that -- essentially, I wanted to take a struct and I wanted to generate an HTML format of that, and I then I wanted to later be able to parse that form whenever a user submitted it, and to basically take all the values they submitted and put it back into that struct, just to make my life a little bit easier, so I could re-share that across a couple different handlers, and just make that sort of thing simpler.
+
+So I created this Form package that used struct tags... And there's other ways to approach this, which I think we should talk about... But it was kind of me just looking at the problem and being like "Is there a way to handle this?" And the struct tags were used for things like if you needed to change the name. If the name in your field had... Say you had a struct, and one of the fields was named "Email", but you wanted it to display "e\_mail" in the actual form, like as the name in HTML - you could use struct tags to change things like that. That's where I was using it. But it also was like an interesting exercise, in the sense that it showed how confusing Reflect can be to write/use in Go.
+
+So I think that's something that a lot of people struggle with when they get there. And part of me feels like that's intentional; it's something that they did -- not like they wanted it to be worse, but they didn't want it to be so easy that people just jumped in there and used it for everything, when they shouldn't be.
+
+**Mat Ryer:** Yeah... Yeah, because there's so much benefit to the type safety. That makes sense, doesn't it?
+
+**Jon Calhoun:** Yeah. And I also even found -- when I was doing this, I probably used struct tags more than I should have in that. Some examples of that were like if you had helper text or default values or anything like that for input fields, I actually had it so you could put struct tags in there that provided those values... And as a result, you could have a pretty crazy, long struct tag that was thrown onto something... And it looked kind of wonky, because you're like "This isn't really code, it's meta-data." But it's providing a lot more than what it seemed like at first glance.
+
+**Jaana Dogan:** Mat, you said something very interesting - that when you first saw it, it looked almost temporary. That was exactly how I felt, because you know, I have these concerns as well... Like, "Hey, Go is a very strongly-type, simple language", but sometimes I feel like I'm over-abusing struct tags... And I was expecting something like an annotation; in other languages we have annotations, and you can have typed annotations, and annotations can handle more complex situations without sacrificing too much from the type safety. And I was expecting something like that, and this was one of the earlier -- a very long time ago, in the early beginnings of the language... But they wanted to keep everything small; the language didn't really grow to have annotations... And one thing that I realized is that I don't see a lot of big mess when it comes to struct tags.
+
+I think people use one struct tag at a time, for very specific things, like for example JSON keys, and that sort of thing. What's your opinion? Do you think it's at a level that we don't actually need anything like an annotation, or do you think that it's a missed opportunity, just because struct tags are hard to maintain, and so on, and we are not doing a good work or we're missing some opportunities to annotate fields in a more richer way?
+
+**Mat Ryer:** \[12:15\] Yeah, it's a really interesting one, because there's definitely value in being able to add a bit of extra meta-data for a particular use case to structs. The alternative has to just be you would describe the same thing just using strong types. So Jon, in your case, instead of having an address struct with different fields and then use the struct tags to add labels and placeholders and help text and things, you'd have a form type and then a field type, probably... And it's quite verbose, but it's very clear. So I guess that's the benefit. But I always was told - and I don't know about this, but I was always told that it's slow, that struct tags are slow to parse. Is that the case still? Has there been any work optimizing that, or is it actually pretty quick?
+
+**Jon Calhoun:** I don't know, but I've never worked on a project where that type of speed has mattered. If I'm rendering HTML and sending it back to the user, chances are sending that HTML back to the user is gonna take significantly more time than whatever the reflect struct tag parsing is gonna take, so it just wasn't a major concern.
+
+And I will say, when you talked about having a form type - I need to get this into a gist and share it, maybe I'll put it in the show notes, but I actually have two versions. One I actually did with a form package that takes something like that, takes a struct and just generates some HTML if you provide an HTML template... And then the other one is more of a -- you describe a form type. I have another struct that would be like "This is my sign-up form struct", but I would write a method on that that used my generic form type and basically spit out what that should look like... And I knew how to render that in HTML with my templates. So I didn't use Reflect at all in that version... And you're right, it is a lot more verbose, but I think that in some ways it's definitely better, because it's much clearer what's happening.
+
+And then in other ways -- it kind of just depends on the type of project, I think is what it comes down to... Because for some quick project where you wanna throw a form up, it's nice to have that "This package just does it." In other cases, where you're going for like "This is gonna be a much longer-lived project", we might need way more customization of stuff. Sometimes it makes sense to go with something a little bit easier to change and a little bit more verbose, but as a result, it's still gets you the same results.
+
+**Mat Ryer:** Yeah, I know that the old data store in App Engine used to use them... Usually, it was around field names, but you could also specify that you didn't want an index to be built on a particular field that you were then gonna put in the data store. I mean, it's extremely powerful to just be able to annotate your structs in that way, and it makes sense, because you really are talking about properties of that field in a very real way. So yeah... And you know, typed annotations - I remember that from C\#. I think Java has it, too.
+
+The idea there is that you have actual types in your code, and you can use those to annotate fields, and then I suppose you can check for the existence of those, and you can probably interrogate them, and programmatically process them. That's a kind of cool meta-programming approach, and you still probably get a lot of type safety with that, too.
+
+**Jaana Dogan:** Yeah. Also, maintainability is higher. You can also easily run queries. You can ask your editor "Hey, show me all the users of this annotation." Or let's assume that you wanna modify a value in an annotation - you can easily search for that and just go in and refactor everything all around. So having some type safety allows you to do that... But again, as I said, I don't think we're over-abusing struct tags in Go. Maybe it's because they're not typed, so everybody's cautious about not overusing them... So I think we still have a nice balance; they're very small. But the biggest problem is people are afraid of -- since they're unstructured and you need to parse them, people are kind of scared of the maintainability, as well as some of the performance implications, if there's any.
+
+**Jon Calhoun:** \[16:30\] I agree entirely with what Jaana said, the sense that it might be because they're a little bit harder to maintain, that people don't use them as much... Whereas if you put those annotated tags in there, I almost wonder if people would use them more than they do now, and use them in cases where they don't make a lot of sense... Because I've even seen that with struct tags. I think there's a class of problems that make sense for struct tags. Encoding with JSON, or really encoding with almost anything that's similar to that... Encoding is a great example where your struct might not match exactly how it needs to encode, so you need to have some way of defining how it should encode and decode. And ORMs are kind of the same, they kind of fall in that same class, where if you're building an ORM where you just wanna quickly say "Put this into the SQL database. Here's what the names of the fields are in the SQL database" - that makes sense.
+
+But then there's other libraries out there, like validation libraries, where you throw things like "This field is required" - and I'm not trying to say people should never use those, but I do see those as potentially being problematic long-term. They could lead to code that has a bunch of different struct tags, all littered in this type, and it's really hard to understand what each struct tag is actually doing, how they all interact together; and there's no compiler safety. If we put annotated tags in there, I wonder if people would just be more willing to do that, rather than look at other approaches.
+
+**Mat Ryer:** Yeah, I've seen linters that check the JSON tag. So if you miss a quote, or if it's malformed in some way, then some -- I had it once where some linter would say "Oh, this is malformed." It's not done by the compiler, so it's not quite the same safety. But I wonder if there is a benefit of it not being a very attractive API, that people tend to avoid it for that reason. And it also does feel a little bit magic. One of the things - particularly for me, and I hear this a lot - that attracted me to Go was that it didn't have much magic in it. It was a very clear and simple language. So now I might even be too far the other way, kind of allergic to magic, despite having the appearance (some have told me) of a magician.
+
+**Jon Calhoun:** Yeah, the magic part -- it's hard, because... I remember the first time I used struct tags, when I was first learning Go. I think I was doing something with MongoDB... And you use (I think it's) Bison to define the struct tags... And I remember when I was doing that, I'm setting up the struct tags, and in my head I'm sitting there thinking "Do I need to import something for this to work? Why is my code okay with this being here without me importing that?"
+
+**Mat Ryer:** Oh, yeah.
+
+**Jon Calhoun:** And it really confused me for a while, because I'm just like "I don't understand how this is compiling." And it wasn't until later when I dove into stuff and sort of understood it, but at the time it really did feel like magic, and that was slightly frustrating when I was first learning about it... Because I'm like "I don't know what's going on."
+
+**Mat Ryer:** I mean, it's literally a string, isn't it?
+
+**Jon Calhoun:** Yeah, but you're like "Surely the compiler is doing something with this. It leads to things being written somewhere", so you're like "How does that work?" and it was just confusing for a while. And then later, when you realize "Oh, they're just parsing this string, and the Bison package later when it's being used is doing it", then yeah, that makes more sense. But at the time I was just very confused as to what was going on.
+
+**Mat Ryer:** Yeah. The Reflect API to interrogate those struct tags actually is quite good. It's quite a simple API. Because some of the Reflect package is -- it's so meta, isn't it? And some of it makes sense. You can get the value of something, and the value is a struct. It's a strong type in the Reflect package. And that describes the value. And then the values, of course, because they can be lots of different types of things, you end up with loads of methods that most of the time are illegal to call.
+
+\[20:19\] If you try and get the length of an integer, of course, the methods are there in the Reflect package to do that. So at compile time you could do it, and it's only then at runtime you're gonna find out you can't get the length of an integer. There's lots of examples of that, so you do end up checking everything. You'd be very verbose when it comes to writing defensive code to make sure that you're not gonna have any of these runtime weird things. And obviously, testing helps, but...
+
+**Jaana Dogan:** Yeah, you mention testing, but it's hard to test, as well. There's no canonical set of tests that you wanna run. I worked on some of the database packages, and there was all this -- like, since Go doesn't have generics, which is something that maybe we can discuss in the context of this talk, we rely a lot on interfaces and type inversion. If you have a slice of interfaces, it could be either a value or a pointer, or a pointer of pointer, and then you have to do all that magic by using the Reflect package, and the reflect package is already a very verbose thing, so wrapping and unwrapping all of those types is very hard. I couldn't figure out an easy way to test, because there was no set of canonical things, like hey, if the standard library was providing maybe some sort of like "Hey, please consider testing these", or providing a list of things to test, that would be so much easier.
+
+**Mat Ryer:** Yeah... Because you have to probably test for all the different possible types, and things like that... And then of course, arrays and slices.
+
+**Jaana Dogan:** Exactly.
+
+**Jon Calhoun:** It's interesting -- Mat, your example of one of the easier use cases was getting the value of something... And the funny part was that was one of the first ones I had issues with when I first used the Reflect library... Because when somebody passes something in -- they pass in a string; you're like "Okay, I'm gonna get the value of this." And that makes sense, your code works, and everything seems great. And yeah, there are things like length, and stuff like that that might not work, but that isn't always gonna work. But then somebody passes in a nil pointer that has a type, but it's a nil pointer, and all of a sudden your code breaks and you're like "What just happened?" and you end up with these weird cases where if the type's kind is a pointer, and if it's nil, then you need to use reflect.new to instantiate a new element, and if it's an interface, you need to get the underlying element type that it's pointing to, because the interface doesn't really help you much...
+
+There's all these weird cases that you get to, where it seems really simple, like "I just want the value of this", but that's really not what's happening. So you end up with all of these edge cases... And even once you get it working and you have some tests for like "Okay, we pass in an empty pointer that has a type. We pass in an empty interface, we pass in an actual value set to that interface" - you have test for all these, and at the very end of it you're still thinking "I don't know what edge cases I'm missing", because there's gonna be one. There's almost no way that there's none.
+
+**Mat Ryer:** Yeah. I mean, in a way it's leaking the internals of how Go actually works. You do tend to learn quite a lot about the types system if you do have to use it... But frankly, I often end up in a kind of trial and error situation, relying on a TDD process to tell me if I got it right or not. I've often written -- if I've used the Reflect package, I often will have code where I call Elem() to get the element, and then for some reason (I'm not sure why) I have to call it again... And it's like "There will be a great reason for that, but I don't know what it is, and I don't really have time", and I know that if I just call elem.Elem(), then I get the thing I need in this case, because the test passes... So I end up being very sort of brute force when it comes to reflection code, which doesn't feel great.
+
+**Jon Calhoun:** \[23:59\] Yeah. I don't use TDD that often, but using Reflect is one of the cases where I most definitely use it, because I'm like "Here are all the different things I know I'm gonna put into this, and they all need to work", and it's just so much easier to start with that. Otherwise, you're just like "Yeah, this works", and then you've gotta run it and nothing works, and you're like "I don't know what's going on."
+
+I'm just looking at some code that I've written with reflect, and I'm seeing the same thing, where it's like .type.Elem() and then like once you create that reflect.New() using that you're like .Elem() for that and you're like "Looking at this code, I have no idea why I did these things. I just know they work", which is really weird. 
+
+**Jaana Dogan:** One thing that I realized - I think the current type system of Go is contributing to some of these problems... Because we kind of fall back to this interface as an argument, or slice of interfaces as an argument, and then all that type inversion, just because we can't limit what the user will wanna do, or what the user will wanna pass... Like, you have to handle all of those cases for your library to work.
+
+One typical example from the spinner Go packages we have is it has to do type inversion from the argument that the user is passing, which is like some interface... And it could be anything; it could be a struct, or it could be a pointer to a struct, or an array, or whatever, but it has to know about the type by doing type inversion, so you shouldn't be passing a regular nil, and you should be passing a type nil. So Go has all this weird stuff, as well as like doesn't have anything for generics, so it invites all this complex stuff to be handled by the libraries by using the Reflect package, and I think it contributes to the problem that we're all experiencing by doing all these elem.Elem()'s, and don't necessarily understand why... The entire language is somewhat contributing to the problem.
+
+**Jon Calhoun:** That's another good point you made - if you're working with Reflect, you're almost always accepting the empty interface. That's almost always what your argument is, and that's usually a bad sign for code when that's what you're accepting.
+
+**Mat Ryer:** Yeah, but like you say, in some cases it's unavoidable...
+
+**Jon Calhoun:** Yeah.
+
+**Mat Ryer:** And one of the things that a lot of us use every day is the JSON marshaling and un-marshaling, and with that thing you can pass in any type, because of course, it can un-marshal into a struct type that you've written... Or a map, actually; a `map\[string\]interface{}`. It can do that, no problem. And actually, the reflect package can instantiate things too, can't it? If you pass in a map, it will create the map for you... And things like that. So it does get quite strange. And I remember in the early days I wanted to -- I was writing a mocking thing for testify, and I really wanted at runtime to create the mocked struct from an interface or from another struct. And at the time you couldn't do it, but since then I've actually seen -- I don't know if it's possible, but I've seen functions and methods that seem like you can actually now instantiate structs, and things; I'd have to check... But that is quite powerful, and if you think about we don't have generics, it is quite tempting to have a look there and see if you could do the hard work and get it done, and then have this extremely intelligent dynamic functionality... Which would be very interesting.
+
+And in test code maybe you're okay with that not being -- it's not gonna be in a tight loop; you don't want test code to ever be slow. But it's not gonna be in a low-latency situation running tests all the time. Of course, we still want test code to be fairly fast...
+
+**Jon Calhoun:** Yeah. Like Jaana said about how the typing system is limited, and then you mentioned the JSON encoding... And I'm sitting here, thinking like - even cases where you know you have to pass in a pointer, you can't just pass in the struct; you have to pass in a pointer to the struct to get the values back... And even having a typing system that sort of allowed you to restrict that would have helped, but because of the way things are set up, it can't do that. Instead, you have to rely on maybe an error, or something... This is not meant to be like bashing Go or something, it's just it's a struggle sometimes when you see that, because I'm sure it's confusing for some people.
+
+**Mat Ryer:** \[28:14\] Yeah. Well, what would the JSON package look like if it didn't use reflect? I mean, you almost certainly would have some kind of callback, but you'd still have interfaces, because you don't know the value type.
+
+**Jon Calhoun:** Yeah, it would almost have to be something like "Encode this", and then instead of saying "Pass in an interface", it would almost have to say "It has to be a pointer." It would almost have to be something along those lines. But even then, that's confusing, because maps don't always work that way, if I recall correctly. I think you can just pass a map in there and it doesn't have to be a pointer, but I don't remember... Does it have to be a pointer?
+
+**Jaana Dogan:** Yes...
+
+**Jon Calhoun:** I haven't passed a map into that in so long that I -- I should go check...
+
+**Mat Ryer:** Oh, no... What have you been passing in?
+
+**Jon Calhoun:** Structs...
+
+**Mat Ryer:** Oh, yeah. It makes sense.
+
+**Jon Calhoun:** I decode intro structs most of the time.
+
+**Mat Ryer:** Well, if you're writing anything where you don't know the data structure -- you know, so many APIs do do that.
+
+**Jon Calhoun:** Yeah.
+
+**Mat Ryer:** It is kind of dangerous territory. But if ever you don't know the actual types... I mean, I wrote a little -- it's not finished; it kind of works, but it's not by any means ready... But it's basically a fake JSON data generator. So you pass in any data -- I mean, you can pass in a struct, in fact... And it will generate lots of examples of that struct, and it uses JSON to do it, because actually just marshaling to and from JSON, at least in the API, is a very easy thing to do. So in that sort of case, yeah. If that was an API hosted on a website, you would want people to be able to pass in any kind of JSON, including an array of objects, as well as a single object... And then it could maybe then generate some test example data from that. That was the idea. And that was very meta.
+
+So those use cases aren't as common, I suppose... But the JSON API I think is great for when it comes to just -- you know, as a user of it. It would be strange -- if you didn't have Reflect, you would end up with some function where you get given the key as a string, and maybe the value as some bytes, and then it's up to you, based on your knowledge of your particular situation, to then unmarshal those bytes. So it is nice that the standard library does that for us, for sure. And by the way, if it didn't have that, I think that would hurt Go's reputation. Imagine the Hacker News article on this, about you have to do JSON marshaling...
+
+**Jaana Dogan:** Yeah, Go wouldn't be adopted as widely if that was how it worked.
+
+**Mat Ryer:** That's right, absolutely.
+
+**Jon Calhoun:** Even with it as is, there's already a couple cases with JSON that are challenging... Like you said, you have structs, but -- I think Stripe is an example of this, where payment sources can be a card or like a bank account, so you have this array of things that can be different, and you kind of need to write your own type to unmarshal it correctly... So you end up having to write some custom stuff for that. And I imagine if you just didn't have the JSON package at all, it would just be a nightmare of people complaining and saying "This is awful." Because even in cases like that, where you have to write custom, I still leverage as much of the JSON package as possible... Like, make a struct with just that one field I want, unmarshal it, figure out what it is, and then pass in the struct for whichever type I care about... And that just saves me the work of doing any of the actual overhead of "How do I umarshal this?"
+
+**Break:** \[31:40\]
+
+**Jon Calhoun:** Earlier, Mat, you were talking about struct tags, and one of the things Jaana and I had sort of talked about a little bit before the show started was, in some ways, I kind of feel like struct tags would benefit being its own separate library, would benefit from that... Because then you can separate this... I feel like struct tags are the safest version of reflection out there, like where you import the Reflect package, and then everything else is kind of the -- maybe not worse, but it's definitely a little bit scarier... So having that edge case, where you could just pull out struct tags - it might be useful to actually have as a separate, like "Okay, I'm just looking at struct tags here."
+
+**Mat Ryer:** Yeah, I see what you mean, so you that you don't have to have the entire Reflect package imported into your code. And I think the Reflect package also has unsafe in there, although I think as a lot of very normal packages also do have unsafe... But yeah, I see what you mean... So that in order to parse the struct tags you would import a different -- like a reflect/struct tags package, or something. Hm, I quite like that. You should tell someone about that.
+
+**Jaana Dogan:** I remember in the earlier days of Go, they were saying like "Hey, if you're importing the Reflect package, that's a no-no." It was almost considered unsafe, because you also are dependent on unsafe for a lot of other reasons, and so on... But you know, it was one of those import lines that you should never see, or you should be very careful; if you're ever using it, you should be very controlled about your usage, and so on. But you know, all of a sudden, everybody starts to import Reflect, because it does a lot of fundamental things, like the struct tags... So I think it just kind of also gives a lot of mental separation to the user, if it was a separate package, so you can write linter tools and whatever to catch for import reflect... But you know, some of the basics or some of the more easier concerns can be living in different packages.
+
+One of the things that I've seen related to this was people -- if they wanna rely on the reflect package, they don't necessarily import it all around; they just kind of like go and encapsulate all the reflect usage in a different package, and then they provide some utilities from that package. Have you seen anything like this, or have you done anything like that?
+
+**Mat Ryer:** No, but that makes sense to me. At least then you've got all of that weirdness in one place... But I don't know if that's a healthy approach, because that's kind of like Kitchen Sink or Utils sort of...
+
+**Jon Calhoun:** I've definitely done the "have one source file, where all my reflect stuff got put", but I've never done anything big enough with Go's reflection that I've had to go that far... Now, I can say that I'm definitely guilty of in like Ruby going crazy with some of the meta-programming stuff, but I kind of feel like when I went to Go I just didn't feel like writing Go code; it didn't seem like the right approach, so I kind of veered away from it where I could.
+
+**Mat Ryer:** \[35:38\] Yeah. I saw an example where somebody wanted to be a very good citizen, and they were gonna put some data in the map, and if the map was nil, it was panicking, of course... So they were actually using (I think) the JSON unmarshaler; in the case that the map was nil, it would just marshal -- and they put the string in-line, like the two little curly braces, to denote an empty object. And it would actually then create a map using that technique... Which means as a programmer you then could pass in a nil map and it would still work. But again, it's a bit too magic, and also just kind of letting it panic, or... Since it was a library, even sometimes I don't mind catching situations that would panic, and then panicking with a better error message, like "You have to create the map before you pass in", or something like that.
+
+But yeah, I've seen a few cases where it's been used where you don't really need it, but people have tried to go the extra mile for their users. So yeah, those kinds of things are quite interesting.
+
+Another form of reflection is with the AST package in Go, and some of the actual code reflection, code analysis packages... And they kept getting better as well. When they started, they were very difficult to use, and there are a few higher-level packages now that makes it a lot easier. We have a project where we actually described our API in Go interfaces, and we use that AST -- there's a packages package which lets you actually open a package and then you can walk through interfaces, and things like this, and then inspect the fields inside the interface, and things... So it does that kind of reflection; it represents that data in its own structure, and then uses that to generate code from a template.
+
+So it's nice, because all of our APIs are described as Go interfaces, which since we're Go developers, that's very easy for us to understand and reason about... And also, it's a real Go package, so it's type-safe, too. You can't use invalid types, so it's a great way to describe an API. You know it's gonna work. And then we can generate the client from that, we generate the server code, the HTTP stuff that stubs all that stuff... Anything boilerplate can get generated, and we even generate another interface, which is actually slightly different from the original one, because it takes a context and it returns an error, and we omitted that from the definitions.
+
+So we then have to -- we can write our definition interface, we run the code gen, and then we implement the interface, and that's it. We've then got a new service that's then exposed in our project.
+
+**Jaana Dogan:** When are you open sourcing this?
+
+**Jon Calhoun:** I think he already did.
+
+**Mat Ryer:** It's open sourced, yeah.
+
+**Jaana Dogan:** Really?!
+
+**Mat Ryer:** Yeah, it's called Oto.
+
+**Jaana Dogan:** Nice.
+
+**Mat Ryer:** It's basically a JSON API at the moment, but actually, since it's just code gen and they're just templates, you could easily write a binary protocol for it, or any other type, actually. Yeah, so it's nice. Somebody's written a server Rust template for it, too... That's quite weird, but kind of awesome also. We'll put it in the show notes. It's github.com/pastedotdev/oto, but I'll put it in the show notes for anyone that's interested. We're using it in production and it just works great. I mean, our use case is somewhat simple, but it's really nice... And it's reflection, really, because we have to programmatically inspect those interfaces, and then do some work with them.
+
+**Jon Calhoun:** So Mat, I'm assuming that you're generating code with Oto?
+
+**Mat Ryer:** Yes, that's basically what it is. It takes Go interfaces, it mixes them with a template, and it generates new code.
+
+**Jon Calhoun:** I was gonna say, I do think -- we talk about how reflection is bad, or you should try to avoid it because it's confusing, it's hard to reason about, and it's just hard to maintain... But I think sometimes that's hard, because we don't tell people alternative approaches... And I do think code generation is one of the big ones out there that can be very useful. Like you said, you're kind of doing that reflection thing where you're actually analyzing the code, and then you generate code from it, and you end up with something that's much easier to manage.
+
+\[40:10\] I've even seen some ORMs that go with that approach; I think SQLBoiler was one of them, where they would scan your SQL database and then they would generate Go structs from that... So rather than using reflection, they're like "Okay, we're just gonna spit out things that match your database perfectly, and you can just use these..." It's a very different approach, but I think having reflection being kind of limited forces people to look at these other approaches and decide "Is this better? Is this easier to maintain?"
+
+**Mat Ryer:** Yeah. There's also the go generate command; you can put a comment in, a special comment... Which again, is a little bit magic, but it's like //go:generate, and then a command, and then if you type that in your project, it will run those commands. They're useful for exactly that kind of thing, where you're then gonna do a sort of pre-build step where you generate the code... And that is a nice approach, because you get the type safety, you get the compiler helping you; maybe not initially, but once it's generated, that code then is part of your project usually, and it's gonna be built... And if it's wrong, you'll find out soon enough.
+
+**Jaana Dogan:** Yeah, that's exactly what I was about to tell... I think the difference between the AST package, or the Reflect package - the AST package is an aesthetic thing; it's not doing things in the runtime. So you generate, you still have the similar level of maintainability, as well as type safety. You just generate some stuff with the compiler. So if you can hand off some of the problems to code generations, that's definitely something to do.
+
+**Mat Ryer:** Yeah, that's a great point.
+
+**Jon Calhoun:** We talked about generics earlier, and I would actually love to see an implementation of generics that essentially just runs a go generate at the start. You write your code as if generics are there, using a proposal, and then it essentially just compiles it using some step beforehand into Go code, and generates whatever it needs to generate, and then goes from there... Because I think it would be possible. It would just take some tricky work.
+
+**Mat Ryer:** I wrote exactly that project with a friend of mine, and it's called Jenny. And it's used, people use it. It uses a special type, which is just an interface type, in a different package, and then -- I think it's the AST stuff again; it goes and finds those instances and looks for where you've -- actually, you list it in a command. You run a command and you list the types you want to support, and then it is just a kind of copy and paste, and it replaces that type wherever it's mentioned. It's not perfect, because you can't do type assertions on it; it stops making sense, because -- well, it just gets quite weird... But for simple cases it works. I think that's what you're talking about.
+
+**Jon Calhoun:** I've used something like that... I guess what I'm thinking is it would almost be nice to take that idea and expand it farther, to be like the current generics proposal; to make it so you can write code exactly like that. Because one of the issues with generics is the fact that it ends up making some of the different steps with compilation and everything else more complicated. So rather than baking that into the compiler, if you could just have a precompile step where it seems kind of like it's already built into the language, but it really isn't, it's like translating it at that point...
+
+**Mat Ryer:** Right.
+
+**Jon Calhoun:** Now, it might end up being so annoying to do that that it doesn't make sense...
+
+**Jaana Dogan:** To me, generics has always been like "Hey, there's this template, and you generate things, and the compiler is handling all of that stuff, because the generated code is too complicated to take a look", and that's why the language needs to provide some syntactic sugar, to be able to engage with those types, and so on. So I wonder -- I think if you expose what is generated, that would be super-scary to the user. You will have all these --types, and all these different cases, and whatever... So I think it's not going to look really good for a large number of cases. That might discourage people to take a look at generics to begin with.
+
+\[44:19\] That's why I'm waiting for the actual generics proposal and implementation, because I wanna see that syntactic sugar, how it's going to look... Even though the hard work is not visible to me, at least -- I just actually am not interested in what is generated under the hood, because I know that it's going to be complicated for a lot of times.
+
+And I think one of the reasons these generators \[unintelligible 00:44:39.19\] didn't really catch up is because you need to have some sort of an officially-blessed generics solution. As a library, I can't really randomly pick one tool versus the other. There's not a lot of experimentation, actually. You can't really expose what's underneath; I just want something that works for everyone, so we can agree on it, and all the library systems switch to it... I don't really care what's generated under the hood, and they can always optimize it, or whatever... There is so much work that has been done in this field, so we're not trying to do this for the first time.
+
+I assume that we should find a solution to generics. It should be in the official language. I don't think we need that much of experimentation... But it will be hard for people, because it's going to definitely complicate the language.
+
+**Mat Ryer:** Yeah, but you know, a lot of JavaScript libraries have this approach, where they have a shim, essentially... I think TypeScript originally was -- or Google's Dart was originally just transpiling to JavaScript, and it did look ugly. But Jaana, you have to just not look at it, mate. Just call the final "Don't look at this .go", or something.
+
+**Jaana Dogan:** \[laughs\]
+
+**Jon Calhoun:** I think you just have to train people that -- it's almost like you build a compiler on top of a compiler, and that's the one that gives you errors that you interact with... And then whatever it eventually compiles into, you kind of have to hide that away somewhere in a bundle folder, or something...
+
+**Mat Ryer:** Yeah, but you think about the IDEs, and everything... Everything breaks when suddenly you've got "Syntax is not valid." None of the tools would work.
+
+**Jon Calhoun:** It'd be much harder now. But with the changes they're doing to how all the different IDEs use the language server - hopefully that sort of experimentation...
+
+**Jaana Dogan:** GoPlease, yeah.
+
+**Jon Calhoun:** Yeah. Basically, the fact that they're all using kind of like a common... I forget what it's called, but basically the language server -- there's a common spec for all the different languages to implement... So hopefully that type of work will lead to more potential experimentation on top of existing languages... So that could be interesting to see.
+
+**Mat Ryer:** Do you wanna just give a quick overview for anyone that doesn't know what a language server is? I think it's LSP, isn't it? Language Server Protocol.
+
+**Jon Calhoun:** That sounds right. So the general idea is that rather than every IDE or editor out there implementing their own implementation of how Go's autocomplete should work, and how JavaScript's autocomplete should work -- I think VS Code was the first one that standardized it, but I think others are using it now.
+
+**Jaana Dogan:** Yeah, I think it came from Microsoft. I'm not sure...
+
+**Mat Ryer:** Yeah, it did. It came from Microsoft's Visual Studio Code.
+
+**Jon Calhoun:** So the idea is that they come up with sort of like (in Go terms) an interface, or "This is what an LSP should spit out for a language." Basically, it should have some methods there that you implement, and it can give autocomplete suggestions, depending on where the user is... And the idea was that you could then implement that for any language, and then any IDE or editor could use it to implement autocomplete inside the editor.
+
+**Mat Ryer:** Yeah, which is amazing... And honestly, I can't actually believe that works, because of how different all the languages are. How on Earth have we found a protocol where you can just describe any of it? I find that to be quite amazing, actually. And no doubt, it's non-trivial as a protocol.
+
+**Jon Calhoun:** It's probably one of those -- there's probably like 1% of edge cases where it's not that great with, but for most developers, that just doesn't matter enough to outweigh the benefits of having the LSP.
+
+**Mat Ryer:** \[48:19\] Yeah.
+
+**Jon Calhoun:** But then there's still the problem -- not a problem necessarily, but different editors use their own approach to this. I think GoLand is one of the ones that doesn't use language servers; they use something completely in-house. And in some ways, that's beneficial... Because when GoPlease first came out, it was pretty brittle. But theirs I think was better at the time with Go modules, but now I don't know if that's necessarily true.
+
+**Mat Ryer:** I've heard very good things.
+
+**Jaana Dogan:** That's usually what JetBrains do. They build everything in-house, that's their niche.
+
+**Mat Ryer:** Yeah. My only thing is a few times that I've had to touch Java, I've had that kind of Eclipse IDE, and it really -- it's an aesthetic thing, actually. I use Visual Studio Code because it looks so much nicer, and you spend so much time in there... I feel like it does matter. I feel like you want it to be a beautiful experience. But I've heard some amazing things about how that GoLand editor, the features it has, the things it can do. I haven't yet played with it, but... Yeah, I would be interested. If anyone would like to tweet at me and tell me about their experience, I'll probably read it.
+
+**Jon Calhoun:** Probably.
+
+**Break:** \[49:40\]
+
+**Mat Ryer:** Guess what time it is...?! \[laughs\]
+
+**Jon Calhoun:** I think we know it's probably unpopular opinion time.
+
+**Mat Ryer:** It's unpopular opinion time!
+
+**Jingle**: \[51:50\] to \[52:08\]
+
+**Mat Ryer:** So does anyone have an unpopular opinion then? Some of these things we've already said probably are a little bit unpopular, but... Have you got anything in particular?
+
+**Jaana Dogan:** I have one...
+
+**Jon Calhoun:** Go ahead, Jaana. I'm gonna let her take the stage.
+
+**Jaana Dogan:** \[52:13\] We need generics. I think this is not a very unpopular opinion, but... I've been saying this since the beginning of the language, and everybody has been hating me... But I think we need generics.
+
+**Jon Calhoun:** See, I agree entirely, but that's because I've done enough -- I think one example is for Go to do well in an educational space, like for people to pick it up in college, they're gonna be dealing with data structures, and it's really hard to do data structures without generics... And I think that's one of the reasons why Java is taught so much in school, is because it does well with that.
+
+**Mat Ryer:** Yeah. What do you think about the recent generics proposals?
+
+**Jon Calhoun:** The most recent one I've seen I liked -- I haven't gone too deep into it, but it seems fine to me... I'm not too particular. My needs are fairly limited.
+
+**Mat Ryer:** I think they're making great progress on the design of it... And of course, you hear a lot that the real concern is implementing it, what that does to the type system in Go, and how easy that's gonna be to maintain, and things like this... Which is great to hear that there are people on the Go team and other contributors that really prioritize that, because it is kind of vital, really. I would hate to get to the point where we can't add any more features to Go, because it's just too complicated now... So I'm with you on that one, Jaana, actually...
+
+**Jaana Dogan:** Actually, I'm super burned-out from this topic, and I stopped following the proposals a year ago.
+
+**Mat Ryer:** Wow. Were you just getting really emotional about it?
+
+**Jaana Dogan:** It's not being emotional, but there are at least 50 concerns I had about every proposal...
+
+**Mat Ryer:** Oh, only 50...?
+
+**Jaana Dogan:** And there are no easy answers... Yeah, I mean, there's high-level ones...
+
+**Mat Ryer:** \[laughs\] You're making other people emotional.
+
+**Jaana Dogan:** Exactly. And I think I was not really contributing to the discussion... And also the points that were in my mind and I was feeling anxious about is you can't truly anticipate what the reality will be like, because it really depends on the people who are going to take the generics and use it... We'll see it over time, how it's actually going to impact the entire library space. So I was feeling like "Hey, I'm not really contributing to this discussion." I'm excited that it's happening. The people - I am almost certain that they care a lot. They care more than me, so what's the point of trying to contribute to that...?
+
+**Jon Calhoun:** One of the things this reminds me of is the type alias, and how much pushback that got, and how it was gonna ruin the language... And ever since it's come out, I don't feel like I've seen it any -- like, occasionally you'll see it, and I've done some weird things, just kind of messing around, to see what was possible... But I don't feel like I've run into libraries that have abused it or anything, which is kind of -- it's funny, because you just saw all that pushback, and then... I get their concerns and I'm not trying to say people shouldn't express their concerns, but it's just funny how that didn't actually come to fruition at all.
+
+**Mat Ryer:** Yeah, but it is in there.
+
+**Jaana Dogan:** I was just about to give that as an example, because that was the moment that I actually started to feel burnout from the project, and I wanted to go and do something else... And everything was becoming this infinite loop of discussions about the possibilities, and so on... So with generics - I didn't wanna contribute to that, because there's already too many voices, and it's hard; it's a hard job to figure out, because you can't predict the future.
+
+\[55:52\] I also trust Go developers, because a lot of people care a lot about simplicity, so they never take a feature and abuse it. I think Go users are pretty informed about the core subset of the language they wanna use... And the language is not huge, but I also trust the larger ecosystem. So I'm not that concerned anymore.
+
+**Mat Ryer:** Yeah, I mean, you can always not use it... And to be honest, I didn't quite realize -- I realized very late that type aliases got into the language. I remember the proposal and the big debate, but... Another one was the Try proposal, with the errors; for me, I was allergic to that, because it felt too magic... And yeah, I felt like that didn't fit in with the philosophy of Go. We have to watch out that we don't just like Go as it is, and we're too rigid not to allow any evolution, but... I do think you're right, Jaana - we kind of are aware of the simplicity and things not being too magic. We are, as a community, quite aware of that. And yeah, I suppose you can always just not use it, if you don't like it...
+
+**Jon Calhoun:** I think this is probably why reflect is such a weird topic, because people come from other languages where it's completely normal to use it. I think I said it before we started recording, but I don't think Ruby would be popular if reflect and meta-programming wasn't such a big thing in that language. Rails and all the crazy things you can do with it are all a by-product of being able to do meta-programming, and in many ways, it makes that language more productive... But none of that would make any sense in Go whatsoever.
+
+So when people come from a language like that to Go and they're like "Well, why is this Reflect library so hard to use? Why is everybody telling me not to use it?", it's a hard mental shift, and I think it's because it's a different approach to solving problems, and a different set of priorities.
+
+**Mat Ryer:** Yeah, I think that's right. I've sometimes just -- instead of maintaining... Like there's a bit of code that's got reflection in it, and instead of maintaining it, I'll rewrite it, because in a way, that process of doing it is how I figure out what's going on. That's the thing - for me it is actually hard to maintain, where I sort of give up, and instead we'll just prefer to rewrite it... Which I tend to do that if I can anyway, frankly, because I always find rewriting is a way to get a better version of what you have... You learn so much doing it that the second time you write it it is always a lot better. But yeah, I don't know, it's an interesting one.
+
+Well, I think that's all the time we have today... Next week we've got a very interesting show. We're gonna invite on somebody who just got their first job doing Go, and we're also gonna invite on somebody who's learning Go and programming in Go, who's still at high school. So looking at the very beginnings of people getting into this crazy world that we call programming.
+
+Jon, thank you very much. Jaana, always a pleasure. We'll see you next time!
+
+**Outro:** \[59:13\]
+
+**Mat Ryer:** That's it.
+
+**Jon Calhoun:** I was gonna play guitar...
+
+**Mat Ryer:** I've got to do air-guitar, yeah...
+
+**Jon Calhoun:** You've got like guitars in the background, but you're like "Nope..."
+
+**Mat Ryer:** Yeah, I'm not gonna play them though on \[unintelligible 01:01:02.08\] Jaana, you agreed once that I do look like a magician. Do you remember?
+
+**Jaana Dogan:** Yeah, you look like a magician.
+
+**Mat Ryer:** Yeah. Do you know how hard it is? You take it for granted, but I had to come out to my parents. I'm like "Mom, dad, sit down. Pick a card."
+
+**Jaana Dogan:** \[laughs\] But you know, it's this facial hair, this style... It's very magiciany.
+
+**Mat Ryer:** It does look like a magician's. It's ridiculous. I should change it, but...
+
+**Jaana Dogan:** \[laughs\]
+
+**Jon Calhoun:** I can just imagine it... You come out to your parents as a magician, and they're like "We've gotta get him a computer, or something... Get him into programming."
+
+**Mat Ryer:** Yeah, get him to be a Go programmer, because there's no magic there. I don't know why a family would be against magic, but... Maybe there's some hidden back-story.
+
+**Jon Calhoun:** Maybe your parents were like "He's gotta move out. That doesn't pay very well."
+
+**Mat Ryer:** \[laughs\]
+
+**Jon Calhoun:** I actually have no idea how much magic pays, but I assume it would be hard to get into...
+
+**Mat Ryer:** Yeah, I think it's difficult... Yeah, I can't think of a joke... It's a shame, because there must be loads of jokes just waiting there to be plucked out of thin air. Well, I was close enough... I like the awkward silence after a joke, that's my favorite bit.

Stop the presses_transcript.txt

@@ -0,0 +1,421 @@
+**Mat Ryer:** Hello, and welcome to Go Time! I'm Mat Ryer. Today we're talking about publishing and programming, and we have a special guest, Peter Cooper, of Cooper Press, who published the famous Go Weekly Newsletter. Hello, Peter. Welcome.
+
+**Peter Cooper:** Hi! Fantastic. I'm looking forward to this.
+
+**Mat Ryer:** It's great to have you. We're also joined by a couple of regulars, the kind that you'd see at your local tavern, for example... Johnny Boursiquot and Jon Calhoun. Hello!
+
+**Johnny Boursiquot:** Hello!
+
+**Jon Calhoun:** Hey, Mat.
+
+**Mat Ryer:** How's it going?
+
+**Jon Calhoun:** Pretty good. I feel like that was directed at me.
+
+**Mat Ryer:** No, no, no... I'm not trying to make any suggestions whatsoever. Peter, thank you very much for joining us. This is actually the first time we've met, but I've been following the Go newsletter since it started, which was - when?
+
+**Peter Cooper:** Well, I didn't start it. A guy called Matthew Cottingham started it back in -- I think it was around the early days, like 2012, somewhere around then... And he did about 40 issues of it before I took over. So I didn't take over until the start of 2015. It was kind of a little bit longer than that.
+
+**Mat Ryer:** Yeah. But it's become such a staple for the Go community, hasn't it? I know it's a place where I go to find out what's going on. It's definitely one of the big resources. How has that happened? How do you think that came about?
+
+**Peter Cooper:** Well, I kind of built on what Matthew had already done. He did the hard work at the start. I've got a history in doing these newsletters for other different topics as well, particularly Ruby Weekly, which is the community that I'm originally from. I'm still very much a Rubyist even now, although I do play around with every language, to a certain extent... But I had to do that work, and I knew the people in the community, and I could build things up, and he was in a similar position with the Go community.
+
+So I managed to put together a deal with him; he didn't wanna do it anymore, and make that sort of weekly commitment, because it is a pretty big thing to publish something every single week... And because I was already familiar with doing it, I got in touch with him, I said "I've noticed you've kind of missed a few issues here and there" and he's like "Yeah, I need to focus on other things in my life...", and I took it over and we kind of came to an arrangement.
+
+\[04:02\] That's kind of where it all just comes from. I kind of built on that initial core of people that he had originally brought together... And I can't remember how many subscribers it had when we bought it, but I think it was like within four digits. It was like 4k, 5k. It wasn't huge, but it was enough that there was something to build upon.
+
+**Johnny Boursiquot:** I'm wondering what got it on your radar. I'm curious, did the newsletter -- you're so into newsletters... Every time I think of you, I think of newsletters. Did the fact that there was another programming language newsletter out there that sort of piqued your interest, or was Go the first thing that drew you in?
+
+**Peter Cooper:** A mixture of both. I always keep an eye on what newsletters are out there, but nowadays I can't keep track, because nearly every single person's got one for their own blog, and everything... But at the time it was a lot less common. You could count like a hundred different programming-related newsletters, and that would have been pretty much all the ones regularly actually doing anything, as opposed to thousands now...
+
+And yeah, I was interested in Go because I've always been interested when there's something new coming along in the programming space. That was also true when I first got into Ruby back in 2004 - I saw something about Ruby on Rails on Slashdot, and investigated, and turned my nose up at it for a bit, and eventually got into it. There's a whole story behind that, but this is true for most technologies that come along.
+
+When Go first came along and I first saw those posts that the Go guys were doing, it struck me... Because if you go back even further from Ruby and all the rest -- I was originally a C programmer essentially; I sort of learned that when I was a teenager in the '90s, and spent a lot of time with C and Pascal, and doing x86, and doing some demo coding, and that type of thing... And Go just really appealed to me because I could see the pragmatism of languages like -- perhaps Python is a better example than Ruby, in this case... Although the way I code Ruby is in a very Pythonish kind of way, I must admit... But I saw that practicality, mixed with the low-level aspects that Go gives you, and it just was so readable.
+
+I know that's one of the goals of the project, to make a readable language, but it just kind of struck me, so I was like "I'm interested in this. I wanna get involved in some way." So I did actually start including things in some of the other newsletters that I was doing, saying "Oh yeah, here's Go, and this is a kind of cool thing. Don't switch to it, because I'm doing a Ruby newsletter, but it's still kind of cool anyway."
+
+Of course, many people - like Mark Bates, for example, who is from the Ruby world - did switch to Go, and now he runs the Buffalo project...
+
+**Mat Ryer:** Yeah, and as I hear, he's not welcome back in the Ruby community at all. \[laughter\]
+
+**Peter Cooper:** Well, I don't know about that, but...
+
+**Mat Ryer:** No, I'm sure that's not really true.
+
+**Peter Cooper:** \[laughs\] Yeah, so I'm a big fan of Mark's. But yeah, a lot of people from the Ruby world did actually come over to Go... Perhaps not as many from other languages. I think a lot of Pythonistas went into Go in particular, but a few Rubyists did... And I was interested, but I wasn't interested enough to become a full-time Go developer, or anything like that.
+
+I did have a publisher approach from the very early on, because they knew about my work writing Ruby books, and they were like "Oh, we want someone to write a beginners Go book." I think it might have been Manning; I can't remember. I think it was Manning, or Addison-Wesley... I went into talks with them and I just decided against it, because I didn't actually want to learn a whole new language and get that deep into it. I was happy with Ruby. But it put it on my radar.
+
+I was learning enough Go that I kind of knew how it operates and what its ethos was, and I knew that outside I was doing in publishing, that it was something that I'd wanna cover at some point or another. Without necessarily being a Go expert, I knew enough about it that I could objectively look at the language and know how it ticked.
+
+**Mat Ryer:** It's funny, I also used to do a lot of Ruby, and I knew of Mark even back then, from the Ruby community. He did a couple of books on it as well, and some posts, and whatever... But it's interesting, there's a lot of people that are interested - or people that were interested in Ruby and also have a shared interest in Go... But it's funny, if you think of the two languages, they're very different, aren't they? They're almost opposites, in some way...
+
+**Peter Cooper:** \[08:12\] In some ways... I think it's because a language like Ruby attracts people that are -- I mean, just discount all the people that are beginners, that have just come to Ruby as their first language. Ruby otherwise attracts people that are kind of looking for something a little bit different.
+
+That was definitely true in 2004, when David Heinemeier Hansson came to it and began working on Rails... Or, as it was, Basecamp, which then he kind of extracted into Rails. But he kind of came to it almost like a progressive developer. I don't know if this is the right world... It might be a little bit politically charged to use, but... I kind of see people like DHH and maybe some of us as being reasonably progressive.
+
+That means that if we see a new language or a new thing, we'll look at it and we'll take it on its own merits and say "Yeah, this kind of fits in for this task, and I would use it if I wanted to do that type of work." That's true of Go for me, for example. I would use Go if I need that low-level access to concurrency, but I don't need something that is quite as low-level as C; if I want some kind of guards around what I'm doing, I would pick Go. I wouldn't pick Rust, because I just don't get along with that language at all. I tried it and I was just like "Nah..." Go feels a lot more C-like to me; C with the good parts, and the bad parts removed.
+
+I think there's a lot of progressive developers who come to a language on its own merits, as I say. So the fact that it's not the same as Ruby, that's fine. If I wanna go back and I wanna build something like a Rails app, I can come back to Rails. I don't have to just think "Oh, I'm just doing Go, and now I have to do it as a Buffalo app." Some people do, like perhaps Mark for example, but I'm happy switching between languages. Maybe that's just an aspect of the work I do... But yeah, I can just take it on its own merits.
+
+**Mat Ryer:** Yeah, I think that's quite healthy for everyone to think like that. It's very easy to get into your little world, in an echo chamber, and then you start seeing tech wars, and things... Yeah, of course, sometimes it's good to choose one thing, sometimes another.
+
+**Peter Cooper:** I come from the world where we used to do in-line code a lot. So this was definitely true in the Perl world, to a certain extent, which was the language I focused on before Ruby. I was very heavily into Perl. And it was quite common to sort of drop into doing some in-line -- you know, putting other languages in, just because it was kind of easy to do... But going back even further, when I was a Pascal developer in the '90s, and I was doing demo coding and stuff, it was really common in that space to drop back and do in-line x86 for setting up your mode X screens blitting pixels all over the place, and all this type of stuff... And to draw lines on the screen, doing sort of the interpolation of the pixel positions and stuff like that, you would drop down into x86 and it just made it a lot more performant on the machines of the day, which were very slow...
+
+So maybe it's something that I would associate even with older developers; I don't know if that's fair or unfair, but... The idea of just seeing new languages and jumping backwards and forwards, and taking on bits and pieces from other languages just seemed very natural to me.
+
+**Jon Calhoun:** So when you're jumping back and forth, especially with like a newsletter, I feel like it would be challenging to pick articles and things like that that you wanna share with people when these languages have very different ideas of what is correct or what is idiomatic for that language. In Rails or Ruby in general it's okay to have this magic and a lot of other stuff, whereas in Go it's the complete opposite. If you publish an article where it's like "Look at all this magical stuff I did", people are gonna react a little bit differently. Does that make it harder to find good articles and things like that to share with people when you're looking at the publishing side?
+
+**Peter Cooper:** Now that you've explained it like that, I get your point... \[laughter\] But I think this is one of the things actually where people talk to me and they approach me at conferences and stuff and they're like "How did you even do this? I can't imagine doing one newsletter every single week, let alone 10-11, or whatever it is." And I'm like "Well, it's just easy. It's just what I do."
+
+\[12:08\] I've just kind of realized it's actually something that I just am innately good at, just through practice and persistence and having done it a lot of times. It's a bit like when you see people that are really good at Rubik's cubes; you know, the cubes with the colored squares on each side. There are people that can look at that and then just go \[unintelligible 00:12:24.11\] with their fingers, and bam! They've solved the puzzle. And I look at it and I don't even understand where to begin, and how you even get to that point.
+
+Maybe to a certain extent - and I'm quite a modest guy - I'm kind of doing that when I'm putting together these newsletters. I'm kind of looking at a lot of stuff, but I can keep those separate narratives in my head and I can just do it like a Rubik's cube, and bam, I produce what I need to produce and it's just correct. Obviously, I'm not saying I'm perfect. I do make mistakes. But as time goes by, I do get better and better, and I just seem to keep it in my head. I'm just saying natural skill might not be the most attractive answer, but it's kind of where I'm at.
+
+**Johnny Boursiquot:** So what would you say if I told you that as a publisher with many readers, many subscribers, that you hold power over the ideas that are disseminated within any given of the communities in which your newsletters circulate? Do you think about it that way, or is this another aspect where we're kind of getting you to sort of realize it?
+
+**Peter Cooper:** I kind of get the point; I probably don't realize necessarily to what extent. Because my interactions with the different communities tends to be reasonably detached. I only attend a certain number of events, just because I travel extremely poorly... So I don't often get to walk the streets, as it were, and hear from readers in the flesh very often. I don't even do calls like this very often, because I tend to find it as a thing that is quite difficult to do... So I operate in a very detached kind of way... But I see other people do very similar things, like industry analysts, for example. That's a really interesting can of worms to dig into if you ever get bored one day.
+
+If you look at what some of these industry analysts do, at places like Gartner for example they write entire papers about how different companies are using JavaScript, or what web frameworks developers are using in the real world type stuff... And then they offer these for sale for hundreds of dollars, or thousand dollar subscriptions to big companies. I'm sure some of you have had run-ins with people from those types of places, where they're grilling you with questions... At least the people I talk to that work at startups are like "Oh yeah, my boss told me I've gotta speak to this analyst, and he just gave me all these pointless questions, and produced this crappy report."
+
+I hear that a lot, but the reason I kind of am okay with doing it is because I do actually code, and I've worked with customers before, and I've built projects with software, and I've sold them, and all that type of stuff... So I feel like I've got enough skin in the game; I've released libraries, I build libraries and I do open source... I've got enough skin in the game that I kind of hopefully understand what that power is and how it relates to newer developers, even though I don't necessarily have the same day-to-day developer experience as everyone else.
+
+**Mat Ryer:** It's really funny you mentioned that... And you mentioned that you are somewhat detached from the communities. It really doesn't feel like that when you look at the newsletter, so I wonder how the quality remains. It's always been very good. Sometimes, some of my articles have slipped through... \[laughter\] But on the whole, they are generally -- I mean, how do you keep that quality? It must be a difficult thing to do in the first place, and also sustain.
+
+**Peter Cooper:** \[15:54\] I think part of it is that if the quality ever slips, I tend to get people telling me it has. So if I've included an article that is just really bad - and this happened a few times; I don't think with Go. I've made a couple of factual errors with Go that I've been pulled up on... Such as when I think I said the Go Modules proposal had been accepted by the core Go team a week before they actually accepted it; I had some people pull me up about that. It just said something like "Oh yeah, we're close to agreeing it", and I kind of went with "Oh, it's all been agreed."
+
+So occasionally I have errors like that, and occasionally I will link to something that isn't quite right... But I tend to get people reach out and say "Why did you link to that? It sucks, blah-blah-blah..." and I tend to learn from that experience. But I don't know... You could say it's luck, you could say it's skill... It's just like trying to ask a footballer "How is it that you score those goals?" It's hard for them to break down and think about the meta-processes that they use.
+
+I do spend time thinking about that type of thing, and I've realized actually when I'm trying to find other people to help me curate the newsletters, it's actually very hard to describe the process to someone else and get the right results back out. And that is actually why I do lead most of the publications (bar one or two that other people edit almost entirely). It's because I've really struggled to find anyone that I can instill those skills into. And maybe that's why I'm one of the very few people that are doing this sort of thing, at this sort of scale, week in, week out. It's just the thing that I'm good at.
+
+**Mat Ryer:** Yeah, you're almost like a search engine that was built by natural selection somehow, and it's highly tuned to that particular thing... \[laughter\] But whatever it is, it does work, and it isn't an easy thing, for sure.
+
+**Peter Cooper:** No... I mean, it kind of relates to a lot of things in my life. I might cover this a little bit more in-depth later, but one of the things that I've discovered in the last few years is I've actually diagnosed as being autistic. And actually, in that process, I've managed to reflect back on a lot of the things I've done in the past, or how I've acted, and how I see the world, and it really fits. I wasn't even going to have a diagnosis for that, it's just something that came out of a separate process I was going through... But I've realized that has actually given me some skills to really pick things up very quickly.
+
+One of the things that I do - and I'm sure many of us are guilty of this - is you'll hear about a plane crash in the news, and you go onto Wikipedia and you read about the plane; it was a Boeing 737 Max, for example. You read about the plane, and you read about the problems it's had in the past, and then you read about a previous disaster that it's had, and then you see what the correlation is between that disaster and the new one. Then you dig into things about aeronautics, and then like two hours have passed and you've tried to learn about everything to do with aviation... That type of thing is just like every single day of my life.
+
+When I see something like an article about a goroutine, or something like that, I'm immediately thinking like "Alright, goroutines. What's a goroutine? How does it relate to green threads in Ruby? How does it relate to a normal thread in the operating system? What's the correlation? How does the runtime fit in with everything?" I'm always trying to build up and deconstruct all of these ideas, on a daily basis. Every time I run into a concept, I tend to reconstruct it again, just because I find smashing things down and rebuilding them helps me get a mental grasp on stuff.
+
+I talk to people about this sort of thing, and some people are just like "Oh, I totally get what you mean. I do the same thing", but a lot of people are just like "What are you even on about? I can't even keep up with what you're saying. You're speaking at 500 miles an hour..." So yeah, it's complicated, is probably what I've realized. But I'm just so big on absorbing concepts and learning things that even if I do make mistakes, I tend to relearn stuff again the next time, so quickly that I kind of make up for my past mistakes. That's probably the best way I can put it.
+
+**Break:** \[19:52\]
+
+**Jon Calhoun:** So when you're looking at all this different stuff - you're working on multiple newsletters and you're exploring different solutions to problems, and as you said, you deep-dive into these things, do you ever look at a problem in one language and think "These idiots... This is solved in other languages. Why aren't they just following their lead, or doing what they're doing?" Because I can imagine looking at different languages; that would be frustrating to see one language struggling with something that maybe two other languages just sort of figured out already.
+
+**Peter Cooper:** I probably see that most on the social side of languages, actually. I do see it on the technical side, but as I mentioned before, I kind of respect the fact that different languages have different wheelhouses, and have different levels of responsibilities and how they interact with the world; that's totally fine by me. I don't necessarily expect a procedural language to take on lots of functional qualities to it, because I'm happy of the fact that there is a separate language that I can write stuff in F\#, for example, and I don't have to think about "Oh, let's take all these things out of F\# and bolt them into another language."
+
+Now, I know the JavaScript world is very keen on "Let's just absorb everything from everywhere." I'm not. But when it comes to the social side, I do see a lot of the same mistakes occurring again and again. That's particularly true in how people interact with each other.
+
+If you were in the Ruby world about 6-7 years ago - and I'm not even gonna recite a lot of the things that happened, because I don't even remember half of them, but there was a lot of drama in the community, and Ruby has kind of gotten this reputation of "Oh, there's always a drama kicking off in the Ruby community. There's always something happening at a conference, or there's always someone having a fight on a GitHub repo." JavaScript developers - or Node developers, more accurately - were kind of laughing on Hacker News and saying "We've got all the cool people. We don't have these kinds of fights." And I said "You will. It's gonna happen."
+
+We had this in all the different worlds; going back where we've had internet communication and we can kind of say stuff quickly, this has always happened. There's always been drama, there's always been problems... And so, of course, it came to the Node world, perhaps in an even more explosive way than the Ruby world. We're now even seeing with Rust, for example, there's been low-level drama; it's been reasonably well handled, because that community has such a good grasp on social issues by the people at the top... But they've had issues. I saw someone today saying they were gonna quit doing documentation, and stuff. It just happens. It's just human nature, and I do see that between the different communities. But I think things actually get better the more we each adopt different languages and we each morph between being different types of developers.
+
+So if you're doing Go, but you also do some C and you also do some Ruby, because you're participating in all those different communities at once, you can kind of bring back and forth some of the values and keep things going. And then really it's just a goal of bringing newcomers into the fold and making sure that they adhere to the community norms.
+
+\[24:06\] I think that's something we're seeing a lot more happening now with codes of conduct, and things like that. People have really had to start thinking about this, and it can't just be some sort of "Old boys club" all the time. Actually, a lot of the most valuable contributions in our different communities come from a diverse array of people. It's taken especially some of the older people like me a really long time to see how that works, but it's really valuable and I'm really happy that that work is ongoing.
+
+**Mat Ryer:** That's really interesting to hear. The way you talk about the drama in the Ruby world makes it sound like I wanna see that on Netflix.
+
+**Peter Cooper:** \[laughs\] Yeah.
+
+**Mat Ryer:** The Ruby Diaries.
+
+**Peter Cooper:** Yeah, Netflix wasn't around quite at the time, but it would have been good.
+
+**Mat Ryer:** Yeah, it would have been. It sounds great. Well, this still could happen, let's see... Yeah, that's really interesting. So the advice then to people is to reach out and join other communities as well, and go and mix with the different people, because you'll be exposed to different ideas, and all that is just good.
+
+**Peter Cooper:** But be happy being a novice. That's one of my big things, I'm really happy to be the dumbest person in the room, because that is quite often the case. Even though I pick up things quickly, which is perhaps my plus point -- actually, that makes me more happy and more confident about being the dumbest person, because I know that I'm gonna look around the room and see all these smart people and be like "Oh, I can learn this from them, I can learn this from them, I can learn this from them." And I think everyone can do that, to a certain extent, whether or not you're a quick learner or not. Just be happy to go into those spaces and those new communities, and be happy to open that first page of a "Learning how to use Rust" book.
+
+I'm not gonna use Rust as a full-time development job at all, but I've spent time learning it, and I've enjoyed it. It did teach me some things. So yeah, just do more of that.
+
+**Mat Ryer:** Yeah, I think that's great advice. I agree with that thing about the joy in being a novice. There's nostalgia in it, isn't there? Because when we were first getting into computers, it was hard. We didn't know what we were doing. So it feels nostalgic to me when I encounter new languages or new ideas that I'm not familiar with, and I get to then be that kind of naive person and have to go and figure it out... Because that's such a fun process.
+
+**Peter Cooper:** You just have to watch that you don't get overwhelmed. That's the real issue nowadays. Because back in the '80s or the '90s if I was picking up a language you'd have a book. That would be it. I would have a book, I would have no one else who knew anything about the topic, and that would be it. I would be treating that like a Bible. I would believe everything that book says. That doesn't happen now.
+
+If you learn about JavaScript, someone will say "Oh, you should use Webpack", and someone will say "No, you should use Parcel for packaging stuff." Someone will say "No, you should use Snowpack", or whatever is this new thing that's come out... If you can't cope with that, you're gonna have problems, and I don't know what the solution is to that, because I've not really thought about it until I just said it. But there is that level of overwhelm when you come into new topics, and that's something that I think with the newsletters I've actually become quite good at seeing through.
+
+If there's a lot of noisy people saying "You must use a certain tool", "You must use goroutines for absolutely everything", I can look through that and see "Well, I know just historically that what everyone is suggesting at one period of time is not gonna be cool in five years' time." So maybe I can get ahead of the curve by just not doing it.
+
+**Mat Ryer:** \[laughs\]
+
+**Peter Cooper:** I'm kind of exaggerating a little bit there, but we've seen it happen with REST, for example. I remember when everyone was like "Oh, REST, REST, REST." All the time. REST everything. And now people are like "GraphQL. Don't use REST. REST sucks. Use GraphQL for this and that." GraphQL is gonna be very uncool in ten years' time. It will be something else. Probably REST again, or REST+, or something... \[laughter\]
+
+I guess just the older you get, the more you see these loops constantly happening... And yeah, maybe it gets easier to be a novice as you get older. I don't know, just throwing that out there.
+
+**Johnny Boursiquot:** \[28:03\] A couple of points... The sheer volume of content coming out... And for me, this is really the value of the newsletters; they provide me with a filter. Basically, rather than paying attention to 300 things for that week, I can wait for the newsletter, whether it be one of yours, or maybe The Changelog on Sundays, or something... And basically just sit down and see what's been going on this week, and then even further do some selection myself and then say "Okay, these are the things that I consider to be important. These are the things that Peter or the Changelog team considers to be important", and then I'll take a look.
+
+There's just too much. It's not a judgment call on the quality of everything going on, it's more like as a human being with a limited amount of time - I have 24 hours in a day, and a good chunk of it is already spoken for... There's only so much I can consume.
+
+So going back to all the different dogma around "Oh, you have to be using this thing", or "If you're not using this thing, you're doing it wrong", whatever it is. To me, that's part of the noise. So I think over time - and maybe this is something that happens to folks who have been doing this for a while... I think over time my desire to limit the number of things I want to master - be it a programming language, or technique, or a particular way of doing something - over time, that desire has become stronger... Because I'm trying to narrow. Maybe this is like career suicide to some people or something, but I'm trying to narrow the things I have to pay attention to, because there's just so much of it out there... And I think that newsletters play a role in that. For me that's the value of these content filters.
+
+**Peter Cooper:** Yeah, I think one way I look at some of that stuff is that I'm trying to be sort of 80th percentile with everything that I can find. I wanna be not the best, but just proficient in everything. I understand the big picture and I can connect a lot of dots together... Whereas I think perhaps you and people who I would consider to be senior developers or career developers essentially want to become 95th percentile or higher at a very small handful of things, especially over time. If you're going into Postgres consultancy, for example, you wanna be in that top 5% of people who know how to use Postgres, because otherwise who's gonna employ you as a consultant specifically about that thing?
+
+I think a lot of developers move down that path and they become very specialized at a handful of things. What that means is that let's say -- and this is where I'm waving my arms in the air, just for everyone to kind of benefit who's listening on the podcast... If you assume that I've got a straight line of my 80th percentile of everything, and then all the different developers have basically got these little peaks that kind of -- you know, they're better than me at a whole bunch of different things...
+
+In aggregate, my readership is better than me at everything, because there's an expert in everything amongst the readership. But the thing is, for each individual person -- if they've got this kind of like mountain amongst the... I'm trying to think about how best to put this... If they've got these peaks, then anything that's outside of their absolute special points, I'm just high enough for them to feel like "He kind of has enough of a feel about this, so I can trust what he's saying, and then build upon that. I can stand on his shoulders on these other topics."
+
+So you will know in Go you will be 95th percentile or higher, on a small group of things, I assume. Like, you are really like "I'm the S on this thing." I won't be the expert on that, but I will have seen enough of the entire community that I know what tools people are using in all different areas. I know how people are interacting with things like Kafka, or Postgres, or what the biggest library for talking to MongoDB is - although actually that's a very small number now, but... I will have an idea of all these things. So that's why when you come to the newsletter, you can be like "Okay, I'm an expert on this thing, and I'm gonna skip all these articles that I know I don't need to read, but on all these other things I'm gonna trust Peter to be good enough at all these other things..."
+
+\[32:15\] I think that's the relationship that perhaps people like you have with the newsletters. You've perhaps not put it into words, but that's how it's kind of seen. I don't have to be an expert at all this stuff, I just have to be good enough that I'm better than you at a bunch of stuff.
+
+**Mat Ryer:** Yeah, that is really interesting. I wonder if you could talk a little bit about some of the technical challenges around doing this. Presumably you've got some tech that you've built yourself, or use, that makes it happen... How does it actually technically work?
+
+**Peter Cooper:** There's a lot of moving parts, and I guess this is one of the good things about being a developer myself - I've managed to just build these parts as I need them. I'm definitely one of those developers who spends two hours writing a script to save myself ten minutes. I'm very much in that group, just because I know what happens over time.
+
+If you have a job that you need to do more than once, you generally need to do it 100 times, so building tools makes sense. So I ended up building tools for doing our booking system for our sponsors, and so on. I've built our billing system, because at the time I built it there was nothing that could handle all the different weird combinations of VAT and stuff like that that we have to do in the U.K. as a business. Those systems now exist, I could move on to them, but at the time they did not.
+
+I built systems for keeping track of projects, for keeping track of different websites, for keeping track of RSS feeds - all of these different things. There's tons and tons of tiny apps that we use here at the company. I mean, I don't even touch a lot of them now. The booking and the billing system and stuff like that is pretty much for use by the people in the company that handle that stuff. Other than that, my curators lean heavily on some of the discovery things I've built, which kind of scrape different websites, and looks at different projects, and things like that.
+
+Perhaps one of the most recent ones I've built actually was a tool that uses the GitHub GraphQL API... It's like a 50-line GraphQL query, or something. I had to learn so much ridiculous nonsense to make it work... But what it does is it looks for projects that match a search query that I might give. I might say "I want all stuff that's to do with Go" or "I might want all stuff that's to do with MongoDB", and then it looks for stuff that has a release or a tag within the past X number of days, and it brings out the information about the star count, who's watching it, what language it's in, or what secondary languages it's in, all that type of stuff, and then works out the version number, and all that type of thing.
+
+So I can find that literally within seconds now what are all the biggest releases of Ruby libraries on GitHub within the last two weeks. I have a lot of work to do to go through those and be like "Well, actually this is just a pointless minor release" or "This is a big deal" and so on and so forth... But at least it gives me the pointers to those things. So I have lots of tools of that nature.
+
+Probably another one actually is the system that sends the email, which is quite an undertaking. The thing runs all the different websites, for all the different newsletters, so it's actually one app. When you go to GolangWeekly.com, or JavaScriptWeekly.com, it's actually hitting the same app... So managing all of that, and all the security certificates, and stuff... That's yet another job.
+
+The editor that we use to put issues together, which is now used by -- well, three of us in the company use it. It's basically a live text editor that we can all use at the same time if we wish, so we can actually work on an issue live. We have a kind of weird version of XML that we use to align all our newsletters on the left, and then on the right it's the live preview, which we can resize to see what it looks like on mobile, and everything... But it's completely live. So as I'm typing, it updates visually what it looks like immediately. That's why we can do in the newsletter some interesting visual things, like we can include breakout boxes and stuff like that, that for other people -- I've seen a lot of people do newsletters, but they're very formulaic; they always look the same, every single week, whereas I can change the layout of the newsletter almost like on a whim. We don't tend to, just because of time, but it's possible to do.
+
+\[36:17\] So yeah, I've pretty much built all the tools for this thing, and maintained them, and that's pretty much what keeps me in the development world for sort of 50% of what I do - just mucking around with these tools... But there's a lot of parts.
+
+**Johnny Boursiquot:** So how much of that is in Go?
+
+**Peter Cooper:** Very little. I have some stuff that does some crawling in Go, because it's just very well suited for that task, and keeping up with the feeds, and stuff... But principally it's in Ruby, because as I said, I am first and foremost a Rubyist. I lie, there's some JavaScript actually as well, but it's principally in Ruby. It's a good fit. These systems are not designed for hundreds of people to use, so they don't need to scale. So when you don't need to scale, Ruby is a great language. It scales with me. It's one of those languages that fits to the developer very well. So yeah, now I'm gonna go podcast, selling the Ruby language to everyone... Yeah, come back to 2004; it's a great language. \[laughter\]
+
+**Mat Ryer:** It is though, actually. I always think it is. I actually do encourage people to do that as well though, to go and check out those other languages. Especially, one of my favorite things in Ruby was method missing, where it's basically a catch-all method that would be called if the method that you'd actually called didn't exist. And so it'd give it a sort of last chance to parse the string of the method name and do something useful with it... Which is just kind of -- when you think of Go, and the way Go thinks about things, that now is so alien. It's kind of gone all the way around through ridiculous, back to being really awesome again.
+
+**Peter Cooper:** Yeah. I mean, I think of Go as being like a cross between C and Python, with a good concurrency story. That's what the story is in my head. That's how I think about it. So it's used for certain things... If I wanted to write a port scanner or something to enumerate subdomains or something like that at scale, I would use Go. It's the natural fit there. But if it was for building the sort of things that I build, which are very non-scaling, kind of just slapping text together, a language like Ruby - or even Perl - is a good fit.
+
+**Mat Ryer:** Once your assets are generated then, are they kind of dynamically generated from a database, the newsletters on the website? ...or are they static files that are accessed?
+
+**Peter Cooper:** When you're looking at the -- let's say if you go to golangweekly.com/issues/209 (or whatever number we're at), that is actually being dynamically generated by a Sinatra app. That's a Ruby app, Sinatra being the inspiration for many a framework actually, in other languages, like Express.js. But I think the actual issue content is cached. So it's not like all of those different items in the newsletter aren't stored in like an items table or something, and they're all being pulled together on the fly.
+
+When we produce a newsletter, it produces a kind of a text and HTML kind of artifact that then remains forever more... So that is then what goes into the website.
+
+**Mat Ryer:** I see. So it generates those kind of static assets then, and then that's what gets shipped around. Is that right?
+
+**Peter Cooper:** Yes. This was a decision I've made quite early on, because I realized that doing it in a formulaic way would end up with very formulaic results... So I wouldn't be able to very easily put interviews in, or just stick a comic in. Front-end Focus, which is one of our newsletters, had a comic just randomly appear in it the other day. Just chucking random things in is very hard when you make it formulaic...
+
+So I was actually thinking more like a publisher than a developer at that point, because the way I look at it is like a clean newspaper page, and I wanna put something here, and put something here, and maybe run a big headline across here... I kind of have that freedom to do that with this. Whereas some of the tools - and I know people that have built some of these curated newsletter tools that are a bit more structured and automated, where you can put items in and it formats, and everything... But you can't just randomly throw different types of content in, or put multiple-column layouts in, or stuff like that, and I can just do that... So I wanted that.
+
+**Mat Ryer:** \[40:20\] And the trick there is that you've got that immediate feedback system where you could see the live preview, isn't it?
+
+**Peter Cooper:** Yeah.
+
+**Mat Ryer:** That is something I've encountered too, and I think it's a good lesson for everyone, actually. When you give users that immediate feedback from whatever it is they're doing, it means that they're kind of free to be more creative, because they can just make mistakes, and see it live. If you had to make those changes and then submit it, and wait for it to be rendered, and then look, you wouldn't do as much... So having that tight loop I think is important. That's a lesson I think everyone can learn with the things that they're building at home, or at work.
+
+**Peter Cooper:** Yeah, I think that actually might be one of the slight downsides to Go as a language, actually - just bringing it back to Go - is it's a difficult language to have super-fast feedback loops with. Now, \[unintelligible 00:41:08.26\] you might disagree with this, it's very fast to compile and run and everything... So in terms of comparing it to a C or a C++ - okay, yes, it's really quick. But it's not quite the same level as like having a JavaScript sandbox, for example, where you can literally change a number in a sandbox and see a different result immediately. There is that extra kind of step. But as I said, it's its own type of language, it's not designed to be used in perhaps a REPL-y kind of way. I know there are some REPL attempts out there, but it's not really designed for that type of use.
+
+**Johnny Boursiquot:** Easy - just don't make mistakes. \[laughter\]
+
+**Peter Cooper:** Exactly.
+
+**Jon Calhoun:** That's one option... But I do think it's useful to look at things like Go Playground and to see that when it was created they did think about this... Because if you have to install Go and do all this stuff before you can even see your first mistake, it's like, okay, that feedback loop is incredibly slow for your first program. Whereas if you can just go to a website, type in some code and hit Run, that actually adds a lot of value. And while it's not the same as a REPL, or as like a Rails app, where you're seeing things visually changing, or like a JavaScript thing where you're in the browser, I do think that it's better than many other compiled, static languages that you might see, in that sense.
+
+**Peter Cooper:** Yeah. For a systems language it is very good.
+
+**Mat Ryer:** Yeah. I mean, I use unit tests for that purpose. That's probably the primary -- when I'm doing something where TDD is a good fit for it... Say I'm gonna write a little package that's gonna parse some strings, and I know what the input strings are (the valid ones), so I can really unit-test that and TDD that quite easily. That's then the feedback loop for me. And it is quick then, when you're doing it like that.
+
+But for visual things or anything - I mean, we don't tend to do too much that's visual... But I've seen some of the graphics things that people are working on, and they do want that feedback loop. We notice it's important to them to have that, because I guess in that world it does make more sense.
+
+**Peter Cooper:** Yeah.
+
+**Mat Ryer:** So yeah, it's an interesting observation there.
+
+**Jon Calhoun:** I think this is also why languages like PHP were so good at attracting new developers. It was like, you know enough to write tests and to use that TDD approach to get instant feedback, but when you're first starting with a language, that's kind of asking a lot, to be like "Okay, think about what inputs you're gonna get, and what outputs you'd expect", and they're like "Um, dude, I'm just trying to write a function that puts a name in a string, that's it."
+
+**Mat Ryer:** I'm just trying to write \[unintelligible 00:43:23.21\]
+
+**Jon Calhoun:** But you know, with PHP and those languages you could literally jump into the server and edit files live, and just see them rendered on the website.
+
+**Johnny Boursiquot:** Oh, good Lord...
+
+**Jon Calhoun:** It's maybe not the best idea for production-type stuff -- I say "maybe"... Definitely not the best idea for that. \[laughter\]
+
+**Mat Ryer:** Hedging...
+
+**Jon Calhoun:** But for whenever you're learning, it's a great way to just jump into something and instantly see that feedback... And I think that's why some of those languages are so appealing to beginners. I think JavaScript has that allure, where you can just open up the console on any website and mess with things; that's very, very appealing, and I think at times you'll probably see that Go doesn't attract as many new programmers as say JavaScript or some of the other languages because of that.
+
+**Peter Cooper:** \[44:11\] Yeah. I mean, editing work in code I think is actually a very underrated learning technique. A way that I have always tended to learn very quickly is if I find a program that works really well. When I was trying to learn modern C -- as I said, I did C in the '90s for quite some time, but I didn't touch it for quite a long time after that... And then I kind of came back to it and I was like "Oh my God, I can understand what's going on, but people are laying it out in a different way, they're putting different structures into their programs..."
+
+So I looked at codebases like Redis - it was a really good example of that, the data structure, the database thing... And the way he wrote that is just very elegant, and I could go in and I could work out it has an "add a command" to it. I think I even did a YouTube video about this - I added a command and showed how that all worked, and everything... And it kind of really -- editing is something that already works and is already... Not perfect, but good. It helps me learn, at least.
+
+That's a way for me to learn, is from examples, rather than going from first principles of saying "Right, we're gonna do #include stdio.h printf("blablabla")" from scratch again, because sometimes you might lack the imagination to think "Well, what is it that I can build with printing stuff to the screen? I want something that works, and then I'll tweak it, change a few numbers here and there."
+
+**Break:** \[45:33\]
+
+**Jon Calhoun:** Peter, when you were talking about using a book, and that being your Bible essentially when you were learning a language - you bringing this up made me think of that, because I remember as I learned, I'd be like "Well, do I really understand this?" and that working program that was in the book was the only thing you had to go from... So a lot of the times you'd go through and be like "Well, let me try tweaking things. If I wanted to do this, is this the right line to change, so that it actually makes those changes?" And I think for a lot of people that's just kind of how they learned. One, you write the code and you read through and try to understand what it's saying, but to verify you understand, a lot of the times it's "Let me change this to see if it actually does what I want."
+
+**Peter Cooper:** Absolutely, yeah. I've seen a few people teach in that way. It's not very common in book form, because the problem with books is that you can't really put a massive program in and say "Here you go, play with this." It definitely suits the modern way of having the web, and downloading a GitHub project, and off you go.
+
+I think Zed Shaw did a little bit of this... He's quite well known in the Python and Ruby spaces. He's learned Python the hard and learned Ruby the hard way. He was very driven about doing everything by example. He's like "Just type this stuff in. You don't have to know exactly what it means, but just do it. And by the fact of doing it, you will eventually learn what it means... Because it's just how it works. It's like learning a foreign language. They don't teach you word-by-word what everything means. You have sentences, and then you kind of derive meanings from it. So I think that's a really good way to go in learning.
+
+**Mat Ryer:** Yeah, and when most people join teams, that's what they're doing as well, isn't it?
+
+**Peter Cooper:** Absolutely, yeah.
+
+**Mat Ryer:** They start completely from scratch, yeah. So it does make sense. I used to copy the code out of Amiga Format Magazine to build little games, and stuff... And then we changed that; we made a pool game once, and then we changed the cushions and just made the pockets really big, so the game was really easy... And just being able to go and figure out how these pockets were described in this weird array, just that process - that's kind of the time when I fell in love with code... I realized we could create things, and make things happen using these strange, magic spells almost.
+
+**Peter Cooper:** \[48:17\] Yeah. There's a lot to be said for watching other people do things... And I guess that is exactly what my job is. I am spending nearly my entire job watching what other people are doing and what other people are releasing, and kind of turning that into personal wisdom that then I can hopefully use to inspire what it is that I choose to go into the newsletters.
+
+**Mat Ryer:** Yeah, that's great. How many people does the newsletter reach?
+
+**Peter Cooper:** We've got a total of about 480k subscribers right now.
+
+**Mat Ryer:** Wow.
+
+**Peter Cooper:** Of course, some of those are subscribed to multiple publications... There's a lot of JavaScript people on the front-end newsletter, for example, and the Node newsletter as well. I don't actually track the unique number of people, just because it doesn't interest me. I'm more interested in looking at it from a publisher's point of view; you know, "10 people bought this magazine, 10 people bought that magazine." I don't care if they're the same 10 people, I sold 20 magazines. That's the way I look at it.
+
+**Mat Ryer:** But what if it was one person buying all 10 magazines? You'd wanna know about that, wouldn't you?
+
+**Peter Cooper:** Well, I don't know... I'd probably be okay with that. \[laughter\] That's the publisher in me speaking. It's all about getting the sales of the magazines. So I don't know exactly, but I know on Go, for example, it's 29,000. We did actually pass 30,000 briefly, but we've recently been going through a process of deleting people that haven't engaged for a certain amount of time, so that process has knocked a bit off, and we're back to 29,000 again.
+
+**Mat Ryer:** Alright. That's not bad. I thought you were gonna say you posted something someone didn't like in the last one, and you're like "Aarrrrgghh!" \[laughter\]
+
+**Peter Cooper:** Actually, that's something you need to watch out for - if anyone is listening and ends up doing a newsletter, don't look at the number of people that unsubscribe, unless you're literally losing half your list overnight. I've found it's actually a way of people that are new to it actually becoming quite depressed very quickly. They're like "Oh, I just sent a newsletter and ten people unsubscribed." But I'm like, "Well, you've got 10,000 subscribers." That's a statistical anomaly.
+
+**Johnny Boursiquot:** Perspective...
+
+**Peter Cooper:** Exactly, like "Don't even worry about that."
+
+**Mat Ryer:** It could have been the same person unsubscribing ten times.
+
+**Peter Cooper:** There you go. Exactly.
+
+**Jon Calhoun:** I think you also have to view it from your personal perspective. You might be interested in the newsletter, and then a couple weeks later, or maybe a year later, you might realize "I've moved on in my roles" or "I'm not doing the same things", and it's not that I don't like the newsletter, it's just that it's not the right fit for me right now.
+
+I think people, like you said, get upset about stuff like that, but realistically, having people unsubscribe is their way of saying "You're not providing me value right now, which is okay." It just means you shouldn't be spending money or whatever your resources you're spending to send them letters are a waste of your time and their time... So it's better than doing what you said, where you manually have to go through and pull people out who haven't engaged in months, or however long it's been. That takes way more effort than having somebody unsubscribe and do it for you, and signal for you they're not interested anymore.
+
+**Peter Cooper:** Yeah. Sometimes it's not even just that; it's also that perhaps they just don't like the way that you present stuff. They don't like the layout of your newsletter; they just have a gut feeling about you, and they're like "Oh, I don't like this guy. I'm gonna unsubscribe." But you have to be okay with that. If you do some tweets and you mention about politics and someone unfollows you and says "Oh, you should do developer tweets all the time." Like, it's actually not wrong for them to say that, but it's also not wrong for them to leave, and it's not wrong for you to be like "Well, I don't care."
+
+I think we ought to be a bit more forgiving of each other and a little bit understanding that we can't satisfy everyone. That's definitely true of us. This is one of the points I put down in my notes - you can never fully saturate an audience. I can never reach 100% of all Go developers. That's just not gonna happen.
+
+\[52:10\] I encounter people that -- say the JavaScript world; we've got 170k subscribers, and I run into people at conferences, and they're like, "Oh, I've never heard of JavaScript Weekly." And it's like, I don't sit there thinking "Well, that's a bit weird... Surely, everyone knows about us by now", because there's millions of JavaScript developers. There's just all these different pockets of different places. JavaScript developers don't even know about each other half the time, let alone all the different resources are in the community.
+
+So there are definitely Go developers who have never heard of this podcast, and it's never even crossed their radar... And I kind of wonder, do we need to actually improve that or not? It's always good to extend your reach, but is it something that's worth worrying about? I don't think it is.
+
+For that reason, that's why I also think it's okay to have multiple versions of the same thing. So you've got Go Time, but then of course - and I'm not gonna name them, but there's other Go podcasts out there; don't listen to them, everyone... \[laughter\]
+
+**Johnny Boursiquot:** That's right, that's right. Yeah.
+
+**Peter Cooper:** Exactly. There's loads of different Go podcasts -- I say "loads"... Like a handful of them. But some of the listeners to some of those other ones won't have heard of Go Time, and some of the Go Time listeners won't have heard of some of the other ones... And people relate to different things in different ways, and that's fine. So if people come along and they try and do another Go newsletter or something like that, it doesn't bother me at all, because I know they're gonna have their own audience. Just the fact that they exist isn't gonna take away from what I'm doing.
+
+The only problem is I think there are some people that haven't had that revelation, and they think that perhaps because I've got 30,000 subscribers, they think "Oh, what's the point of me doing something, because he's already doing it..." and "Oh, he sucks" and blah-blah-blah. I've had this run-in with a guy before, who does a Go newsletter and he was kind of like having a go at me on Reddit, and saying "Oh, don't subscribe to it. Subscribe to mine, because it's better." And I was just like "You don't need to be like that. There's room for everyone."
+
+If your friend says "I'm gonna go and write a novel about a political scandal" or something, you're not gonna go "Oh, I've got a friend who's also writing a novel about the same thing. You must go and get in touch with them."
+
+**Johnny Boursiquot:** "Don't write yours!"
+
+**Peter Cooper:** Exactly. "Write the same novel together. You can't compete." It's just not the case. So I guess the point I just wanted to get across is that if anyone wants to do the same sort of things that I'm doing, then by all means, go ahead... As long as you're not using my name, or copying every single thing that I do, then fine. There is enough audience out there to warrant it.
+
+**Mat Ryer:** Yeah.
+
+**Johnny Boursiquot:** Do you have non-English newsletters currently, or are you planning on having anything like that?
+
+**Peter Cooper:** No. I struggle enough with English, let alone other languages. I must admit, I'm definitely a single-language person. I know a tiny bit of French, and that's all I know. I know there's a huge Go community in China. I come across some of their output quite often actually; they have GitHub repos that are just like long lists of links that they've found in the community, and stuff... And I can't understand a single word of what they're saying, of course, but I know they've got their own things going on over there.
+
+We've had people approach us, and say "Oh, I'd love to translate the newsletter for such-and-such language", but it's literally been a handful of inquiries over the past however many years... So if someone came to us and said "Oh, we wanna just copy what you do and just translate it into Russian..." That's probably a good example, because that's a culture that's just different enough from our culture that they're probably not gonna read my newsletter en-masse in Russia. Because it has its own alphabet, it's own culture, and everything. And if someone came to me and said that, I'd probably be like "Yeah, just do it." Because it's not going to affect me in any way, and it just will help the community, which is partly what this is about. It is my business, but at the same time, if I just wanted to make money, there's a lot of things I could be doing that I don't do, just because it's not right to do... So I am very keen on helping the community.
+
+\[56:00\] So if anyone is listening and you're in a different land - I'm not gonna say Australia or something like that, of course, because people from Australia or America or England or Ireland and whatever will all subscribe to the English one... But if you're in a different culture - Poland, Saudi Arabia... Or I don't know - where is Go popular other than China...? Yeah, just get in touch, because I'd be interested in seeing you do stuff like that.
+
+**Mat Ryer:** Great. Well, first of all, I don't know if we thanked you really yet for the work that you and the team put into the newsletter... It really is a resource that is valuable for everyone. I'm sure I speak for everyone.
+
+You mentioned not satisfying everyone and not being able to satisfy everyone, and that brings us neatly to our new regular slot, Unpopular Opinions.
+
+**Jingle:** \[56:55\] to \[57:12\]
+
+**Mat Ryer:** So Peter, maybe you could kick us off and give us an unpopular opinion.
+
+**Peter Cooper:** Well, I've spent a lot of time on Hacker News - that's probably an unpopular opinion straight away... \[laughter\] I've spent a lot of time on Hacker News, and Twitter, and stuff like that, and I always see, whenever the topic of advertising comes up, people will absolutely lose it. They're like "Oh, I hate being tracked! I never click on a banner, I've never clicked on an advert in my life. On the TV I close my eyes when the adverts come on." \[laughter\]
+
+Now, I think this is an unpopular opinion amongst geeks, but it's actually not a controversial opinion, in that this is actually how the world works, and that is that advertising is generally (or can be) a force for good. And I know that's a slightly controversial statement, considering just how badly it's been used by so many companies and so many media... But I think that if you are an ethical person and you approach it ethically, then advertising is actually better than most of the alternatives.
+
+Just to give a quick example of that, I've had people come to me and say "Oh, it really sucks that you have to run sponsors in the newsletters. You should let us pay whatever per month, and do it that way. It'd be much more honorable. You're a bad person for having adverts in there." And I've researched it and I've looked at case studies, and other industries where people have done this, and the best I can make out - and I've looked at other developer newsletters that have done this as well - is perhaps I would convert 3%, 4%, maybe 5% of the audience to paying me a pittance each month.
+
+And it would pay quite well, because 3%-5% of my subscriber base is quite a lot. But. My newsletters would then only be reaching 5% of the people they do now... And I don't think that has a good enough effect on the world or the community as a whole, than if I let everyone have what I have - and that happens to be some sponsors which are clearly marked, don't do any kind of onerous tracking or third-party whatevers, all that type of nonsense. We don't sell our lists to them, all that type of stuff. So I actually think the way that -- if you can do advertising ethically, it's better than the alternative, which is paying for stuff.
+
+**Mat Ryer:** That is very interesting. I think of that sponsorship model as well when it comes to conferences. It's very common for conferences - people pay for the tickets, but the prices would be a lot higher if it didn't have any sponsorship there. So it's actually a good one, because you're right - I think if you ask people generally, they'll say "I hate ads" or...
+
+**Jon Calhoun:** \[59:51\] See, it's a challenging subject, because generally speaking, I think people want things as free as advertising makes it, but they don't want the advertising. And it's like, you can't have your cake and eat it, too. You've gotta kind of choose. And I get that there are cases where people are overly zealous and they track you, and they do all these things that you don't want them to do, and that can be troublesome... But then there's also cases, like you said, where conferences are cheaper, newsletters are made available... This podcast is possible because we have advertisers. Otherwise a lot of this stuff would be very hard to do. We wouldn't have people to edit the show and do different things like that.
+
+So I can imagine this podcast would be a very different thing if we had to charge everybody, and it would also make the people that you can reach - it would limit it so much. You could no longer help people who are struggling to get into tech from a rougher background, or maybe they have a job that doesn't pay that well... And I think being able to reach those people is worth it, in many cases. But I do agree with you, Peter, that you kind of have to approach it ethically. You can't just be like "Let me just track everything, do whatever, and just abuse all of my readers/listeners", or whatever they happen to be.
+
+**Mat Ryer:** Yeah, you don't support advertising smoking to kids, do you? \[laughter\]
+
+**Peter Cooper:** Yeah, I mean - we've had a few interesting run-ins, actually. The only one I can think of that we had a bizarre run-in with was we had a company - that I'm not gonna name, because their name is actually a little bit suspect... But they were basically a company that offers sexual services online, and they were advertising a job with us. Now, I was a bit like "Hm... I wouldn't link to them as a service, because it doesn't make any sense. But since they're trying to hire a JavaScript developer, let's go with it." And they were quite subtle in the way they worded it, and everything... And one of the people in the JavaScript world just went absolutely bonkers, because he's like "Oh, this is encouraging..." - and it was actually about a particular fetish, so... That's just to kind of give you an idea. He's like "Oh, this fetish is absolutely wrong, and even if people consent, it's actually not good etc. You shouldn't be running their ads" or whatever.
+
+**Mat Ryer:** And you should have strong types in your language, and that sort of thing.
+
+**Peter Cooper:** Exactly.
+
+**Johnny Boursiquot:** It's the not-a-number fetish, yeah. I've heard of it. \[laughter\]
+
+**Peter Cooper:** So we do tend to steer clear of that sort of thing now. We've kind of learned a few things. Even if I agree with it, the optics can be a little bit weird on things like that... But yes, there's definitely a lot of ethical issues that you have to figure out. And one of the things that I think we do - yeah, we obviously abide by all the laws, which now there's quite a lot of them, with the whole GDPR thing, especially here in the EU... Oh, sorry, I say "Here in the EU", but I'm no longer in the EU; I keep forgetting, but... \[laughter\] Yeah, don't get me started on that one.
+
+**Johnny Boursiquot:** ...too soon. \[laughs\]
+
+**Mat Ryer:** You can tell by the screams in the street. \[laughter\]
+
+**Peter Cooper:** Yeah. So we abide by stuff like that...
+
+**Johnny Boursiquot:** That's what that is... \[laughter\]
+
+**Jon Calhoun:** So you're trying to tell me that the U.K. was just like "You know what, we don't wanna do this GDPR thing. We're gonna go ahead and just leave."
+
+**Mat Ryer:** Yeah, we're gonna purge instead.
+
+**Peter Cooper:** Well, it's still the law, but yeah... So we abide by all that sort of stuff, but the other thing that I've always been really keen on doing is making sure that we aren't out-pricing the startups and we aren't out-pricing the small sponsors... And that's been really important to me, because we've had people like -- and this isn't so true in the Go world, this is more true in the JavaScript world... We've had a couple of independent companies that were very small kind of build off of the back of our audience, and become popular with our audience, and then they've grown and grown and stayed customers just by virtue of that...
+
+Frontend Masters is a good example of that; they're kind of like a screencast type training website... And there's a JavaScript tool called Wallaby.js, which is like a live coding debugging thing - they've had the same experience. In our early days we had IBM come to us - their marketing agency came to us and said "Look, we wanna buy all the inventory on all of your newsletters, but we have to spend at least $100,000/month. We can't spend less than that, because otherwise it's not worth our time." And I looked at the inventory I had and it was worth a percentage of that. And I'm like, okay, even if I marked it all up and I said "Yeah, it's gonna cost you $100,000/month", do I want IBM Cloud as every single sponsor of every single thing? For me, that's unethical. That's the sort of ethical decision I have to make.
+
+\[01:04:09.29\] I have to think "I could sell this inventory for tons and tons of money", say double what I make now, and have one company plastered everywhere, that -- who cares about IBM Cloud? IBM Cloud is a good thing, but it's not something you wanna see week in, week out. It's just one service among many, that people wanna consider. And that's what I want the newsletters to be.
+
+So when sponsor go in, I want them to be sponsors that people can look at and say "Actually, I am interested in checking that out", because it's a different thing each week, and we try not to let people run sponsorships back to back... And because it's relevant to me, because it's teaching me something, or because it's a service that's like a hosting company that - maybe I need a new host, I'll check them out. I want them to be things like that. I want them to be indie developers, I want them to be startups, and that is definitely an ethical decision I've made.
+
+Separate from all the tracking, and all that sort of stuff, it's that the sponsors are actually good. And I know that's also something that's happened here with the Changelog - you've got sponsors like Linode, for example; they're good companies, that are worth checking out, and are actually relevant to our day-to-day jobs. They're not just some enterprise play that we're only gonna recommend if we've got like a million-dollar budget to play with.
+
+**Jon Calhoun:** Whenever you say you try to keep it approachable for the smaller players, like the indie developers or the startups, do you have to do anything else? Let's say you have a mailing list that's 200,000 people. You'll probably have some set price for like a job listing. Do you have to do anything special to try to make that accessible, or is it just certain types of spots?
+
+**Peter Cooper:** We've not really got to that point. JavaScript Weekly is our biggest publication, with 170,000, so that does have a price premium upon it... But I would say actually - and this is part of the whole pricing thing I mentioned before - if you scaled up one of our smaller publications to the same size as JavaScript Weekly, the amount you'll be paying if you apply that multiple would be a lot more than we actually charge. So we undercharge on JavaScript Weekly, and I don't wanna say we overcharge on the others, but we charge more healthily on the smaller publications.
+
+We've got a database publication, for example, that only has 15,000 subscribers, but because the database space is awash with money, let's say, and it's a relatively small space (kind of) in terms of the sponsors, there's more competition for wanting to appear in the newsletter, so we can charge a premium there... Whereas with JavaScript weekly, less so.
+
+**Jon Calhoun:** One of the things I guess I'm thinking about is that you have the added benefit of you send letters out to people uniquely, and you have sort of control over what you're sending to each person... Whereas for a podcast, for instance, they insert the ads, then they put it out to distribution channels, and you can't really dynamically insert an ad for each person. And I don't mean tracking, I mean like - let's say you have 175,000 people and you wanted to sell "Okay, we'll give your job ad to 25,000 of our subscribers", and you can sort pool that ad listing with maybe five other advertisers... Does that make sense?
+
+**Peter Cooper:** \[01:07:14.08\] We have the technology to do that. It's only ever come up in one context, with a sponsor, and that was where they're saying "We're an Israeli company. We wanna advertise that job just to the Israeli people." The only problem with that is that in that situation we looked at the numbers and we were like "Okay, 1% of our audience is from Israel", or it was probably less than that. And it's like, "Do I really wanna charge --", "Even if I charge them double, do I wanna charge them 2% of our normal pricing, to go through all that effort and then not have it appear in the main one?" It just seems pointless. So we're not at the scale to do that.
+
+**Jon Calhoun:** For something that small you almost have to charge them the regular price and just say "We'll limit it to maybe like half-price" or something, because of the amount of effort and overhead. I guess I was just thinking, you know, keeping things accessible and making it so startups and smaller indie people can access it, looking into those pooling options and things like that are unique approaches, I think
+
+**Peter Cooper:** Yeah, we're not quite there with that. I mean, we have the tech to do it, but we're not quite there. If we had millions of subscribers to a single newsletter, then obviously that would be very pressing... But then part of the problem with that is that people are gonna start wanting targeting at that point; so they'll say "We wanna reach certain companies." It's very easy nowadays to actually take email addresses and turn those into names, occupations, and stuff like that.
+
+There's companies out there like Clearbit, for example - you throw emails at them and you get data back... And we began doing some of that. This is pre-GDPR. We began doing some of that, so that we could kind of build audits of who our subscribers are - what countries they're from, what companies they work for, and that type of thing... But we've since discarded that data, because it just doesn't fit ethically with how I see things going data-wise. It's nice to know that we have a bunch of people from Microsoft subscribe to us, but you can kind of tell that from the email addresses. I don't need to run all this weird, clever stuff over it.
+
+So yes, it's possible to do that slicing and dicing, but both for demand reasons and ethical reasons it's not something we really do. And I wouldn't wanna run a different ad to different blocks of the subscriber base... Because I think a bit like a magazine, like a copy of Vogue, for example - you'll open it up and there'll be a big thing in there for PRADA, let's say... That's part of the experience. The fact that they're there, it says something about PRADA; they wanna be in there. So the fact that you open up a Go newsletter and you see a certain company is the prime responder - I actually want that almost to be a talking point in its own right now. I'm not big-headed enough to think that people are sitting on Twitter and going "Oh, did you see this hot sponsor in Go this week? It's fantastic", but there is still kind of a narrative there you can have with sponsors, and say "Look, you are the sponsor." And that's actually worth it to them.
+
+**Mat Ryer:** That's really interesting to learn about. It's the sort of thing I never really consider. Well, Peter, we could talk to you all night, but unfortunately that's all the time we have for today. Thank you so much to our special guest, Peter Cooper. If you wanna get the Go Weekly newsletter, go to golangweekly.com. You can sign up there. And we'll see you next time on GoTime.
+
+**Break:** \[01:10:30.13\]
+
+**Jon Calhoun:** I feel like I'm a tech support person, where like "Have you tried restarting your modem?" \[laughter\] It's like, "No, the phone line has a tree in it. That's not the issue."
+
+**Peter Cooper:** Everyone's a tech support person when it comes to video conferencing.
+
+**Mat Ryer:** Yeah.
+
+**Johnny Boursiquot:** And families.
+
+**Peter Cooper:** \[laughs\]
+
+**Mat Ryer:** It's like other people's showers, as well... \[laugh\]
+
+**Peter Cooper:** Oh, very true.
+
+**Mat Ryer:** I always notice that those are so complicated.
+
+**Jon Calhoun:** Every Christmas I get mad at my brother-in-law, because he comes home, he gets his mom some fancy new technology, then he goes back to California.
+
+**Johnny Boursiquot:** \[laughs\] Leaving you to deal with the wreckage.
+
+**Jon Calhoun:** He'll get it set up, sort of, and then she calls me "How do I do this?" and I'm like "I don't have that thing. I don't have a clue." \[laughter\]
+
+**Mat Ryer:** "Where do I get games for my Switch?"
+
+**Johnny Boursiquot:** \[laughs\]

Telemetry and the art of measuring what matters_transcript.txt

@@ -0,0 +1,305 @@
+**Johnny Boursiquot:** Hello, and welcome to this episode of Go Time. I am Johnny Boursiquot, and joining me today are Jon Calhoun, Jaana Dogan, and special guest, Dave Blakey. Welcome to the show, Dave.
+
+**Dave Blakey:** Thank you. Thanks for having me.
+
+**Johnny Boursiquot:** Jaana, it's good to have you back. You seem to have been traveling the world, and trying this whole--
+
+**Jaana Dogan:** That's not true...!
+
+**Johnny Boursiquot:** No?
+
+**Jaana Dogan:** I only took a week of a break. I'm serious, yeah...
+
+**Johnny Boursiquot:** Okay, okay... The pictures - I'm like, "You're right. Mm-hm."
+
+**Jaana Dogan:** But next week I think I will be in London, so if you see me joining with Mat from the same room or something, don't get surprised...
+
+**Johnny Boursiquot:** Oh, that would be kind of cool. But you see, you are traveling the world. You've been in more places in the last month than I have this year. Actually, it's still a new year, so... \[laughter\] Jon, how are you, man?
+
+**Jon Calhoun:** I am doing well.
+
+**Johnny Boursiquot:** You've been busy lately. You've been releasing courses, and trainings, and everything.
+
+**Jon Calhoun:** Yeah, I've been busy... I mean, the truth is I've been busy for a while; it's just everything gets finished at the same time, so then it looks like I've been especially busy now.
+
+**Johnny Boursiquot:** It's good to ship, so... Yeah. Good stuff. Dave, we know little about you, but we're about to fix that. We know a little bit about what you do and who you work, or should I say whom you work for... But in today's episode, we're actually gonna cover a topic that's special, and near and dear to my heart, as a cloud engineer or operator... As I sometimes call it, telemetry, and you're gonna have to forgive my accent here. The French is trying to come out... Telemetry... Yeah, you can correct me \[unintelligible 00:03:13.11\]
+
+**Dave Blakey:** I think that was perfect.
+
+**Johnny Boursiquot:** Okay, good. Thank you. Telemetry is something that we rely on quite a bit if we're doing cloud operations work, but it's not just for that. The use case for telemetry is much broader, and you actually are working on something that actually involves quite a lot of that... But before we dive into that, I'd like us to level-set a little bit; let's talk about what telemetry is, what it's used for, and who is best positioned to leverage it.
+
+**Dave Blakey:** Absolutely. Telemetry is an extremely broad term, as you said. Obviously, here we can narrow it into at least computing and modern computing, I suppose, but at its core, it really means collecting and storing, and I like to think using, but not necessarily... But collecting and storing data from remote sensors, or remote machines, remote computers. So from the left side being how much electricity or gas is your water heater using, to the right side being what's the response time of an API back-end that you have. It's around getting that information, and all of that kind of stuff.
+
+\[04:26\] I think more recently it's around a lot of the techniques and the areas in which to add telemetry to applications while you're building them. As you all know, when you start to scale things out, it's often too late, if you haven't done it already.
+
+**Johnny Boursiquot:** So it's one of those things where it may not feel important in the start of a project, but when you need it, you're wishing you had put it in from the start, kind of thing... Yeah?
+
+**Dave Blakey:** Exactly. And sometimes you don't know you need it, until you're way too far gone. When you start having bottlenecks, you start having problems... I mean, that's the traditional problem that telemetry tries to solve, but now that is with security consents, and with scaling, scaling out, scaling in... Things are not static anymore; telemetry has started to play a much bigger role than just saying "Why is my webpage slow?" It's much more than that now.
+
+**Jaana Dogan:** Is it because there's not enough established practices around this? I think a lot of companies I've seen - especially when they're first bootstrapping - they don't necessarily care about anything around production excellence, or SRE practices... And telemetry plays a big part in this, but they're maybe thinking that it can be such an afterthought, and then eventually feel very overwhelmed by the amount of work they need to do.
+
+**Dave Blakey:** Exactly. And the worse that becomes, it's like it's this kind of snowball effect... Because you just start randomly adding telemetry, and it's not really -- you're ultimately trying to solve a problem, and I think the best way to look at telemetry is to try and store all of the important components, and even some of the ones you might not think are important yet, for an application. Not to solve a crisis now, but to shine a light on this process as a whole.
+
+A lot of the time, if you add -- you say "Well, why is my website slow?" A simple solution would be to say "I need more web servers." But it might be the database server that's slow. That's a common example, but you get the idea. It could be something that's not immediately obvious to you, and the more data you have, the easier it is to track that kind of thing down... And that's just around scalability.
+
+**Jaana Dogan:** What is the best approach in terms of planning? Should we start thinking about this at the design process, or what is the best time to start thinking about telemetry?
+
+**Dave Blakey:** Absolutely. I think when someone is launching a large-scale project, it's probably something that they're all considering already... But I think maybe it's more appropriate now to say "What about a small one or a medium one that's got some growth?" Obviously, if you're doing an internal wiki for your own five-man office it's probably not a big problem... But if you're building a project, I think it's important to start from the first line of code, basically. And it can be very simple - you can just have a class in your system that you can send random gauge information to or metrics to, or whatever... And once that exists, it's very easy to just parse that information there.
+
+That's what I would advise - make it super-easy to send just a slug, and the value, and that it's a gauge or a metric or a point in time measurement to a response time. And even if you don't actually send those anywhere on day one, at least you're starting to put it in the code.
+
+At our business we have what we call our code contract, and it's this set of nine rules for everything that we write... And one of the rules is that everything has to use this telemetry helper that we put in, because we knew at some stage it would become a problem. And it uses very little development work, you know?
+
+**Jaana Dogan:** \[07:58\] Yeah, I've seen a lot of cases where people are debating what to collect, and how to collect, and so on... I think there's also some sort of confusion around what matters for the success of the project, and so on. So you have to be more holistically maybe thinking about all the specs - availability, debuggability - in order to at least have a better understanding of what you wanna collect and how you are going to be utilizing it.
+
+A lot of times, small companies end up failing because they start too late, and so on... But it's very important to start thinking about this at the very early stages.
+
+**Dave Blakey:** Exactly. I'd rather be in a situation where you are collecting some information that was useless, or you were collecting something in not the most efficient format or something, than you either were collecting nothing or had this telemetry paralysis, where you feel like "Well, we've gotta put so much time and effort into this." Ultimately, I think just do what suits the project and the business, and just make sure you're doing something, and it'll evolve.
+
+**Johnny Boursiquot:** I'm interested in pulling that thread a little bit... Jaana, you kind of touched on it when you mentioned basically trying to keep track of what's important. When I think of the things that matter to me as somebody who's looking after infrastructure, versus something that's important to perhaps a back-end developer or a front-end developer, and ultimately the end user, who has to use whatever it is we're putting in front of them - there are different things that are important to us at those varying levels.
+
+I'm assuming telemetry is useful in all these areas, but ultimately, the business cares about the end user experience... So how do you approach -- when you gather a team and you're about to start doing the work, at what point do you start carving out the things that are important for the different teams and the different stakeholders?
+
+**Dave Blakey:** I think it's iterative. Again, I would say that a very large project would function quite differently. That would be part of the design decision, and it would be built-in from the foundation, and it would be quite a complicated approach to telemetry, because it would need that... But in medium to smaller projects - and by medium, I mean it still could be a large project. I consider our product to be medium in size, and it's six million lines of code. In that type of project, I think you can iterate. So we start by saying "Okay, let's make sure from the code point of view developers are storing the metrics that they think are important, and we can always add more. Let's make sure from a metric performance point of view and from a systems and scalability point of view we're storing the things that we think are important." And then when we have problems, it becomes immediately obvious where telemetry is missing or where telemetry is useful. Because that's the funny thing, you don't always know.
+
+Let's say you're looking at a cloud engineer point of view, and you say "Okay, my telemetry is showing me that my CPU usage is 80% on my ten cloud instances at Amazon, and I probably need another 20 instances." But you might not notice that there's 200% or 2,000% more failed logins per second than there normally are... And actually, what you've got is a brute force attack. Now, if you've got all of these metrics - this is jumping forward a bit, but I think the best thing with telemetry is to store as much as you can, and have somebody look for anomalies.
+
+If you have that type of setup, where you're saying something is a statistical anomaly, then when you go to say "Okay, what's going on here?", if those things pop up - far more requests per second from a certain country, way more failed logins than usual - and then all of a sudden you realize that the problem is not what you thought it was... Or you don't find it, and you have to start digging and digging, and then you implement a way of tracking that in the future.
+
+**Jon Calhoun:** So if you're starting off with your telemetry, and say you don't have a clue what to start with, like you're somebody who just hasn't gone about doing it, what are the first few metrics you would suggest they try out?
+
+**Dave Blakey:** \[11:48\] I would say it's probably broken up into three areas. The first area you've got is your actual server. Whether it's a cloud instance, it's a VM, it's a container, whatever it is, the actual system that's hosting it - most people don't realize how far down the journey of telemetry they are... Because they can tell the CPU usage on there, they can tell the memory, or they can tell if it's online or offline. That's a data point, right? Like, is the server working or is the server not working? So you start to monitor things like that, and you start to have some basic understanding of your server, obviously, and servers.
+
+The second thing is your network. That's where most scaling and telemetry information and data becomes very useful. The time of up and down sites is long gone, but what if your website's or your API's response on average is 200 milliseconds, and there was a deployment last night and now it's 400 milliseconds? This is very important information to have. Simple things like HTTP reply times, and your HTTP reply statuses, for example. How many 200 codes are there, 400 codes, 500 codes... Just picking up that there's 5% of responded pages are errors, versus 0.1%, can really help you to shortcut an issue.
+
+And then the final one is the real key, the fundamental area of telemetry, which is in your app. That would be starting to track the stuff that's important to you. If you've got a key-value store that you use for caching, track what's your cache hit rate. If the cache hit rate hits the floor, then you know something might go wrong. What's your database's response time like? How much cache are you storing? How many logged in users are there? All the components that make your site work, you just start tracking and tracking and tracking. And it's so easy... You literally have a function called stats.Gauge(10) users.Total, stats.Gauge(12) users.Total. You just start to track that stuff, and you find out that actually it's not hard to implement. The much harder part is taking responsibility for that data and using it. But that can come. The first part is storing it, having it available and understanding how it impacts your service or application.
+
+**Johnny Boursiquot:** Whose responsibility is it to care for that data, as you suggest? Is it the operations team, is it the engineering team, is it the product manager, is it everybody?
+
+**Dave Blakey:** In maybe traditional structures, in a traditional way you've got IT operations, you've got security and you've got development, kind of separate houses. The more siloed it is, the more likely it is that they will have pieces that they care more about than others do. But that's kind of like a crux of telemetry, because like I was saying, if you don't see the whole picture, you might not see -- if you were just in IT ops, you'd launch ten more servers, instead of realizing that you've got a security problem.
+
+We work primarily with what you would call more modern types of deployments, I guess; a lot of Kubernetes type stuff, cloud-native people, things like that... And interestingly, the use case there is quite different. It's a word I hate using, because everybody has made it just mean everything, but it's like a DevOps type of role. What that means to me is someone that cares about the application as a whole. So they don't care about the code, or the server it runs on, or the cloud they use, or the firewall, or the load balancer, they care about the whole application... And that team will normally be in charge of the telemetry and monitoring of it, and everything... At least in our experience.
+
+**Jaana Dogan:** I think one of the other questions is - you know, you mentioned a bit about anomalies, or some teams, some organizations prefer to set some SLOs, and they produce some alerts as soon as some of the metrics are out of the boundaries... And I think each organization has a different strategy. Some organizations prefer a monitoring team or an SRE team to be reactive to the alerts, and then they escalate it or delegate it to other teams - to the first-responders, versus other folks, and so on. It has a lot to do about the organization and the way the company/organization works, right?
+
+**Dave Blakey:** \[16:02\] You make a good point, because I was talking almost from the angle of saying "There's something wrong. Let's look at the telemetry", but the next kind of natural step from that is exactly like you're saying... It's to rather have the data be presented to people when things are picked up, like anomalies and that. And yeah, the bigger the business, the more likely there is a team that is responsible for that... But that doesn't mean that smaller businesses can't use open source free tools to achieve very similar types of results.
+
+**Jaana Dogan:** We talked a lot about metrics, but you specifically mentioned that our systems are getting larger, and there are a lot of different components... Recently, in the last decade or five years, distributed tracing and logging especially, correlated with a trace IDE or a request IDE has also become very popular in terms of collecting signals. Some organizations at least use them as another source of telemetry... What do you think about that?
+
+**Dave Blakey:** Yeah, I think it's critical the larger the organization is. The reason why I am kind of choosing my words carefully is because it can be quite difficult to achieve in an early project, or to add to an existing project. You'll often find that level of scrutiny is quite challenging for a smaller business (or a medium size business even) to achieve. We're jumping forward a bit, but if you take a look -- we ourselves could have 50 devices at a client, and each device could be generating 100,000 lines of logging a second... And for a company to actually store that information is often beyond their ability. That's the nice thing about -- if you have all the hooks in to get this information, then when you need it, you can grow into it.
+
+**Johnny Boursiquot:** I'm interested in understanding the telemetry landscape a little bit right now... You mentioned obviously at your company, at Snapt, that's the business you're in, so you likely have an understanding of the landscape right now... We hear about these projects, but we don't really quite know where they fit in. I'm thinking of things like OpenTelemetry, OpenTracing, OpenCensus... There's a lot of these open source projects that all seem to have overlap in terms of the problem they're trying to solve... But to me, it seems like some teams decide "Okay, we're gonna adopt OpenCensus", whatever that means; then they go find the clients, they find the severs, and they do their thing... And now you wonder "Okay, when there's a standard, if there is a standard - do we retro-fit everything?" It seems like right now there's a lot of churn in that space... Can you lay out the landscape for us here?
+
+**Dave Blakey:** Yeah, there is a lot. It suffers from that same DevOps state where people have wound up building their own in a lot of situations. I don't mean building the entire stack, but a lot of tooling and custom work to get things to work the way they want. By far, what we see the most are people using things like Prometheus and Grafana and stuff like that to dashboard and visualize stuff... Because most of the companies we work with, it will be mostly internal, their collection of the information and their ability to send it somewhere... Because it will be from different apps, different stacks. It could be some data coming from Microsoft servers, some coming from containers, some from Amazon... But they'll often have a single source of dashboarding and reporting and analysis for that, so that will usually be something like Prometheus, or something like that, where then they can automate a lot of the anomaly detection, and visualization of that data, and stuff like that.
+
+So it's a pretty developed space in terms of how you see that information once you start to store it and keep it in a time series database, and all these kinds of things... But it's really up in the air with how you track, how you communicate. Probably the biggest thing we see are people that are just using StatsD to stream telemetry data to something, and then collect it and ultimately output it into some sort of dashboarding solution.
+
+**Jaana Dogan:** \[20:05\] Yeah, as a person who has some experience in this field - I used to work on OpenCensus, and I think we were trying too hard to maybe unify the approaches; unify the export types, the exported data, or unifying the library space, or trying to establish standards... But it seems like the field is very crowded, and it's just hard to -- maybe it doesn't make much sense, because at the end of the day, all you care is getting the data to a dashboard and be able to utilize the data... And I think that's primarily what the organizations care about; they don't necessarily care about the export format or the library they're using to instrument...
+
+**Dave Blakey:** In a lot of cases they don't even care about the reliability of it, and that's one of the challenges with that space as well. If your telemetry data is something that you're collecting every second or multiple times a second, losing some of it doesn't matter, in most cases. If we for example are writing the response times of an API the whole time, we stream that information through UDP, and we don't even check if the destination got it... Because we'll pick up that node data has been plotted for five minutes, but if one packet drops, a lot of the time with telemetry that's not a big problem. That's often internally developed, how people get that data out... And much more, the standards seem to be on the display of it and the detection of it. But like you guys mentioned, there are a lot of projects starting out there, so maybe it will clear up on that.
+
+Often, when people build their own things it's because there is a need, but you also have to deal with the fact that there are so many people that build so many things now that it's -- it is a bit of a web...
+
+**Jaana Dogan:** True. Also, there's a lot of pre-packaged software and cloud platforms that can export a lot of telemetry, and there's no standard around where they would export, or what data format it would be... It would be nice to have some sort of standard at least, so we can go and talk to all these open source projects or the cloud providers to export some telemetry out of the box... Because everything is a black box when you have a prepackaged something, or like a vendor solution, right?
+
+**Dave Blakey:** Exactly. We have that problem with our product... We do our own dashboarding for our servers and systems and things, but when we ultimately wanna let people integrate that into their DevOps tooling or their environment, it's like how do we get that information out? So you provide a REST API, then you provide a webhook URL... Because you're trying to find some way to fit into what they do, and there's no standard... That's 100% correct.
+
+**Johnny Boursiquot:** If you've been in this space for any length of time, you're gonna hear the term "observability" quite a bit, right? And we know that telemetry plays a part of that, but oftentimes it feels like it occupies a very large slice of the pie. I've heard people talk about the pillars of observability, and metrics and tracing and logging, and all that... What are the concerns that one has in terms of observability? When I say I want observability, what am I really asking for here, and how does telemetry help answer these questions for me?
+
+**Dave Blakey:** I think the term observability -- like you say, there's pillars of it, there's all these things... But to me, it has seen a rise in popularity lately because of exactly what we were just saying, this black box effect that things have. So really what it is -- let me give you an example in our world... You've got one web server that runs your API, and then you have to scale that out and you've now got two. Then imagine you scale that out a lot, and instead you've got 30, and they're in multiple data centers... And it's all going through some load balancer, and someone says to you "Oh, every time I use my Android phone, if I'm in South America, when I try to log in I get a 500 error." That's to me observability. It's like a needle in the haystack. The problem just becomes so compounded when everything is being funneled through one point and then split off into all different directions.
+
+\[24:23\] The rise of observability I think actually comes out of trying to problem-solve, trying to debug issues, and not being able to see them... Where telemetry came into play and you said "Okay, you know you have an issue. Let me look at the general health and well-being of my system at large in order to be able to see where I should focus down."
+
+If you were looking for that problem, perhaps you will notice that Azure data center has 5% errors on requests, whereas your Amazon one has 0.2%. So you know "Hm, it seems like something is going on in my Azure data center", and I can start to drill down there. And that's where then the rest of observability comes in, like how accurate can your logging be; can you actually look for all 500 errors that went to all the web servers in this data center, and then find the web server that it went to and dig into that...?
+
+But at its core, observability to me is just really being able to see through that veil, to actually see what's really happening, what's the traffic look like, what are the valid requests, what are the invalid requests, where are things breaking, and not have it obscured by a cloud service, or a firewall, or a load balancer, whatever it might be. It's almost like a simplification of the complex system that things run on now.
+
+It's even worse when you get into things like Kubernetes, and the pod you're trying to see (that the error was on) has been destroyed and it's just gone now, and where's that data... It really starts to get hard. But that's really what I think it is - it's just about being able to see in a simple fashion, and as simple a fashion as you can what's going on... Because either you want to prevent something going wrong, or you're trying to discover what is wrong.
+
+**Jaana Dogan:** Yeah, one of the definitions that I heard and I liked is "Observability is more about asking questions that you are not prepared to ask."With typical sorts of metrics and so on we basically know what we are looking at. We plan so we collect metrics around it, or eventually we learn over time that "Oh, these are some of the failure modes, so we should maybe better collect more metrics around that."
+
+Observability is a broader approach to be able to utilize whatever you collect in order to be able to answer some of the questions that you're not prepared to answer.
+
+**Dave Blakey:** Exactly. My simple example in the beginning that you don't need more servers, you need to stop the brute force login attack - it's that kind of full visibility of the system... Because what you think is wrong may not be what's wrong. And if you can see all of the moving pieces and components, then you can hopefully see what's actually happening on your system, and ideally prevent an issue, but also debug an issue.
+
+**Break**: \[27:12\]
+
+**Johnny Boursiquot:** Let's take it down one level a bit... So if I'm a Go developer - obviously, we have a lot of Go listeners on the podcast; I'm not sure if you realize that, but... They are going to want to understand not only basically "Hey, I'm a Go developer. Where do I get started with telemetry? What do I measure? How does Go make it easier or harder, or simpler?" Basically, they have these concerns... But in all of our collective experience, does Go make the job of collecting or emitting or whatever we do around telemetry in our projects - does Go make that harder compared to other projects? I'm curious...
+
+**Dave Blakey:** I don't think so. About 50% of our stack is Go. We're using it exactly in the way that I described to you, and developing products for clients that do it in the same way... I think it's actually quite easy. It's very easy to get that data out in an efficient way. Obviously, that's one of the easiest, the nicest things to do - you can just dump that data out and you don't have to worry about it affecting the performance of your program; that's also really nice when you look at things like telemetry... Because you don't want the telemetry to ultimately become a bottleneck in your platform. That's why I said UDP, for example, is very popular, because you can just fire and forget. And it's very easy to do that with Go.
+
+But Go itself in everything has telemetry. When you look at telemetry, we often think "Okay, it's very advanced measurements around very specific application-focused things", but your garbage collection is telemetry; how much memory have you freed, how much memory have you allocated, what's your current usage... All these kinds of things are telemetry, and once you start to monitor that stuff, you start to think of things that you might also want.
+
+We have a client server app, so we output from our Go server system; well, how many people are connected right now? Is that changing? How many requests per second are those people creating? And that's all just simple telemetry; we don't even use a third-party library, or anything... We just (like I said) fire and forget a UDP send out of it.
+
+So in my opinion, I would say it's very easy, but then I think it's easy to do it in any language. I certainly don't think Go hurts, and it's very easy to do it in a performance-sensitive way.
+
+**Jaana Dogan:** I personally wish that there was an easier way to export... You know, if the runtime was writing to a UDP port by default or something, that would be much easier. A lot of times people learn to care about telemetry at a later time, as you said, and it's really significant if they were able to just turn on something and collect that data in production, or sometime when they need it.
+
+There's been a lot of discussions around the standards, I think primarily for this reason, because we wanna be able to address "Oh, how can we make people turn on maybe collection at a later time, and collect as much as possible and utilize it when the user needs it?"
+
+So I think there's one particular thing that we may take care in the long-term, and that's this - being able to collect at a later time.
+
+**Dave Blakey:** It's so difficult... Because I agree with everything you said, and then at the same time, it's a hard problem to solve, because the important metrics in one app are totally different from another. But I do agree that if there was a very easy, accessible, well-documented -- you know, the lines of code for the project would probably be very small... But a well-documented source that people could use just as the book on what you should store from a Go app, and what foundation you should start with... I think that would encourage people to not have to go back in time, like you said, and add to it.
+
+**Jaana Dogan:** \[32:23\] Yeah, I altered a page on the [golang.org/doc/diagnostics](https://golang.org/doc/diagnostics.html) but it's never a document that people read through before they push something to production... So maybe we should do a better job explaining the whole production-related issues.
+
+**Dave Blakey:** Often a popular package does a better job of getting a readme across than a page...
+
+**Jaana Dogan:** True. \[laughs\]
+
+**Dave Blakey:** A package that has a lot of stars, that a lot of people use, you see "Oh, everybody's using this..." And it can be 50 lines of code, but if it just sets the standard for what you think, then that's quite a good way of getting that.
+
+**Jaana Dogan:** That's such a really good point. The number of times that I just published some packages, very small packages or tools - it's because it was hard to give the user an entry point... So you just make it a small project, and then people start to like it, and share it, and it becomes more of like a de facto thing. It's a really good point, that presenting it as a project or some utility tool is a really good way to spread the word.
+
+**Johnny Boursiquot:** Speaking of packages, Go has - curiously enough - an *expvar* package that's built into the standard library. If one's curious, or if one's kind of scratching their heads wondering "Well, that looks awfully like some sort of mechanism where I could be collecting metrics and instrument my Go code, and expose that to something that's gonna come scrape it", or something like that... Should folks be looking at that as a starting point for instrumenting their code in Go? What are your thoughts?
+
+**Jaana Dogan:** Can I explain something about this?
+
+**Johnny Boursiquot:** Yes, please.
+
+**Jaana Dogan:** Basically, that package has been modeled after *varz*. *Varz* is a convention at Google where you have some keys and values, and you can basically in the binary register any key, and then set a value. The *expvar* package was very identical to the *varz* libraries at Google; I think they needed it because some SRE folks demanded it when they were first going to production with Go... But over time, *varz* turned out to be like "We think that it's not very scalable", because people just dump a lot of random things, and then the name space is becoming very complicated, and so on... So they sort of like deprecated varz and switched to a different model.
+
+I think in 2.0 there's a topic around this, that they're thinking about deprecating *expvar* and maybe replacing it with something better, especially if there's an established standard; or they're going to reconsider it for Go 2.0. That's the background story... But you know, we can still discuss if it's useful for end users.
+
+**Dave Blakey:** Yeah, we are by no means the authority on collecting telemetry information. We focus on a very specific sector of application telemetry, and then we process it and report on it all ourselves. But in my personal development experience - not from a large-scale project or anything like that - I've found that it's better to fire and forget telemetry than to expose a telemetry collection point. I don't know if that's really where the standard will go... Maybe people will point to this podcast as where I was wrong about what the future of telemetry in Go would be... But you know, exposing a bunch of almost what I would call debug stuff as the solution to telemetry is a bit of a slippery slope... As opposed to saying "This is a metric that we care about for reason X, and we're gonna send it to location Y, and in the future we'll use it for various things." Because one of the biggest parts where you start to learn what telemetry you need and how to use your telemetry is when it actually either helps you solve a problem or doesn't.
+
+\[36:14\] If you've got an issue and you're able to see where that issue is through your telemetry, then you learn something... And especially if you cannot see where it is through your telemetry, you learn something. We've had that, where we've said -- you know, we've had this performance problem that we've ultimately found, and our telemetry didn't find that, and so we've added more tracking in that piece of the code.
+
+I think it's almost just like a dump out on some HTTP GET that people need to then collect data and pretend to process it in place... It probably doesn't actually solve the developer problem of making sure that the things get used... But that could just be my personal opinion.
+
+**Jaana Dogan:** This is actually a very good topic, and it's still a very relevant thing - what is the best way to pull metric data, or to make the process push. We currently think that scheduling the pushing is better, because at least the process knows "I can schedule the push." Even if it's not just like a UDP fire and forget type of a push, the process has a better chance to run this in the background and just do the push whenever it's better.
+
+In the pull model, imagine that a server is receiving a lot of traffic, and there's already a huge workload on the server, and then your monitoring system comes in and tries to pull, and doing a bunch of work in order to just be able to generate all the values of the metrics and present it as an HTTP endpoint, in the Prometheus endpoint fashion - it's just kind of overloading the process. So instead of that model, it's much better to push... But you know, this is still a controversial topic, because it also depends on how you deploy your monitoring stack.
+
+I think the pull model came from -- Prometheus' pull model is coming from Borgmon, because at Google initially everybody was deploying their own Borgmon instances... So they'd kind of have more of an overall control. They shifted to more of like a central, globally-scalable type of monitoring stack. The requirement - it's almost like you don't have to care about the availability of your monitoring stack at all, and you don't have to strictly position your monitoring stack or collector with the processes you have... So they had more flexibility in terms of pushing. But they didn't hit this as a bottleneck initially, because there were other problems such as maintaining your Borgmon instance, and so on.
+
+But if you have a globally available collector, pushing is much easier, because at least the process can tell "Oh, I don't have much traffic right now. Maybe this is the better time..." Because you know, exporting metrics is important, but it's not as important as serving the user traffic, right? So giving that flexibility to the process is really important.
+
+**Dave Blakey:** Yeah, I couldn't agree more. That's the nice thing about pushing - you can go all the way from fire and forget, like I say, which is really nice, because then there's no headaches around that... But if you go further up, you give the process the ability to decide what's important and what's not. If it's about to fail, it might block to send that message, to say "Listen, we've got a serious issue here." But on the other hand, if it wants to decide that it doesn't need to store telemetry information right now because the system is overloaded, then it can do that as well... Whereas with collection it's just a static -- it's almost like you've got a cron job, which is \[unintelligible 00:39:52.22\] and gets a whole bunch of pages, regardless of what's happening, and you just dump stuff onto those pages.
+
+Yeah, if that answers your question -- I think our approach has always been to push the stuff out where possible, and to let the app decide what's important and what's not, and how it wants to deliver those messages.
+
+**Jaana Dogan:** \[40:12\] Since you're talking about the UDP, do you have an agent that collects...? What is the collection model like?
+
+**Dave Blakey:** For us -- so we've got two sides of telemetry, really... We've got our product, which collects specific telemetry for our [ADCs](https://en.wikipedia.org/wiki/Application_delivery_controller) and load balancers and things like that... But then more so I'm talking about for our own internal use, like for our code, and our hosting systems, and all that kind of stuff... And for that, we just have our own -- again, that DevOps, hacked everything together... But we have our own collector thing in the middle, that does a whole bunch of various things with that data. And the reason that that happened was because we use it for some of our actual applications, our clients' telemetry as well, specifically for anomaly detection in it... So it does all of that stuff for us.
+
+But then some of our data - we stream directly out of that UDP fire and forget, and we send straight to Datadog, for example. So we even explored off-platform, some of our shared, SaaS-based hosting things, and then other stuff we keep in product. So we're exactly that bad example where we kind of built it ourselves.
+
+**Johnny Boursiquot:** It's interesting that a company whose product is collecting and exposing some of that data is actually using another company who's able to display that... So is this a case of -- I'm wondering if this is a symptom of sort of "Basically no one tool or platform that does it all, or that answers all the questions you might have", so you end up having to pull in a bunch of different things in order to get an overall observability answer?
+
+**Dave Blakey:** Exactly. Because we don't answer all the questions. Our product is sitting at the entrypoint to the network. It's an ADC, so you've got load balancer, security, firewall etc. for the traffic that's coming in. So we're reporting very specifically on that information. And that means that we also then need to offer that information out to our clients to integrate with other things... Because if they've got a problem in that space - yes, they'll come directly to our platform and look at their reports on their data, and things like that... But if they've got a problem with the app at large, we need to just contribute our small piece of information to their overall telemetry. So it's quite common for us to ship information off our platform to theirs, or expose it in some way.
+
+Generally, we've tried to be as open as possible, especially when we deal with larger enterprises. They have almost all got their own use case, and as wide open as you can make your platform, I think ultimately it's the best. To the points earlier though, there's not a lot of standards, so we wind up adding seven different ways of getting data out of our platform, because that's what's needed... \[laughter\]
+
+**Jaana Dogan:** One of the interesting things that we realized when initiating OpenCensus was a lot of our large customers were dependent on multiple products... And sometimes this is about really trying to get some additional stuff, additional feature from a vendor, and sometimes it's about the team preferences. In a very large organization, a team is like -- they like Datadog, they wanna use Datadog, some other team wants something else... So we thought that having something vendor-agnostic is really the key. You can't really lock that type of data to a provider; that's not going to be useful for anyone... So being able to export to multiple vendors was also very important in our case.
+
+**Dave Blakey:** I think that's 100% true. When you look at the more traditional model, you've also got multiple stakeholders, who only want certain pieces of the data on their certain platforms. You've got IT ops and you've got security, and they can run totally separately... So I think that it's critical. The way we have wound up having to do that is by building it ourselves.
+
+**Break:** \[44:15\]
+
+**Jon Calhoun:** So you talked about on the ADC side of things you're collecting certain telemetry... Can you share some of the more important ones you feel like you guys are collecting, and where customers have found them to be useful?
+
+**Dave Blakey:** Yeah, absolutely. Our newest product is called Nova, and it's our kind of cloud-native-focused scalable ADC. An important component of that is that we run many ADCs centrally, so it's like a control plane/data plane model; we are collecting a lot of data from the data plane to display on the control plane... But we had a lot of learnings in our traditional product sample, which is like a standalone ADC.
+
+But what's interesting is that we've tried to tackle it in a very different way. We collect mostly the same data - how many of every type of HTTP reply code are you getting? How many requests are you getting? How many TCP connections? How many TCP connection failures? How many timeouts are there? What's the reply time? And when you look at the response times, there's a lot of information there. Like, what was the TCP connect time to the server - is there a network issue? What was the HTTP reply time from the server - is there a back-end issue? What was the response like to the client? How long until we closed that session with the client - is there a front-side network issue?
+
+There's all of these metrics, but what we've tried to do - and time will tell if our approach is interesting enough or right enough now... What we've tried to do is not put any hardcoded values in for any of those, but rather to do just like anomaly detection and predictive profiling of what we expect the data to look like. Because one of the things is our system autoscales, so it will pre-scale, so it needs to do a lot of prediction off of those numbers. So we've wound up in this system where we collect a huge amount of telemetry and we set no hard lines for what should be alerted, but rather just if it changes too much... And so far that's going well, but I think it's a little bit odd for some people, because they wanna say "Well, I expect my website to respond in 200 milliseconds, so if it's ever more than 250, please tell me." And instead, we're saying "Well, if it always responds in 200, then we will tell you if it's 250. But if it doesn't, then we won't."
+
+So all of that type of stuff is your traditional things that you expect, like what's throughput of the collect, or the request rates, the response codes... Because you can pick up a problem long before by saying "Oh, I normally generate 0.1% errors, and now I'm generating 0.5% errors." You might not notice that, but it means something's changed, and it could mean that something's about to get a lot worse; it could mean that there's a security issue, and it could mean all of those things.
+
+But by the same token, we will also check for variances between two things. For example, if the average user sends far more GET requests than POST requests, but one user is sending far more POST requests than GET requests - is this a security issue? Are they trying to brute force a password, is this something weird? Is a specific user getting way more 404 errors than everyone else? Why is that? It's probably some script, or something. So telemetry is often a combination of two values, like "What is this value versus that value?" as opposed to just a single value. So that's a lot of the stuff we focus on.
+
+\[48:09\] The client connects to us, we connect to the web servers and we send their data back. That's our model... So everything in that communication chain is the telemetry that we care a lot about, because it could mean that there's a problem with the client servers, it could mean that there's latency or issues that are affecting the user, or it could mean a security issue... So that's the type of stuff we need to obviously track for scaling up and scaling down, as well as for alerting the user to problems with their service.
+
+**Jaana Dogan:** Yeah, it sounds like a very difficult field, especially given the trends of traffic can change, the usage can change... You have to incorporate all of that in order to actually be confident about the detection, right?
+
+**Dave Blakey:** Exactly. You know, people tend to use the word "ML" here, right? Machine learning. That's what they tend to say. But really, it's just a statistics problem at its core. You're really just evaluating numbers against other numbers. We do work with some ML type of stuff because of exactly what you've said - traffic patterns can change very rapidly. In one minute you could have ten times the traffic than you do in the next minute, but they change in a way that makes sense if you look at all of the data instead of one data point. For example, your throughput will go down in a predictable fashion with your request per second, as will your HTTP 200 replies, as will your POST requests, as will your CPU usage on the sever, as will your network latency... And if something there doesn't decrease at the same pace that something else decreases, then anomalies become very obvious to a system that's looking at the data as a whole.
+
+Where we've had a lot of difficulty is weeding out all of the trash that it picks up... But that's kind of our value-add to that, I guess... But really trying to find the balance of saying -- you know, because the worst thing about a telemetry and analysis and visibility/observability platform is if you generate so many alerts that people start to ignore them; then that's a total loss. Rather have too few, so that you get the really important ones through... So it's quite hard; it is difficult, and also especially as we are scaling. We're trying to pre-scale based off of that information. So it's quite a balancing act.
+
+One of the biggest learnings for me personally with telemetry, something that I've learned from our team is that things start to make a lot more sense when you're looking for anomalies in sets of measurements, instead of individual measurements. I think that's been a big, core design factor for our platform.
+
+**Jon Calhoun:** I think that all makes sense though, because anybody who's ever been on pager/had a pager or anything for a product knows that when you get paged for too many anomalies, you're basically to this point where you just assume "I'll wait till it does it again, to see if it's actually a real thing." And when that's happening, it's like "Okay, that defeats the whole purpose of the system we have in place", because people are ignoring things... But then you also mentioned, you will get some -- if you're working on anomalies, you can get some spikes in traffic...
+
+I remember one of the ones that stuck out to me was I was helping with Google Code Jam, and it's one of those competitions where everybody logs in at the exact same time, because that's when the competition starts... And I believe at the time the way that they were doing some of their monitoring stuff was basically the same thing - look for anomalies. So the guys who were setting it up basically knew that you sort of had to warm up the servers ahead of time. So it was this weird thing where you're like "What are you doing?" and he's running a script to sort of get the server used to this request load coming in... And it was just because that was the simplest way to ignore that anomaly, because you knew it was coming... But it really tended to happen during very specific things.
+
+If you have a timed event, and that sort of traffic spikes, then it becomes very challenging... And I think that's probably also -- you see video games and stuff like that that have a launch date, and I think they have to deal with that type of problem pretty heavily, where it's hard to detect an anomaly when everything just skyrockets all of a sudden.
+
+**Dave Blakey:** \[51:56\] Yeah, exactly. But a lot of the time an anomaly can be informative. I think that's also up to the team that gets them, to make sure that they do the right thing. If our website gets ten times the views after this podcast, I'm happy to be told. It's not gonna go offline... But you know, sometimes informative telemetry is not necessarily a problem, but yeah, it can reach the point of spam, which then people start to ignore, which is a big problem... But you know, it's a balancing act. With alerts and with anomaly detection it's all about balancing it; you wanna make sure you pick stuff up.
+
+The problem at our scale becomes so vast, because -- let's use a use case. Let's say a banking client of ours - they might have systems in 20 different countries. Now, how many failed logins per second do you think they get? It could be 500, it could be 1,000 that they get per second... So if they've got 10 more, or 20 more, or 50 more, it might not detect an anomaly. But what if they get 50 more in all of their locations around the world, all from the same country? Is that a problem? Probably it is. So sometimes -- that's the funny thing about telemetry, people tend to zoom all the way in.
+
+We were talking about trace IDEs, and "What is the individual request?" A lot of the time that's very important, but sometimes it's actually really important to have that 10,000-foot view, where you're just like "What is the lay of the land? What does it all look like as a picture?" And that's also something that's not that easy to do now. There's not a lot of standard stuff for that, or just best practices, like "How do you set up your dashboarding, if you're using Prometheus, or if you're using Datadog, or whatever it is...?" Using that is like a big failing, I think, in DevOps teams and traditional teams today. It's making sure that you always go back to your telemetry and say "Why didn't this tell us about this problem before it happened?" There should be almost like a root cause analysis of the issues... And it doesn't have to be this fancy process, but just going "Why were we not aware of this, now that we understand what it was?"
+
+**Jaana Dogan:** Yeah, that's such a really good point... Especially large teams, large companies - if they haven't thought about telemetry in the beginning, they wanna introduce it at a later time, but they don't know where to begin. Anomaly detection really helps them to explore the area as well. It's not super-obvious to you, but you can maybe run it just to see and explore all these edge cases, and some of the critical things, the correlations... It may actually help you to explore what you need to take a look at, even if you end up having an SLO type of approach in the end.
+
+**Dave Blakey:** Yeah, you said the most important word, which is correlations... And a lot of the time it's not obvious to the human eye, but it can really help when you're trying to scale systems. The nature of scale has changed so much now. You can scale up easily nowadays in the cloud, or in containers, or whatever, but the difficulty and the challenges at the languages that we write in are so high up the stack that a lot of the time the difficulty in diagnosing the bottlenecks or the performance issues of things can be very hard... And being able to put two data points together and understand that that's why something is slow, or that's why it's never gonna scale, even if you put 3,000 servers behind it - that can be helped a lot by anomaly detection.
+
+**Johnny Boursiquot:** Yeah, this is more art than science, it sounds like...
+
+**Dave Blakey:** It really is. I read this funny thing the other day about the difference with developers - some are artists and some are engineers, or whatever it was... But it is like an art, because you really need to say "What are the what-ifs, and what can I just store and see what I get out of it?" It is experimental, I think.
+
+**Johnny Boursiquot:** Okay. One last question for you before we go to our next segment... Based on what you're seeing so far and how your customers are, or what you're seeing in terms of the data that you can glean from how your customers use your product, are most of the things that trigger something to look at, something that's important to look at, of the anomalies - are most of those triggered from internal sources, meaning that the developer is pushing new code, making changes that's causing issues, or are those coming from the outside? Maybe there's somebody who's trying to brute force their way in, or maybe the company just got listed on some popular website or something, and maybe there's a surge in traffic. Generally speaking, where are the biggest sources of problems?
+
+**Dave Blakey:** \[56:25\] Generally speaking -- the answer I wanna give you as well is "Well, it's both." But let's pin me down for an actual answer...
+
+**Johnny Boursiquot:** \[laughs\]
+
+**Dave Blakey:** If my back was against a wall, I'd say usually it's the servers and the apps that fail, in a condition which the team has struggled to test. Testing things nowadays can be very hard. Take our platform - we need to test ten million active connections, ten million active devices connecting to our platform. How do we do that? We've got six Kubernetes servers that are running on 2,000 machines, and it's still a nightmare. So you get these systems where people scale things up, and where people put things in auto scale groups and everything, and ultimately there's still some bottleneck and things fall over that they just couldn't test. It's like Black Friday; if massive e-commerce sites can fail, I assure you yours can too, in this unpredictable load.
+
+So the reality is that most of the time it's that that's failing... But what's interesting is that it's often easy to predict that it's gonna fail, and allow them hopefully time to correct for it. To predict that page load times are slower than normal, that traffic is higher than normal at this time, on this day, we'll go all the way down to DNS queries. If there's way more DNS queries coming in than we normally have coming in for the site, versus the requests per second - all these kinds of things...
+
+So usually an issue is downtime on the upstreams; the actual origin servers for the API, for the website, for the e-commerce store, whatever it is. But the cause, the reason that it happens - it will often be a burst in traffic, or something unexpected, or some new feature that gets rolled out, or a change in the database system. Someone upgrades from one SQL to the next SQL version, and the query cache is now no longer one gigabyte default, instead it's now zero by default, and the whole system falls apart... You know. But you can start to see that, because like at two in the morning actually the page load times got worse. And if someone could see that and say "Hang on, at six the page load times are worse. What happened last night? It won't fail over at 9 when the traffic starts."
+
+I think that's the beauty of telemetry, is understanding those unknown changes. And you upgrade your SQL server, you go to the website, everything works, and you think "Phew... It's working." You don't know that there's been a 25% page load time decrease, because you can't feel that... But when you get hit by 100,000 requests per second, you feel it big time.
+
+**Johnny Boursiquot:** Jon, would you like to introduce our guest to our next segment?
+
+**Jon Calhoun:** Sure. Mat started this segment called Unpopular Opinion... And I think right about here they put in some little riff...
+
+**Jingle:** \[59:10\] to \[59:27\]
+
+**Jon Calhoun:** Basically, the idea is we want you to share an unpopular opinion you have, preferrably in tech, but it doesn't necessarily have to be... With the goal being to just share with listeners that not everybody agrees with the really popular opinions, everybody has different things that they disagree with, and wanna share.
+
+**Dave Blakey:** Yeah, absolutely. Well, I would be remiss if I didn't say this, but I'm vegan, and that's pretty unpopular... \[laughter\] But if we're talking about tech, the biggest thing that I have made the mistake of myself, and that I see a lot of small companies doing - you know, I work with a few startups, or helping people, a lot of our earlier stage companies that join... We have a community edition, which is free, so we get to communicate with a lot of these guys pushing the boundaries of things that they are doing, and get in touch with them... One of the things that is probably an unpopular opinion is that I think that startups and a lot of people are writing code in the wrong languages, almost all the time.
+
+**Johnny Boursiquot:** \[01:00:27.06\] So they should be writing in Go, is what you're saying. They should be using Go.
+
+**Dave Blakey:** Well, it depends. How many Go devs do they have on their team? So that's the point, right? That's the point. I think people at young companies choose the language based on how trendy and how cool and how high-performance it can be. But no one really wants to maintain a wiki that's been written in Erlang. And a lot of the time, people are not worrying about how easy is it to hire talent for this, how easy is it to scale this? How well-known is this in the developer scene and in the market?
+
+I think much more likely your app is slow because your code is bad than because the language you wrote it in is bad. When you get to that point, then you're past that struggle... But this tendency to always chase the latest language I think gives people business scaling problems, and it's very difficult to get talent for it, and it's very difficult to build an engineering team around it... So yeah, I would say that I think people are often choosing the language that they use incorrectly.
+
+I don't actually think that Go is an example of that, because it's one of the ones that I think is very easy to pick up, and to learn, and to get resources on, and to find people that are playing with it, for whatever reason... Like, it's done well to get a community. But a lot of people will just write in whatever the last podcast or webinar they watched was using... And I think that's a mistake.
+
+**Johnny Boursiquot:** Awesome.
+
+**Jon Calhoun:** I can definitely relate to that. I've seen it go the opposite direction too though, where the general advice is "If it's gonna be three of you, you're probably gonna work on this thing with just the three of you for six months to a year before you can really afford to hire." Maybe not always, but a lot of times that's the case... So it's like, in that six months to a year, how are the three of you gonna be the most productive? So you kind of pick a language based on that. And I say three - it could be one person, two people, or however many people.
+
+I've seen companies that start with really old languages as a result, and -- what was the one...? I think it was Perl; they used Perl, and the only real issue with that... They were productive and they got a lot of stuff done, but I think they struggled later, like you said, with hiring, because when it comes to hire later, you're like a trendy startup, but everybody looks at the language you're using and they're like "Yeah, that's not exactly my first choice..." Nobody really wants to spend time learning a language that is probably not gonna benefit their career in the future.
+
+If you're learning Go, you're like "Okay, this is gonna at least benefit me in the future if I find other companies that are picking it up." But if you're learning a language that's not dead necessarily, but it's not growing, then it's a little bit different, too. So I guess I see what you're saying, but I guess I'd also take caution on the other side of it and say "Don't use something that's also gonna cause problems because it's so old, or you just know it so well. Even though you know it that well, it might still present issues.
+
+**Dave Blakey:** Yeah, you're exactly right. The crux of my point is make sure that it's easy to hire people with that language... Which I think is exactly what you're saying - there is a balance between something that's growing steadily, that's got a lot of acceptance and people talking about it, and it's also actually being used at companies, that people have production experience with it... Because just because someone understands a language doesn't mean that they know how you build a massive app and keep this thing online and actually deliver it in that language, and maintain it.
+
+Frameworks are a good example. "Stay away from frameworks, because we must write this thing as bare as possible." Write it in Assembly then. If a framework is gonna make your team of three get five times more work done in the first six months that you've got to get your MVP out, then use that. I would say find that middle path, but I think people are far too far forward and you need to caution them to go backwards.
+
+**Jon Calhoun:** \[01:04:15.26\] Yeah, unfortunately I think that's a problem that comes with all -- I'm trying to think of how to word this... Basically, how you deploy things can also present issues, where everybody wants to use Docker and Kubernetes, and stuff... But I remember when Google's App Engine was fairly new; I knew a couple of people who wrote a lot of stuff in Python, and they were like "Well, let's go to App Engine. It's going to auto-scale for us." And their first project they really struggled, because there's a lot of specific things you kind of have to learn about App Engine. Once they figured that all out - their second project, they would have flown on App Engine, but their first project was really a pain in the butt to deal with all these blockers that really shouldn't have been there.
+
+**Dave Blakey:** Yeah, you bring up a second unpopular opinion of mine. I mean, maybe it's not unpopular, but... I don't think that containers and Kubernetes and cloud-native is a destination. When we were all on tin, everyone said "Okay, everybody is gonna be on VMs." And then were all on VMs, and now everybody is gonna be on the cloud. And that's where it started to get shaky, because not everything did move to the cloud... And now this idea that the next step in that evolution is containers and cloud-native I think is wrong.
+
+I think there are workloads that are excellently suited to that, there are workloads that are suited well to serverless, but there will always be workloads that are suited to tin that is within a mile of your house. I think it's a spectrum now. It's like, stop trying to make a round peg fit in a square hole. Not everything has to be deployed into containers, which is exactly to your point. It's often the easiest thing, the thing that is used by the most people, that turns out to be the best decision.
+
+**Jaana Dogan:** I feel like the cloud is becoming more like the programming language industry; you have to introduce a new product or a new abstraction layer in order to get the attention of people. Maybe I feel like partially why we had so many different solutions was just because people wanna make some noise about it.
+
+**Dave Blakey:** Yeah, you couldn't be more right... And it's interesting how the simple clouds are doing better, are growing quickly now. It's something that we actually see in our business, because we compete with commodity load balancing in clouds, with ELB, or LB, or Azure's Gateways, or whatever it might be; every cloud's got a load balancer, so we compete with them. But what's interesting is that people wanna be cloud-neutral now. So they wanna be able to say "Yeah, I'm in GCP now, but I can shift that to Azure etc" and they actually wanna use less and less of the commodity, proprietary cloud stuff, and try and stay neutral. So they're delivering more and more features to keep on everyone's tongues and keep talking about it, but I think people are steering more and more away from using one specific infrastructure provider's solution.
+
+**Johnny Boursiquot:** Hey, new features bring new sales, man... You know, you've gotta factor that in. \[laughs\]
+
+**Dave Blakey:** Look, we add new features all the time. I feel you. \[laughter\] Do what I say, not what I do.
+
+**Johnny Boursiquot:** \[01:07:13.03\] \[laughs\] Awesome, awesome. So in general then, we can say that orchestration and scaling have generally become easier. Telemetry is one of the hard problems still remaining, that a lot of people are trying to solve, your company included... And as we explored today with you, Dave, we know that there's no one-size-fits-all. But at the very least, it's advisable that everybody starts with something; that should be considered -- having some form of telemetry that provides some form of insight into your workloads is at minimum required to be considered production-ready, to some degree.
+
+**Dave Blakey:** Yeah. I mean, when you start working, if you put in a comment that says "To-do: add telemetry to this, because it's a bottleneck", then that's fine, too. Just start thinking about it, and the rest will come.
+
+**Johnny Boursiquot:** Nice. Hopefully, you do a little more than just think about it, but... \[laughs\] Yeah, indeed. Thank you for joining us, Dave, and thank you to my co-hosts, Jon and Jaana. I am Johnny Boursiquot, and I'll catch you in the next Go Time.

The Zen of Go_transcript.txt

@@ -0,0 +1,421 @@
+**Mat Ryer:** Hello, and welcome to Go Time. I'm Mat Ryer. Today we're talking about the zen of Go. On today's show we have Johnny Boursiquot, Carmen Andoh, and the one and only, Dave Cheney. Hello, everybody!
+
+**Carmen Andoh:** Hello!
+
+**Dave Cheney:** Hello, friends!
+
+**Mat Ryer:** How's it going, Dave?
+
+**Dave Cheney:** It's going well. How's everybody doing in radio land?
+
+**Mat Ryer:** Yeah, not bad. Carmen, it's been a while since you've been on actually, hasn't it?
+
+**Carmen Andoh:** I know. I've missed you all, it's nice to be back...
+
+**Mat Ryer:** Yeah, and we've missed you. I was about to say that.
+
+**Carmen Andoh:** Dave is fave! I didn't come back for you, I came back for Dave, sorry...
+
+**Johnny Boursiquot:** Ouch.
+
+**Mat Ryer:** None taken. None taken. \[laughter\] Johnny, how are you, mate?
+
+**Johnny Boursiquot:** I'm doing well. Not too bad.
+
+**Mat Ryer:** Good. What's been going on?
+
+**Johnny Boursiquot:** You know, not much... Just crackin' on. Just trying to be nice to everybody. Unlike you, I think.
+
+**Mat Ryer:** Oh, that's hard work. It's too hard.
+
+**Johnny Boursiquot:** From what you've told me, yeah. It's hard. \[laughter\]
+
+**Mat Ryer:** This show is called The Zen of Go, and Dave - I wonder if you could just kick us off and tell us what does that actually mean, and where does this come from?
+
+**Dave Cheney:** Okay. Well, first things first - I have to give credit where credit is due. The idea for the Zen of Go came from many places, including here in Go Time. I think the last time when I spoke, we were talking about try. One of the things that Peter Bourgon said - and maybe for him it was just a throw-away comment, but for me \[unintelligible 00:03:21.28\] was when he was talking about error handling should be explicit. He said "This should be a core value of the language. We should write it down, we should make this explicit." So that was one of the ideas.
+
+In some of the other extended discussions about try - this isn't all about error handling - Daniela Petruzalek said other languages have core values, and she kind of pointed to the zen of Python. So that was what had kind of got me percolating for a year. Then a couple months ago I was lucky enough to be invited to go to the second GopherCon in Israel. They said "You have an hour" and I was like "Well, what have I got that would fill an hour?"
+
+\[04:03\] So very much the talk that came out of it and the ideas that came out are kind of an hour's worth of ideas... As well as recognizing these really are not my ideas, these are just things that I've pulled together. I also need to apologize for everyone having to read an hour's worth of text. That's the reason why that document is so long; if they gave me 30 minutes, it would have been much shorter.
+
+**Mat Ryer:** \[laughs\] That's great. There is a kind of mini-site as well, where you've written out a set of these rules, or these points... How can we find that?
+
+**Dave Cheney:** I guess we'll put it in the show notes, but it's thezenofgo.netlify.com. The main reason I do that - and this is perhaps a little kind of presentation inside judo is when I'm giving feedback to people on their presentations, I always say "Look, I have a superpower. I can control what people in the audience are thinking. And the way that I do that, is I put the words really big behind me." So if I want them to think something, I write those words -- like, if I want a very important thing out of a sentence or a paragraph to be the key message, I put that behind me in a million-point font.
+
+**Mat Ryer:** Dave, I'm gonna write this down. I've been doing my talks wrong, for sure.
+
+**Carmen Andoh:** Yeah. Total Jedi mind trick right there. \[laughter\]
+
+**Dave Cheney:** Also, if you want people to quote you well on Twitter, like with a photo, make sure that the words that you want them to repeat are behind you.
+
+**Mat Ryer:** Oh...! That's a good idea.
+
+**Carmen Andoh:** I tried that, but it was a Dave Cheney quote, so the picture is of me and a Dave Cheney quote behind me... Just - sorry.
+
+**Johnny Boursiquot:** \[laughs\]
+
+**Mat Ryer:** Yeah, I do that sometimes... And then there's not enough space to fit the credit on, so I just take that off... \[laughter\]
+
+**Carmen Andoh:** And just take credit yourself?
+
+**Mat Ryer:** Well... It's default, isn't it? I can't help it. That's the default value of that.
+
+**Carmen Andoh:** You know, even though thezenofgo.netlify.com are the-zen-of-go.netlify.com, I actually quite like the blog post better, because I don't mind reading an hour's worth of text. I find it quite enjoyable, all of it.
+
+**Johnny Boursiquot:** It's more contextual, yeah.
+
+**Carmen Andoh:** Yeah.
+
+**Dave Cheney:** So a part of it was to have something snappy to take away. One thing you always do in marketing on a trade show is make sure you give somebody something with your name on it, so when it's on their desk, they look down at it and they're like "Oh, I remember that."
+
+So part of the little mini-site was a thing to give to the audience to take away... And there's always a tension between a 6,000-word blog post and people who don't have the time to read the whole way through... So I think there's kind of a notion of giving them something more concise. Again, to give props to all the inspirations, Rob Pike's Go Proverbs was another big inspiration there. Maybe we'll get into this later...
+
+I have this notion that the proverbs are kind of like advanced level, maybe... Because popped out of the idea of proverbs in both the way that Pike said them and the original gang of Go inspired ones - they were these little kind of these whimsical observations... Which kind of make sense, like "Oh, now I understand the \[unintelligible 00:07:03.17\]" But is that really useful to someone who's never seen the game before?
+
+So that's why each of them comes with at least a little paragraph to describe them... Because ideally - this is kind of my take on it, that they should be usable especially by novices, because this comes back to my whole rant about idiomatic Go actually being quite an exclusionary term. I didn't want them to be vague in that way, that you could have a long stand-up argument about it, in the way that the zen of Python is actually a -- it was a thought devised to prompt people in the Python community to really say "Well, hang on... That doesn't seem to be how I'm actually writing Python." Tim Peters was obviously being like "A-ha! That's the trick."
+
+**Mat Ryer:** You're right, the titles of each little point are quite poetic, so it does leave a lot of room for interpretation. They aren't code, are they? So yeah, I appreciate that, and it's a good point to make... And I've seen this actually happen where someone will give a talk, and they'll kind of skip over some of the fundamentals just to get to the higher-level ideas, and then somebody that perhaps didn't have those fundamentals - their experience of it is then very different... And I've seen that happen a couple of times. So I think that is a thing that we have to look out for. But the little descriptions on the site I think for sure help that.
+
+**Dave Cheney:** \[08:30\] To give some background - in a growing community like Go's growing community, which we all hope it is, most of the people in this community will have joined recently. We know this makes sense mathematically, and it also makes sense anecdotally. If you think of all the conferences you go to, like \[unintelligible 00:08:46.01\] you might do a show of hands of like "Who's been here a long time? Who's their first time here?", the majority of the people at conferences and at meetups are newcomers... So we should optimize for the newcomer rather than over-focusing on the old hands who've been here a long time.
+
+**Mat Ryer:** Yeah, I think that's a nice decision if you have to make that decision... Because the thing about people with more experience is they have more experience. So they are more kind of equipped to be able to find the right information, and things like this. I do feel like whenever I give a talk, I always try and give it to someone that's new. I always want to make sure they're at least covered... That's why sometimes - even on this show - I sound like an idiot... It's not because I am one, it's because I'm being nice.
+
+**Johnny Boursiquot:** \[laughs\]
+
+**Carmen Andoh:** Well, you mentioned new, and I think one of the beautiful phrases for making sure that you have that perspective is Zen Mind, Beginner Mind, which is why I love the title of your talk and of the blog... Someone who's never heard of it - "Well, what is Zen Mind, Beginner Mind?", it says "In the beginner's mind, the options are many. In the master's mind, the options are few." So sometimes the proverbs that Rob has - the brevity and the wit comes from layers and layers and layers of meaning... So when you get it, that is the delight of those proverbs, but it also is exclusionary of those who haven't yet gotten there... So I'm glad that you did this and made it more explicit.
+
+**Johnny Boursiquot:** I'd like to perhaps briefly touch on the fact that you consider the term "idiomatic Go" to be exclusionary, in some way... I kind of take a slightly different view of it, in the sense that I think it's an aspirational goal almost. So when you join a community, there is going to be an idiomatic way of writing, be it Go, or Rust, or Ruby, or whatever. You want to fit in, in a way, and to write the programs like most others are doing... At least the correct way that most others are doing it.
+
+So to me, it's an aspirational, less so an exclusionary term... But I can definitely see where you're coming from when you say that. To me, when I juxtapose the Go proverbs and the Zen of Go, I see the Zen of Go as something you understand and take to heart on your way to the Go proverbs. Once you reach the Go proverbs, you're like "Oh, okay. Now so much starts to make sense", because you've been working with Go for a while, you understand the layered meaning in the proverbs. The Zen of Go is more explicit, it's deliberately an explanation of how you should approach your Go work.
+
+So I see writing idiomatic Go as aspirational, something that everybody should be striving to do... Not necessarily shutting the door on folks who don't quite know how to do it yet, whether they know if they're doing it right.
+
+**Dave Cheney:** Yeah. I think you touched on my key uncomfortableness with this idea of idiomatic Go. It is not that we shouldn't be writing code in the idiomatic style, but actually the word "idiom" itself - I went to the dictionary, and the definition of "idiom" is "The way that we do things around here." It's sort of the accepted norm. And any kind of culture that has that "This is the way that we do things around here", it can be groupthink. "We write code like this because we write code like this."
+
+\[12:14\] Now, the reason that I said idiomatic Go can be exclusionary is not the ideas of "We should write good Go code in the style that we've all become accustomed to", it's the way that that's actually administered. Scott Meyers, the C++ author, has this amazing blog post on writing The Effective C++ books. He totally breaks the fourth wall about how he went about writing the Effective C++ books, and the thing that I took away from that is -- well, first of all, avoid absolutes. You have titles like "You shouldn't say always and never, but say prefer or avoid." So first of all, moving away from absolutist language is important.
+
+The second one was that, in my observations - and I really wanna ask the rest of you how they perceive it... Most of the times, the way the idiomatic Go has been used, it is usually with a negation in front of it. That's not idiomatic. You did that wrong. And that's the key thing which I'm concerned about idiomatic Go as a teaching too.
+
+**Carmen Andoh:** Good point.
+
+**Dave Cheney:** Does that seem fair? Like, when you've seen idiomatic Go used as a justification or as a goal... I've generally found that it's very close on the page with words that imply that the thing that the person did is not idiomatic, it is not in the tribe, it is not following the accepted standard. They did it wrong.
+
+**Mat Ryer:** Yeah, it somehow encodes a suggestion of naivety, doesn't it?
+
+**Dave Cheney:** Exactly.
+
+**Mat Ryer:** ...in that criticism. And we don't need to do it. You can probably get the same ideas across without resorting to that... Which kind of makes sense. I've found that it's sometimes useful to follow patterns that other people are doing, even if I don't quite understand why they exist... And then I can sort of learn later. I've found that happens too, do you know what I mean? Sometimes I think if you expect everyone to learn everything straight away -- it is a lot to learn. Sometimes we say "Oh, it's easy", and that's a mistake. This stuff is not easy at all. In fact, it probably wouldn't interest us if it was that easy.
+
+So yeah, I feel like if you expect people to know everything too quickly, it can really put people off. So I like the idea that people can follow these ideas around, even if they don't have a full grasp of their deep, textured meaning, but it can still be a good shortcut for people.
+
+**Dave Cheney:** To come back to your point... The concern I have with idiomatic Go is that everybody is well-meaning, it's just the way that I've generally seen it expressed in code reviews, in comments, online - it's always with that "The thing that you did is not idiomatic." And what struck me was that negative feedback right at the point when the person is most excited about trying Go, most interested in investing, most interested.
+
+So this whole exercise was really a question of "How can we reframe that commentary or that feedback to be just in a positive frame?", so nobody has to hear "Don't do it like that, do it like this." And to go back to Meyers' work, rather than saying "Never use some of the millions of \[unintelligible 00:15:23.14\] in C++" make a suggestion that is "Do do this. If you have a world of choices, start with this one." That was really how I wanted to reframe it.
+
+**Johnny Boursiquot:** Yeah, that makes sense.
+
+**Carmen Andoh:** Getting to the heart of what idioms are for in programming - many ways for the code to compile, but in order for understandability to increase, in order for the code to grow well, age well, be able to be refactored, we have to layer the shared meaning on top, which is idiomatic Go.
+
+\[15:52\] So what I'm hearing is when you say "No, that's not idiomatic", there are certain contexts for which that might be okay... If the person is already at a certain level of understanding of Go. But we should also be very explicit in saying "The reason why we like idiomatic Go is because it helps us later on for X, Y and Z." I just love that we're even talking about this, because this transcends just Go, to any programming language.
+
+**Mat Ryer:** I like that idea that you've just brought up, which is instead of saying "Oh, this is rejected because it's not idiomatic", even if you don't say "idiomatic", you could say "This is a way to do this." And it's a sort of teaching opportunity, isn't it? ...which we don't always feel like we have time for, but sometimes it can be just a different way of framing the same thing you were gonna say before... And suddenly, like you say, Dave, it's a positive experience then, instead of a negative one.
+
+**Dave Cheney:** Precisely. To take my favorite one, which is \[unintelligible 00:16:51.03\] to not use a negation in there... "When you start a goroutine, always know when it'll finish." It's a much better way of saying that. Imagine if you hyperlink that in a code review. Like, thumbs down, hyperlink to something that says "When you start a goroutine, always know when it'll finish" is much better than someone writing "Hey, you've got a leak there."
+
+**Carmen Andoh:** Yeah. Being able to describe it by saying "Here are the consequences, all the way through to the way that the data transforms" shows a level of understanding and awareness... But I often have caught people who say "That's not idiomatic Go" with just kind of cargo culting, and not knowing why. So I love that if you're ready to give a constructive criticism to somebody, or some feedback for "Let's do it this way", I would love that whoever is doing that feedback be able to give some time to be able to talk about why that is. And that's hard to do. I think when we are asked to do that on the spot, very few people can actually do it. So we kind of then default back to "Oh, that's not idiomatic", without understanding why.
+
+**Johnny Boursiquot:** I was going to ask you at some point, which one was your favorite of these sayings, and you sort of mentioned that before... One thing I've noticed of the set - most of them (at least eight of them) are very explicit. You can literally read it and you know exactly what it's telling you to do or to avoid. At least two of those, maybe three, can be interpreted more subjectively... I'm thinking perhaps the most abstract one is moderation. "Moderation is a virtue." Obviously, in my mind I see moderation as a virtue as the top most abstract one; the one that requires a bit more experience, like having seen a few things to kind of know what the right amount, quantity is... And then obviously, maintainability, which different people will have different ideas about what maintainability means... And then the third one in my mind is each package fulfills a single purpose.
+
+So again, these three... Depending on who you're talking to, the teams that you're in - these things are gonna mean different things to different folks. I'm wondering, what is a tool that you would employ, especially for the first one, especially for the moderation part. What is a tool that you personally employ to know how far is too far, or how much is just right.
+
+**Dave Cheney:** Yeah, an excellent question... And I need to give some editorial context here. To talk about public speaking - I'm sure everyone's been through that wave where you're like "This is a great idea. I'm gonna propose this as a conference talk." And then even better, you get accepted and you're elated; you're like "This is fabulous!" And then you're making some notes, and you're gonna drink some coffee, and you write in your book for a little bit, and then in about a month you come back and you say "Right. Time's ticking, I've gotta go and write that talk", and you sit down and write it, and you're like "Crap. This idea is not as good as I thought it was. This isn't gonna work out as well as I thought it was. I've made a terrible mistake."
+
+**Carmen Andoh:** \[20:01\] \[laughs\]
+
+**Dave Cheney:** This is a process that I go through every talk that I write. And then it's a process like pulling yourself out, saying "Okay--", you've promised to do this talk, on this subject, you probably missed the point to change a topic, so you've just gotta grin and bear it... From talking to other speakers, I think this is actually a very common thing, that happens to everybody.
+
+So to break the fourth wall a little bit, as I got into it, I ran out of similarities I could find between the zen of Python and a potential zen of Go. The talk -- I thought there was gonna be a lot of cross-over, but I actually found that I had to reinterpret what Peters wrote quite heavily in some points. Towards the end of the talk I said "Look, I've mined as much as I can from this." They're fundamentally different languages, although Go is roughly 30% Python anyway, if you look at its lineage... Some of those points worked really well with the material, others were a bit of a stretch, some just have no lineage at all.
+
+So moderation is a virtue. There were some in there that I felt had to be said. Like simplicity - we cannot talk about Go without saying simplicity. Simplicity is a core value, and if this is a list of things that Go programmers as a tribe/group hold dear, simplicity has gotta be in there. So no matter how handwavy and hard the final conclusion is, you've gotta say it's simple. And that's totally subjective.
+
+Moderation is a virtue... I thought about that, and I thought about my own experiences a lot. When I came to Go, I was very excited about the things in Go that were different from the languages that I was using before... Like concurrency, the fact that it was compiled - that was great, compared to the JVM languages... So those things that were unique about Go - I wanted to use all of them. Concurrency for everything, channels for everything, and everything is gonna be its own goroutine. "There's a reason these special features are in the language; there's a reason, so therefore I should use them" was the theory that I had. And I don't think this is uncommon.
+
+In another point I wrote that Go has reasonably performance so people will use it for the things that it's for that. Go has good concurrency support, so the programs that'll be written in it will use concurrency heavily. So we can't deny that the things which are done well in the language will be the things that people want to use. So there's a tension there between "Well, I want to use all." This kind of novice thing of "I want to use all the features. They're there for a reason. I should use them."
+
+And the realization which I think each of us is going through in our journey through learning Go and becoming proficient in it - realizing that, just as in life, overuse of any one of those things just leads to code that is either hard to read, not maintainable, or fundamentally just kind of too clever for what it's doing. That was certainly my experience when I went overboard with these goroutines. I broke the program into tiny little pieces, and then missed the bigger point that actually they were all waiting on each other, so it's fundamentally sequential, but I'd just made it harder to follow.
+
+One of the big pieces of code that I contributed to the Go project was working with Adam Langley on the SSH package. That initially had channels for everything. Every possible thing, every message -- SSH is kind of a stacked protocol, so at every level they communicated through channels... And again, very complicated, as well as checking the ownership of those channels, closing them, all that stuff... We fundamentally missed the point that this is a stacked protocol where nothing happens at the higher level until something's happened at the lower level... So those are just function calls.
+
+At the end of moderation is virtue. I was left with this unsatisfactory thing of saying "Do you feel that you've done the least you could? Have you used the features in Go in the least?"
+
+To point to maybe some guidance here - does everyone remember the blog post a couple years ago called "Choose boring technology"? This idea of innovation tokens... Has anyone heard of innovation tokens?
+
+**Mat Ryer:** Yes. But maybe some of our listeners won't have, so maybe we could -- for their benefit...
+
+**Johnny Boursiquot:** \[laughs\] Sure...!
+
+**Dave Cheney:** \[24:06\] We'll definitely put it in the show notes, but the idea of innovation tokens is that every time you start a new project, you bring with you all the knowledge that you have. So you're doing a new web application - well, you're gonna pretty much start from the old application, and maybe you'll change a few things. Maybe you'll be like "Oh, that database... That was a poor choice. For this I think we need a different kind of database." But then in taking on something new, like changing your database, you've just spent one of your innovation tokens. You've taken on a big risk, because that's a unknown. The main push of this "Choose boring technology" blog post was you don't have unlimited innovations tokens. You have two, maybe three. More like one.
+
+So if you're thinking about trying to apply moderations of virtue to your Go code, say "Well, this code sure does use a lot of concurrency." Maybe that's the one innovation token. Using a lot of goroutines to kind of break up this problem, then maybe we shouldn't think of something hypothetical; maybe we shouldn't have a lot of complicated data structures as well. It's really about recognizing "Hm, I've used a lot of a certain part of Go. The goal is not to us all of the parts at the same time." I've pushed my stack in on that one idea.
+
+**Mat Ryer:** Yeah, so it's sort of taking that minimalism a little bit further as well into the choices that you make about stuff. It's interesting, because that kind of thing is at risk of becoming one of those headlines, like "Don't use channels." That could easily fall into that same trap. And the thing is - I'll admit - I have over-indulged on channels in the past... And I actually end up tying myself in such knots. But that learning process I've found to be probably quite valuable, so I wouldn't necessarily want to deprive anyone of that. You're dead right - on a real team, in a real job, in a real situation, of course, there's limits. There's only so much you can do. And I like that idea of having a "Don't over-indulge on something." Having a healthy kind of use of these things. It's kind of a nice idea.
+
+**Dave Cheney:** I'll give you another example of where something could be overused. In my first talk at dotGo I wanted to talk about Rob Pike's functional option pattern, and how I applied it in the design of some code that I was working on. Two years later a number of people had said to me "Look, I really like that talk, but I'm not sure that to other people I can propose these ideas. It's too weird. Like, this idea of anonymous functions that return anonymous functions - it's too weird." That was kind of the impetus for another talk about "Hey, free functions are a thing. We should use them."
+
+But it's a realization that that is probably -- like, if everything in your codebase is functions that return functions that call functions, that's a really big innovation in tech. You're mixing that with heavy channel use and heavy goroutine use - that's a pretty pro level move. That would be one of those cases where you seem to have used all the features of the language. Isn't that gonna be a little bit hard to follow?
+
+**Carmen Andoh:** Yeah...
+
+**Mat Ryer:** Yeah, if you use all of the 25 Go keywords in the program, then yeah, I reckon you probably are. \[laughter\]
+
+**Break:** \[27:17\]
+
+**Carmen Andoh:** There was a recent post about the six stages of a Go programmer, and it was the evolution in a small, quippy GitHub gist... Which is basically this - you start out simple, but "No, now I wanna use all the features." Then you realize over time you go back to the very beginning, where you started. Then there was another cheeky Rob Pike homage without syntax highlighting. \[laughter\]
+
+But again, it's not just programming or even Go; it's human nature that we want to be able to try to be clever. There was a reason why the word "sophomoric" exists. "Oh, that's so sophomoric." It's when we're just starting to master a thing, but we're overdoing it, we're overcooking it... And you start to see time and time again, when you look at the masters, they have -- I think there's a reason why the simplicity matters, or simplicity is complicated, from a person who had to go through the hard knocks. Or some of the quotes that we hear time and again from -- like Brian Kernighan, "If you are not smart enough to debug the code when you write it, you won't be able to debug it later."
+
+So this is all part of the journey, and there's just not fast way through it. You have to let the learners abuse some of these features, and then come out the other end. I feel like it's just like Dante's Inferno - you've just gotta go through that hell; descend deeper and deeper and deeper, and then you'll come back out the other end. You'll be fine. \[laughter\]
+
+**Dave Cheney:** Can I ask a question of the group...? Have you joined a project - this could be joining a new company, or just moving between teams, or being assigned a new project - and you thought "Ugh, this code... Ugh. I need to rewrite this." \[laughter\] This isn't gonna look great on radio, but... Show of hands - who's done it? \[laughter\] The records show that I'm raising my hand, Johnny put his hand back down, Mat's is raised as well...
+
+**Carmen Andoh:** You know what - I internalized it as the impostor syndrome. Like "Maybe I'm not that good after all. Maybe it's me." I didn't have the courage to be able to sort of \[unintelligible 00:31:05.07\] my elbows and have the confidence that what I knew was probably a good way. Or equally a good way.
+
+**Mat Ryer:** Yeah. It definitely happened more for me in languages where there were more choices to be made. Having go fmt do all of that for us and take that decision away kind of removed--
+
+**Carmen Andoh:** Oh, okay, so formatting...
+
+**Mat Ryer:** Yeah, sometimes I think even just stylistically. Not just the style, but even if it's an OO programming language, the class hierarchies. There's lots of different ways to slice that, and sometimes you think "Oh, there's a kind of clear abstraction here", and there's a few weird abstractions that you can tell have got dust on them now... And yeah, of course, you just think "Now this could be written so much better." That is what we always think, isn't it?
+
+**Johnny Boursiquot:** The reason why I raised my hand sort of sheepishly - it was because I was like--
+
+**Mat Ryer:** A coward? \[laughter\] Oh, we'll cut that out, for sure.
+
+**Johnny Boursiquot:** \[32:02\] Like, I realize I've done this -- so you do this enough times in a project and you realize the ego that you're bringing to a project. It's almost like you're dismissing months or years' worth of people's hard work and effort... And requirements are never clear the first time around. So there's just so much you don't know about a project before, and then here you are, you just show up and you're like "Oh, this is nonsense. I'm just gonna rewrite the whole thing over the weekend." The ego about that, it's just like "Get over yourself." \[laughter\]
+
+**Dave Cheney:** This is fascinating. I think as we grow in our careers and put on some hat or some little lapel pin that says "Senior" or something like that - maybe we don't say it out loud anymore. Maybe we say it in our heads. We're a little more cognizant of "Okay, so the code is written in a time and a place." But can I ask you this question - so that's the project you've been assigned, you've got a group of humans you're working with... But what if it's a library that someone comes along -- you have a vendor library, and you're asked to integrate with that; I'm not gonna name any names, AWS...
+
+**Johnny Boursiquot:** \[laughs\]
+
+**Dave Cheney:** And you look at it and you're like, "Ugh, this doesn't look like any other code I've ever written in this language." And it's not just that it looks weird/strange, but you think "Well, if this breaks, I'm on the hook to fix it. I integrated this into my codebase, I'm taking responsibility for this." And wouldn't it, on balance, be safer if I just wrote it myself?
+
+**Carmen Andoh:** This is against the grain of being a software developer in 2020, when you can just npm-install. Like, you're actually reading through every line of dependency that you're pulling in... So bravo to that, at least!
+
+I try to explain to my kids what is it about programming, and I said "Well, if you like writing, programming is really writing, because what you're doing is you're taking other people's paragraphs and you're putting it into your book, from all over." Peter Bourgon - he once tweeted out "My Bash is like Hemingway. It's short, it's brief, it's clear." And I'm like "My Bash is like Shakespeare - kind of obtuse, and..."
+
+**Mat Ryer:** Like tragedy? \[laughter\]
+
+**Carmen Andoh:** "...I don't know if people understand it." Kind of like tragedy... \[laughter\]
+
+**Mat Ryer:** Lots of people die in it...?
+
+**Carmen Andoh:** That is what you're criticizing, Dave, I think, when you're seeing these libraries. They probably work perfectly well, but they're just not you, they're not your voice... I think there's a level of acceptance of pulling in -- barring bad, insecure code, and monkey patching... That's kind of what I tell my kids you have to live with when you program.
+
+**Dave Cheney:** Guilty as charged on all those counts, but the point I was trying to lead to was the one that I closed on in the Zen of Go, which is that maintainability counts. Peters wrote "Readability counts", and I've thought a lot about this over the last couple of years in the writing I've done and in the talks that I've done, and in my experiences, being tech leader across a bunch of projects - the honest truth is that I've written Go for about 5-6 different companies, and I've left all of those companies, which means somebody else has to maintain this code after me.
+
+I have felt anxious about that... Not as in "Oh, is someone gonna mess up my work?", but literally like "I'm leaving this debt for other people. And I've said it a couple of times now, if code can't be maintained, either because there isn't -- to give an example, the first Go code I wrote Atlassian was rewritten after I left, because I was the only Go programmer. It was reasonable that they rewrote it. But that is not a strategy for the growth of a language. Code that we leave behind as we move between jobs has to be maintainable and has to be sustainable.
+
+\[35:51\] To summarize the whole thing, these are suggestions for maintainable Go code that other people coming into the project are going to be able to understand and work with, and also Go code that you'd be happy to work on yourself. In a way, it is trying to find a way of explaining idiomatic Go without saying "Don't do this. Don't do that", because idiomatic Go can also be quite positively absolutist, in like "Always do this, always do that."
+
+The real take-away I want from this is that, for the success of our language and for the success of any project, it has to be maintainable, and that exists beyond any one individual.
+
+**Mat Ryer:** Yeah. It's important even for yourself. When we write code, we often think "How are we gonna feel about this when we've forgotten about it next month and we come back to look at this? How obvious is it gonna be then?" It's obvious now, because we're in the weeds, but once we've lost all that context and we look at this later when we eventually find there's something wrong with it, how do we feel about it?
+
+So when you think like that - imagine that you're gonna forget about this completely and come back in a month, if you think about writing it for that person, or for those people instead, I feel like that helps a little bit, rather than... When you're in the weeds and you're writing it, everything is very clear, because we've just spent hours trying to figure it out. So you've got so much extra information that \[unintelligible 00:37:14.13\] it's gonna go. Unless you can capture it somehow, you will lose it. So yeah, it is important to think about that, even for yourself.
+
+**Dave Cheney:** I think the top -- Google C++ Group have a style guide; there's a version on the web. And the first item of that is "Write for the reader", or something like that. Obviously, an idea that has infused in Go, which is that the act of writing is immaterial compared to the cost of reading tens or thousands of times. It's probably a Kernighan or a Strung and White quote, "Write for the reader."
+
+**Carmen Andoh:** Ha-ha! Kyle Simpson in The Economy of Keystrokes - another wonderful talk. He talks about the taxonomy of needs in the JavaScript world, which is "How do we rank the needs?" The needs of the reader trump the needs of the writer, trump the needs of the maintainer... And he goes on to explain in that talk - some of these idioms that we have, what are they for? Is terseness really better if you can't read it?
+
+I really love the ideas coming out of that talk there. They're very similar to your words, Dave.
+
+**Dave Cheney:** To give a shout-out to the author of Bitbar here - your ideas of "keep to the left", "return early", the patterns of guard clauses, and not putting successful code inside the indented block totally speaks for the reader.
+
+**Carmen Andoh:** Yeah. You've talked about glancability before, Mat, and I think this is a different flavor. It's a visual flavor of this very topic.
+
+**Mat Ryer:** Yeah, you're right. If we think about a database and we're gonna write some data and we wanna mix it with some of the things, we might choose to denormalize that... And that's a bit expensive at write time to do. But if there's so many reads happening on that, then of course, it makes sense. When it's that world, it feels okay, it feels normal. But when it's code, I don't know that we think like that, and maybe we should... Because like you say, David, it gets read if it's successful. It gets read a lot more times than it was written.
+
+**Carmen Andoh:** Yeah. The one thing I keep thinking about though as Go enters in two million people, and new people coming in all the time - most of the new people are coming in with previous ideas, previous assumptions, previous idioms, previous culture... And that is the thing, when you ask questions like you show up to a company and "Do you wanna rewrite everything?" or what do you do when you want to try to get to some sort of consensus about culture idioms or whatnot... So how does that clash when you either are getting a team to try Go from another language, or writing a new service, and it's traditionally been thought of as a Java shop or a Python shop...
+
+\[40:02\] In the same way that an American is going to maybe make some faux pas in another country, what are the faux pas that you get when you come to Go from other languages?
+
+**Dave Cheney:** That's the $64,000 question.
+
+**Carmen Andoh:** Yeah... \[laughter\]
+
+**Dave Cheney:** To bring it back to where I think idiomatic -- the idea of idiomatic Go not particularly a good teaching tool is because it's literally saying "We don't do that around here. You're not fashionable. It looks weird." To take some examples, as Go is taking off inside Canonical, almost to a person, the first comment was like "Where are my list comprehensions? What is this rubbish? I have to write a for loop like an animal?"
+
+**Carmen Andoh:** Yeah.
+
+**Dave Cheney:** But this is just the starter time. It's important to note that at that point where this new programmer has been put on a team, and like "I've been told I can't use Python on this. I have to use this other language; I'm already feeling a little bit anxious", and you're told "This is just the way we do things around here" - that's the wrong message to send.
+
+**Mat Ryer:** It sounds like you're in a Western, doesn't it?
+
+**Carmen Andoh:** Yeah...
+
+**Dave Cheney:** It is really a "Put up or shut up" kind of thing... That's not gonna lead to success for anybody. Either people are resentful that they don't have the things that they like from another language, or they're just confused, or they're embarrassed... There's a whole lot of social pressure involved there, of like "I've been put on this team. I don't wanna fail." So I just dislike that whole notion of the way that idiomatic Go is pointing to what other people have done and tapping the screen and say "Make it look like this." It might be a very didactic teaching tool, but it's not very humanistic.
+
+**Carmen Andoh:** Yeah.
+
+**Mat Ryer:** Another one of them on the list that I like is "Leave concurrency to the caller." This idea that as a package author, you really do want to deliver this amazing package... And especially if it's doing something and you know that it's gonna be used in a concurrent way. It's almost the fun bit, isn't it? It's the fun bit of the package, sometimes, and you have to maybe give that up for the sake of simplicity. What are the other benefits of leaving concurrency to the caller? Why is that important?
+
+**Dave Cheney:** To give credit where credit is due, this is another one of Peter Bourgon's fantastic observations. I'm probably forgetting other people have contributed to this debate, but you're absolutely right. If you're like "I'm writing a find function, or a search function. I can bring all the power of the computer to bear on this. I'm just gonna start a few goroutines, we're gonna do this in parallel..." And that's great. You're like "That's the fun bit. I get to write the fun bit. I don't have to do the API call, or write the documentation etc. I get to do the fun bit."
+
+But the difficulty with that - and I'll give a link for the show notes - is Peter talked about... I think he had a presentation at GoFest (the one that happens in San Francisco), called "The way I do things", which is about his experiences running Go. And one of them was this idea of "It all comes down to not losing track of that goroutine you started." A part of it is the very kind of dogmatic way I put it, of "If you start that goroutine, make sure you know where it's gonna finish." It's probably not one of the better ones on that list... But Peter's observation of "There needs to be some high-level concept of keeping track of that goroutine." If that goroutine is one that you started and holds resources, you need to know when it stopped, because you might need those resources back.
+
+How many times have you fought with tests that are flaky because they're hitting little residual parts of the previous test case, that haven't quite shut down yet? That's the fundamental problem of concurrency - it's not that we can't start and use it very effectively, it's that we have to know when it's all shut down and finished... Because if you want to do it again, you can't run over yourself with the previous instance.
+
+To bring it back to "Leaving concurrency to the caller", one of the best ways to do that is (unfortunately for the author) to give up the responsibility of "Who's gonna start that goroutine?" to the core, to somebody else. To provide them hooks... I mean, the easiest way to provide a hook for someone to run your function concurrently is just write a function. They can put it in a goroutine if they want.
+
+\[44:14\] Those are some of the ideas that Peter has, these more complex notions of when we have workers. You think of the workers that collectively represent providing a service, like an index, or a search, or a thing that accepts connections - they kind of all conceptually work together as a group; you need a way of managing them as a group. So Go Kit's run package has some good ideas in there.
+
+**Mat Ryer:** I was gonna say, one of the other things - if you do the concurrency inside your package... You know, in Go it's very easy to make something concurrent. It's not very easy to do the other way around, to take concurrent code and synchronize it, especially if it's running inside some package, and there's other channels and things that exist that you don't have any access to.
+
+So that's another good point that people should remember - it's easy for people to make their own concurrency, and then they know what's going on. They're responsible for it. It is just a shame.
+
+But I'll tell you one trick - and I've actually had this very recently... One trick is to write a little example that does it. You still get to do it, you get to prove it, you still get to see it working... You get the fun bit, if that is a fun bit for you... But it doesn't have to be a part of the package API. It could be an example that ships alongside the project.
+
+**Mat Ryer:** I think the testable examples that we have, that just fall out of go test, are one of Go's undiscovered gems. They should be an example for every single API function. There isn't, because, basically, the documentation... But the power that they give to someone reading GoDoc - there's literally a piece of sample code. That isn't just from a code comment; it's tested, it works with the example that you're seeing, right there on the screen. It was such a straightforward thing, such a simple thing, but incredibly powerful. I can't think of another language in the contemporaries that has that baked in... Not just in the documentation, but the integration through the testing package and through the way that they're highlighted in GoDoc is just a hidden gem.
+
+**Break:** \[46:15\]
+
+**Johnny Boursiquot:** I would like to touch on the bit where if you think something is slow, prove it with a benchmark. That's actually one of my favorite features in the language. You don't have to bring in third-party tooling or anything like that to test the efficiency of your code; it's built into the language. It's built into the test framework. So if you think something is slow, don't just eyeball it; actually write the benchmarks to actually prove it, and then do the work.
+
+\[48:00\] And to that point, I think there's this sort of -- early on, when I first started doing Go, there was this constant buzz around the community about "Oh, Go is fast, Go is fast, but you have to know how to use goroutines, you have to know how to program it just right." There was this constant murmur, a focus on pure performance. If this thing wasn't doing zero allocation, speedy... This was a time when we were coming out with that HTTP router every other week... Maybe this was a period where we as a community had to mature a little bit, get to know the language, and grow as developers...
+
+But this whole notion that "Oh, if it's not fast (by some arbitrary measure), then you need to go back to the drawing board" kind of thing. So yes, please, do touch on that... Because to me, Go already solves a vast majority of the problems I have from a speed standpoint. Very rarely do I have to actually try to optimize just one part or one area of my code. Maybe that's just me, but... I don't find myself constantly having to worry about optimizing my Go code.
+
+**Dave Cheney:** Yeah, so if you think it's slow, cards on the table, that was a little bit of my opinion doing this workshop on performance I come head-first into that Is-this-the-fastest-it-could-possibly-be-mentality, and unsurprisingly, I have had opinions about that.
+
+The two things which I think are important in this discussion - there is a continuum between readability, between maintainability and performance. Does everyone agree that to make the fastest possible code sometimes the readability and the maintainability of the code has to suffer? And if the fundamental goal of this work is maintainability is the most important theme, then when you make that decision to de-optimize the readability of the code for performance, optimize performance by making it more subtle, harder to maintain... And it doesn't necessarily need to be even like "That code looks like crap on the page", it can be "This has very subtle invariance", like that part having been niled, or "That can only be said once." Very, very fragile things, which hurt more than just the readability and the maintainability. People are like "That code is very sensitive."
+
+There are some examples in the standard library where a function is split into two, so that the simple case which usually is in-lineable, but the exceptional case is not - very subtle code, that if there wasn't a comment saying "Don't move that line", because literally \[unintelligible 00:50:30.06\] would never know.
+
+The other point I was speaking to is dogma. Defer is slow. Always use atomics, never use a mutex. Those absolutist statements which drive that kind of thinking. So how do we know when is the right time to invest in more subtle, harder to maintain code? Use the benchmarking tools that Go comes with. Don't optimize for -- I'll give you an example... In Go 1.14 a deferred mutex unlock is almost as fast as a non-deferred version. When I say "almost", on my machine it was like 0.7 of a nanosecond slower.
+
+So in the absolutist view you can say "Oh look, that's still slower", but the point is that's an uncontended mutex. That's literally one goroutine taking a lock and putting it back. Once you have that atomic actually contended, the cache line is moving between CPUs. Let me tell you, that 0.7 of a nanosecond is not going to count.
+
+**Mat Ryer:** You're right. It's a good point as well, because as soon as you have a few things in your program interacting, things get very unpredictable, very quickly... Way too quickly, in my opinion. But... Physics, I suppose. Yeah, that's another reason to prove it, isn't it? That's why you sort of say "Prove it", because actually, it might be wrong, and you might be really surprised by how your code actually runs in the wild, or even just at some kind of realistic scale.
+
+**Carmen Andoh:** \[51:59\] Yeah. And the one thing that I would say in terms of idiomatic Go is that you have to know when to also break the rules, and that's hard too, especially when you're a beginner. And when you talk about dogma - I mean, you can't have when each release is fixing previous problems that you were dogmatic about... Defer is the perfect example, with 1.14.
+
+**Johnny Boursiquot:** It's stressful as a beginner who's learning the language, whether because you have to fork, or out of desire to learn something new, learn the language... It's frustrating to always be sort of second-guessing yourself "Am I doing this right? Am I doing this the idiomatic way?" or is somebody more experienced than me going to look at this code and tell me the very thing that Dave was advocating before? ...basically, don't say "Hey, we don't do things like this around here." So it's stressful.
+
+To anybody out there who's learning Go, don't worry about whether it's idiomatic or not. Just learn, play. Hopefully, somebody will say "Hey, there's a better way to do it" and they'll do it the nicer way, rather than just say "Hey, this is not idiomatic" and move on. But even if you have those detractors, that's okay. There's no shortage of detractors, right? Just focus on actually just playing and learning. The idiomatic part - that's just gonna come...
+
+I didn't come to a threshold in my Go journey where I just walked through and I'm like "Oh yeah, now I'm writing idiomatic Go!" That's not gonna happen. There's no marker, there's no milestone there. Your code increasingly just becomes what we subjectively can call idiomatic, but there's no official threshold. So just play, just learn.
+
+**Carmen Andoh:** Yeah. My sister is a university Spanish professor, so she teaches language, just not a computer one... She's done it for over two decades now, and she can pretty much predict who's gonna learn Spanish and who's not by who's willing to get messy within the first week... Because learning is messy. So what you've just said, Johnny, is exactly that - just get messy. Do it wrong.
+
+**Mat Ryer:** Sacré bleu.
+
+**Carmen Andoh:** Yeah, there you go.
+
+**Johnny Boursiquot:** \[laughs\]
+
+**Dave Cheney:** One thing I think about with idiomatic Go - initially, it starts with well-meaning, like "Hey, you're new here. Let me show you around, let me show you the ropes, let me show you where the coffee machine is..." But at some point it can also become gatekeeping. We only do it this way. We're not going to change. And that's another important thing to realize, that in any healthy community -- spoken languages and the use of computer languages evolves. Think of memespeak; 20 years ago no one said 'lol'. That's a word that we just invented, and it changes the discourse.
+
+So to put the final nail in the idiomatic Go coffin - the last thing I want for it to be is gatekeeping. "We've always done it like this." It should be totally right to question "Why do we always do it like this?" "Because we've always done it like that" can never be an answer in a sustainable community.
+
+**Mat Ryer:** Well said. I think that's a really nice summarization of the show, actually. Okay, looking at my watch, I think it's time for our regular slot, Unpopular Opinion.
+
+**Johnny Boursiquot:** Wow, you actually butchered that one up... \[laughs\]
+
+**Jingle:** \[55:32\]
+
+**Mat Ryer:** So does anyone have an unpopular opinion? It can't be about my timing this time... Dave, have you got one, please?
+
+**Dave Cheney:** Do I have an unpopular opinion? Well, I've given you 11. \[laughter\]
+
+**Johnny Boursiquot:** Well, according to Hacker News, you have. \[laughter\] What a fun bunch...
+
+**Carmen Andoh:** \[56:10\] It doesn't have to be about Go... It could be about anything. Yeah, about anything. Here's your moment.
+
+**Dave Cheney:** So... Think about your day. Think about the things that you've done in the day, and think about the things that have interrupted you during that day. Think about the things that have interrupted your focus. A lot of them are probably unavoidable, like "My dog just came downstairs", or your family, or there might be someone at the door... But the thing that occurs to me is that many of these interruptions are actually self-inflicted. Mobile phone notifications, pop-ups, chat messages, the inevitable clamor of Slack... All of these things interrupt your focus.
+
+The thing that it seems to me is that as knowledge workers, we have three skills. The first one is our experience; the experience that we bring to our task. The second one is your ability to communicate. And the third one is your ability to focus. These are in order of difficulty and priority. Experience - we can always just google things. There are people you can talk to, you can get experience. Your ability to communicate kind of gates that; how well you can communicate, how well can you ask questions, how well you can you formulate your thoughts, how well can you convince other people to do things. But I think the most important one - and the one that I struggle with the most - is focus.
+
+Think of all the things throughout the day, "Oh, I'm just gonna pick up my phone and have a look at Twitter." I've deliberately let myself be interrupted. And as a group of knowledge workers - and maybe this applies to everyone - your focus is your super-power. If you can't focus on anything, you can't get anything done.
+
+**Mat Ryer:** That's a great one, actually. I don't think that's gonna be unpopular. I thought you were gonna go on to say you're now selling Dave Cheney straight jackets for developers. \[laughter\]
+
+**Dave Cheney:** No, I think the unpopular thing is that there's a lot of things in this world which compete for our attention. No matter how much money you have in the world, you can't invent the invent the 25th hour in the day.
+
+**Carmen Andoh:** Yeah...
+
+**Dave Cheney:** So your attention is a commodity which is constantly being competed with -- and there are a lot of things in the modern workplace actually quite negative towards that. Being always available via notifications in Slack, and chat, and things like that... Not to say that chat is not fine, or that group chat is not fine, but can you honestly say -- to the radio audience here, can you honestly say that if you turned off your chat notifications for a day you wouldn't either a) feel guilty that you're ignoring your co-workers, or b) feel a little bit worried that your boss thought you were slacking off?
+
+**Carmen Andoh:** Yup.
+
+**Dave Cheney:** Yet, how are we as programmers supposed to work if we cannot create any space to focus in?
+
+**Mat Ryer:** Yeah, this really speaks volumes to me, Dave... Actually, I'm working on a new project that kind of is founded on those exact principles. I think you're dead right. If there was a room with 100 developers working in there, you wouldn't walk in and just shout "HERE!" really loud, wouldn't you? \[laughter\]
+
+**Carmen Andoh:** Right. \[unintelligible 00:59:11.24\]
+
+**Mat Ryer:** And yet, that's what is happening on Slack. And it is a cultural thing too, because it's instant messaging. You expect that instant reply.
+
+**Carmen Andoh:** Yeah. If you don't reply right away, "Why didn't you?"
+
+**Mat Ryer:** Yeah, and you do feel like you're letting people down... So then you do react immediately to get it out of the way, so you can get back to focusing, and then you've just sort of rewarded that behavior and reinforced it a little bit.
+
+**Carmen Andoh:** Yeah.
+
+**Dave Cheney:** I'll ask anybody, in a day that they're at work, to just be cognizant of if you Slack, or Discord, or any chat system; it's not against any one of them... But do you reflectively Alt+Tab and just check the window? Just check to see if there's anything you should respond to.
+
+**Carmen Andoh:** Oh... Guilty as charged.
+
+**Dave Cheney:** \[01:00:01.15\] Because you're actively destroying your ability to focus. And if there's one thing that's gonna determine from two individuals who is gonna be more successful fundamentally, it's their ability to focus... Because that will preclude them from being effective enough as a communicator to reach out and get assistance from everybody, that will preclude them from being able to do any serious deep thinking, because they're either being interrupted by their circumstance or just have trained themselves to interrupt themselves. So this is the challenge that I've set for myself late last year. Anytime that I will reflexively think to pick up my phone to check my phone, I'll pick up a book. I always have a book with me, and I'll try and read a page.
+
+I love reading Twitter when I get up, I love catching up with the world, just like I read the news over breakfast, but I realize just how easy it is for myself to get distracted by it... It's just an Alt+Tab away, or it's just literally sitting next to me, because we have to have these damn things next to us for our 2FA tokens. I can't leave this downstairs, or else I can't log in and do work.
+
+**Carmen Andoh:** Or if you're an on-call rotation... Yeah.
+
+**Dave Cheney:** So just think about how the circumstances have conspired to actively keep you in the state of always being interruptible, and how that fights against your ability to focus. I think Nate Finch had a tweet a long time ago saying "Where are we supposed to focus in open-plan offices?" There's my unpopular opinion.
+
+**Carmen Andoh:** That's another Dantes Inferno's hell - open-plan offices are the real Dantes Inferno. \[laughter\]
+
+**Johnny Boursiquot:** So all this interruption - we should note that you are excused from everything that we've just said about interruptions when you join the \#GoTimeFM Slack channel to participate and willingly choose to be interrupted with our fun discussions. So please, do keep doing that.
+
+**Carmen Andoh:** \[laughs\] Well said.
+
+**Mat Ryer:** Yes. But I fear our time has come to an end today... Dave, thank you so much. Anyone interested, check out the show notes--
+
+**Dave Cheney:** Thank you for having me.
+
+**Mat Ryer:** No, absolutely, it was a pleasure, as always.
+
+**Johnny Boursiquot:** Thank you. Indeed.
+
+**Mat Ryer:** Someone said in the Slack channel \[unintelligible 01:02:15.26\]
+
+**Johnny Boursiquot:** Take it up with them. \[laughter\]
+
+**Dave Cheney:** I barely know myself...
+
+**Mat Ryer:** \[01:02:28.22\] But no, it's been great, and I think that is a lovely lesson as well that we should take away from this, which is if you do find yourself about to say something's not idiomatic, think instead about why we think that's idiomatic. What are the positives, what are the pros for that? Talk about those ideas instead, and be open to people challenging that. I love that idea. I think we should all actually spend a bit of time thinking about that.
+
+**Dave Cheney:** Let me leave everybody with a challenge. This was the challenge that I set for myself - I became tech leader of this project that I've been working on at VMware for a while... As it's happened in many jobs as I've kind of been hired as the only Go programmer that they can find, I'm kind of seen as the expert. But I set myself -- as tech lead, you're expected to really love code, give guidance, things like that. But the rule I set for myself is that I was never allowed to use this "Just do it because..." "Do it this way because..." "Do it this way because I know better. Trust the experts, trust your father. Trust your betters. You'll understand."
+
+I said to myself I was never allowed to use that excuse. So in a way, many of the things that I've been writing and working and thinking on are I can't use "Just do it this way, because trust me" as an excuse. I have to be able to justify everything that I'm saying, like "Thank you for this code review. I have some concerns about this API because of this second parameter. It doesn't seem to be used all the time, and that means it's confusing for a newcomer." Those are statements you have to make when you don't let yourself say "Just do it this way. Trust me. Do it this way. You're wrong."
+
+So the challenge I'd leave to everybody is when you're having a discourse with other people, when you're giving feedback on code review, if you find yourself saying -- especially if you feel frustrated and you're like "Oh, we've had this discussion hundreds of times. Why can't you just do it the way that I said?", maybe it's because you haven't actually justified it. You're leaning on your own history, rather than having the open discussion about it. Rather than saying "Do" or "Don't", say "I think it would be better if..." and those kinds of statements.
+
+So taking Scott Meyers' words to heart, never have those absolutes. Say things like "In general, prefer...", or when you do make an exception, you talk about it." That is the challenge I leave to everybody. When you're talking about code with your co-workers, don't just say "Do it because...", say "I think that this would be better, and these are my reasons." Because it is fundamentally your opinion at the end of the day. But instead of making it absolutist, you say "Here are my suggestions. Do you agree?"
+
+**Mat Ryer:** Well, challenge accepted, and I hope all the listeners do, too. Dave, thank you so much. Please come back. And for everyone else, we'll see you next time.
+
+**Break:** \[01:05:32.09\]
+
+**Mat Ryer:** Hello! \*coughs\* That's a bad start. Sometimes, kids, things don't go to plan... And that was one of those. Just a little cough at the exact wrong time. But don't worry, through the power of editing, this will definitely make it into the end thing.
+
+**Johnny Boursiquot:** Yeah, it's definitely staying in, yeah.
+
+**Mat Ryer:** Yeah, isn't it? This is gonna be like a DVD extra, for some reason... \[laughter\] We don't have many moments to use, but little things like this... Someone coughs - that's gold; get that on the DVD extra. Okay, I'll do it again...
+
+**Carmen Andoh:** "Dad, what's a DVD extra?" \[laughs\]
+
+**Johnny Boursiquot:** "What's a DVD?" \[laughter\]
+
+**Carmen Andoh:** "Dad, what's a DVD?" \[laughter\]
+
+**Mat Ryer:** Oh yeah, I forgot the world's moved on...

The engineer who changed the game_transcript.txt

@@ -0,0 +1,443 @@
+[0.00 --> 7.06]  Hey friends, we're helping out Red Hat to promote season six of Command Line Heroes.
+[7.46 --> 12.60]  If this is the first time you're hearing about this podcast, Command Line Heroes is a podcast
+[12.60 --> 17.42]  that tells the epic true tales of developers, programmers, hackers, geeks, and open source
+[17.42 --> 20.36]  rebels, people like you who are revolutionized in the technology landscape.
+[20.62 --> 23.54]  And today we're sharing a full length episode for you to check out.
+[23.62 --> 26.22]  In fact, I helped to handpick this episode for you.
+[26.22 --> 31.28]  It tells the story of Jerry Lawson, the engineer who changed gaming forever by inventing the
+[31.28 --> 33.14]  cartridge-based system for gaming consoles.
+[33.28 --> 36.20]  Before Jerry, a console could only play one game.
+[36.60 --> 38.42]  Jerry quite literally changed the game forever.
+[38.90 --> 40.70]  Search for Command Line Heroes, email us in the podcasts.
+[41.22 --> 43.88]  We've also included a link in the show notes that makes it easier for you.
+[44.16 --> 46.86]  Many thanks to our friends at Command Line Heroes for their support.
+[47.36 --> 48.08]  All right, here we go.
+[51.56 --> 54.38]  Jerry Lawson, I said, who is Jerry Lawson?
+[56.22 --> 63.40]  It's 2011, and Joseph Salter, CEO of Entertainment Arts Research, receives a call.
+[63.92 --> 66.56]  It's from journalist John William Templeton.
+[67.02 --> 70.16]  And his call has left Salter shaken.
+[70.36 --> 74.70]  He said, Jerry Lawson was the first game developer, cartridge game.
+[74.76 --> 76.02]  I said, what are you talking about, man?
+[76.02 --> 79.34]  I said, I just finished a book, Introduction to Game Design and Development.
+[79.34 --> 84.74]  And all of a sudden, you tell me about this guy that I didn't put in the book, man.
+[85.90 --> 91.20]  Salter was about to discover that the history of video games, the one he had in his head,
+[91.52 --> 93.00]  needed to be rewritten.
+[93.58 --> 98.18]  And when he realized why, Joseph Salter broke down in tears.
+[98.18 --> 105.50]  So, why did learning about Jerry Lawson mean so much?
+[105.86 --> 107.30]  We're about to find out.
+[108.40 --> 113.50]  I'm Saranya Bark, and this is Command Line Heroes, an original podcast from Red Hat.
+[113.50 --> 122.34]  In Season 6 of Command Line Heroes, we're getting personal, learning about the lives, the motivations,
+[122.76 --> 129.40]  and the extraordinary innovations that defined eight powerful but not quite famous inventors.
+[129.66 --> 137.06]  All of them integral to the advancement of tech in the 20th century, and all of them less known than they should be.
+[137.06 --> 146.22]  We're going to discover the man who brought PCs to IBM, the woman who helped put us all literally on the map,
+[146.38 --> 150.46]  and the man who made Silicon Valley more accessible for everyone.
+[151.10 --> 157.96]  It's a whole new history celebrating Command Line Heroes who deserve a lot more credit.
+[159.18 --> 161.56]  This time, it's Jerry Lawson's turn.
+[167.06 --> 170.90]  When Jerry Lawson came on the scene back in the early 70s,
+[171.10 --> 177.42]  the few video game consoles available had a one-to-one ratio between machine and game.
+[178.36 --> 183.06]  You'd lug home a console, plug it in, and pretty much that was what you were playing.
+[183.44 --> 186.68]  That one game they'd burned onto the machine's memory.
+[187.20 --> 192.00]  The idea of snapping in the new Zelda, and then trading that out for some Mario Kart.
+[192.00 --> 197.30]  Really, the whole idea of a software-centered gaming industry hadn't happened yet.
+[197.84 --> 198.82]  But it was coming.
+[199.30 --> 204.50]  And that revolution started in a little beer hall down in Sunnyvale, California.
+[208.04 --> 212.48]  Jerry Lawson walked down El Camino and into Andy Capp's Tavern.
+[212.94 --> 214.80]  A crowd was playing an arcade game.
+[215.22 --> 217.24]  One of the first arcade games anywhere.
+[217.84 --> 218.78]  It was Pong.
+[218.78 --> 221.96]  Lawson was blown away.
+[222.60 --> 225.28]  For the first people to see video games come to life,
+[225.56 --> 228.66]  that bouncing ball might as well have been Grand Theft Auto.
+[230.10 --> 234.96]  Pretty soon, Lawson got it into his head that he'd like to build a game too.
+[235.50 --> 237.42]  But here's the thing you have to remember.
+[238.08 --> 240.20]  Just a year earlier, in 1971,
+[241.02 --> 244.10]  Intel had released its 4004 microprocessor,
+[244.10 --> 246.86]  the first commercially available microprocessor.
+[246.86 --> 252.28]  And Lawson was convinced that he could use a microprocessor in the game he wanted to build.
+[252.78 --> 253.70]  Only problem,
+[254.70 --> 256.78]  that had never been done before.
+[257.48 --> 262.06]  The machines being built back then all used dedicated logic circuitry.
+[262.66 --> 264.68]  They were what you'd call a state machine.
+[265.42 --> 268.72]  Code and computers weren't really what drove gaming back then.
+[269.20 --> 273.44]  So Jerry Lawson's peers told him that just wouldn't work.
+[273.68 --> 274.42]  It couldn't be done.
+[274.42 --> 279.44]  Luckily, Lawson was used to getting done what other people called impossible.
+[279.98 --> 281.74]  He was used to being underestimated.
+[285.74 --> 289.34]  In fact, being underestimated ran in the family.
+[290.12 --> 292.36]  Lawson's grandpa dreamed of being a physicist.
+[292.90 --> 296.30]  But that was decades before the Civil Rights Act of 1964.
+[296.80 --> 299.88]  And being black, he could only get work at the post office.
+[299.88 --> 305.16]  Raised in Queens in the 1950s, options were still limited for Lawson.
+[305.60 --> 309.38]  But his mom was determined that her son would have a different fate.
+[310.14 --> 312.50]  She watched young Jerry build ham radios,
+[312.90 --> 314.94]  watched him make his own walkie-talkies,
+[315.34 --> 316.58]  watched him repair TVs.
+[317.50 --> 323.08]  So she decided to drive her son each day into a white neighborhood where they had better schools.
+[323.08 --> 326.32]  She even became president of its PTA.
+[327.44 --> 330.86]  I wanted to know how that kind of history shapes a family,
+[331.36 --> 332.26]  shapes their ambitions.
+[332.78 --> 335.96]  So we tracked down the perfect person to ask.
+[336.54 --> 337.96]  Hi, my name is Anderson Lawson.
+[338.08 --> 339.80]  I'm the son of Gerald Anderson Lawson.
+[340.46 --> 345.70]  So I imagine when it comes to the career aspirations of your great-grandfather
+[345.70 --> 347.52]  and those people who came before you,
+[347.58 --> 351.76]  I imagine racism probably had a huge effect on their goals, on their aspirations.
+[352.36 --> 357.44]  I know that he wanted to be a, I want to say, a physics teacher.
+[358.20 --> 365.90]  And at the time, he wasn't seen as fit for that role and actually became a mailman.
+[366.66 --> 368.90]  Now, that's a stretch, right?
+[369.52 --> 373.22]  But I think, you know, we've come a long way, but we've still got a long way to go.
+[373.22 --> 378.32]  And how did that story of your great-grandfather influence your father's own pursuits?
+[379.04 --> 385.24]  You know, I really don't know because my father was so, his attitude, he was so freaking stubborn.
+[386.06 --> 391.06]  He was, for lack of a better term, he was the type of person where if, you know,
+[391.10 --> 395.30]  no matter what's going on, if somebody just said, hey, you're not able to do that,
+[395.48 --> 396.24]  he would have found a way.
+[397.06 --> 399.16]  He was different in that way.
+[399.16 --> 403.10]  And it was more to spite the person that said that.
+[403.22 --> 404.18]  Than anything else.
+[408.34 --> 414.00]  That determination to build his own reality, to literally craft it with his own hands,
+[414.18 --> 416.60]  was something lost in past onto his children.
+[416.92 --> 424.74]  When I was about, actually I was 10, my cousin was with us and we were playing games.
+[424.74 --> 431.18]  And I guess my father had enough of seeing us play the video game for the day.
+[431.50 --> 434.32]  So he goes into the garage and he comes back out.
+[434.52 --> 435.40]  He shuts off the game.
+[435.48 --> 436.52]  He doesn't let us play anymore.
+[436.52 --> 440.28]  And he takes a book and throws it onto the counter.
+[441.00 --> 445.46]  And it is 101 basic video games.
+[445.86 --> 450.70]  And he's like, if you're going to play games, you better learn how to make them.
+[451.48 --> 452.54]  And walked away.
+[452.54 --> 458.30]  Set up a portable IBM computer in the room.
+[459.30 --> 461.34]  And said, figure it out.
+[461.82 --> 462.44]  Turned it on.
+[462.92 --> 463.86]  Now, we're kids.
+[464.50 --> 471.28]  We had to figure out how to get in, log in, find the basic compiler, and start making games.
+[471.28 --> 472.52]  But we did it.
+[473.80 --> 479.04]  In the 1970s, Lawson got himself a job at Fairchild Semiconductor,
+[479.32 --> 482.64]  working as a field applications engineer out in California.
+[483.16 --> 486.72]  That basically meant he was a traveling troubleshooter for clients,
+[487.22 --> 490.52]  driving all over in an RV full of demo products.
+[491.62 --> 493.04]  Lawson had other plans, though.
+[493.40 --> 498.36]  For starters, he became the only Black member of the famous Homebrew Computer Club,
+[498.36 --> 502.86]  where Steve Wozniak and Steve Jobs used to show off their latest ideas.
+[503.54 --> 505.52]  More important than the Steves, though,
+[505.74 --> 510.54]  was the fact that Lawson began assembling the tools he needed to build something of his own.
+[511.06 --> 516.88]  I guess let me give you kind of a breakdown of what a Saturday afternoon would be like with Jerry Lawson.
+[517.40 --> 524.80]  I could spend three or four hours with him going to different electronic stores and even flea markets,
+[524.80 --> 529.42]  looking for parts and things of that nature or talking to people in the industry.
+[529.64 --> 532.54]  Like, his life was electronics.
+[532.94 --> 533.78]  That's all he did.
+[534.38 --> 535.52]  Computers, electronics.
+[535.90 --> 536.96]  Computers, electronics.
+[537.46 --> 538.46]  All day, every day.
+[539.92 --> 545.26]  I mean, the man had a PDP-11 inside of the garage.
+[545.26 --> 551.20]  So if you know anything about computers, Digital Equipment Corporation put out that computer,
+[551.68 --> 558.06]  and it was about as big as a refrigerator today, sitting there in our garage, among other things.
+[558.40 --> 559.50]  It was really different.
+[560.28 --> 564.76]  Side note, when the team at DEC found out Lawson had a PDP-11,
+[565.16 --> 569.56]  they paid for $10,000 in upgrades in exchange for access.
+[570.14 --> 572.94]  And that garage became Lawson's place of zen.
+[572.94 --> 579.92]  Sitting in the middle of it all, surrounded by computers, manuals, capacitors, tape spitting out of a printer,
+[580.28 --> 583.32]  and his young kids marveling at the magic machines.
+[583.88 --> 586.58]  When Jerry Lawson came home after seeing Pong,
+[586.94 --> 591.54]  he walked into that garage and began willing into being a game of his own.
+[591.92 --> 596.56]  A coin-operated arcade cabinet he called Demolition Derby.
+[597.46 --> 598.62]  Back to Anderson.
+[598.62 --> 605.96]  So he was doing all this building and creating all these, you know, games and electronics and all that on the side.
+[606.06 --> 608.08]  He was still working at Fairchild full-time.
+[608.48 --> 614.14]  So did his bosses find out, or how did they find out about his garage project?
+[614.56 --> 618.60]  Yeah, so someone got wind of that and got around.
+[618.98 --> 622.44]  Hey, you know, Lawson has a side project.
+[622.62 --> 624.02]  He's been working on this game.
+[624.02 --> 633.72]  And so he got called into an office and was told about this kind of skunkworks project that they had going on.
+[634.58 --> 636.30]  A gaming project.
+[637.16 --> 640.06]  But Fairchild getting into gaming was a weird move.
+[640.76 --> 644.54]  Fairchild was this dry, industrial semiconductor company.
+[645.30 --> 649.46]  Journalist Jenny List from Hackaday.com explains why it made sense.
+[649.46 --> 655.92]  They had a microprocessor system and they were looking for markets for it.
+[656.10 --> 657.90]  They were looking for things they could put it into.
+[658.26 --> 665.04]  They saw consumer microprocessor products as an up-and-coming thing.
+[665.70 --> 670.72]  And they weren't a gaming company, but there weren't any gaming companies because there wasn't really a games industry at the time.
+[670.72 --> 679.40]  And for a semiconductor company to suddenly realize, here's a consumer business coming up and we've got to be part of it, it's actually quite an adventurous thing.
+[679.58 --> 687.58]  And so it must have been an absolute godsend for them to have, as I say, a very talented engineer who actually had some experience in it, when probably few others did.
+[687.58 --> 697.60]  By this point, that coin-operated arcade game Lawson built in his garage, Demolition Derby, had been tested out at the local pizza parlor.
+[698.16 --> 703.68]  It was living proof that a game could run on Fairchild's F8 microprocessor.
+[704.32 --> 710.06]  The Fairchild execs were intrigued and invited him to make a home game system for the world.
+[710.34 --> 714.44]  A chance to finally prove what he was capable of building.
+[714.44 --> 717.52]  Oh, well, he worked night and day with that.
+[719.16 --> 727.24]  Lawson left his RV days behind and became Director of Engineering and Marketing for Fairchild's brand new video game division.
+[727.94 --> 732.70]  But what Jerry Lawson was about to build for Fairchild wasn't just a new game console.
+[733.26 --> 736.70]  What Lawson was about to build for them was a game changer.
+[736.70 --> 746.88]  In the early 1970s, a home video game console didn't offer you much.
+[747.88 --> 754.78]  Some systems came with a vinyl overlay you'd attach to your TV screen so that the same game would look different.
+[755.32 --> 761.08]  Then there was the Magnavox Odyssey in 1972, which sort of used cartridges.
+[761.08 --> 763.36]  But they had no memory of their own.
+[763.86 --> 766.74]  They just reconfigured the console when you plugged them in.
+[767.54 --> 769.80]  Those predecessors were super limited.
+[770.48 --> 772.06]  The gameplay was limited.
+[772.60 --> 780.40]  But two employees over at Alpex Computer Corporation, Wallace Kirshner and Lawrence Haskell, had cracked something new.
+[781.04 --> 782.70]  They saw what Lawson saw.
+[783.20 --> 787.72]  The potential of microprocessors to revolutionize the gaming industry.
+[787.72 --> 790.86]  Tech historian Benj Edwards explains.
+[791.26 --> 794.74]  The guys at Alpex said, hey, we could do this with more intelligence.
+[794.88 --> 796.86]  We can make it a software-based system.
+[797.10 --> 800.44]  So they developed a prototype with very primitive graphics.
+[800.64 --> 802.04]  I think it was black and white at first.
+[802.52 --> 805.50]  And a very low resolution that would hook to a TV set.
+[805.72 --> 807.58]  And they could program it in software.
+[807.74 --> 810.24]  They could make a simple game and play it on the screen.
+[810.38 --> 813.72]  And then switch it out by loading a different program.
+[814.52 --> 816.92]  But things were still largely conceptual.
+[816.92 --> 823.88]  Back then, removable storage devices meant magnetic hard drive disks or even paper tape.
+[824.20 --> 828.84]  It would take several innovations to turn that idea into a consumer product.
+[828.84 --> 837.44]  These guys thought, let's just take the EEPROM chip that we're using for the development kit of the Intel processor they're using.
+[837.84 --> 839.66]  And let's just put that in a box.
+[840.14 --> 843.10]  So they put the chip on a board, a circuit board.
+[843.42 --> 844.50]  They put the board in a box.
+[844.62 --> 850.58]  And the box had a connector on it that had a more durable connector, like a 25-pin D connector.
+[850.58 --> 856.74]  So they could plug that module into their game system and change games out.
+[856.92 --> 860.06]  They could play that game and unplug that box, plug in another one.
+[860.14 --> 862.20]  And that was the first game cartridge prototype.
+[862.20 --> 868.20]  Kirchner and Haskell had the seed of something great.
+[868.20 --> 872.86]  A computer-controlled game console with removable software cartridges.
+[873.38 --> 876.06]  But it was a long way from becoming a durable product.
+[876.62 --> 880.28]  That would be Lawson's job once Fairchild had licensed their work.
+[880.28 --> 885.40]  The path from prototype to store shelf would be long and complicated.
+[886.24 --> 892.96]  For starters, Lawson translated the hardware circuitry from the 8080 to Fairchild's own F8 chip.
+[893.46 --> 896.96]  But meanwhile, fundamental questions were cropping up everywhere.
+[897.58 --> 901.70]  How do you fit all the circuitry in a box that can sit on top of your TV set?
+[902.34 --> 904.96]  And what does that box look like?
+[905.44 --> 906.88]  Everything was a first.
+[906.88 --> 912.60]  Kirchner and Haskell, for example, had players using a keyboard to play their games.
+[913.20 --> 915.74]  This was a computer-powered experience, after all.
+[916.06 --> 917.86]  Why not use a computer's normal interface?
+[918.78 --> 920.78]  But Lawson was set on building a hybrid,
+[921.20 --> 927.50]  where microprocessors lived alongside the tactile experience of a joystick you'd find at the arcade.
+[927.84 --> 930.88]  A joystick with eight axes of movement.
+[931.26 --> 934.74]  So you could move your paddle around, up, down, left, right.
+[934.74 --> 940.56]  You could twist the top of the joystick to rotate the paddle's angle.
+[940.76 --> 944.72]  And then you could move your goalie piece like a different player
+[944.72 --> 948.50]  by pulling up and pushing down on the plunger of the joystick.
+[949.14 --> 952.64]  And that was Jerry Lawson's creation, that idea.
+[953.28 --> 957.86]  They also had to design a contact system between the cartridge's pins and the console.
+[957.86 --> 964.72]  An edge connector that wouldn't get wrecked after being rammed into the console thousands of times by some clumsy kid.
+[965.48 --> 969.34]  And inventing something that could survive rough use was only half the battle.
+[969.86 --> 973.16]  They also had to make it survive the regulators at the FCC.
+[974.38 --> 977.66]  Lawson became famous for wrangling these technical issues.
+[978.26 --> 981.12]  They encased the console's motherboard in aluminum.
+[981.80 --> 986.56]  They stopped leaking radiation by covering the cartridge adapter with a metal chute.
+[986.56 --> 991.16]  Al Alcorn, the creator of Pong, walked us through the red tape.
+[992.70 --> 995.02]  The Federal Communications Commission said,
+[995.16 --> 1003.98]  Wait a second, FCC had to make regulations to limit the radiation of these new devices coming out
+[1003.98 --> 1006.60]  so as not to interfere with other devices.
+[1006.60 --> 1013.02]  And so they made these regulations based on the only device that was out there at the time,
+[1013.10 --> 1014.84]  which was the Magnavox Odyssey.
+[1015.28 --> 1019.32]  And because it was an analog device, not digital,
+[1020.00 --> 1023.34]  it had very little radiation inherently coming out of it.
+[1023.74 --> 1028.34]  So the regulations were very strict and without getting too technical,
+[1029.00 --> 1034.56]  frequencies like 10 megahertz, the radiation limit was absurd, absurdly low.
+[1034.56 --> 1036.98]  The regs were overly tight.
+[1037.18 --> 1037.86]  It's very, very hard.
+[1038.20 --> 1044.78]  To get approval, you had to make a prototype device and all the connectors, cables, everything,
+[1045.16 --> 1050.02]  and physically bring it to the FCC offices in Maryland.
+[1050.70 --> 1056.82]  And they had this big building there and sit in the lobby and give them the product and hope for the best.
+[1056.82 --> 1067.18]  To move things forward, Lawson pretty much camped out at the FCC's offices for days pushing for long-delayed approvals.
+[1067.82 --> 1071.76]  You might have noticed those were hardware problems Lawson was tackling.
+[1072.30 --> 1076.86]  That's because gaming was still a hardware-first industry back then.
+[1077.44 --> 1083.54]  The software-focused industry we know today was, in fact, the one Lawson was about to create.
+[1083.54 --> 1088.76]  At last, it was ready for market.
+[1089.26 --> 1091.70]  They called their machine the Channel F.
+[1092.84 --> 1095.08]  Channel F, the one with all the fun.
+[1095.36 --> 1098.58]  The Fairchild Video Entertainment System, just $169.95.
+[1098.98 --> 1101.18]  Video card cartridges, $19.95 each.
+[1102.10 --> 1107.70]  Channel F hit the shelves in late 1976, just in time for the Christmas rush.
+[1108.30 --> 1112.86]  Most people didn't notice it was the world's first ROM programmable video game, though.
+[1112.86 --> 1115.38]  Its genius was its simplicity.
+[1116.26 --> 1120.90]  You could use this fantastic software innovation without understanding it at all.
+[1121.72 --> 1127.48]  These were programmable cartridges, each bearing a game of their own that even a child could manipulate.
+[1128.26 --> 1133.68]  There was hockey, space war, blackjack, a whole catalog of programmable fun.
+[1134.94 --> 1139.26]  Turned out, though, their innovation was a little too much for some.
+[1139.26 --> 1144.42]  Christmas morning, Lawson was at the office and got stuck answering the helpline.
+[1144.82 --> 1149.58]  He found himself a troubleshooter once more, this time for the general public.
+[1149.92 --> 1151.74]  And they had plenty of questions.
+[1152.58 --> 1156.36]  One customer had taken the console apart looking for batteries.
+[1157.06 --> 1162.30]  Some kids had put grandpa's dentures in the machine, thinking they'd show up on the television.
+[1162.30 --> 1173.52]  We forget today how strange a cartridge system would have been when it first arrived.
+[1173.92 --> 1180.34]  But for the few engineers tackling the problem of reimagining the industry, his accomplishment was clear.
+[1181.22 --> 1185.10]  Alcorn told us that Lawson had done something truly remarkable.
+[1185.10 --> 1195.70]  He really was a pioneer in coming up with a cartridge concept and built the first microprocessor-based video game system.
+[1196.70 --> 1200.92]  To build the first video game, that's engineering.
+[1201.44 --> 1202.94]  And that's what Jerry did.
+[1203.32 --> 1211.68]  He certainly opened the door to a lot of what video game became by virtue of simply doing it, of making it work.
+[1212.06 --> 1214.24]  And making it a credible system.
+[1214.24 --> 1221.66]  And we were surprised to discover Lawson even had a hand in creating Alcorn's own breakthrough, Pong.
+[1222.46 --> 1223.34]  Oh yeah, yeah, yeah.
+[1223.34 --> 1228.04]  I mean, he was very helpful on parts that I was designing for the Pong game.
+[1228.40 --> 1231.04]  Certainly helping design the power supply.
+[1231.28 --> 1233.06]  And then it had a score.
+[1233.16 --> 1235.88]  And that score was very difficult.
+[1236.26 --> 1240.04]  How do you make numbers appear on the TV screen?
+[1240.42 --> 1241.96]  It's an engineering challenge.
+[1241.96 --> 1245.22]  And Jerry said, all I got, I got this new part from Fairchild.
+[1245.58 --> 1247.58]  It's a CMOS device.
+[1248.26 --> 1253.20]  And it would put score digits up on the screen on a digital system like that.
+[1253.32 --> 1256.46]  And he gave me a few sample parts and I hooked it up.
+[1256.66 --> 1266.22]  And in fact, the very first Pong prototype that is today in the Computer History Museum on display,
+[1266.22 --> 1268.84]  it has those Fairchild parts in it.
+[1268.84 --> 1276.74]  The more we researched, the more Jerry Lawson's little-known contributions grew in scope.
+[1277.26 --> 1286.00]  I'm Jeremy Saussier and I'm Assistant Vice President for Interpretation and Electronic Games at the Strong National Museum of Play.
+[1287.00 --> 1291.80]  Saussier says it's critically important to feature Lawson's work at his museum.
+[1291.80 --> 1299.28]  Here's someone whose story really hasn't been shared as widely as it should.
+[1299.28 --> 1306.28]  Really one of the few Black engineers working in this area, certainly in the 1970s.
+[1306.28 --> 1321.34]  And someone whose impact really in helping to create essentially a platform for what games would become.
+[1321.34 --> 1332.48]  And if we think about what video games were before the video game cartridge, before the interchangeable cartridge console,
+[1333.54 --> 1339.62]  you had a lot of versions of Pong, a simple tennis game.
+[1340.12 --> 1341.44]  You had some racing games.
+[1341.48 --> 1344.60]  You had some other themed games.
+[1344.60 --> 1350.86]  But they were generally on single systems, or they were arcade games.
+[1351.24 --> 1353.78]  And so this was really, truly revolutionary.
+[1354.64 --> 1358.84]  But, you know, revolutions don't always benefit the first heroes through the gate.
+[1359.58 --> 1362.94]  Initially, it's going to sell pretty well.
+[1363.30 --> 1365.06]  A few hundred thousand units.
+[1365.74 --> 1372.34]  But what you see over time is that it will be overshadowed.
+[1372.34 --> 1373.22]  Thanks, anyway.
+[1373.22 --> 1377.68]  After a family bought an Atari video game, they had no trouble getting babysitters.
+[1377.78 --> 1378.60]  Oh, my God.
+[1379.02 --> 1379.30]  Hello.
+[1379.70 --> 1380.06]  Kate.
+[1380.30 --> 1382.08]  No, we don't need a babysitter tonight.
+[1382.26 --> 1382.60]  Bye.
+[1382.98 --> 1384.36]  Everybody enjoys Atari.
+[1384.68 --> 1390.68]  Just a year after the Channel F was released, the Atari console arrived, selling millions.
+[1391.54 --> 1396.76]  Key to Atari's success was their ability to take Lawson's philosophy and put it into action.
+[1397.30 --> 1399.86]  Their focus was on the games themselves.
+[1399.86 --> 1407.22]  Al Alcorn was with the company back then, and he remembers how Lawson's work paved the way to Atari's success.
+[1407.22 --> 1413.16]  We determined at Atari that clearly a cartridge-based game was the way to go.
+[1413.76 --> 1416.42]  And Jerry had put one out well before us.
+[1416.42 --> 1420.88]  And he tackled some issues and problems with that.
+[1420.88 --> 1427.34]  And that we determined that we're going to, in our game, we were not going to architect it that way.
+[1427.56 --> 1430.14]  We wanted to get the cost way down.
+[1430.64 --> 1434.48]  There were too many chips in the Channel F Fairchild system.
+[1434.48 --> 1440.14]  But it was interesting and very important because he blazed the trail.
+[1440.36 --> 1446.92]  I mean, at that point, when we were doing a cartridge-based game, we sure as hell couldn't say you can't make one because Jerry had done it.
+[1447.22 --> 1448.70]  And it was on the market.
+[1448.80 --> 1454.74]  And the problem was that the product was a bit expensive because of all the parts that it used.
+[1454.74 --> 1462.14]  And frankly, Fairchild was not very good at marketing a consumer product, you know, at that point in time.
+[1462.46 --> 1463.94]  These were semiconductor companies.
+[1465.80 --> 1471.32]  When you bought an Atari, you got access to Space Invaders, Asteroids, Pac-Man.
+[1471.70 --> 1473.62]  These were people who knew games.
+[1474.12 --> 1478.64]  They also had third-party game developers like Activision to boost their catalog.
+[1478.64 --> 1488.02]  The world Lawson helped to usher in, where game makers were king, ironically spelled doom for a hardware company like Fairchild.
+[1489.88 --> 1494.46]  Lawson's son Anderson has a great way of thinking about his father's legacy.
+[1494.96 --> 1496.50]  Here's a bit more of our chat.
+[1497.86 --> 1506.00]  So when you think about the gaming industry after the Channel F console, what impact do you think that console had made?
+[1506.00 --> 1519.60]  I mean, you know, looking back, I think that the, like, again, that's probably the legacy is, hey, we, you know, successfully decoupled the software from the hardware.
+[1520.04 --> 1523.50]  And that's probably the biggest piece right there, right?
+[1523.58 --> 1526.20]  I mean, that's the model that's still followed today.
+[1526.34 --> 1530.72]  I mean, it's, you know, we went from cartridges to DVDs, right?
+[1530.72 --> 1534.32]  And from DVDs to, you know, streaming.
+[1534.80 --> 1541.56]  But there's still, even some games, like, you know, my son has a Nintendo Switch, right?
+[1542.22 --> 1545.90]  They have those SD drives they use as cartridges.
+[1546.20 --> 1547.80]  So it's the same thing.
+[1547.94 --> 1550.82]  It's just using different mediums to house the games.
+[1550.94 --> 1554.16]  So I just think that's kind of like the biggest piece right there.
+[1554.68 --> 1555.00]  Absolutely.
+[1555.00 --> 1563.78]  So when you engaged in a cartridge game, when you used one or when you blew into it to make it work, did you think of your father?
+[1565.26 --> 1565.66]  Yeah.
+[1565.90 --> 1570.86]  So it's funny because, you know, blowing into the cartridge.
+[1572.32 --> 1577.58]  But it was really about the cartridge sitting in the carriage the right way.
+[1577.58 --> 1580.10]  And it's funny, it spread like wow far.
+[1580.16 --> 1583.08]  People were doing that, but I don't think it was really doing anything.
+[1585.02 --> 1588.68]  Well, you just probably blew the minds of many of our listeners.
+[1588.96 --> 1589.86]  So thank you for that.
+[1592.04 --> 1600.06]  My name is Karen Lawson, and I am the daughter of Gerald Jerry Lawson.
+[1601.90 --> 1606.60]  We asked Karen about her father's life after he left Fairchild in 1980.
+[1606.60 --> 1615.72]  He created a company called VideoSoft, and his headquarters were right down the street from us, had an office space.
+[1616.54 --> 1621.96]  And he designed games for the Atari 2600.
+[1622.82 --> 1627.28]  So it was a story of, you know, you can't beat them, join them.
+[1627.28 --> 1639.44]  In time, Atari was itself surpassed by Nintendo and Sega, which then gave way to Nintendo 64, to PlayStation, to the Wii and Xbox, and then to Nintendo Switch.
+[1639.44 --> 1647.66]  And as Karen's brother Anderson pointed out, they all have in common this decoupling of the game from the system.
+[1648.22 --> 1650.48]  The priority of offloaded memory.
+[1651.14 --> 1657.08]  When Karen looks over the whole history of gaming, she takes pride in knowing her father shaped that industry.
+[1657.44 --> 1661.82]  We just didn't really know what the impact was at the time.
+[1661.82 --> 1667.02]  Later on, it became apparent that, hey, well, why don't they talk about this?
+[1667.06 --> 1669.10]  They talk about all these other gaming systems.
+[1669.60 --> 1671.16]  Why don't they talk about this one?
+[1671.98 --> 1681.80]  Things we do know that he was a little disappointed about how that ended up, but not ever disappointed about the accomplishments that he'd made.
+[1682.48 --> 1682.96]  Never.
+[1683.82 --> 1684.32]  Never.
+[1684.32 --> 1687.30]  Because everything that came after had a cartridge.
+[1688.68 --> 1699.06]  No longer were there systems being developed and made where the games were embedded in the console.
+[1699.90 --> 1703.96]  So when you look at it, it was mind-blowing.
+[1704.56 --> 1706.04]  It really, really was.
+[1706.04 --> 1716.30]  Joseph Salter, who we met at the top of this episode, was the diversity chair at the International Game Developers Conference.
+[1716.86 --> 1720.62]  And it would be his job to give Jerry Lawson his moment of glory.
+[1721.20 --> 1727.12]  Once he understood everything Lawson had done, he invited him to be honored at GDC 2011.
+[1727.12 --> 1738.46]  I picked up Jerry in the morning and we drove him to the session and he was in a wheelchair because he had lost one of his legs through diabetes.
+[1738.92 --> 1747.12]  So I pushed the wheelchair into the session and everybody was like, whoa, who is this coming in here?
+[1747.12 --> 1758.70]  You know, I said, I'd like to introduce you to Jerry Lawson, the man that created the first cartridge game for the Fairchild F Channel.
+[1759.20 --> 1761.44]  And people were like, well, this is amazing.
+[1761.58 --> 1762.36]  This is a blessing.
+[1762.50 --> 1763.84]  This is so great.
+[1764.02 --> 1765.70]  We really didn't know who you were.
+[1765.70 --> 1775.08]  It was like all of the colleagues that were there were completely stunned at the fact that this man had done what he did.
+[1775.48 --> 1783.94]  He had been a part of the life of the game industry from the beginning, right from the beginning.
+[1784.14 --> 1789.94]  He was the one who set the cartridge game into existence.
+[1789.94 --> 1802.12]  One month after the conference, Jerry Lawson, 70 years old, passed away, having just barely lived to see himself recognized by the industry he helped to create.
+[1802.78 --> 1806.50]  And why was Joseph Salter brought to tears when Lawson got his due?
+[1807.24 --> 1813.70]  So many times in the game development arena, as an African-American, you're very lonely.
+[1813.70 --> 1826.80]  And anybody that's in the industry who is African-American knows what I'm talking about in terms of being lonely to the point of, you know, really not being accepted, really being looked at as an outsider.
+[1826.80 --> 1839.46]  And to know that Jerry had gone through all the things that he had gone through and had survived into a great human being.
+[1839.80 --> 1844.46]  It was so important for me to see that.
+[1844.66 --> 1846.14]  And it just got to my heart.
+[1846.24 --> 1847.62]  And I actually started crying.
+[1847.94 --> 1852.12]  Usually black man doesn't want to say I started crying, you know.
+[1852.12 --> 1865.52]  But it was the sheer knowledge of his participation in an industry that I know is not very accommodating to African-Americans.
+[1866.66 --> 1868.24]  So the tears were real.
+[1868.80 --> 1870.54]  The tears were real.
+[1872.76 --> 1876.16]  And so was Jerry Lawson's place in history.
+[1877.24 --> 1879.18]  Lawson's son Anderson remembers.
+[1879.18 --> 1884.08]  The very last conversation I ever had with my father was about that.
+[1885.16 --> 1893.26]  You know, I had asked him, I said, don't you find it kind of odd that near the end of your life, you're getting your roses?
+[1894.22 --> 1898.32]  And he said, yeah, you know, I do find that kind of odd.
+[1898.38 --> 1899.66]  But he wasn't upset.
+[1899.78 --> 1907.44]  He was just, you know, I think that he understood that it had finally come, that, you know, people were recognizing him.
+[1907.44 --> 1908.96]  And he liked that.
+[1909.52 --> 1917.78]  I think he just liked to talk about, you know, a lot of the things that he worked on and inspire younger people to get interested.
+[1918.06 --> 1919.30]  That's really what he was all about.
+[1919.90 --> 1921.72]  And it wasn't about the game per se.
+[1922.08 --> 1927.06]  Like, he was more into the scientific aspect of the endeavor, if that makes sense.
+[1928.14 --> 1929.34]  He was a true engineer.
+[1929.34 --> 1934.98]  Today, there are maybe three billion gamers in the world.
+[1935.34 --> 1940.46]  And they're spending about $200 billion each year on their games.
+[1940.80 --> 1945.24]  In just a few decades, we've gone from this to this.
+[1945.24 --> 1958.42]  Even as we were working on this episode, Lawson's contributions were featured in a new docuseries from Netflix called High Score.
+[1958.42 --> 1969.00]  The world is waking up and realizing it was the cartridge revolution that jumpstarted the industry, paving the way for game development to become a juggernaut.
+[1969.00 --> 1978.14]  The birth of cartridge games shifted the focus from hardware to software, from the console to the game itself.
+[1978.88 --> 1981.84]  Pong was made by one guy alone in just three months.
+[1981.92 --> 1991.94]  The gaming world that Jerry Lawson helped deliver is one where huge teams spend tens of millions of dollars to build games over the course of years.
+[1991.94 --> 1998.64]  And those games are now reviewed in the New York Times with the same reverence as operas and symphonies.
+[2000.00 --> 2010.70]  So, even if cartridges have largely been replaced by the cloud, we're all still living in that rich gaming landscape pioneered by people like Jerry Lawson.
+[2011.22 --> 2014.78]  Innovators who saw that games were ready to level up.
+[2014.78 --> 2022.44]  Command Line Heroes is an original podcast from Red Hat.
+[2022.44 --> 2029.40]  For some great background material on Jerry Lawson, head to redhat.com slash command line heroes.
+[2031.04 --> 2035.62]  Next time, we meet the woman who helped make GPS a reality.
+[2036.34 --> 2039.12]  Mathematician, Dr. Gladys B. West.
+[2040.32 --> 2042.40]  I'm Saran Yitbarek. Keep on coding.
+[2044.78 --> 2050.52]  I'm sorry, I'm still thinking about how blowing on game cartridges doesn't do anything.
+[2051.02 --> 2052.80]  My world is upside down right now.
+[2052.80 --> 2053.58]  My world is upside down right now.
+[2053.58 --> 2054.04]  I'm an astrologer.
+[2054.04 --> 2054.58]  My world is falling.
+[2054.58 --> 2055.16]  My world is upside down right now.
+[2055.16 --> 2056.10]  My world is upside down right now.
+[2056.14 --> 2056.84]  I'm still laying on your iPad.
+[2057.02 --> 2057.72]  I'm solid.
+[2058.46 --> 2067.30]  impostor, Amy glaube, traffic, traffic, traffic, traffic, 있어, what ever務 way with the newентов on
+[2067.30 --> 2067.96]  radio sound
+[2067.96 --> 2068.34]  Today isconds hurricanes.
+[2068.34 --> 2069.92]  Before I start to notice a whole lot of limits.
+[2070.00 --> 2071.20]  Give me an example to fearsome,yson,趁 steady, still are Toyota's copyright,
+[2071.20 --> 2071.54]  the animated 1986 psychologically.
+[2071.56 --> 2072.64]  Unfortunately sadlycock andlamaكم out the reason I am lost trying to figure out into things about.
+[2072.64 --> 2073.66]  But I really can与 it more depth and feel safer to understand olabilir anyறத.
+[2073.66 --> 2074.14]  Since Unfortunately, we are quite olmaz no doubtful idea.
+[2074.14 --> 2075.66]  But I'm so careful about you from Kendyl Windows and friends as well Rip,

The future of Testify_transcript.txt

@@ -0,0 +1,559 @@
+**Mat Ryer:** Hello, and welcome to Go Time. I'm Mat Ryer, and today we're talking about testing frameworks. Some subjects which aren't controversial in other languages are sometimes controversial in Go, and this is one of those... And we're gonna find out why that might be. Today I'm joined -- it's only Mark Bates; hello, Mr. Bates.
+
+**Mark Bates:** Hello, Mr. Ryer. How are you doing today?
+
+**Mat Ryer:** Good, how are you?
+
+**Mark Bates:** Alright... \[laughs\] I really thought I was gonna have something there and I just didn't.
+
+**Mat Ryer:** Yeah, you were just waiting, and nothing came.
+
+**Mark Bates:** Yeah. I haven't been on the show in a while, so I'm just a bit rusty, that's all.
+
+**Mat Ryer:** That's alright, mate. Don't worry, you're in good hands. We're also joined by special guest Boyan Soubachov. Boyan, did I welcome you? Welcome.
+
+**Boyan Soubachov:** Yeah.
+
+**Mat Ryer:** Okay. Let me do that again--
+
+**Mark Bates:** Did you welcome him? Yes. Did you say it correctly...?
+
+**Mat Ryer:** Boyan Soubachov.
+
+**Boyan Soubachov:** That's perfect.
+
+**Mat Ryer:** Okay.
+
+**Boyan Soubachov:** Hey, Mat. Hey, Mark.
+
+**Mat Ryer:** Welcome to the show.
+
+**Mark Bates:** Hey, Boyan. I've gotta ask a question here; the folks at home can't see this, but I think you're trying to pretend like you're in some sort of fancy Mediterranean paradise, where I have a feeling it's a beige wall behind you. How close am I to the truth?
+
+**Boyan Soubachov:** Pretty damn close, because it's five AM in Sydney in winter, so it's anything but sunny and warm. \[laughter\]
+
+**Mark Bates:** Fair enough...
+
+**Mat Ryer:** Well, thank you for joining us so early.
+
+**Boyan Soubachov:** Not a problem.
+
+**Mat Ryer:** It's great.
+
+**Mark Bates:** Or maybe it's so late, we don't know. Maybe he's been up all night...
+
+**Boyan Soubachov:** \[laughs\]
+
+**Mat Ryer:** Just getting in?
+
+**Mark Bates:** He's just getting in, he's been in one of those all-night raves...
+
+**Boyan Soubachov:** Thankfully, our on-call schedule isn't that bad... \[laughter\]
+
+**Mat Ryer:** Fair enough. Well, so today we're talking about testing frameworks...
+
+**Mark Bates:** Okay, I have a question, Mat, before we go any further, about testing frameworks.
+
+**Mat Ryer:** \[04:11\] Testing frameworks -- what?
+
+**Mark Bates:** Are we talking about frameworks that help us test, or testing frameworks like Buffalo?
+
+**Mat Ryer:** Oh, no, no...
+
+**Mark Bates:** Sorry, I just had to get the Buffalo plug in there early, and then that way -- it's like a Hitchcock thing; people are just waiting for it. Get it out quickly. Buffalo. Then we get to move on to the rest of it.
+
+**Mat Ryer:** Okay. Well, we can talk about testing testing frameworks later as well, if you like... We could really get creative.
+
+**Mark Bates:** Yes, I like to use a different testing framework to test the testing framework, just to be on the safe side.
+
+**Mat Ryer:** You actually have to, because the testing framework doesn't exist. And what if there's a bug in it?
+
+**Mark Bates:** That's not true. Go is written in Go. Come on...!
+
+**Mat Ryer:** Go is written in Go, yeah...
+
+**Mark Bates:** \[laughs\] Everybody knows. First one's in C, and then you bootstrap. \[laughter\]
+
+**Mat Ryer:** What are testing frameworks then, and what does it mean in Go? Maybe it's worth just a bit of an overview for junior developers or people that aren't familiar with this yet. What are we talking about?
+
+**Mark Bates:** Well, that was actually -- I was making a bad little joke there about the testing frameworks, but I do have a legitimate thing... I think we need to talk about the language. Are we talking about assertion libraries, which are this .. false, true, not nil? Or are we talking about frameworks like Ginkgo and Gomega and stuff, where you have to buy in to their entire process. GoConvey has its own scripting language if you want a testing/scripting thing, if you wanna go through that... So when we talk testing frameworks, are we talking one, are we talking the other, or are we talking both?
+
+**Mat Ryer:** We're talking all of those things today, I think...
+
+**Boyan Soubachov:** I think that it's a spectrum, effectively. There's trade-offs as there is to everything, where you have the lightweight stuff like Testify is, which is almost like syntactic sugar... And you have, as you mentioned, Ginkgo, where it's on the complete opposite end, where this is how you will do it, and it's almost like a fat middleware layer, and the right tool for the job, I guess.
+
+**Mat Ryer:** Yeah, so for anyone at home then that's new to this, testing generally in code is a way to really improve the quality of the code by writing automatic tests that can run, and they basically become a user of your code, and then they work as though your user of your code is gonna do the work, and then they can make assertions about what happens. So that's generally the process of testing. And then, of course, if you make little changes, you can run your tests and make sure you haven't broken things... So testing frameworks help you with that process in various ways, and there are different types.
+
+For example, the simplest case, like the syntactic sugar case - and I think Testify at one point was the most imported package in Go. I don't know if that's still true, but it's certainly a popular choice, so let's start with Testify. What is Testify, what does it do, and why would you use that?
+
+**Mark Bates:** Well, you wrote it, Mat. Why don't you tell us?
+
+**Mat Ryer:** Well, I've just been talking too much...
+
+**Mark Bates:** A lot of people don't know that Mat was one of the original authors.
+
+**Boyan Soubachov:** Yeah, so Mat and I share a common thread on that one. Mat, you know the history way better than I do, and I have a bit more context on the status quo, I guess...
+
+**Mat Ryer:** Yeah. I'd like you to give the history though, please...
+
+**Boyan Soubachov:** Sure, of course. Obviously, Testify -- well, Mat, I believe you were one of the original authors...
+
+**Mat Ryer:** \[07:52\] Okay, yeah... So this was way back -- I think even before the first GopherCon. So this was like 2014 time. And we came from Ruby, and I'd done a lot of JavaScript and C\#, and a lot of these languages have testing frameworks, or even some of them are just built-in, where you can use language like assertEqual and you give it two things. And if those two things aren't equal, that's considered then a failing test. So in Go, if you don't use a testing framework, you end up writing long-hand really that process. So you'll use an if, and check two variables; if they're not equal, or whatever the condition is, then you use the testing.T to do some work to either report the error, or fail the test, or whatever it is you wanna do.
+
+We did Testify just because we were used to writing in that format, that's all. It was purely kind of syntactic sugar. And then because it was so early, it just kind of became part of an easy choice. It was the oldest, probably, at one point, and probably always is...
+
+**Mark Bates:** No, an even older one came out...
+
+**Mat Ryer:** Yeah, an older one came out after--
+
+**Mark Bates:** Just last month, actually... \[laughter\]
+
+**Mat Ryer:** Yeah. But I think it's that - it got in early, and... People were complaining about writing tests in this long-hand form, and some people weren't writing tests for that reason... So that's why we did Testify. And we love testing and unit testing. It was me and Tyler Stillwater that created the original, and then lots of contributors... The project has grown and taken on a life of its own, as these things do... And that's really it, I think, for the history.
+
+**Boyan Soubachov:** Right. And just to add there, actually I would disagree on a small point... You did mention, obviously, because you were early; I think that helps, but I also think the majority of it is just -- it saves the community time, and the community finds this valuable... Because it's a simple problem, that's solved with a simple solution, in my opinion anyways. So exactly that... And now, the current state is almost like a "Oh, time to write some tests... Quickly, import Testify." It's almost second nature to a lot of gophers... Which in my opinion is a great thing, but it's also a big responsibility. Because if something is shared by all gophers, it has the pressure of it being all things to all people.
+
+**Mat Ryer:** Yeah, absolutely. It's funny, because some of the innovation that happened over the years with that - there is a Require package, which has kind of a slightly different behavior. I think one of them lets you carry on...
+
+**Mark Bates:** That's the one I use. Assert lets your test keep running, and just logs the failures. Require stops the test. For me, I love Require. I use it for all my tests. As a matter of fact, my \[unintelligible 00:10:48.26\] Go snippet when I type "test tab", it fills in the require.new testing.T thing right there. It's just part of my tests.
+
+My problem with the Assert is typically, if something fails, I want the test to just stop, and not keep throwing panics for the next five minutes, which is usually what happens... Because something that wasn't supposed to be nil is nil now, and your test is just panicking all over the place... \[laughs\] So I like that.
+
+For me, like I said, I'm a huge, huge fan of Require. I probably used at the most ten or twelve of the assertions. You know, error, no error, nil, not nil... Really basic assertions. I don't typically use some of the really grandiose or odder assertions. I'll occasionally maybe grab a contains, or something like that.
+
+What I like about it is - just what Mat was saying earlier - all that repetitive code, just all those if checks, now I've gotta come up with a good log message... And when I'm writing tests, I just wanna write my tests and I want it to be fast. I just need to get tests, and I want to spend more time writing more tests, and not more time writing less tests. And just pulling out those if statements, and pulling out that logic... You know, in my Go code that I'm testing, of course I'm gonna take care to craft better error messages, and have better syntax, and all that sort of stuff. But when I'm writing my tests, I just need it to say that these two things weren't equal, and I don't need to keep repeating that constantly... And things like Require just kind of do that.
+
+\[12:33\] And I also quite like the way it reads, too. My tests are a lot smaller. Maybe they're now a half dozen or eight lines long, as opposed to three times that, with all the if checks that are going on in there... And I can just read it; it's like "Do this. Okay there was no error. This was not nil. These are equal. Done." I just read it quickly down...
+
+I remember, Mat, you and I had a conversation around this offline one day, about is.
+
+**Mat Ryer:** Offline, you mean in real life?
+
+**Mark Bates:** Yeah... \[laughter\]
+
+**Mat Ryer:** Remember offline?
+
+**Mark Bates:** I do remember that... Sorry, I don't mean to digress, but GopherCon UK is supposed to be next month, and we were gonna co-host together in London, as always...
+
+**Mat Ryer:** \[laughs\] Yeah, it's not gonna happen.
+
+**Mark Bates:** I was so looking forward to it. Anywhoo, back to the subject. So yeah, I love Require, and I like what it does for my tests in terms of just making them small, quick, clean and readable.
+
+**Boyan Soubachov:** It's an interesting point, because as I mentioned all things to all people, we've had so many requests, I get so many GitHub notifications in issues saying "Please remove Require. It's useless. We just use Assert." \[laughs\]
+
+**Mark Bates:** It is not useless; it's more useful than assert.
+
+**Boyan Soubachov:** Exactly.
+
+**Mark Bates:** If anything, I would say remove assert. \[laughter\] Like, why do you want your tests to keep going? It just boggles my mind.
+
+**Mat Ryer:** It's too late though... You can't just remove Assert now, because of what would happen...
+
+**Mark Bates:** I know, but it's just like... Why would you use Assert over Require? I don't know...
+
+**Mat Ryer:** Yeah, I always wondered that, actually. In fact, I remember in the original design we had the assert methods return bools as a way to get around it. So you could do an if block and say "If this is not nil", and then it returns false if that failed, I guess...
+
+**Mark Bates:** That's so weird...
+
+**Mat Ryer:** Right? It's confusing.
+
+**Mark Bates:** I can see why you got rid of that.
+
+**Mat Ryer:** I think it's still there...
+
+**Mark Bates:** I don't know, I don't use Assert.
+
+**Mat Ryer:** Yeah. And Require doesn't have it, I think, as well. Actually, Ernesto Jimenez was one of the maintainers, who's a great engineer...
+
+**Mark Bates:** Lovely chat, too.
+
+**Mat Ryer:** ...automated the generation of Require by inspecting Assert. So it was actually an automatic process. They didn't have to keep maintaining the two codebases.
+
+**Boyan Soubachov:** Correct. To this day it's a great addition. We stole autogen require from the Assert code.
+
+**Mat Ryer:** Yeah. That's cool, ain't it? \[laughter\] I think that's cool.
+
+**Boyan Soubachov:** It's awesome.
+
+**Mat Ryer:** Anything that we can do to get computers to write code for us, I'm all over it.
+
+**Boyan Soubachov:** Yeah, exactly.
+
+**Break:** \[15:21\]
+
+**Mat Ryer:** Well, we talked about the way it reads, so you can say "require not nil", or "assert equal", those kinds of things...
+
+**Mark Bates:** I prefer the variable version, to be perfectly honest. So I'll do require.New(t), and assign that to R, usually, is what I do. So I just do "R no error", whatever. Again, it's all about that quick, "I just wanna write stuff." Typing require.=t, two things - it's so much more code than just r.=two things. I don't know, that's just my preference.
+
+**Mat Ryer:** Yeah. And Testify wraps the testing tier, as well. I don't think that's true for all of them. There are some of those that do the same. And I think that's quite a nice little design thing, because I think originally, again, they were just global functions. Boyan, are they still global functions?
+
+**Boyan Soubachov:** You can still access them package-level.
+
+**Mat Ryer:** Okay, so that's originally what it just was...
+
+**Mark Bates:** But you have to pass t in to do it.
+
+**Mat Ryer:** Yes, you pass t in as the argument, that's right. So it's a different flavor of that. Yeah, it's funny, because it really is an example of a project that has evolved in quite an interesting way over the years.
+
+I wanted to just talk about the format though, the readability of it, because there is another kind of style of this, which are behavior-driven tests. Does someone want to just tell us what's a behavior-driven test and how is the language different?
+
+**Mark Bates:** Yeah, so if you're coming from Ruby, like a lot of us did - so many people I know from Ruby in Go over the years - that was one of the big things, our spec being one of the biggest testing frameworks in the Ruby world... And in Ruby, it works really well because of the nature of Ruby being a scripting language. And you can write these really beautifully syntaxed sentences where it says --
+
+**Boyan Soubachov:** Almost prose.
+
+**Mark Bates:** ...describe "As a user, I want to do..." and then X, Y and Z. And you know, log in, and then here's a little thing; and then "I want to be able to forget my password and have it reset", and this more of a natural language style, where you're telling a story with it... So that kind of "As A, I want to..." kind of story that we all know of, and we all hear that someday the other side is gonna write, and then we can just test, and that never happens... \[laughter\] But that works really well in those languages. And most of us in Ruby, I think, lived in that world. I don't know about you, Mat. Did you write your tests that way, or did you use more of a unit test style when you were doing Ruby?
+
+**Mat Ryer:** I always used the unit test style. I found it to be quite unnecessarily verbose... I know that sometimes they -- because you literally put those things as strings into the test code, don't you?
+
+**Mark Bates:** Yeah.
+
+**Mat Ryer:** And then from that, you can generate some quite nice-looking failures. If something fails, it reads quite nicely... But I've found it to be too verbose actually, and just saying "Not equal" and then showing you the two values or something was just easier to see.
+
+**Mark Bates:** The thing I like about BDD versus unit testing actually has to do with the names of the tests... You know, when you're writing a simple test - you know, test that create does something; that's a pretty simple test name. But when you start having all those weird variants, then the string-based text names become really useful. When you just need a little bit more description as to what it is you're trying to test... And that's harder to do in a units type of a test, where you have a function name.
+
+\[20:05\] For me, I get around that a little bit in Go with underscores in my test names. So I have a different approach to how I do my test names. So it obviously starts with "test", because we all have to start there, and then usually it's an underscore, with the next thing being if I'm testing, say, something on a struct, or a type, then it'll be the type. So it's test\_type\_method, or something like that, where I can kind of look at the tests, read them a little bit better, try to figure out what they're doing... And also, when I'm trying to do run, it helps me, because I have a bit of a pattern. So if I wanna test all the file stuff, my types called "file", and I wanna run all the file tests, I can do test\_file\_ and it's gonna get me all of the tests for the file type when I do the run.
+
+So that's how I'm getting around it in Go, is with using that... But there are frameworks like Ginkgo that let you experience that a bit more in Go.
+
+**Boyan Soubachov:** I think a key thing there - and at least my takeaway of it - is because the philosophies behind BDD and TDD, at least the way I understand it, is that BDD should be more focused on the user or the consumer that we're writing this code for, whereas TDD is a lot more technical or engineer-focused, personally, I tend to hybridize, where I'll do them TDD style, because I find them simpler and less verbose; I'm not expecting the user to read my tests... But I'll try and order and make the tests actually focus on user behavior in more complex functions, rather than just proving that the function works, or stopping it from regressing, in a way.
+
+**Mat Ryer:** Yeah, have you seen property-based testing as well? This is another kind of style.
+
+**Boyan Soubachov:** Yeah, it's almost like fuzzing for your functions, where you specify not what values to test with, but what types of values your function takes as input. And obviously, also what kind of output you expect. Then the property-based testing framework, of which one example is Gopter, will then just generate - whether it's random, or in some specific sequence - plenty of values to then test your function and try and find edge cases for you that do not conform to the specification.
+
+**Mat Ryer:** That's really cool, isn't it?
+
+**Boyan Soubachov:** Yeah. The first exposure I had to that was with Hypothesis in Python, and it was pretty awesome. An obvious problem there is you now have -- especially if your functions are a bit more complex, that it can take a while for tests to run, because you're now running the same function 10,000 times, rather than once or five times. But for pure functions, where your code doesn't have side effects, it's very good at finding edge cases and little behaviors that you don't anticipate.
+
+**Mat Ryer:** Does it explicitly have edge cases programmed into it? So if it's like an int, it's gonna try an extreme, like max int, and minimum int, and things like this... Or is there an element of more randomness, like fuzzing?
+
+**Boyan Soubachov:** I think it depends on the framework. I haven't personally used Gopter just yet in Go, but in a lot of them you can actually specify where I want to range from min int to maxim, so whatever the minimum and the maximum value is... Or you can just say "Give me this distribution." I believe it can be customized. But if I'm not mistaken, the default is to just basically sample from the minimum to maximum, and try and find pathological cases of the input values.
+
+**Mark Bates:** \[24:00\] Do either of you do fuzz testing/fuzzing?
+
+**Mat Ryer:** I have once, but it was really to play with it. But I think it depends on the kind of code you're writing, because it's not suitable for everything, is it?
+
+**Mark Bates:** Yeah, I think I'm the same as Mat. I've done it once to try it out... Yeah, Damian Gryski is all on it. Did that sound about right?
+
+**Mat Ryer:** I don't know, I let you do it. \[laughter\]
+
+**Mark Bates:** Thanks. I apologize. But you know, he's always talking about fuzzing... So I do try to play with it every once in a while. I think you're right, I feel like there's a type of code that it works really well for... At least I haven't figured it out well enough to see it beyond, to see it in cases where it's useful to me. I couldn't figure out how to get it to work, say, in a web environment, other than to pass in random JSON blobs...
+
+**Mat Ryer:** Yeah, but one example is the Plush library in the Buffalo project, actually... Because that takes user input, and it parses it, doesn't it? It literally breaks it into tokens, into an AST, and all that... So that's probably a good candidate for fuzzing.
+
+**Mark Bates:** That was the one I tried it with. \[laughs\]
+
+**Mat Ryer:** Is it?
+
+**Mark Bates:** Yeah.
+
+**Mat Ryer:** Yeah, because essentially, fuzzing - for those that don't know (and do check it out, because it's cool) - if you've got a string as an input to some method, it'll just make loads of bonkers strings, as crazy as it can, including massive ones, and all sorts of random nonsense, in an attempt to break your code... And if it can, it tells you. Its input will break your code. So I guess the point about that is, ideally, Mark, nothing should make that Plush panic. It should always be a meaningful error. And you might find some panics or something that you would just never think of because of fuzzing... But who knows.
+
+**Mark Bates:** Not me, apparently... \[laughter\]
+
+**Mat Ryer:** You just do it manual fuzzing, don't you, Mark?
+
+**Mark Bates:** We've just established that I don't know. \[laughter\]
+
+**Mat Ryer:** Mark just does manual fuzzing; he just writes loads of tests, but he just does it manually, like mashing the keyboard...
+
+**Mark Bates:** No, my dog takes care of it all.
+
+**Mat Ryer:** Pardon?
+
+**Mark Bates:** My dog takes care of it all. Ringo is an amazing fuzzer. \[laughter\]
+
+**Boyan Soubachov:** I've heard of cats walking on keyboards. I haven't heard of dogs walking on keyboards yet... \[laughs\]
+
+**Mark Bates:** Well, Ringo is a special dog. You know...
+
+**Boyan Soubachov:** Things you see on Zoom nowadays...
+
+**Mat Ryer:** Yeah.
+
+**Mark Bates:** So it's good. Are we done?
+
+**Boyan Soubachov:** \[laughs\]
+
+**Mark Bates:** No, I'm just kidding. Do we have questions from the audience?
+
+**Boyan Soubachov:** Actually, I've seen the GoTimeFM channel, and there's one about mocking.
+
+**Mat Ryer:** Mocking...
+
+**Mark Bates:** Mocking...
+
+**Mat Ryer:** Okay, so who wants to tell us what mocking is?
+
+**Mark Bates:** Not me...
+
+**Boyan Soubachov:** I'll give it a shot.
+
+**Mat Ryer:** Go ahead, Boyan. Boyan to the rescue.
+
+**Boyan Soubachov:** So we unfortunately don't live in an ideal world, where all our functions are pure and have no side effects... Sometimes, for example, they need to interact with databases or external services... And when we run tests, especially unit tests, or specifically in the tests, we want them to run quickly and just test that function, rather than the side effects it has on these external resources. So a general and a common practice is to mock out those resources, so effectively return some blank, known value for the test, or preset that we know for the test, and we then assert that with these values the function did what it was supposed to do, given these inputs from the external resource.
+
+**Mat Ryer:** Right. Yeah, so that's great. The specific question from Aditya was about database mocking... So how do we feel about mocking databases?
+
+**Boyan Soubachov:** \[28:00\] Myself, personally, I find it quite useful a lot of the time. An approach I actually like to take is to try and abstract our database interface as an interface... And then just spawn a new stub or mocked version of that interface and pass it in.
+
+**Mat Ryer:** What does the interface look like for a database?
+
+**Boyan Soubachov:** It's highly dependent, obviously, on what you're using or what you're needing... But you obviously have your basic CRUD operations, if your function is using them; and then anything more specific, in the abstract sense, it's difficult because it becomes incredibly context-specific, given what your code does.
+
+**Mat Ryer:** Yeah, it does. And it is also possible to create tightly-coupled test code as well, where it becomes so brittle that any change in the implementation will break something... Because sometimes things are over-tested, I think. I actually think 100% code coverage, depending on the project, could be a mistake... Because essentially, you have described probably (well, almost certainly) everything in that package, and any change to that will then break the tests... And that can, from a practical point of view, get really annoying when everytime you make little tweaks, you have to then go and also update tests.
+
+**Mark Bates:** I always tell people, try to hit for around 90%.
+
+**Mat Ryer:** Right.
+
+**Mark Bates:** 90% is a good number. If you can get it around there, you're doing alright. Because you know what you're not gonna get at 90%? You're not gonna get all those "If error equals nil, return nil" errors... \[laughter\] That's what you're gonna get if you hit 90%; you're just gonna skip all those particular ones, because you can look at them and validate that they're returning an error. You don't really need -- and sometimes it's more complicated to try and get that error inserted, just to get the test to pass, when it's more just like "Hey, I don't really care if any error comes back from this. Just pass it back up." I don't need to force an error there just to trigger that one code path for a test. But to get back to the database thing - I do not mock out my databases.
+
+**Mat Ryer:** Right. Do you use a real database?
+
+**Mark Bates:** I use a real database, yeah. In Ruby, it was a real problem. And that's nothing against -- it's not Ruby, it's not Rails, it's just... The work of something like Active Record taking all those database records and breaking them into all of those Ruby objects, and all the work that it did in that highly-dynamic interpreted language meant that to hit the database takes a long time, and uses a lot of resources. It really slowed your tests down after a while in Ruby.
+
+In Go, it's not my problem. \[laughter\] My tests are still plenty fast. They're hitting Postgres locally; how much of a slowdown could that be...? And honestly, I find that it's usually plenty good enough. My tests are fast enough. I don't have to deal with trying to keep all those objects -- and I find when I'm dealing with database stuff the amount of things I need to mock out, to make it usable, is sizeable.
+
+**Mat Ryer:** Yeah, you end up building a database emulator in memory even sometimes...
+
+**Mark Bates:** Yeah, and for me it's like, if I'm gonna spend -- I don't need to spend all that time. I'll just hit Postgres. It's right there; it's gonna be there. My server is gonna be there, my CI, it's local, it takes me 30 seconds to install... I'll just hit it. And I know that all my queries are correct, I know that my marshaling to and from the database is correct... I feel like there's a whole -- and this is my unpopular opinion. Mat wanted me to come up with an unpopular opinion, and I was gonna say that I like assertion libraries... But I think this is more unpopular.
+
+**Mat Ryer:** \[32:12\] Okay, well save it for the official segment.
+
+**Mark Bates:** Well, you've got it now, Mat...
+
+**Mat Ryer:** Don't preempt the segment, Bates...
+
+**Mark Bates:** Well, maybe that's an unpopular opinion...
+
+**Mat Ryer:** That is an unpopular time.
+
+**Mark Bates:** ...that I should have to wait till the segment... It's a second unpopular opinion. \[laughter\]
+
+**Mat Ryer:** That's very meta. You're right though... That's kind of an integration test, in a way, what you're describing then, Mark...
+
+**Mark Bates:** It is, in some respects. We could talk about this for a long time, and the distinctions between unit integration, and even the distinctions between mocking and stubbing in interfaces...
+
+**Boyan Soubachov:** I completely agree - the right tool for the job, if it works fantastic... But that starts being a problem when you have 5,000 other engineers and you need to do this on services with a ton of complicated DB interactions in code, I guess...
+
+**Mark Bates:** Yeah, I'm not saying it's for everybody... \[laughter\] I'm saying for my stuff, I see it as something that I would rather try to come up with a couple good interfaces where I need them. And typically, a couple good interfaces are all I really need; a couple one-method, two-method interfaces are usually good enough... If I write my code well enough, to fully isolate kind of - or at least for the most part - the things that I really wanna tweak and change. And I try to allow myself those insert points in code, where they might be in the form of, say, a plugin shape, where I can add certain plugins to my web server in testing, that mutate the request, or do stuff that lets me fake out third-party requests with transports, and all sorts of other stuff. I can't remember what the point was, but... It sounded good.
+
+**Mat Ryer:** It did.
+
+**Boyan Soubachov:** \[laughs\] Spend time on what works, on the important things.
+
+**Mark Bates:** Oh, that was it, yeah. And then, you know, abstract -- and then when I hit those problems, like you were just describing, with all these other engineers, and a bigger abstraction is needed, then I put the time into that level of abstraction. But right from beginning it's just -- you've got bigger fish to fry than that abstraction, I think...
+
+**Boyan Soubachov:** Exactly.
+
+**Mat Ryer:** Yeah, context probably is very important. I have seen as well -- one simple way to mock...
+
+**Mark Bates:** Context the interface, or context knowledge?
+
+**Mat Ryer:** Oh no, I just meant the English word "context". \[laughter\]
+
+**Mark Bates:** Okay. Sorry, I was legitimately confused for a second.
+
+**Mat Ryer:** No, it's fine... \[laughter\] Yes, there are some simple ways to mock out databases as well that I've seen, and used in the past. Sometimes it's a case of just have your own interface that you describe in your code, and your implementation of it touches a real database and has the database code in there... But that could be quite easy to mock. And then you aren't making database calls, but they might be something like *loadUser* or *loadLesson*, where in production that's gonna be a database request; you could mock out that interface at that level. So it's a kind of slightly higher-level wave of doing it. Have you done things like that before, anyone?
+
+**Boyan Soubachov:** Personally, at least in the last few years, where I've been working in larger Go teams, definitely that has been the way we've done things...
+
+**Mark Bates:** Right.
+
+**Boyan Soubachov:** \[35:49\] We're spending more time on getting the right interface and abstraction going. But that's, as we've been talking, purely because of the team size and the codebase size. It becomes more practical to do that.
+
+**Mat Ryer:** Yeah. The other thing about mocking - if you want to mock something out, even if it's a database or some third-party service, or whatever, it is useful in test code to be able to do that so that you can return realistic data, so that your code can do it's thing, so that it's predictable, and all that stuff. I have, in the past, done assertions inside the mock almost, and I feel like this might be a step too far, in most cases. The idea is there's a method -- in fact, there's a project called Moq, which is one that I wrote with some friends of mine, I can't remember who. It's in the repo though; credit where credit's due.
+
+**Mark Bates:** The show in which Mat tells us all about his former projects he no longer uses. Go on, Mat...
+
+**Boyan Soubachov:** \[laughs\]
+
+**Mat Ryer:** No, I do use this one. Actually, yes -- this was from David Hernandez, and he came up with this nice pattern for doing mocks, which was essentially you create a struct, and then for every method there's a field that matches the signature, and then you implement on that struct those methods as well, which just call those functions... So that gives you almost like a transparent layer of that type, which you can then in test code create and give it little functions, which will be the things that get called. And you can do things like simulate errors, like we were talking about earlier. More importantly, they're good for returning useful and predictable data.
+
+I think it's better not to make assertions about what was called in that, because it's kind of implementation detail potentially there... But it probably does, again, depend on each case, doesn't it?
+
+**Mark Bates:** I think it's okay to assert errors, and stuff like that in there. If you're getting a new UUID and it errors out, or something, I think it's okay to do that. But yeah, asserting on the data there... I often do that field function thing not just on test structs, but regular types too, where there are areas where I want to provide maybe a default implementation and then allow people to replace it. So one might be around ENVs for example. I've been kind of recently playing around with this idea of using zero-value structs for my configuration, where the methods on that struct - there's no real field, it's just methods, and those methods will look for an ENV, and then if they find it, return it, or otherwise return maybe some default. And I like that you can just kind of grab a config, just var it up, and boom, all your stuff's there, you don't need to really initialize it.
+
+But obviously, when you're dealing with environments, it's really hard to test. And sometimes - not even that, but you want to, maybe on a specific request as you're passing maybe a config down, you wanna change a value. And you can't do that if they're all functions. So one of the things I do there is have kind of a *getEnv()* function that matches *os.getEnv()*... But you can set it, and all those functions that are on my config will check and say "Oh, if there's this function defined, let me use that", and that could just be returning a test value. Or use the default, which is *os.getEnv()*.
+
+I use that pattern quite a lot too (I like that), of "Here's a default way of doing things, and then here's a function you can implement as a field", same signature, and it'll call that instead, if it has it.
+
+**Mat Ryer:** Yeah. A good use of that is if you need to control time, because the *time.Now()* function just returns the current time. Obviously, in real code, that's different every time... But you can have a function field that matches the *time.Now()* signature - which I think is just empty func - and then assign it by default to *time.Now()*, and then control that in your test code. So yeah, I love those sort of mini-mocks almost...
+
+**Mark Bates:** \[40:03\] In Ruby we had a great thing called timecop; it was such a random Ruby thing... It would override basically time.now to be whatever value you wanted it to be \[laughter\] You could just override anything; you could say like "I wanna be three weeks into the future", and time.now would return three weeks in the future.
+
+**Boyan Soubachov:** Great Scott! \[laughter\]
+
+**Mark Bates:** It was spectacularly awful, and good, and fun... \[laughs\] It was one of those things you can only do in a dynamic language though. It's also the reason I'm not using a dynamic language anymore...
+
+**Boyan Soubachov:** The way I look at it, at least - if it's something that's... How likely is it to change? Basically, how likely is your database state to change? How likely is your time to change? If it's anything other than low, take it out, mock it out, make it as a dependency that's injected. But that's a rule of thumb that I follow anyways...
+
+**Break:** \[41:03\]
+
+**Mat Ryer:** Talking about Testify then, Boyan, what do you see are the challenges with maintaining a project like Testify? What's easy and what's difficult, and Great Scott?
+
+**Boyan Soubachov:** Great question, Great Scott! \[laughs\] So one, due to obviously its size and popularity, a lot of people wanted to work for them, and that's fair enough. I think our goal is to have a framework that makes life as easy for as many people as possible. That being said, you have infinite different use cases for it.
+
+A big difficulty there is just how many different requests we have to add this specific feature, so that it works in my case, and this specific feature in my case... And we try and be as fair as possible and consider each case, but ultimately we're guided by what the community as a whole or as a greater have benefit from this. Also, a big thing is to always be vigilant of starting to couple it to be dependent on either some other technology or some other proprietary standard, or whatever... Because I think a great example was we've had a request to change how the framework works, so that it makes it easier to run in GoLand... Which would break it forever on who's not using GoLand... So that was a no-no in our books.
+
+**Mark Bates:** \[44:40\] One of the things that I often hear people say about Testify, obviously (because it's one of the big ones), but about any of the assertion libraries, is "Well, occasionally I'll hit a rough spot, or I'll hit something that doesn't quite work." Equal is always a great example of that. There's some really edge cases there. I ran into just yesterday an edge case with Testify zero, where I had an expectation that if I implemented *isZero() bool* which is kind of like an unofficial interface you'll find across a standard library (Time has it), I thought that if I did that and called not-zero, or zero on Testify, that it would look at my type, see that it implements the zero, returns false, and then is not -- you know, whatever. And that's not the case. And I even dug deep into Reflect, because it uses Reflect zero, and even that doesn't call that interface.
+
+So I hit one of those edge cases, and it took me a few minutes to dig through the code, figure it out, and then I said "Oh, man - okay, I really thought it worked that way. It doesn't. I'll just change the way I'm running my tests a little bit." And I just moved on. I don't know, I don't put a lot of credence into the "I hit one random edge case, so they're all terrible" argument. I find that if you hit one edge case, you've hit one edge case. If the rest of the time it's working just fine for you, that's definitely a baby with the bathwater kind of a thing.
+
+If I threw out all my tests yesterday because the Require zero did not follow the is-zero interface, then that would just be stupid on my part. That would be such a waste of time and effort and energy, because I hit one edge case...
+
+**Mat Ryer:** Yes. It's kind of like "Perfection is the enemy of good", isn't it?
+
+**Mark Bates:** Yeah.
+
+**Mat Ryer:** And I understand, if you're using a technology and it lets you down in some way like that, I completely understand the credibility gets lost a little bit. But Mark, your attitude with 90% is actually quite a healthy perspective to have, which is - it's good enough, isn't it?
+
+**Boyan Soubachov:** Agreed.
+
+**Mark Bates:** Yeah. And I'll keep adding the tests as things break, and I find those new edge cases; I'll keep adding new tests to cover them. But yeah, you've just gotta move on with your day at some point. And like I said, there was no way I was throwing out Testify because *isZero()* didn't behave quite the way I thought it did... Despite the fact that -- and don't forget, the documentation makes no claims to respecting that pseudo-random interface anyway. I made an assumption on my part, proved out to be wrong, and I just changed my test to \[unintelligible 00:47:23.02\] And that was fine; I just made that little change, but...
+
+**Mat Ryer:** \[47:33\] Yeah. I think that's right. When Testify was in its early stages we had this policy of anybody that contributed a PR was invited to become a maintainer. This was an idea that -- because it's a community-owned thing; it was completely open source, so it's kind of this community spirit and inclusiveness, and stuff. The effect of that, I think, in Testify's case, was that the API grew quite a lot. And if you look at the index page in GoDoc for stretchr/testify/assert package, it's massive. And of course, the Require is too, yeah.
+
+So what's the downside to having that much stuff there? There are pros and cons to it, so what are the pros? What are the cons?
+
+**Mark Bates:** I don't know that there are that many pros, but go on...
+
+**Mat Ryer:** Well, the pros are it does all those things that people wanted to do.
+
+**Mark Bates:** Yeah, but the cons - I don't know, I'm not gonna speak here for... I would feel the cons far outweigh the pros of a large API in this case, but...
+
+**Mat Ryer:** Boyan, what do you think?
+
+**Boyan Soubachov:** Exactly, obviously the pros are it covers a few specific use cases that some people have... But the cons are you have this code that needs to be maintained, and we need to make sure it doesn't regress... But also a big one is people just end up importing and polluting their package namespace.
+
+If you have a test and you want to assert, but you now want to figure out "Which assertion should I use with Testify?" I think that's why people just default to equal and start doing a lot of the hard work to mangle it into an equals assertion problem.
+
+**Mat Ryer:** Yeah... But that's quite a good strategy, because then at least the complexity is with the users, in their code; it's in userland. And that's something that I have learned over time, which is that if you can solve this problem easily enough in userland, i.e. outside of this package, then that's a good signal that it shouldn't be something that goes into the package. That wasn't a policy or anything, as you can tell by the size of Testify, I think... But I do wonder if we could use that BigQuery dataset to find out in open source Go code which methods are used and which ones aren't really used? And Boyan, what about the future of Testify? What about if we find out a lot of these methods aren't used ever, or are very rarely used, what about phasing methods out and tightening up that API?
+
+**Mark Bates:** Or doing a v2 that has a much tighter API would be the other idea. Like I said, I personally used maybe 10-12, and those are a mix of their pro and con methods, they're true/false variant. So error/no error. Nil/not nil. Equal/not equal. Maybe contains, if I'm looking at a big blob of HTML that gets dumped out and I wanna assert that various elements are in there. I'll use length...
+
+**Mat Ryer:** The contains one you can do by doing "assert true"...
+
+**Mark Bates:** *strings.Contains()*... Yeah, I could do *strings.Contains()*; I could do *equals()* on all of these but I don't wanna write the error message.
+
+**Mat Ryer:** Right, yeah. That's it, yeah.
+
+**Mark Bates:** That's why I said contains. But things like the delta methods and all that stuff, I'm like "That's too much for me", and it's very overwhelming and intense. Ultimately, my tests are so small that those 10-12 basic ones - true/false, nil, error, equal - they cut it. They do 90% to 95% of my heavy-lifting for me. And they're \[unintelligible 00:51:30.22\]
+
+**Boyan Soubachov:** Yeah, exactly. What we now discussed has touched on a lot of very good points, which is - we've heard a lot of feedback, we've seen a lot of people ask us "Why do you have this? Why do you have that?" and functions that we've even never seen before are there, people submitting pull requests on... So the future for us is -- I mean, myself and my two other co-maintainers, George and Martijn, we think it's about time to look at a version two. We've had so many breaking change requests, where we would break the API... We've got a nice, long list that we could put in a whole separate version, so that's why we started a trial for version two... And because it's community-first; we do this for the community, not for some ideals we have. If I were to write Testify myself, I know what it would look like... But I'm not writing it for myself.
+
+\[52:36\] Mat, as you pointed out, maybe we can do something with a dataset, but we mostly want to hear what the community's opinions are, rather than look at what -- because we can gauge what's used and what's not, at best, even if we could... But we can't gauge what's missing, or what the community would like to have changed. That's why we have a survey out, a simple Google form, if I may plug it...
+
+**Mat Ryer:** Please do, yeah.
+
+**Mark Bates:** Where can we find that survey?
+
+**Boyan Soubachov:** It's on cutt.ly/testify. And we would love to hear the community's feedback... And if it's time, and if it is according to the community to move forward on to a new breaking version two, what they would like it to look like. We're doing it for the community.
+
+**Mark Bates:** Can I ask you to put that link in the chat room for the show? As well as, if you haven't already, on the readme, for Testify. Right at the top.
+
+**Mat Ryer:** Yes.
+
+**Mark Bates:** Make it available there, because that's the place I'm most likely to go and see it.
+
+**Boyan Soubachov:** Awesome, thanks. Yeah, of course.
+
+**Mark Bates:** And that'll also end up on the package.go.dev as well too, if you put it in the readme.
+
+**Mat Ryer:** Yeah. So Boyan, thank you so much for that. I'm very excited about what a version two could be, because of course, it means you can have breaking changes, we can take things out, we can simplify things...
+
+**Mark Bates:** Just keep Require. Get rid of Assert. \[laughter\] Get rid of Assert; it's useless.
+
+**Mat Ryer:** I've just called time.now, and it is now time for the officially sanctioned Unpopular Opinion.
+
+**Jingle:** \[54:22\] to \[54:42\]
+
+**Mat Ryer:** So who's got an unpopular opinion for us today?
+
+**Mark Bates:** I've given you two today. I'm done.
+
+**Mat Ryer:** Come on then... \[laughter\] Mark, you've given about 12...
+
+**Mark Bates:** I have nothing but unpopular opinions. This segment is my life.
+
+**Mat Ryer:** \[laughs\]
+
+**Mark Bates:** I'd be harder pressed to find a popular opinion.
+
+**Boyan Soubachov:** At the risk of going a bit philosoraptor and meta - and this is tangentially related to testing, but frankly, the code that we've had the least issues with, that's the most reliable, that has the most positive feedback, is the one that rarely changes and that we haven't touched in months, if not years... Well, years I mean. So frankly, we do this because change, because the industry is growing and moving and progressing, but sometimes it's just best to leave things alone. \[laughter\]
+
+**Mark Bates:** Would it be easier if we slowed the industry down, and stopped progressing? Would that make maintaining the library easier for you, or...?
+
+**Boyan Soubachov:** It actually would...
+
+**Mark Bates:** Okay. Mat, can you get on that?
+
+**Boyan Soubachov:** I'd have my nights free. I can watch movies again. \[laughs\]
+
+**Mat Ryer:** This is not Ruby, Mark. You can't just choose what time it is. \[laughter\]
+
+**Mark Bates:** It's not even choosing what time it is, it's just overriding whole methods, and types, and fun stuff... \[laughs\]
+
+**Mat Ryer:** \[55:57\] Yes... Well, I think that's a great one, actually. Boyan, there's definitely something in there. I think a lot of what we do is quite weird as software people, but it's best not to think about it too much and just get on with it.
+
+Thank you so much. We have reached the end of today's show, but thank you... And anybody at home that's interested in talking about Testify, go to the project, check the show notes, have a look in the GoTimeFM Slack channel... We'll be leaving a link to the survey there for you to do. Boyan, thank you so much. Especially, thank you for getting up so early.
+
+**Mark Bates:** Or going to bed too late.
+
+**Mat Ryer:** Thank you for all the work you do to support the community.
+
+**Boyan Soubachov:** It's a pleasure.
+
+**Mat Ryer:** Sorry, Mark.
+
+**Mark Bates:** No, I was just making a joke about him going to bed late... \[laughter\]
+
+**Boyan Soubachov:** I had to make a special concession and go to bed early last night... Because I'm in the future, it's Friday for me.
+
+**Mat Ryer:** Oh, so it is Ruby!
+
+**Mark Bates:** \[unintelligible 00:56:57.25\]
+
+**Boyan Soubachov:** It is Ruby land in Australia right now... \[laughs\]
+
+**Mark Bates:** I bet if we all did time.now, we'd get different answers... Someone clearly didn't write a unit test for timezones...
+
+**Boyan Soubachov:** That's another unpopular opinion... \[laughs\]
+
+**Mat Ryer:** Get rid of timezones. We don't need them now! Get rid of them, I think...
+
+**Mark Bates:** Timezones? Who are they for? What's the deal with timezones?
+
+**Mat Ryer:** I don't mind waking up at 20 past G, or whatever...
+
+**Mark Bates:** 20 past G?!
+
+**Mat Ryer:** ...if we wanna change it and have a new system.
+
+**Mark Bates:** Did you say 20 past G?
+
+**Mat Ryer:** Yeah, yeah. I think we should use letters for the hours.
+
+**Mark Bates:** I think we should move to a 100-hour based day, and go full on metric with the decaday...
+
+**Boyan Soubachov:** Mat, is your unpopular opinion today forever on old time measurements to move to Unix time?
+
+**Mat Ryer:** \[laughs\] That would be good... I'd love to see someone with a Unix time watch. That's something I've not seen, and I would like to see that.
+
+**Mark Bates:** When the Apple Watch first came out, my immediate thing - I jumped into Xcode, which I'm not an Xcode programmer, and I was like "Oh, let me make an Apple Watch face that's just the Unix time. I'm sure some geek would buy that for $1.99 in the app store." You can't do that. You can't make custom watch faces... Or at least you couldn't then.
+
+**Mat Ryer:** You might be able to now.
+
+**Mark Bates:** I think you can make a complication.
+
+**Mat Ryer:** You can do the complications, the little things on it...
+
+**Mark Bates:** Yeah, but that's not the same...
+
+**Mat Ryer:** It'd still be good. You could have the Unix time on your watch.
+
+**Boyan Soubachov:** It's only useful for another 18 years though... \[laughs\]
+
+**Mark Bates:** Exactly.
+
+**Mat Ryer:** Thank you very much everyone for listening... See you next time!

The latest on Generics_transcript.txt

@@ -0,0 +1,257 @@
+**Carmen Andoh:** Welcome, everybody, to this edition of Go Time. We have a very special episode for you today, the latest on generics. Here with us are Robert Griesemer and Ian Lance Taylor of Google's Go team. Good afternoon!
+
+**Robert Griesemer:** Hello, everybody.
+
+**Carmen Andoh:** And also with us are our hosts with the mosts, Johnny Boursiquot...
+
+**Johnny Boursiquot:** Hello.
+
+**Carmen Andoh:** ...and Jon Calhoun.
+
+**Jon Calhoun:** Hey!
+
+**Carmen Andoh:** And I'll be running point today, I'm Carmen Andoh. Welcome, again, to Go Time. So I guess let's start with the updates. We have Ian on here. I think it was about last October, discussing what was then the GopherCon presentation given on templates... And since then, a new draft proposal has come out, so maybe you can talk a little bit about what that update is, Ian?
+
+**Ian Lance Taylor:** Sure. So Robert and I released the updated design draft for moving forward with generics. The biggest change was that we dropped the idea of contracts, and just decided that instead of having a separate syntactic construct, which was a contract that we could just use interface types to describe the contract between the type argument and the type parameter. A lot of people looking at contracts had seen that they seemed a lot like interfaces, and people had trouble separating out exactly when you would use a contract and when you would use an interface... So we simplified this - and this was, I should say, almost entirely due to Robert - to just use interface types.
+
+And then the second big step we made was we've released a translation tool and a type checker. So we have a type checker that works for the design draft, the description of generics in the design draft, so that gives us some confidence that what we have written about can actually work...
+
+\[04:11\] And we have a translation tool which translates code into ordinary Go. The translation tool is not, by any means, a final thing. There's cases it doesn't handle. It's just an experimental tool, but it lets people actually write code that can actually run using generics. So we can get a feel for whether generics actually works for people, and whether it actually addresses the issues that they have.
+
+**Carmen Andoh:** Can you talk a little bit about what your understanding is based on the feedback so far of what people's issues are when it comes to either lack of generics or this current draft proposal?
+
+**Ian Lance Taylor:** Well, a lot of the feedback has actually been about the syntax, which is sort of simultaneously the least interesting and the most accessible part of the proposal. Obviously, it's really important to have a good syntax, and we're really paying attention to the feedback that people have given. We actually now have two possible syntaxes implemented in the translation tool, which Robert did...
+
+And beyond that, there's the semantics, of course... And I think the feedback on the semantics has been quite positive so far. People have tried out generics, they've written some pretty extensive pieces of code, and I think the feedback we've gotten there has been uniformly positive. Robert, do you remember any real concern to this point?
+
+**Robert Griesemer:** No, I think most of the feedback was really based on syntactic issues, and we tried to address this with this alternative that we have now... And we need to play with this a little bit more. And then on the semantics side - of course, there's not everything ironed out quite yet. We have even mentioned that in the design draft, especially when it comes to type lists and exactly what does it mean to have for instance type parameters in the type lists, and things like that.
+
+So we are in the midst of refining that, but I don't recall off-hand right now that this was a primary sticking point with feedback from the people.
+
+**Ian Lance Taylor:** We're trying to sort of really pin down the real semantics of type lists, as Robert said. But we haven't had much feedback. There's been some people wondering about the exact details of embedding a type parameter inside a struct or an interface, which we need to decide precisely what it means... But as a matter of fact, almost no one, in terms of actually using generics in practice.
+
+**Carmen Andoh:** Speaking of semantics, one of the things that you did in order to inform this new draft proposal version is team up with type theory experts, including Philip Wadler, and you came out with a paper called Featherweight Go, which I don't understand a word of... \[laughter\] And I actually am giving a bounty to the entire Go community for anyone who wants to try to demystify that. I know that there is a panel that you were on with Phillip, trying to talk about that paper and pairing up with that... But maybe you can try to demystify Featherweight Go maybe in its essence, for our listeners right now. Do either of you wanna try to take a stab at that?
+
+**Robert Griesemer:** I can try. Maybe not a real stab, but maybe a little explanation. So this cooperation, I should say, happened because Rob Pike actually reached out to Phil Wadler. They knew each other from way back, and Phil Wadler was interested, and then we started talking with Phil Wadler and with Rob Pike... He didn't really have time to produce a beta in this, but then Phil Wadler, Ian Taylor and I started talking about what we wanna do... And they have, of course, a strong background in type theory. Phil Wadler has done this -- not the same work, of course, but similar work, many years ago, for Java, so he's really an expert.
+
+\[07:56\] So now we have a whole team - it's not just him - that have been working on this Featherweight Generic Go... Which is based on Go, but very much slimmed down. So we now have a language that really only has type declarations and methods. And those type declarations are only interfaces and only structures (structs), and the only thing you can have is methods and interfaces, of course, and methods associated with structs... And inside those methods you can only have basically single functional expressions. So it's a very, very simplified language, but what you can do is you can invoke a method.
+
+In this paper, he explores two situations this Featherweight Go, which is like the basic Go, simplified Go, without any generics \[unintelligible 00:08:45.28\] which is that basic, simplified Go, extended with generic features. And those generic features are very much modeled along the draft design, with the exception of type lists. So those type parameters, as in the draft design - the type parameters have what in the paper is called type bounds (we call them now constraints), and there are interfaces, there are also interfaces in the paper, and they basically model in a very simplified fashion what the design draft is trying to do with real Go.
+
+The goal of this paper is to prove, first of all, that this is a sensible design, in the sense that the type system that we're creating here is sound; you cannot create situations where you could write a program that would be unsound in terms of the type system. And then also, they tried to prove and have proven that it is possible to translate such a simplified generic Go program into a regular Go program, through a process which is called monomorphization, basically expanding everything for every possible instantiation of these generic functions and types... And then they prove that these programs are basically \[unintelligible 00:10:06.22\]
+
+That's the gist of this paper, and this gives us very strong confidence that they're a) not designing somewhere into the blue, and b) that what we're designing actually makes sense from a type system point of view. We're not gonna hopefully find problems down the road, where we have some internal inconsistency. So I think that's really the benefit...
+
+I think it really helped us also understand a little bit better what it means to have interfaces as constraints, and how we need to type-check this. So I think there is a real synergy here.
+
+**Carmen Andoh:** Very cool. And would you say because of the partnership it helped you understand some of the semantics better in terms of how it informed the newest draft proposal?
+
+**Robert Griesemer:** Oh, absolutely. I think we had part of a prototype working before this paper was complete, and in the process there's of course all kinds of questions, and what we sort of invented ad-hoc, they basically did in parallel, independently. Then when we started talking to each other, especially when we went through the individual steps in the paper, we could basically verify that our thinking was matching their thinking, and vice-versa. So our ad-hoc design that was maybe more based on what seemed to be the right thing to do - and not so much driven by a mathematical background - matches... And so that's great; that means we're not doing something weird.
+
+Phil Wadler actually took the time himself to walk us through the paper in detail, and this is how our understanding of the paper came about. I'm not a type theory person... So I now feel like "Okay, I have some sort of idea how to read the math", but I --
+
+**Ian Lance Taylor:** \[12:04\] Actually, my name is on the paper, but I have no claim to understanding the paper at all. \[laughter\]
+
+**Carmen Andoh:** Okay, you've heard it here, folks - it's not just you; even Ian and Robert had a really hard time with the Featherweight Go paper and all that math notation. This is funny.
+
+**Ian Lance Taylor:** Let me add that they really helped us with the move away from contracts and towards interface types. They pushed us in that direction, as Robert did as well... So it helped make clear that it would be equally powerful, and it would be usable.
+
+**Johnny Boursiquot:** It's worth noting that there is a YouTube video of Phil actually walking through portions of this as well... And I attempted to read the paper; I couldn't get very far, but I did watch the video and he did an excellent job walking through some of the key concepts there. I believe there is a plan for an expanding on the ideas of Featherweight Go, and I believe there's a \[unintelligible 00:13:00.06\] or some other implementations these I imagine are going to build on the foundation you have now, to try and figure out "Okay, what else haven't we thought of, with the Rust generics, in Go?" Is that correct?
+
+**Ian Lance Taylor:** Yeah, that's exactly right. They're gonna try to add -- I mean, as Robert said, Featherweight Go really is very limited, so they're gonna try to add other features of Go in and make sure that this type system is still sound, and the generic system is still sound. Which we think it will be. But you know, it will be very interesting to see what they come up with.
+
+**Johnny Boursiquot:** Something you mentioned earlier was that some of the feedback was somewhat superficial, in that it was limited only to the syntax; obviously, there's a lot more under the hood that really must be solved in order to really have a consistent implementation. But to do that some justice, you have folks that are sort of familiar with a common utilization of angled brackets as the way of specifying the generic types and whatnot... And people were scratching their heads, "Well, first it was in parentheses, now they're thinking about square brackets... Why can't we just do angle brackets? What's the problem?" So maybe if we can just kind of explain the dangers of trying to put that into Go right now - maybe that addresses somewhat some of that early feedback.
+
+**Ian Lance Taylor:** So Robert had a really good example in the email he sent out about square brackets, showing a case where you really can't parse angle brackets if you don't know whether you're looking at a generic function or type, or whether you're looking at a pair of expressions that happen to have a comma in the middle because it's some kind of multiple assignment. So that's just sort of a fundamentally ambiguous syntax. Robert, do you wanna talk about the importance of parsing without type-checking?
+
+**Robert Griesemer:** Sure. So even in existing Go we have some situations where we do not know at parse time what we have. The classical example is if we have a conversion or a function call with exactly one argument. So if you say f(x) is this a function call, or is it a conversion? We do not know at parse time. But it doesn't matter, because we can build the syntax tree at parse time. That's the only thing that matters. And that syntax tree that has a functor - maybe it's a type in a conversion... And a list of arguments. That's all we need. And that's the same for a function call or a type conversion. And then at type-checking time we can look at this functor and see "Is it a type? Well, then it must be a type conversion." Or if it's a function - then, well, it must be a function call. So that's all jolly.
+
+The problem with the angle brackets is that at parse time we cannot even know how to create a syntax tree in this specific case, especially in that example that we've given in the mail. We don't know how to parse this, so we don't know how to build a syntax tree, and that means there's just no way forward to resolve this.
+
+\[16:10\] One way to go forward would be if we had type information at parse time. In languages such as C++ where they use angle brackets for templates, there is symbol information at parse time, and the parser needs that to make the right decision. But that also means that everything that you will use at a particular place needs to be already declared at that point. And so in C++ you need to make sure that everything that you're using in an expression has been declared before, some way or another; maybe with a forward declaration of sorts. In Go we cannot do this, because -- well, we could, but we don't have forward declarations in Go, and we don't want them. In the very first version of Go, that's never seen the day of light, there was actually forward declarations, but we got rid of them pretty quickly.
+
+In Go you can have a package that is spanning multiple files, and so if you refer to a function in one file, that function may not even be declared yet. It may show up at the very end of the last file that the parser is gonna see. So there's just no way to have this information available. So without that information we cannot -- it's just not parsable, and there's no way around it... So angle brackets, as it is right now - it cannot work. So it's not like we don't wanna do it, or we don't like them, it's just it cannot work with the Go as it is right now.
+
+**Break:** \[17:37\]
+
+**Carmen Andoh:** So what Robert is referring to is rather a discussion thread on Golang Nuts, and it is one of the most recent, and it's an addendum to their draft proposal, where syntax feedback had been received, and a new addition was adding parentheses. So now it's a case of getting feedback from the community about whether the preference is for square brackets, versus parentheses.
+
+Robert, or Ian, what are some of the trade-offs for either/or of these, in your mind?
+
+**Ian Lance Taylor:** The advantage of parentheses I think is that type parameters really are parameters, and type arguments really are arguments... So it makes sense to use a syntax for type parameters that's similar to the syntax for regular, non-type parameters. \[unintelligible 00:19:16.07\] And I find the result to be -- it sort of feels natural to me, at least... And it reads well. The disadvantage is in a complex generic function you can have lots of parentheses flying around, you can have type parameters, you can have regular parameters, you can have result parameters; they're all parenthesized lists, and it can get a little hard to see exactly what's going on.
+
+\[19:47\] Also, in a call, at the call site, sometimes you pass type arguments, sometimes you don't, and it can be a little unclear, again, exactly what's going on. Like, if you had a new function, the new function might take a type, and then you have to have another set of parentheses for regular parameters... So there's some potential confusion there.
+
+We've also discovered certain ambiguities with parsing when using parentheses. Not common cases, but cases that do arise in real code. There were cases where it was ambiguous when you referred to an instantiated type, or an instantiated function, and it was hard to know exactly what was going on.
+
+A simple example would be an embedded field inside a struct. You can embed and instantiate a type in a struct. It's not really clear whether you're embedding a type in a struct, or whether you're doing some other kind of operation there.
+
+So square brackets, by comparison - they're still a list syntax, so now type parameters and parameters don't look the same, which for many people is an advantage, but for some people it's a disadvantage... And then there seem to be at least so far fewer parsing ambiguities when using square brackets.
+
+I don't have a good feel right now for the sentiment of the broader Go community... There are definitely people who like parentheses and definitely people who like square brackets. I don't have a clear sense as to one clearly being better than the other, but I have seen a lot of people saying that either could work; that they don't have any major objections. Do you have anything to add, Robert?
+
+**Robert Griesemer:** No, I think that's an accurate description. I think we can confidently say that the square brackets don't have the ambiguities at this point that we've seen with the parentheses. We did not know this in the beginning, we only found out after writing that code, where we ran into problems... But we decided to stick with the parentheses because we wanted to make progress on all the other fronts. The reason for this alternative now is that we're revisiting this decision so we can make sure that at the end we have looked at all the alternatives.
+
+**Johnny Boursiquot:** Anecdotally, from what I've seen in the Twitterverse and the feeds and all these things, I think there is a penchant towards the square brackets. From what I hear, most think it's more readily apparent what is going on when you look at the code. You don't have to do a double-take - "Okay, what's applying to the parentheses here?" We can more easily, very quickly figure out "Okay, this has to do with the generic type, and everything else is what I would expect." Basically, that's my two cents there, from what I'm seeing from the feedback from the community so far.
+
+**Robert Griesemer:** Great. Thanks.
+
+**Jon Calhoun:** So I do have one question on that... Have you guys talked with developers of IDEs or syntax highlighting tools that people use, to see if any of them have feedback on any of this sort of thing? An example is JetBrains - since they have GoLand, I assume that maybe they'd have some feedback on which one is easier to make obvious inside the editor, where people are actually coding... So have you had a chance to talk to people who are developing tools like that, to get their feedback?
+
+**Ian Lance Taylor:** We have talked to the people who have developed the Go Please (gopls) language plugin... And from their point of view, I don't think it matters that much, because they're just hooking in for the parser, and the parser does support both cases, and the parser is just gonna feed back to them what the code looks like... So they didn't have much trouble adding parentheses support, and they've just recently added square brackets support to Go Please as an experimental thing... So I think at least at that level it hasn't been a problem. We haven't talked to JetBrains, though; that's a good idea.
+
+**Carmen Andoh:** So more practically, for the Go community - when do you think that you're going to get enough feedback to move forward with moving from a draft proposal to actually putting it forth as a proposal to change in the language?
+
+**Ian Lance Taylor:** \[23:51\] We don't have any timelines in mind, I'd say. As we mentioned earlier, we're still trying to pin down some of the precise semantics, which I don't think is going to affect any existing code; in fact, I'm sure it's not gonna affect any existing code. We wanna make sure that we understand and we wanna make sure that the multiple Go compilers will all implement the same thing. We're gonna have to have some sense of how to add to the language spec. So those are the steps we're looking at now. We're certainly gonna move forward as fast as we can before making a formal proposal. Of course, at that time none of it will be a surprise. People will have seen all of the ideas already, and we'll just have to see how it flies. So far, I feel like the reaction has been largely positive, which is encouraging... But I don't know exactly what the timeline is going to be.
+
+**Carmen Andoh:** What kinds of feedback at this point are you, Robert and Ian, looking for?
+
+**Robert Griesemer:** I think we wanna see things that don't work, and that you would expect to work. We definitely have seen some of these things, which tended to be just bugs in our prototype... And we have spent some time fixing those bugs, and we have slowed down a little bit on that, because it's just a prototype, and at some point you have to make a call as to how much time you wanna spend on that, and making progress elsewhere... But yeah, generally I think we would like to know "Can you write the kind of generic code that you expect to be writing with this design?" And if not, why not? Let us know. Is there things that you expect to be working, but they don't? Are they fundamental to our design? And if so, is there something we need to do?
+
+These are the important questions that we should try to answer ASAP, because once we have something more firm, it's gonna be very hard to make these changes later, if not impossible.
+
+**Ian Lance Taylor:** Yeah, I totally agree. And I'd say - is there anything you find surprising? If you're looking at code that's written using generics and you read that code, is there anything that you just say "I don't get it", or "It just doesn't act the way I expected it to act" - that kind of feedback would be extremely useful.
+
+**Jon Calhoun:** Were there any specifics that you guys wanted to avoid supporting? Like, do other languages have something in generics that you looked at and decided "This isn't something we want to support?" or maybe it's just like a really obscure use case, or something...
+
+**Ian Lance Taylor:** Well, my favorite example is that C++ templates are in fact a Turing-complete language in and of themselves, which is really cool... And we decided we absolutely did not want to support that, in any way whatsoever. \[laughter\]
+
+**Carmen Andoh:** One of the things that I happen to know about both of you is that you have extensive experience in generics coming from other languages... Ian, you were on the C++ readability team at Google, when you famously saw a spec for the Go language and wrote a compiler for it, I think... And Robert, you were on the V8 team that was writing a VM for Java... So are there any things from those lenses of experience where you kind of want to make sure you prevent -- and I'm specifically trying to get the angle here of the naysayers who say generics are gonna add complexity to the language. What is your experience about keeping that complexity at Bay in this proposal?
+
+**Robert Griesemer:** So just to clarify - I was working on V8 in the very beginning for maybe barely a year... I didn't really do anything with generics in V8. I was on the implementation side. My experience with generics was maybe C++, with templates, and probably the highest point there was when I was able to - as Ian alluded to before, it's Turing-complete - write a program using C++ templates that would decide whether a constant was a prime number or not... And the compiler would decide it at compile-time. So that's not the kind of thing we would like to support.
+
+\[27:54\] With respect to what I'd like to see or not see - honestly, I'm worried about the kind of code that people are gonna write. There's no question about that. And we see some of the examples that people send us, that cause crashes in the prototype, and they're just unbelievably convoluted and really hard to decipher... But as other people have pointed out, those people are really pushing the envelope; they're trying to just see "What can I do with this thing?" And I hope this is not gonna be the kind of code that people are gonna write down the road.
+
+I think one of the first things we need to do if we have this for real - we need to come up with a kind of best practices guide, that guides everybody a little bit as to how you should use generics, and when you should use them and when you should not use them. Very similar to what we developed for goroutines and channels; in the early days of Go everybody was using goroutines and channels for everything, and it took a little while for us to learn where it was appropriate and where it wasn't.
+
+**Ian Lance Taylor:** I agree completely. And I'll add that -- I think that one of the things we've been really focusing on during this whole multi-year process is to avoid the complexity that exists in C++, and to a somewhat lesser extent in Java... Because those languages are very powerful, and at the same time, they can lead to code that people find difficult to understand. And that's just not a good fit for Go. And partly, it's because they are languages that are much more object-oriented than Go is, that have inheritance built into the language, and therefore they have to reflect inheritance in their implementation of generics... Which then leads to considerable complexity in understanding "How do you choose the exact type that's going to be used to instantiate this C++ template?" And then they also have overloading, so we have to do overload resolution...
+
+These are all really powerful techniques that let people write quite compact code, that can be extremely effective... But at the same time, novices come in and they just don't understand exactly which type is gonna be used... So we wanted to make sure that we avoided that as much as possible. We wanted to be very clear exactly how a generic function or type was going to be instantiated, and what the type arguments were going to be.
+
+**Johnny Boursiquot:** It's just like we saw with the goroutines, and the overuse of channels for everything... It was kind of crazy there in the beginning. But I'm letting you know right now, you are gonna see that surge; there is gonna be that spike where everybody wants to use generics for everything under the sun... And then we're gonna start walking that back and developing best practices.
+
+I'd definitely like to see some leadership from the Go team on that, maybe an expansion of effective Go on the Go blog, and adding some of the ways to caution, to provide along with that, that says "Hey, this is really the best use for this", and whatnot. And obviously, I think a lot of the community members are gonna be stepping up and writing blog posts and showing the do's and don'ts, and that kind of thing... So it will definitely not be all up on the shoulders of the Go team but it is something that I think is to be expected; like any new toy, everybody's gonna try and abuse it... But it's all in the spread of implementation, right? I think we are going to develop best practices around what should your generic code look like for a production system, for when you're the one who steps away from it and somebody else has to step in and understand and read what is going on... I think that's something that is going to come... I hope.
+
+**Ian Lance Taylor:** Yeah.
+
+**Carmen Andoh:** Yeah, I think one of the things that I really like about the Go culture is that we have managed complexity, our idioms through the culture. We have absorbed values and simplicity, and we can continue to do so for generic Go. I think that will be a big part in whether we keep complexity at bay. It's not necessarily the technical enforcement, but the idiomatic/cultural enforcement. We've talked about idioms as culture on this show before, so it's kind of interesting to see it play out with generics.
+
+\[32:02\] One of the questions that we had was a forward-looking question, and it's assuming that this is gonna be a proposal that gets ratified and put into the language... Are there any plans for managing the surge of feature requests for the standard library, now that generic data structure and algorithms are possible? Are you gonna let that happen in the ecosystem?
+
+**Ian Lance Taylor:** That's a great question. Are there any plans? I'd say no. There are not at this point any plans... But there will be plans. The experimental translation tool does come with a tiny little set of sample libraries. When I wrote those, I viewed those as kind of prototypes for what we might wanna add to the standard library going forward. I don't think they're great examples or anything, but I think that they can sort of show areas where we might wanna add new standard library packages, and show possible implementations as subject to people really looking at them and making sure they make sense. I don't expect there to be a lot of additions to the existing standard library packages. There might be a few, but most of the existing standard library packages were written without generics and they work fine.
+
+So yeah, there may be a lot of people saying "What's that? Generics?", but you know, the truth is they work already, and we don't need to add generics there. I think it's more gonna be a matter of adding some new packages to really take advantage of generics. The translation tool, for example, has a slices package, which has various functions that operate on slices of any type... And it has a chans package that operates on channels of any type. This is the kind of code which we aren't able to write with Go today, but we are able to write with generics... So I feel like that's gonna be where we're gonna add to the standard library. We're not gonna be moving fast on any of this, for sure... But you're right that we should develop some kind of framework for how we're gonna add packages.
+
+**Break:** \[34:00\]
+
+**Jon Calhoun:** One of our listeners in the Slack had asked "While you're collecting feedback, is there a good time or a good expectation for measuring how build speed changes over time?" Especially now that things are experimental, I assume that it's not really fair to assume that that's exactly how things are gonna be whenever it actually ships... So what should people expect and when is the right time to give feedback on that?
+
+**Ian Lance Taylor:** \[36:11\] Okay, that's a great question. Yeah, the experimental tool has no similarity whatsoever to any real implementation. We know it's slow, and it's gonna be slow, and that's just inevitable... If this does move forward to become a proposal and it gets accepted, then most likely the implementation will be to start with a branch of the main Go toolchain, and we'll start adding generic support on that branch, which will involve changing the compiler mainly, and any other changes to other tools that are required... So that'll be the time to start giving feedback about changes to build speed.
+
+We've talked about it with some of the compiler developers, like Keith Randall especially, and we think we can do it without a significant increase in build speed. I mean, there will be some increase in build speed; we don't think it's gonna be a huge increase... But you know, this is really speculative at this point. So the time to give that feedback is when we're able to start doing development... And hopefully, people will also be able to contribute work when we start doing that work on the public branch.
+
+**Carmen Andoh:** We can ask other questions there on the Slack channel... "When generics are released, would it be released together with the idea of best practices?" What they call it here is a standardized collections interfaces, to avoid fragmentation... "Or do you think this will not be a problem, for example size versus count versus length?"
+
+**Ian Lance Taylor:** Yeah, it's a little hard for me to visualize exactly what a standard collection interface will look like in terms of Go. Clearly, it exists in C++, and maybe we can just borrow what they do. I'm not quite sure exactly how that's gonna work. But certain things are pretty straightforward. I think iteration is the more complicated one. C++, in part because of the sophistication of the template capabilities is able to have a generic iteration interface that works for any collection, and I'm not sure that we're there yet with our less powerful and less complex... On the other hand, I don't know that anyone's really looked at it in detail, so maybe we can make some progress in that area.
+
+**Robert Griesemer:** There's also a difference in the sense that in Go we already have, for instance, maps, which of course is not the whole collection hierarchy, but it's a significant chunk, which doesn't exist in C++, so there is a temptation there to build this very complete hierarchy... While in Go in many cases we just use a map for various things, and that works fine; and the map isn't sometimes generic already. But of course, that's not to say that this would replace a more comprehensive package perhaps...
+
+**Ian Lance Taylor:** It's a great question though.
+
+**Robert Griesemer:** I think it's gonna be interesting to see what people will do with this. The design - we tried to make it as orthogonal as possible to the rest of the language. That really means adding generic type parameters somewhere adds a new dimension to the kind of programs you can write. So it really opens up an entirely new dimension of possibilities, and it will be very interesting to see what people do with this.
+
+**Carmen Andoh:** One other comment from the Slack is a technological guard rail, if you will... So for the language, currently govet and golint function as sanity checks we depend on. What is your opinion on adding new checks added to discourage the use of generic code in this toolchain, if all you need is a non-generic version?
+
+**Robert Griesemer:** Well, we can easily make the compiler two times slower for each additional type parameter; that would limit the complexity quite quickly...
+
+**Johnny Boursiquot:** \[laughs\]
+
+**Carmen Andoh:** It's really gonna have to probably be enforced through culture, or these best practices and idioms, right?
+
+**Robert Griesemer:** \[39:47\] It's quite easy to see that programs have one or two type parameters and functions... But if it goes over two or three, then you are starting to wonder "What's happening here? Is this really necessary? Is this really good?" So I would say there's some immediate questions right there, when you look at code like that. But I'm shooting from the hip here; I'm just guessing here. And I suspect that there may be things we can say "This is not good." And maybe such things can go into a vet check eventually, but I don't know what that would be at this point.
+
+**Jon Calhoun:** I suspect things like the Go Proverbs will sort of help on this front, just because -- like, we have the ones like "A little copying is better that a little dependency", I think is one of them... And I think if people stick with that mindset of copying an entire type -- like, if you're only using two different versions of it, or an entire function or whatever it happens to be, it's better than writing the generic version for just two different types. But if you do find yourself in a situation where you need to use it for 5-6 different types, then generics might actually be the right solution.
+
+**Robert Griesemer:** Mm-hm.
+
+**Johnny Boursiquot:** The rule of three still applies here... I think we've discussed this previously on the show; usually, I don't even start to think about abstracting away or making a generic version of anything until I've seen at least three instances of it, if you will. Then I start to say "Okay, I'm starting to see the edges here. I'm starting to see what is the likelihood of this thing being used in the future." Because any code you write - again, we write the code just once, but we have to maintain that code, possibly indefinitely... So the same rule around those decisions we're making today, asking ourselves these kinds of questions - I think it's the same thing we should apply here.
+
+But again, you are gonna see folks who are anti-generics in our community. I'd say it's like a 50/50 split. We have people who love Go because of the lack of generics... Which is a very specific reason to love a language, but hey... And then you have others who are like "Hey, if Go had generics, I'd be using it." So it's kind of an odd mix, but again, I think it's gonna be up to the community as a whole to come up with what are the idiomatic ways really we wanna treat generics.
+
+**Jon Calhoun:** I think those would be nice too, because if we have a concrete implementation to point to, even if you just copy it to the docs, for the comment, and say "This is a generic implementation of this concrete thing". I think anybody who is unfamiliar with generics, or a little bit newer to programming, could look at that and have a much better chance of understanding the code.
+
+**Carmen Andoh:** One of the things I think about for generics if this does get put into the language is whether or not it's going to bring on what I would call the would-be gophers. Right now we have the die-hard gophers, people who love Go and use Go and maybe have internalized the no-generic philosophy of Go... And having generics or implementing generics into the language successfully may bring new gophers to the table, and those new gophers are gonna be your .NET and your C++ and your Java gophers. Robert and Ian, do you have any thoughts or suggestions on how to help them keep Go Go, while also giving them this tool that they've leaned on so heavily on coming from their other language?
+
+**Ian Lance Taylor:** Yeah, I hope they will be able to carry their programming practices over where those practices make sense... Where they make sense for Go, I should say. I can't tell whether adding generics will make it that much more appealing to C++ or Java programmers; I certainly hope it will... But whether it really will in practice - it's just hard to know. It's still gonna be a different language, of course.
+
+Certainly, there have been people in the past who have just rejected Go outright because it doesn't have generics... But I don't think there have been that many people who take that point of view, and I hope that those people will take another look when generics comes out. But from my point of view, I want Go to be open to everyone. I want all these people to find Go to be a productive language. It's not really a matter of hauling people from the other language communities... I think generics is interesting mostly because, as Robert says, it's orthogonal; it lets us write code that we couldn't write before, in Go. It lets us write code that solves problems in a way that we couldn't really solve before, at least not without going through type reflection, or massive copying, or whatever.
+
+\[44:12\] So it'd be great to open more people; I hope that more people keep writing Go... But I think our main interest is to give people another powerful programming tool. At least my main interest.
+
+**Robert Griesemer:** Yeah, I'd add on that that, again, Go is not now becoming the generic language. It's just yet another mechanism in the language, like we have interfaces, we have methods... It doesn't mean now you have to write everything in an object-oriented style. Go has always been multi-paradigm; we enable different ways of programming, and we encourage people to choose the right approach for the problem at hand... So in a situation where a generic approach might be the right approach, then by all means, go for it; and if it's not, don't do it.
+
+Of course, there's gonna be people that really like that playing with types, and more so maybe than getting that code to just do what it's supposed to do. For them it's, of course, appealing to have this new mechanism. But again, if your goal is to get something done at the end of the day, use the mechanism, the tool that fits your needs and solves the problem in the greatest way.
+
+**Johnny Boursiquot:** Along those lines, that's actually a good segue for the segment of the show where we talk about unpopular opinions.
+
+**Jingle:** \[45:37\] to \[45:52\]
+
+**Johnny Boursiquot:** So what I'd like to do is ask each one of you to spill onto the mic your unpopular opinion. It could be related to what we've been talking around generics, or it could be something that you're seeing out there that perhaps you have a different opinion from.
+
+**Jon Calhoun:** Or if it helps, some people have just told us they think buses are more efficient than other forms of transportation in New York City. \[laughter\] So it can literally be anything.
+
+**Robert Griesemer:** Well, let me start on that... So I cannot opine on the buses in New York City; I'm a big fan of public transportation, even though this is perhaps not the right time to advocate for them, for other reasons... But unpopular opinions - I don't know if it's an unpopular opinion, but I like short identifiers. I do.
+
+**Carmen Andoh:** \[laughs\]
+
+**Robert Griesemer:** And I feel like the closer they are to where you use them, the shorter they can be. And the further away they are from where you use them, the longer they should be. And then there's some exceptions, like when an identifier is really important in your package and prevalent, then it can be one letter, even if it's a global. And the most prominent example for that is perhaps testing.t.
+
+**Jon Calhoun:** I'm not sure if you're allowed to be a teacher now.
+
+**Johnny Boursiquot:** \[laughs\]
+
+**Carmen Andoh:** Uh-oh... Why, Jon?
+
+**Jon Calhoun:** Every teacher expects you to write really long, self-explanatory variable names, regardless of where you use them -- at least that was my experience. I felt like every teacher wanted long variable names.
+
+**Robert Griesemer:** So I will comment on your change list if it uses -- in a simple for loop, if the iteration variable is called "index", I will probably comment on that. Call it i, or j, or whatever... \[laughter\]
+
+**Johnny Boursiquot:** Ian, what have you got?
+
+**Ian Lance Taylor:** \[47:48\] Okay, I don't know if this opinion is unpopular, but I feel like I write it a lot, so there's certainly people who don't seem to grasp it... And that's that the language is not perfect, but every change to the language carries a heavy cost. So when you wanna come and argue for why the language should be changed -- and we see that a lot; I'd say that there's probably one a day suggestion for some way to change the Go language... Don't just talk about how it makes the language better, but also spend some time to talk about how it makes the language worse. Because there's no such thing as a 100% good change to the language. Or I shouldn't say there's no such thing; maybe it's out there. Maybe no one has thought of it yet. But probably, it's a good bet that all the 100% good changes to the language have already been made. So when you wanna change the language, spend some time to think about how it makes things worse, as well as how it makes things better.
+
+**Johnny Boursiquot:** I think I just saw Ian drop a metaphorical mic... \[laughter\]
+
+**Robert Griesemer:** We're gonna go back now and think a lot about how generics make things worse... \[laughter\]
+
+**Ian Lance Taylor:** I think we've got plenty of people telling us that already...
+
+**Jon Calhoun:** I was gonna say, if you haven't gotten that feedback at this point...
+
+**Carmen Andoh:** Well, it has been a pleasure... On behalf of my co-hosts, Jon and Johnny, we thank you, Ian and Robert, for taking the time to talk about the new draft proposal on generics with us.
+
+**Ian Lance Taylor:** Yeah.
+
+**Robert Griesemer:** It's my pleasure.
+
+**Ian Lance Taylor:** It's been fun, thanks for having us.
+
+**Outro:** \[49:30\]
+
+**Jon Calhoun:** Carmen, I've heard a lot of if's from you today.
+
+**Carmen Andoh:** Some what?
+
+**Jon Calhoun:** A lot of if's from you today.
+
+**Carmen Andoh:** What kind of if's?
+
+**Jon Calhoun:** "If this makes it into the language..."
+
+**Carmen Andoh:** Well, okay - I think partly because of what I have learned with try last year... In that a lot of time was spent really thinking about this and trying to solve a good problem, and then a lot of the community felt like it was a fait accompli. And I probably am speaking for Robert and Ian here when I say if a lot. I think that the draft proposal is explicitly meant to not seem as a fait accompli.
+
+**Robert Griesemer:** Yeah. I mean, if there's something we learned from that is that we cannot just come out with a it doesn't matter how well thought out idea, and put it out there and say "So, yes or no?" It's just not gonna work. There needs to be an educational process, as we had to go through with too, and that we're trying to achieve now through being more open and going in smaller steps and getting everybody on board.
+
+It's kind of strange, because programming language evolution is really a social process. It doesn't actually matter if you have -- if you have seen the light and you know exactly the perfect language, you would just put it out there. And you know, maybe it's 20 years ahead - nobody would even buy it, because people would not see the reasoning why you got to that point.
+
+So you really have to get everybody along -- and some people may already be where you are, and some people may not, but you have to get everybody along in little steps, and that's how we eventually end up where we wanna be... And we can see this with all kinds of things, like garbage collection. Garbage collection was invented in 1950-something, with Lisp. The first Lisp had garbage collection; it was in 1958, I believe... And it's taken forever before it became accepted as something that a mainstream programming language should have. Maybe Java was the first one that really made it mainstream... And now this is not something that is still disputed or debated, I should say, but it's not as outrageous anymore. I think that's true for a lot of things.

The monolith vs microservices debate_transcript.txt

@@ -0,0 +1,579 @@
+**Mat Ryer:** Hello, and welcome to Go Time. I'm Mat Ryer. Today we're talking microservices versus monoliths. My co-host today - it's Jaana B. Dogan. Welcome back, Jaana!
+
+**Jaana Dogan:** Hey, how are you?
+
+**Mat Ryer:** Good, how's it going?
+
+**Jaana Dogan:** Good. Just self-isolating. What about you?
+
+**Mat Ryer:** Same, same. Self-isolating, and practicing -- I've been practicing for a few years now, so I'm really getting good at it.
+
+**Jaana Dogan:** \[laughs\]
+
+**Mat Ryer:** You know, this subject, microservices and monoliths often generates a lot of heat... But not today, not on this show, because Jaana and I have sourced two extremely cordial and polite young men. The first is Matt Heath, from Monzo. Matt is a senior staff engineer at Monzo, which is a bank here in the U.K, where he works on Monzo's microservices platform. Hello, Matt.
+
+**Matt Heath:** Hello! Thanks for having me.
+
+**Mat Ryer:** Thanks for coming, I really appreciate it. Don't worry though, you're not on your own... We're also joined by Tom Wilkie, from Grafana Labs. Tom's one of the Prometheus maintainers, and author of the Cortex and Loki projects. In his spare time he brews beer and tinkers with his 3D printer. Hello, Tom.
+
+**Tom Wilkie:** Hi, Mat. How are you?
+
+**Mat Ryer:** Did you like my sort of local radio intro?
+
+**Tom Wilkie:** It didn't sound scripted at all.
+
+**Mat Ryer:** It's not at all, no. What sort of tinkering do you do on your 3D printer?
+
+**Tom Wilkie:** Well, for your listeners, I can actually switch the camera over and show you right now what's printing...
+
+**Mat Ryer:** Great, yeah. The listeners are gonna be loving this.
+
+**Tom Wilkie:** They're gonna really enjoy the video of "I'm not quite sure what it is yet."
+
+**Mat Ryer:** The video doesn't go out, Tom, so this is just --
+
+**Tom Wilkie:** This is just for me, is it?
+
+**Mat Ryer:** \[04:08\] Yeah. The video doesn't get broadcast.
+
+**Tom Wilkie:** Okay... I'll put it back to my face then.
+
+**Mat Ryer:** No, don't worry. \[laughter\]
+
+**Tom Wilkie:** I don't know how much you know about 3D printing, but mostly my 3D printer prints more 3D printers.
+
+**Mat Ryer:** That's asking for trouble. I'm sure that's how Terminators started.
+
+**Matt Heath:** That sounds dangerous!
+
+**Mat Ryer:** Yeah... \[laughter\] I've got a 2D printer, and I was looking at it the other day and thinking "You know what - there just aren't enough D's."
+
+**Tom Wilkie:** Yeah...
+
+**Mat Ryer:** I'd love to have a 3D printer, mate.
+
+**Tom Wilkie:** I could print you one.
+
+**Mat Ryer:** \[laughs\] I love the idea. You can just print them. So I know both of our guests, and Jaana does, too... Some people have probably seen you doing the circuits in conferences and online, on videos, and doing talks... But for anyone that hasn't come across you yet, why don't we kick things off and get to know each other? Does anyone have any working from home tips? A lot of us are finding ourselves working at home these days... Has anyone got any tips for it?
+
+**Matt Heath:** I think mine are the fairly simple ones I've heard people say recently... Turn Slack notifications off; like, strongly off. But I can't describe how much more relaxed I feel now there's not a red bubble on my Mac. I can still check in, and I can still do all this stuff, but I'm just a lot more relaxed. I was really surprised by how much of a difference that made.
+
+That and just having a nice place to work. Get a plant, that kind of thing.
+
+**Mat Ryer:** Yeah, that's nice. I actually think you could go a step further. Since we're not gonna have any guests for a while, any of us, just do whatever you want with your furniture. You don't have to follow any of the simple, normal rules, I think.
+
+**Tom Wilkie:** What do you have in mind?!
+
+**Mat Ryer:** I mean, literally, move the bath into the living room. Imagine that. Just watch telly while having a bath, for example. But where you put your desk, where you work... You could carve out a new little corner... Sometimes you have to get creative with the space you've got, especially if you're living in the cities.
+
+You're not gonna have any guests for a while, so turn yourself around, if that helps. I don't know if that helps, but...
+
+**Tom Wilkie:** Face the wall.
+
+**Mat Ryer:** Yeah, maybe. But you can just do what you want, can't you? You're kind of free.
+
+**Matt Heath:** I mean, there's not much happening outside the windows these days, so...
+
+**Mat Ryer:** Yeah, that's true. So Tom, do you have a working from home tip?
+
+**Tom Wilkie:** I do, yeah, but it doesn't work anymore.
+
+**Mat Ryer:** Oh, really?
+
+**Tom Wilkie:** Yeah, so I've been working from home for 3-4 years now, and the trick was making it look like you weren't at home. Otherwise you took the password deliveries for the entire street.
+
+**Mat Ryer:** Oh... \[laughs\]
+
+**Tom Wilkie:** But now everyone's at home, it doesn't work.
+
+**Mat Ryer:** Well, you don't need also that service anyway, because everyone's in. They can just get their own passwords, too.
+
+**Tom Wilkie:** Exactly. That's what I mean. I'm glad you understood that.
+
+**Mat Ryer:** \[laughs\] Okay, great. So microservices versus monoliths then... Let's get into that. For anyone that doesn't know what these two things are, does anyone wanna give a broad overview, in highlight of what we're talking about here?
+
+**Matt Heath:** Yeah, I can give a brief intro. I'm sure you can all correct me... The way I say it is a long time ago, when I started writing software, I'd write fairly straightforward web applications, and I've have one application, one codebase, and I'd deploy it in one go. And that was really quite nice and simple, and you have one component. And that has lots of pros and lots of cons.
+
+And in my experience, as the applications got larger and larger, sometimes you want to break those things into smaller components. You can do that through classes, or various other approaches, depending on the language. Modules, obviously, in Go. And then I think the real difference is if you want to deploy them independently. That's certainly the approach that I've seen quite a lot, where you have a number of applications that are either relatively simple servers, but you can have many of them, and they all talk to each other through HTTP, or some other mechanism. You might have a small number or you might have a very large number of them, that would depend... \[laughter\]
+
+**Mat Ryer:** \[08:21\] Yeah, I think size is gonna be a recurring theme on this show, because I have a similar thing to you, Matt - I started off, I would just build little monoliths, really... And in fact, sometimes that's kind of where I start everything, actually. I start with just keeping it really simple, and then at some point you can start to feel some pain, can't you...? And you can see how microservices came to address that. What sort of pain points does it actually help with?
+
+**Tom Wilkie:** I think from a microservice perspective there's really a couple of key things that make it attractive for someone like myself. One is isolation. In the service that we run, it's very useful to isolate the read path and the write path. We run hosted monitoring services, so being able to put the right path, which is super-predictable, which is very steady and stable, and stateful, so pretty critical - being able to put that in one set of components, one set of service (for a lack of a better term), and then put the query path, which can be maybe dynamically scaled, can be a bit more risky, a bit more stateless... Putting that somewhere else allows you to separate out the blast radius of bugs, for instance.
+
+**Jaana Dogan:** Matt, you mentioned several pain points, and I'm seeing there are two different strategies. Some organizations or teams actually try to start thinking about microservices at the design time, whereas other start to introduce them as they see pain points. Do you have any experience in terms of when did you start seeing microservices coming around when you're just bootstrapping a new project?
+
+**Matt Heath:** Yeah, so I think starting with microservices at the beginning of a new project is probably quite a risky strategy... And that's actually what we did at Monzo. Rather than building on monolith, we started by trying to break the problem apart into lots of areas, and building services from that. That meant that we avoided - hopefully - many of the problems that we'd seen at other companies, around scalability, both from the organizational perspective and also the technology. But it means that it's quite hard to do if you're attacking a new problem domain and you don't have a really deep understanding of it. It's quite hard to draw those boundaries.
+
+In comparison, the previous place I worked, which was Halo, we went through a monolith to microservice migration. So we hit loads of the scaling problems, and microservices was an approach to getting out of some of those problems.
+
+**Jaana Dogan:** Yeah, it seems like there are two main concerns - scalability and organization scalability. Those are the two critical times that people start thinking about microservices.
+
+**Matt Heath:** Yeah, I think so. Certainly when I was working on a monolith, the main problems we had were many different teams working on the same codebase, and quite often not even isolated sections of the codebase, but working as a cross-functional team, on an area of the product... And that required you to change lots of different bits, and you kind of ended up changing the same bits as other teams. That really ground our development to a halt in some areas.
+
+**Mat Ryer:** But does microservices solve that, or does it just move that problem somewhere else? Because if you do have a component and you need to add some capability to a dependency, you kind of need to do that anyway, even if it's a microservice or if it's in a monolith. Of course, if it's in a little monolith, then you can get things like type safety... We can describe our boundaries with interfaces and things in Go, for example, and it's very clean... But when we deploy it, or if we build that as a big microservice architecture, does it really solve these problems, or does it just change them?
+
+**Tom Wilkie:** \[12:15\] I mean, there's a strong argument that changing things in a monolith is actually somewhat easier. It's deployed atomically, you deploy the whole thing at once, so interdependencies between the services are not something you really have to worry about... And you've got kind of control over the whole codebase. At any one point you can kind of atomically snapshot "This is the code at this point in time. This is what I'm gonna deploy."
+
+So there's an argument, especially with small teams and simple development processes, that monoliths will allow you to move quicker. I don't wanna be the one fighting the corner for monoliths - we also run microservices - but I feel like they get a bad rep sometimes.
+
+**Matt Heath:** Yeah, I definitely agree with that. If you have an interface that you want to refactor, it's much easier if you have one codebase, one application, and you can change both the implementation of that interface and also all the places that use it, in one go. And that's much harder to do with microservices. Unless you take very specific approaches, I think that's quite hard to do.
+
+**Mat Ryer:** So you mean roll your own tooling, or some other -- whether it's code generation, or some other mechanism that allows you to solve those kind of common problems; that sort of thing.
+
+**Matt Heath:** Yeah, so I think there's two aspects to it. One, if you have a monolith, you generally have one Git repository, or some repository of code, and all the code is in one place, so it's easy to change. And I've definitely seen lots of places where people use microservices where they have a different repository for every single service. We don't do that at Monzo. We have one gigantic monorepo. And that means that because it's in one repository, we can have a single pull request that does that same kind of change that you would normally do in a monolith, but we can do across 1,600 microservices.
+
+**Mat Ryer:** So the monorepo gives you that anyway.
+
+**Matt Heath:** Yeah.
+
+**Tom Wilkie:** Who had ten minutes before we started talking about monorepos? \[laughter\]
+
+**Matt Heath:** Yeah. It's on the Buzzword bingo, right? \[laughter\] And I think it gives you the type safety aspect as well. So if you're just implementing services with HTTP, you're not gonna get that type safety. But a monolith does give you that, potentially, and you have to put extra effort to get that with a microservice architecture.
+
+**Mat Ryer:** Yeah. Kelsey Hightower was on the show, and his unpopular opinion was actually that he prefers monoliths... And he was talking about -- because Tom, you mentioned it's deployed as a monolith, and he was talking about actually that is another choice you get to make. It is possible to -- you could imagine, even if you've never done it, some simple techniques. You could deploy -- it is the same thing, potentially, getting deployed, but with different parts of it switched on, or whatever... So I suppose you end up doing things like that if you don't wanna incur the cost every time of setting everything up, if this thing is only gonna be used in one context. Things like that.
+
+**Tom Wilkie:** I think the distinction between microservices and monoliths is something that's kind of overblown, to a certain respect. Like, you can definitely take a lot of the deployment aspects you're referring to and apply them to microservices. When we do a production deploy of -- I don't know for a lack of a better term, a macroservice, we tend to update all the microservices to the same Git SHA, so we know they all work together.
+
+But the nice thing about the microservice approach is that in between those big deploys - you know, we maybe do these big deploys once a day, or once every few days - you can tune individual services; you can maybe deploy a bug fix that only touches one service, on a case-by-case basis... Which is flexibility you don't really have with monoliths.
+
+But then just to throw a spanner in the works, you used to be able to do this with Erlang, right? Erlang arguably was a monolithic architecture, but you could easily just hotfix in a single function, right?
+
+**Mat Ryer:** \[16:09\] Could you? I don't know. Could you do that?
+
+**Tom Wilkie:** You could, yeah. This is when languages were powerful and let you do things like that. \[laughter\]
+
+**Mat Ryer:** Yeah. Well, Ruby lets you call any method you like. If that's power, that's where you wanna go.
+
+**Tom Wilkie:** Ultimate power.
+
+**Mat Ryer:** Yeah. Just as a catch-all method that gets called...
+
+**Tom Wilkie:** Yeah, what's wrong with that...?
+
+**Mat Ryer:** Yeah. So you mentioned tuning, and I think a lot of this as well we think of microservices, because you do have a lot more control over the deployment and the situation that they run in... And even if it's just kind of simple configuration changes, if you have an auth service that's constantly getting hit every time, you might choose to have some of those constantly running, where you might have some profile picture services that don't really get hit that often, and they can spin up and down as needed. So you get to make those sorts of different decisions as well, don't you?
+
+**Tom Wilkie:** I think at the end of the day, if you think about the stream of instructions flowing through the processor - you know, the processor doesn't really know the difference between whether you're running isolated microservices, or one big monolith, right? You know, maybe there's a bit of extra RPC overhead and a few more context switches, but realistically, this is all gonna be one big system that we're thinking about.
+
+So I like to see it more as a spectrum. I know that's a bit of a copout. You could even think of functions as a service as like the ultimate -- like, take every single function and wrap it in its own service and deploy it in an autoscaling fashion, and only run them on-demand... And you can think of that as like one far end of the spectrum.
+
+But I think it's more about the different techniques and workflows that these enable, and even how you can apply some of these techniques to places that kind of seem a big weird.
+
+To give you an example, one of the projects you mentioned at the beginning, Cortex, which is this horizontally-scalable version of Prometheus we've built - so we started off as a set of microservices, like 15-20 microservices. And it's fine, we run on Kubernetes; the marginal cost of another service is trivial. It doesn't matter; it's easy to add more services. But as we saw more and more people try and deploy this thing -- it's an open source project; you can go and download it and deploy it yourself... And it's just about to go 1.0 in fact in a few days' time.
+
+**Mat Ryer:** Oh, congrats.
+
+**Tom Wilkie:** Thank you. It's just a number. \[laughter\] But yeah, as we saw more people use it, we realized actually there's a huge barrier to entry, to have to deploy 15-20 different services to run this thing. So what we ended up doing was compiling all of the different services into a single binary, and having a little way in which they could run in a single process mode, and wire them all together with some internal RPC -- actually, they still just use the gRPC to local host.
+
+And now you can actually -- a single binary, a single process, a single command can run a Cortex node... And it's still horizontally scalable, and it's easy to deploy. And of course, we still run it as a microservice internally, because that gives us the control knobs and the flexibility and the ability to go in and set flags on this, and roll patches out to that, and so on. But I don't think you have to start from a point of view of like "I'm gonna just do microservices" or "I'm gonna just do monoliths." I actually think it's way more nuanced than that.
+
+**Mat Ryer:** And you get the best of both worlds.
+
+**Jaana Dogan:** Isn't this also super-useful for development time lots of people are having trouble just running things locally, or in a staging environment...? So maybe we can provide several ways of deploying things for people who want all the flexibility. They can maybe go and deploy microservices individually...
+
+**Tom Wilkie:** Exactly, yeah.
+
+**Jaana Dogan:** But at least there should be a simple, more like a monolith way of deploying things, at least, for people who don't need that flexibility and simplicity?
+
+**Tom Wilkie:** Yeah, I 100% agree. We call that Airplane Mode. You're flying transatlantic, you don't have a Wi-Fi and you need to run it on your laptop. We call it Airplane Mode.
+
+**Jaana Dogan:** Yeah, that's a cool name.
+
+**Tom Wilkie:** Yeah. It's a shame it's already taken.
+
+**Mat Ryer:** I like it. Well, you like to think different, don't you?
+
+**Tom Wilkie:** Oh... \[laughter\]
+
+**Matt Heath:** \[20:04\] I think we're definitely seeing that with other projects, as well. Istio, for example - I think version 1.5 is combining the various services into a single deployable binary, and that just makes it much simpler for people who are going to operate that... But they still have the separation of those components.
+
+**Tom Wilkie:** Yeah, and how many builds of Kubernetes are there that deploy as a single process, right? Like K3s is one, minikube... There's just so many where "Oh, we'll just embed it, CD in, and we'll--"
+
+The other thing that occurred to me is like name a successful, widely-deployed open source project that's microservices-oriented. Kubernetes doesn't count.
+
+**Matt Heath:** Yeah, I think that's fair. I think the overhead to deploying those kind of systems is just quite high... So it requires quite a lot of investment into that ecosystem, for you to want to get past that hump to actually benefit from it. And even if you want to them try it out locally-- I mean, this is a problem that we have experienced at Monzo with local development. We have 1,600-and-something services right now; 500-600 of them connect to Cassandra. It turns out a single Cassandra Docker container does not like 500 or 600 binaries connecting to it at the same time. \[laughter\] So that requires you to optimize those things a bit better locally. Yeah, it's not something that's easy to run on a single local machine.
+
+**Mat Ryer:** So 1,600 services... What are some examples of those services?
+
+**Tom Wilkie:** When you say 1,600 services - what are you measuring there?
+
+**Matt Heath:** Yeah, it's actually a lot simpler than people might imagine. One thing that I've noticed is there's a lot of sticker shock... Whereas if you described any complex application being made up of 1,600 classes, people would be like "Oh yeah, of course. That's software."
+
+So they vary quite a lot, but there's generally -- when we say "services", we mean each one of those is an independent Go HTTP server that has a number of functions, handlers that it will respond to. Many of them connect to a database, they will have their own databases when they do... But they range quite a lot in size. I think the difference is we have quite a lot of high-level systems, which you could probably model as a much smaller number of monoliths, but it would still be many monoliths... And within those, we've divided them up quite granularly.
+
+As an example, we've built our own Mastercard processor. So Monzo is both an issuer of cards - many services that make that happen - but also a processor directly connected to the Mastercard network. There will be one service somewhere that probably validates the signature of a CVV3 on a Google or Apple Pay transaction... And that's relatively complicated. So having it as an isolated chunk of code is a useful thing. So I think there's about a hundred or so Mastercard services that make up a Mastercard processor. And then you can repeat that until you get to the number that we're talking about... \[laughs\]
+
+**Tom Wilkie:** Sure, yeah. Are you talking about like a hundred containers, a hundred pods, or like a hundred deployments, a hundred stateful sets?
+
+**Matt Heath:** So in Kubernetes that would be a hundred deployments. Each one of those will be independently scaled with a number of pods, at least three; some will have - probably not in the Mastercard flow, but some will have a hundred plus pods spread across different availability zones. But yeah, so there will be 1,600+ deployments. I think 6,000-7,000 pods, something like that... And yeah, some of them will only have three, because they're really straightforward; some will handle emoji, so they will be less critical... \[laughter\] Or maybe more critical, who knows...? But they vary in that size, and importance, basically.
+
+**Tom Wilkie:** Do you run regionally within Monzo, and would you count the same service in two regions as two different services?
+
+**Matt Heath:** \[24:13\] So currently we run across three availability zones, and some physical locations, for a variety of reasons. Those would still be -- we treat that as like one deployment. I think as we move into having many Kubernetes clusters, that deployment that many of our engineers will see may translate to many Kubernetes deployments, but we would still see that as a deployment of a service, just running in different geographic regions.
+
+But yeah, the thing that we're trying to do is provide isolation, which we kind of talked about, and the ability to independently tune them, but also reducing the blast radius. As an example, we don't really have one API. If we built many services that broke up our domain logic, we would have just ended up with one huge application that was an API that provided that to our apps, and various other parties. So instead we follow a kind of similar pattern to Netflix, where we have a gateway, a bit like Zuul, and then behind that there's 200 or so APIs. So every path is a different binary, which is a different deployment. So lots of moving parts...
+
+**Mat Ryer:** Yeah. If that was a monolith then, calling a function to say "validate that CVV number" - that is kind of obvious what you would pass into a function, because it would be in the argument, and it's kind of obvious what you'd get back. How do you describe those ins and outs of each service? Do you have a consistent or standard way of doing that by now?
+
+**Matt Heath:** Yeah, I think you need those consistent approaches. If you're going to have a lot of microservices, I think you need a lot of consistency. We use protocol buffers, and that allows us to define an API in very concrete terms for each one of the services... Similar to gRPC allows you to define a service that has many endpoints within that definition, and each one takes a very concrete request and response type.
+
+So you get some type safety... Unfortunately, without extensions to the protocol buffers; it doesn't support the breadth of types that you would get if you had one application... But in many cases we can support that, because we control the whole codebase.
+
+I think one example of that would be a money type. If you're ever using money in an application, please don't use 64-bit integers to store pennies, or cents. \[laughter\] It turns out some numbers can get really big, and some numbers - you need a lot more granularity than two decimal places. So we have an internal money type that's very well tested, and we can pass that across the boundaries by unmarshaling and marshaling.
+
+**Mat Ryer:** Is it like a string then? How have you done it?
+
+**Matt Heath:** We pass it on the wire as a string, so it's passed as a decimal format, but then that's converted into a type.
+
+**Mat Ryer:** Into a special type.
+
+**Matt Heath:** Yeah.
+
+**Break:** \[27:22\]
+
+**Mat Ryer:** So we talk a lot about teamwork, and microservices helping larger teams... And I suppose the argument goes that essentially you can have teams that are somewhat isolated, and they sort of map neatly to the services that they're providing, and they kind of don't need to worry too much about other parts of the system. They can get to focus in their little boundary. I've heard that said as an argument. Are there teamworking advantages to microservices, or can you get the same thing with monoliths?
+
+**Tom Wilkie:** I think that's really the big advantage of microservices, is the isolation extends all the way down to the development process. My experience tends to be that teams look after sets of microservices, and not just individual services. I always normally find that that set of microservices could easily have just been a micro-monolith, or microlith, or whatever catchy phrase you wanna use for today... But I still think dividing code up along team boundaries - Conway's Law - is super-important. It's both a kind of necessary evil, like you have to do it, otherwise people's productivity is terrible, and it's also kind of like -- it's really detrimental when you actually wanna get cross-service flows in place; when you want to do something that's gonna end up touching tens or hundreds of services, you have to talk to tens or hundreds of people.
+
+**Jaana Dogan:** You mentioned that there's always an affiliation between the team and the services, or like a service group... I'm seeing some common services end up being in a bad spot in terms of lots of people are trying to contribute small fixes and so on, especially in organizations where there's a single monorepo, and it's so easy to just push things... \[laughter\] What do you think about that? Maybe it's not a super-critical thing, but I'm seeing some organizations tend to just enable that culture... Which might be good or might be bothersome.
+
+**Matt Heath:** I think ownership is definitely very important. I think the best way to enforce that is every service needs to have a clearly-defined owner. We use GitHub for code review, which means we can use code owners within the repository... So it means each service can have a defined owner. I think any of the engineers in the company can propose a change, but the code owner has to approve the change. I think that's really important. It gives people the flexibility to fix problems that they experience, rather than the problems being hidden behind this API that they find frustrating. You can see what the problem is, you can potentially fix it... That means people are really happy, because you fix bugs for them.
+
+The flipside of that is you still have the safety and control where people who have the expertise and deep understanding of how those systems work can review and approve the changes, and also make sure that it fits in with the longer-term direction of how that system is going to evolve.
+
+**Jaana Dogan:** I think this might be an advantage of microservices compared to monoliths... Because if there's a monolith, people just are all over the place, because there's no one person who can control the deployment, or has an authority to say no in some cases. You can have the same sort of ownership in a monolith, but I think with microservices you have more authority, because the deployment is really dependent on a particular team, so they can just say no; they don't have to accept the change... Or they have more power not to push things.
+
+**Tom Wilkie:** Right, but that could be super-frustrating, right? When something that's really important to you is not important to them.
+
+**Jaana Dogan:** Yeah, yeah...
+
+**Tom Wilkie:** When they just refuse, like "Oh, that feature doesn't belong here..." I've found in organizations that have adopted microservices sometimes you can get that kind of a bit territorial. "This is my microservices, it's done in my style. It's done with what I think should be the right way of doing it."
+
+\[31:58\] What I've always found works very well to diffuse that - and it works in open source similarly - you've always gotta have the big hammer of like "I'm just gonna fork your service if you don't wanna do it this way. I'm just gonna run my own."
+
+**Jaana Dogan:** Yeah, completely. I'm seeing lots of people that are also proxying things; that's also a really good way, because you can just slightly tweak the behavior, and then you can fall back to the other service to do the real thing. So yeah, there's all these different cases that microservices actually enable.
+
+**Tom Wilkie:** Yeah. As you said, you can always do code owners in a monolith. The flipside of drive-by commits in microservices is "Oh, I've gotta go and redeploy that. Oh, I've gotta go and test it, and drive it through staging", and so on. So I think we could all agree that microservices mean you've gotta invest a lot more in tooling, a lot more in automation to get rid of some of that toil around the deployment and the CI and the CD and this kind of stuff.
+
+**Matt Heath:** Do you think that differs from a monolith, or do you think at a certain point you just need a lot of tooling when a development team hits a certain size?
+
+**Tom Wilkie:** I mean, it's a fair point. I think in the monolith world it's much easier for an organization to invest in release engineering... Because it's like "I'm going to have a person whose job it is to push out the release every day/week." In microservices, it's like "Okay, now I've got a hundred teams that all need a little bit of release engineering, but none of them quite have enough of a problem to really justify it." But you know, you can make the argument in both directions, for sure. It's not clear-cut.
+
+I would like to change the subject though, because mainly -- so I wanna make one observation... One thing that hopefully we can all agree on is figuring out what's going on in a microservice is much harder.
+
+**Mat Ryer:** Really? Why? In a microservice?
+
+**Tom Wilkie:** In a microservice environment my request is gonna ping between (let's say) hundreds of processes, on hundreds of different machines. Like, figuring out why it's slow.
+
+**Mat Ryer:** Right.
+
+**Tom Wilkie:** Matt, you're smiling, because I think you know exactly where I'm going with this, right?
+
+**Matt Heath:** I'd say yes and no... I think it just depends on the tools that you have. If you have a monolith, then we have pretty good tools, and you can run them on your computer. You can run debuggers... We have a toolset that's evolved, and is quite easy to access. Whereas to get that same level of tooling and insight, you have to deploy a lot of stuff, potentially. And those tools exist now, and I think getting really good metrics and understanding of how things are performing, and then using tracing and various other systems can give you that insight... But it's harder, I think that's fair. And because it is harder, I think it's a trade-off of "Is that the most effective use of our time, or would it be easier if we just had a smaller number of bigger applications?"
+
+I think in my case, my personal experience has been with tracing and relatively simple metrics, if every server has metrics on its request/response times, and the clients that are sending requests to those servers have request/response times instrumentation on your databases and that kind of thing - I think I find it quite easy to pinpoint which service has a problem, and at that point then you're into profiling, and various other problems.
+
+**Tom Wilkie:** But there's a whole class of problems that exist, independent of a single service, right? There's a whole class of problems that extend from the composition of tens or hundreds of services.
+
+**Matt Heath:** And you've put the network between all of them as well, right?
+
+**Jaana Dogan:** Exactly, yeah! Compared to like a function, very fundamentally speaking -- I mean, microservices has a lot of layers to go through, and being able to pinpoint a single caller, like a single service plus a call...
+
+**Matt Heath:** Especially as soon as you layer that on top of an orchestration platform, and if you have...
+
+**Jaana Dogan:** Yeah, Kubernetes.
+
+**Matt Heath:** \[36:06\] ...an overlay network maybe, that's moving things around, if you actually want to trace where a packet has gone, between your function calls essentially, that's quite hard sometimes.
+
+I think the thing that makes that okay is you have a lot of layers of abstraction, and most of the time those abstractions are pretty good... So the majority of cases you don't need to dive too deep into how those abstractions work. That's not to say that we've not had problems where we've had to go really deep on particular network problems. I think we've created our overlay network at one point, and the MTUs were subtly different, and... Yeah, that caused some problems that we weren't expecting, and it took a long time to find what they were. But yeah, having the right tooling I think is essential.
+
+**Jaana Dogan:** Yeah. As a cloud provider, our main job is (I sometimes feel) really navigating all these problems and trying to figure out whose problem it actually is... Because from the customer's perspective, it could be anything. It could be their problem. They sometimes don't know if that's their problem. So all those problems are leaking to us, and we have to have the right tools and the capabilities to be able to navigate the problem, and be able to pinpoint to certain things.
+
+**Mat Ryer:** Yeah. Well, Matt, you mentioned these abstractions... Common services can be little abstractions of their own, and sometimes - in monoliths, or anytime you have common services or common functionality - it doesn't quite serve all of its masters very well, so you end up with a kind of Frankenstein service... How do you fight against things like that happening, and do you have any particular strategies for it? If there's a service that does almost what I want, do you do that thing that Jaana was talking about, where you could potentially proxy? That's almost like a sub-class kind of thing, but - then how do you keep track of where that happened?
+
+**Matt Heath:** I think in general we try and avoid abstraction where we can, because most of the times we've seen a problem and thought that it was some sub-case of a much larger problem. We've actually been wrong, and we've not been able to predict the future... So when we've added those abstractions prematurely, we've just added complexity, and those things have made it harder.
+
+So the trade-off there is effectively efficiency. What I mean by that is -- I was talking to someone from Amazon... You could potentially end up with many teams that have essentially built something that's very similar. And that feels like a really bad trade-off of time... Except what we often don't factor in is the trade-off of talking between all these teams and the communication overhead, and actually trying to build one universal product that solves all of those use cases... Whereas it actually might be better to just have a couple of similar kind of things, but that are very tuned to that specific case.
+
+So yeah, we have many core services that provide common functionality. Those evolve over time, so we refactor them occasionally and we change some of those models, effectively... But yeah, we try not to prematurely optimize those, because there's quite a high cost to it. We definitely do have some other proxy things we've talked about earlier...
+
+**Jaana Dogan:** I'm also seeing proxying is a good way to sometimes debug things... Like, people just proxy to be able to collect more data over the thing. So that's a legit case, which is a separate topic, but... You know, I just wanted to interrupt the discussion. \[laughter\]
+
+**Tom Wilkie:** That's the whole service mesh idea, isn't it? Proxy everything.
+
+**Matt Heath:** Yeah, I've definitely seen that with service messages.
+
+**Jaana Dogan:** Yeah.
+
+**Matt Heath:** We used to use Linkerd, and one of the--
+
+**Tom Wilkie:** Famously...
+
+**Matt Heath:** \[39:59\] Yeah, and one of the cool things with Linkerd was you could use a dtab as like routing rules that could decide based on certain parameters where your request would go... And a common use of that, which isn't something that we used, was to route very specific requests, say for a particular user, into a proxy, so that you could debug log the stuff between services. I mean, it adds extra hops and complexity, but yeah... From the sounds of it, people used it; it was super-useful.
+
+**Mat Ryer:** That is a cool idea. And you can deploy that service live and get it inside live, can't you? So you struggle to do things like that with monoliths without deploying the whole thing, probably...
+
+**Matt Heath:** Yeah, I think if you're turning on CPU profiling, do you turn it on for a particular code path, or for the whole application? Or you could just turn it on for one application, because it's subdivided so much that you're only affecting a small percentage of your overall application.
+
+**Mat Ryer:** Cool. So Tom, when you think about monitoring, and things like this, does anything change when you have microservices versus monoliths, or do your needs essentially stay the same?
+
+**Tom Wilkie:** No, loads of stuff changes. The simple, first-order stuff is just the sheer volume of moving parts. You've gone from effectively having a single major process per machine, to hundreds. And each one is going to export its own set of metrics. So the sheer volume, just from a metrics perspective.
+
+But we've already talked about the level of complexity you really necessarily need to debug performance issues in large microservices architectures, like tracing. This is necessary. You have to have this.
+
+**Matt Heath:** You say that? We've got a really long way at Monzo, without a good tracing system...
+
+**Tom Wilkie:** I mean, I didn't say you have to have a good one...
+
+**Matt Heath:** We had quite a bad tracing system that I wrote a long time ago...
+
+**Mat Ryer:** Was it just you running around? \[laughter\]
+
+**Matt Heath:** \[laughter\] Probably about as good.
+
+**Tom Wilkie:** I mean, most people's tracing systems is grepping a unique identifier through a log, right? ...which actually gets you most of the way there. You don't need fancy visualization, you don't need a lot of things... But that's still tracing, in my opinion at least. And yeah, you've got a long way there, but you still need that. It's the first thing people introduce, in my opinion, and it's -- you know, especially, there's a big difference between latencies, effectively... Like, if your latency class is like a couple hundred milliseconds, then you can get away with a lot of brute force and ignorance... But if you are doing high-volume, if you're doing trading, if you're doing anything where you're expecting to handle a huge amount of small things - you know, low latency applications and so on - tracing is just 100% necessary.
+
+And then, you know, the sheer volume - we've talked about volume as one of the main things, but also you have this extra kind of dimensionality to your data... We've already kind of mentioned Kubernetes, but no one really does microservices without some kind of orchestration system, and that orchestration system has information about the things that are running. So this is, you know, any kind of observability system... Whereas previously, you know, I remember the billing system is on a server called John; I'll just go and look at that server. You don't necessarily know that anymore when you're in a microservices architecture with some kind of orchestrator.
+
+So you have to have a way in your observability stack of incorporating a lot of this metadata and this extra dimensionality. And I think this is why we've seen the rise of systems like Prometheus, because they support this multi-dimensional data, and they support very rich integrations with things like Kubernetes. And I think that's really one of the drivers behind the demise or the lack of popularity around things like Graphite.
+
+**Matt Heath:** Yeah, I think those things have been really essential for us... Both being able to just pump a lot of metrics into a system and then go and introspect it later, and both to use that for investigations later, but also monitoring in real-time... And then - yeah, the tracing mechanism; we use Jaeger now, and for a while we had a system which would use structured logging to propagate a request ID...
+
+\[44:12\] So you could look at all the log lines across a hundred different servers to just pick that needle out of that haystack and combine them all together. And I think those were the only real ways we could understand those kind of things. But with those tools in place, I think it's quite useful, really beneficial.
+
+One of the questions I was gonna ask, actually - I think one of the things you get for free if you have that approach is... We have many services, and a call graph crosses many of these service boundaries. And each one of these boundaries has all these automated breakpoints effectively that are instrumented. So every call between a service has the perceived time from the caller and the server. So you have all these checkpoints all the way down your call graph, where you can see the performance of different things, which engineers don't need to think about, they just get for free, because the client that we use adds them. How would you do that if you were building a monolith? ...because you'd have to remember to add them, presumably at all these different places. And then that depends on the team, or is there a common approach?
+
+**Mat Ryer:** We're building this, we're building something. I'm building something. We're a tiny team, and I think team size definitely plays a big part in this decision-making process around this subject. So we're a tiny little team; there's just two of us building a thing, and we still have a kind of service abstraction a little bit, because we have a frontend client that through APIs is interacting with the backend.
+
+The way that we've done that - we've used Go interfaces actually to describe our interfaces, and then we've used the reflection tools in the standard library to look at those interfaces and then generate everything we need from that. So we do have an opportunity from that tooling to generate middleware that matches even -- it could be strongly-typed too, because it's generated dynamically from these templates. But I think you'd end up with something like that. Otherwise - yeah, it's a case of just a bit like we do with errors in Go, where you just have to remember to do it every time. You have to do it.
+
+**Tom Wilkie:** I mean, other languages exist, and other languages have more elegant solutions to this...
+
+**Mat Ryer:** What do you mean, "Other languages exist"? Not on this podcast, baby.
+
+**Tom Wilkie:** Aren't I allowed to use the J word here?
+
+**Mat Ryer:** Yes, please. Of course. \[laughter\] Everyone's welcome, Tom; you just won't necessarily be asked back.
+
+**Tom Wilkie:** Are you gonna edit this bit out, aren't you, Mat?
+
+**Mat Ryer:** It gets blipped out automatically if you say that word. \[laughter\]
+
+**Tom Wilkie:** You were railing on Ruby for having an accept all method when you call it on a class, but that pattern is super-useful for building this kind of middleware you're describing. Suddenly, I don't have to run a code generator; I don't have to interface everything out. I can just add a proxy class in, it intercepts all messages and instruments... And this is how people did it in monoliths for a very long time.
+
+**Jaana Dogan:** Austin Clemens has an idea to automatically maybe intercept things for instrumentation purposes... This idea has been around for a long time, but nobody actually did anything, yet, for Go...
+
+**Mat Ryer:** What is it, Jaana?
+
+**Jaana Dogan:** So we wanna be able to just arbitrarily intercept all the function calls in order to collect some data, so we can see all the execution trees, and potentially generate some sort of visualization, or run some analysis to see where are some of the hot paths, and so on... So he was interested in doing some work around this, but there's a lot of other things going on, so this hasn't been a high priority for a while.
+
+**Tom Wilkie:** And you can always do this with profilers, right? Profilers -- you pause a thread, sample a stack, done.
+
+**Jaana Dogan:** Yeah, yeah.
+
+**Tom Wilkie:** So there are techniques to do this. But you could actually argue that the technique of automatically instrumenting RPC boundaries is the newer one.
+
+**Matt Heath:** \[48:08\] It's a profile. \[laughs\]
+
+**Tom Wilkie:** Yeah, it is.
+
+**Jaana Dogan:** Yeah, yeah.
+
+**Tom Wilkie:** If you were wondering what I was 3D-printing, it's a clamp.
+
+**Mat Ryer:** What's that for, mate? It's like a Monzo card.
+
+**Tom Wilkie:** Yeah, because I'm just a big fan of Monzo, Mat...
+
+**Mat Ryer:** Yeah... \[laughter\]
+
+**Tom Wilkie:** But yeah, it's a clamp.
+
+**Matt Heath:** Nice.
+
+**Mat Ryer:** Yeah, but it's a podcast, mate, so you've gotta describe it. What do you mean, a clamp? That's not gonna make any -- imagine someone is listening to this on a podcast.
+
+**Tom Wilkie:** How do you describe a clamp...? It's a thing that you use to clamp things with...
+
+**Mat Ryer:** Fine, but why?
+
+**Matt Heath:** It's got a screwy thing...
+
+**Tom Wilkie:** Mat, you clearly don't understand 3D printing if you're asking why.
+
+**Mat Ryer:** Okay, fair enough. That's a good point. \[laughs\]
+
+**Break:** \[48:47\]
+
+**Mat Ryer:** So testing is another area that I think is quite interesting. One of the things that appeals to me with the microservices is this idea that you could really well test it. You could test that service almost every possible input and output. You could test it that much if you're so inclined... And because it's so focused, then hopefully you create good tests as well. Are there other things to consider when it comes to testing? Does it get easier or harder, have you found?
+
+**Matt Heath:** A controversial opinion, perhaps - is that testing always useful?
+
+**Mat Ryer:** Right. That's actually a fair question.
+
+**Jaana Dogan:** Oh, yeah.
+
+**Matt Heath:** So it definitely is, I should probably clarify that... And testing an individual service gives you really good insight into how it works and how that chunk of domain logic works. And you can test it to infinity... And that's really good, unless it's actually the boundaries between your services that the problems occur at, which from my experience is usually the case. Testing across those boundaries - that's quite hard. I don't think I've seen a silver bullet there, really.
+
+**Tom Wilkie:** And honestly, how is isolated service testing any different from unit testing?
+
+**Matt Heath:** Yeah, it is just unit testing.
+
+**Tom Wilkie:** Right. So we already know how to test stuff in monoliths. It's called "design your code properly, in well-isolated chunks, and unit-test it." So the nice thing about microservices is it kind of forces you to do that. You can't just call random methods in a microservice, unless they're exposed. So that's a nice thing, but it's just enforced like boundaries, that's all.
+
+**Mat Ryer:** Yeah, you could just do that. You could do that yourself too, even in a monolith, and actually should be. Similarly then, the integration tests, which - they're quite difficult to do in the microservice world, they're also difficult to get right in the monolith, too.
+
+**Tom Wilkie:** Sure. The only argument I'd make in the monolith is actually orchestrating a set of (let's call them) services within a monolith is easier than within microservices, where potentially they live in separate repos, built in separate languages... At least within a monolith you can wire them together in process, potentially.
+
+**Matt Heath:** \[52:13\] Yeah. I think that integration testing is where a lot of the value lives... And some of the problems that I've seen with the unit testing is if you have well-defined interfaces within your service, that's great. But as soon as you're testing the API part between them, one of the patterns I've seen is where we might stub the API response from another service, and in order to do a test, you actually have to stub many services.
+
+**Tom Wilkie:** Yeah. The entire thing.
+
+**Jaana Dogan:** Mm-hm. So I'm seeing a pattern that lots of service owners are also providing their stubs, or mocks, and so on... Do you follow a similar strategy? How do you deal, if you wanna test some service that depends on another one? Are you just generating those payloads yourself, as a developer on a different team, or are you just taking someone else's mocks, or anything that is already available?
+
+**Matt Heath:** I think in an ideal world teams that own services would provide mocks which other teams could use. I don't think at Monzo we have many examples of that, which means we -- we have request/response types that are defined in protocol buffers, so we can define those, but you have to effectively stub the response yourself... And that gives you some protection, because if the interface changes, it won't compile anymore, but it doesn't give you all the protection we would really want there... And that's where you then have to fall back another level to full-stack integration tests, where you're testing many different permutations of a request, all the way down the stack.
+
+**Jaana Dogan:** It's so complicated, especially given the orchestrators, and so on. I'm seeing a huge trend, people just wanna push things to production, or to Canary; that's like their almost primary testing environment sometimes. That's a bit sad, but that also captures the complexity of the reality, especially when you have Kubernetes, and different deployment problems, as well as replication problems, and that sort of thing. In the end, you just end up having one final test in production, by pushing to Canary.
+
+**Matt Heath:** Yeah, that's the kind of thing that we're looking at now.
+
+**Tom Wilkie:** Yeah. I mean, I would take a country view there... Like, to what extent is monitoring just extended integration testing? If you know -- and you should be running any sufficiently complicated system, you should be injecting kind of artificial load in your system and measure its behavior, right? So why not just call your integration tests "Actually, it's just a staging environment, and we're just gonna pummel it with artificial load, and use the same production alerts, the same monitoring and playbook and everything we use on prod, we're just gonna use it on integration."
+
+**Jaana Dogan:** Yeah, that's true. It's also common to actually mirror the incoming traffic... You know, set up a staging environment or a testing environment and use the same load, with similar payloads... That's also very common.
+
+**Matt Heath:** That's something that we're building at the moment.
+
+**Tom Wilkie:** That's a better quality test than any kind of artificial, contrived integration test.
+
+**Jaana Dogan:** Yeah, definitely.
+
+**Tom Wilkie:** You know, it does, again, come back to that point of "You need better tooling", because the only way that's ever gonna work is continuous deployment, is techniques, config management, ways in which you can ensure your staging environment is incredibly similar to production.
+
+**Matt Heath:** Yeah. Another way is to do it in production, where you have guarantees it won't affect your customer data, or various KPIs that you have.
+
+**Mat Ryer:** You mean test=true? \[laughter\]
+
+**Tom Wilkie:** \[55:57\] A lot of people will argue for testing in production... And I'm not one of them, but they've got some very good points. If you can arrange a system... You know, I've never been a huge fan of service meshes, but this is one of the things they can do which is really kind of nice... Is partition off a separate area of effectively production, that you can use for testing. That is kind of a nice system.
+
+**Matt Heath:** Yeah, those are the two main areas that we're looking at at the moment... One of them - we ran our last crowdfunding through the same platform, which at face value might sound insane, but it allowed us to use many of the same systems to make things reliable, but it requires to load-test lots of things. So we built a shadow traffic system there, where we can multiply traffic at our edge, and we can replay that traffic through the production infrastructure, so that we can effectively get the same usage patterns, but just dramatically scaled up. We use that as one kind of load testing system, and we run that periodically.
+
+The other aspect is the service mesh approach that you mentioned, Tom. One of the things that we've added recently is if you're propagating a context all the way through your request, which allows you to pass trace IDs, and various other useful stuff, we can propagate the environment through that as well, and we have mechanisms to make sure they're not modified. But that allows us to mirror production traffic into another environment, or mirror staging traffic into another environment. Those are the things we're looking at at the moment... Not in production yet, but we've got prototypes of those running in our staging environments at the moment, for testing, and isolation between different teams, to be honest, which is a big benefit there.
+
+**Mat Ryer:** That's so cool. Well, actually, it's time for our very regular part of the show... It's time for Unpopular Opinions!
+
+**Jingle**: \[58:03\] to \[58:19\]
+
+**Mat Ryer:** So, anybody is free to shout out... Do you happen to have an unpopular opinion? It doesn't have to be tech-related. I think our first one was Julie Qiu saying she likes taxis in New York.
+
+**Tom Wilkie:** What?!
+
+**Mat Ryer:** Yeah, she loves them. Absolutely loves them.
+
+**Tom Wilkie:** It doesn't have to be tech-related? You didn't tell me that...
+
+**Mat Ryer:** Well, if you've got a tech one, then that's alright, mate...
+
+**Tom Wilkie:** \[laughs\]
+
+**Mat Ryer:** Do you have any popular opinions, Tom?
+
+**Tom Wilkie:** Um, no.
+
+**Mat Ryer:** Okay. Well, then do you want to do an unpopular one?
+
+**Tom Wilkie:** I think I'll choose configuration management. I'm gonna say -- obviously, no one loves YAML. That's not an unpopular opinion. But I really despise Helm... And for good reason. I don't wanna belittle the work that the engineers have done, and it's incredibly popular, so they are clearly onto something... But the level of boilerplate templating and repetition that's been encouraged by Helm is something that really grinds my gears.
+
+We at Grafana Labs - and we've been encouraging anyone who will listen - use something called Jsonnet. Jsonnet is this config language that extends JSON with functions, and comprehensions, and all the jazz. From there, it has this really nice operator that allows you to compose together two dictionaries, like merge together two dictionaries.
+
+So my unpopular opinion is Jsonnet is awesome, and is the future... And I know - and this is a callout to Bartek; I saw your thread on Twitter last weekend... I think Jsonnet is awesome, and we use it for all of our Kubernetes config management through -- it used to be a project called Ksonnet, and now it's Tanka; we use it to distribute all of our alerts, our dashboards, anything... The Prometheus mixin, the Etcd mixin - these are all written in Jsonnet, and I think this is super-powerful.
+
+Yeah, so my unpopular opinion is that Jsonnet rocks, and that we should all use that.
+
+**Mat Ryer:** Great. And we'll put the link to that in the show notes. Mister Heath, do you have an unpopular opinion, mate?
+
+**Matt Heath:** \[01:00:09.06\] Yeah... I think the microservices one is quite unpopular, to be honest... \[laughter\] Judging by the reaction on the internet most of the time... Yeah, I don't wanna -- I don't know, it feels a bit of a cheat to continue the conversation we've had.
+
+**Mat Ryer:** No, not at all.
+
+**Matt Heath:** I think the unpopular part is when people see a number that describes something that you don't have context for. It feels at face value -- like, rather than finding out more information, usually the sticker shock value is the thing that feels quite unpopular.
+
+And I think the part of that that is really "What even is complexity?" - I think it's really easy to look at a system that has many components and see it as being very, very complex... But if you don't actually get exposed to all of the complexity, all of the time, and you only deal with a small subsection of it, and by breaking apart systems into many smaller components, each of which is very simple, I think actually that complexity aspect can kind of go away. Also, just different teams can work on the different aspects of the complexity.
+
+So yeah, that feels quite unpopular, whenever we discuss it. I think the microservices thing works really well for some organizations, but it doesn't work for everybody. It depends on the problem, really.
+
+**Mat Ryer:** That's a great one. I think that it's fine. Yeah, think about beginners with this... Because as we've kind of started, we've sort of said that when you're starting out, or if it's a small project, you don't have necessarily the investment needed to build out the big microservices thing... And that might leave some people in a situation where they only work on small teams, and don't ever get exposed to any of those ways of working. Are there any things that beginners can do to get a bit of exposure to microservices? Or is the answer to that "You have to be on a big team", because that's really when things get hard anyway?
+
+**Matt Heath:** I think a little bit... If you're building a product, then scaling or the need to scale is a problem you only have if the product is successful, right? You only have to scale things if people like your product, and they want to use it. But yeah, the scaling part has lots of challenges. So it depends, really.
+
+We started right from the beginning with microservices, and we based that on prior experience, having worked on those kinds of systems before... And it definitely slowed us down at the beginning. I wouldn't recommend diving straight in on a new project with that kind of approach. So I think there's like a natural transition point where a project/product goes through that bit.
+
+**Mat Ryer:** And it's probably a better way to do it, because it's more obvious when you are feeling the pain that your monolith is not doing what you want it to do, and you're feeling that pain, you're in a much better position to design a solution that's gonna work. If you try and do that in advance, some things I think are obvious, and they seem obvious, but you can't really beat actually building it and living with it and seeing this thing in the wild, to then get a better sense of how it behaves.
+
+**Matt Heath:** Yeah, and understanding the problem space. For example, we have a transaction service that represents what a money transaction looks like... And it's not the accounting concept, it's the visual representation to a customer. We built the core part of that product before we released prepaid cards.
+
+**Mat Ryer:** Oh, wow,
+
+**Matt Heath:** Our first prototype was P2P, like Venmo, or something, and we built a transaction service as a part of that. Now, we've evolved it a bit since, but clearly we made -- I made a load of assumptions five years ago based on very little knowledge of how that might work
+
+**Mat Ryer:** Some of them have stood the test of time, by the sounds of it.
+
+**Matt Heath:** \[01:04:04.19\] I mean, some of them, but not a lot... \[laughter\] Yeah. So I think you learn stuff as you go along, and being able to change stuff rapidly is probably the most important bit.
+
+**Mat Ryer:** Yeah. Jaana, did you actually have an unpopular opinion? I don't know if you've said one in a while.
+
+**Jaana Dogan:** Sure... I actually have an unpopular opinion about microservices. Even if you're working for a small company, it's a microservices environment; even if you have a monolith, you still have some external dependencies... Or it's just like really being -- you're not exposed to a lot of scale-related issues, but just being in a large organization, working on a single service is not quite different than working for a monolith. You're just very focused; that's the main reason microservices exist, as a way to scale the organizational problems. Most of your time is just really isolated and focused on a single service.
+
+And yes, you have some external dependencies, and so on... And at a small company, you're likely to have an external service from a different provider probably, so you can still put some of these practices in your daily routine, like the stuff that we talked about, monitoring and debugging things... Maybe you won't be able to touch all the aspects in terms of building this internal tooling that works very consistently, or having some core services, but you can still get a good feeling of what it feels to work in a primarily microservices environment.
+
+I mean, everything is a microservice at the end of the day. When they get bigger, we just split them more.
+
+**Mat Ryer:** That's great.
+
+**Tom Wilkie:** Yeah, and I think that reinforces the idea that even when building a monolith, you should think about how you may separate it out in the future... And that's just good engineering. Build nice interfaces internally, because they might become external interfaces in a month or a year.
+
+**Mat Ryer:** That's great, yeah. Well, this has been awesome. I've learned a lot, and I'm sure our listeners have, too. That's all the time we have today, so thank you very much to--
+
+**Tom Wilkie:** Oh, no...!
+
+**Mat Ryer:** I know. It's sad, isn't it? You'll have to come back and join us another time, Tom. Are you gonna do that?
+
+**Tom Wilkie:** Yeah, am I gonna be invited? \[laughter\]
+
+**Mat Ryer:** Yeah, I don't know...
+
+**Matt Heath:** Only if you 3D-print something again...
+
+**Mat Ryer:** Yeah, not a clamp though...
+
+**Tom Wilkie:** What can I 3D-print next time? Not a clamp... What's wrong with my clamp? My clamp's lovely.
+
+**Mat Ryer:** Something more boring.
+
+**Jaana Dogan:** \[laughs\]
+
+**Mat Ryer:** Yeah, it is lovely. Can we take a picture of it for Twitter?
+
+**Tom Wilkie:** I think "lovely" is the right word to describe this clamp.
+
+**Mat Ryer:** Yeah, let me just do a screenshot...
+
+**Tom Wilkie:** I didn't design it, by the way.
+
+**Mat Ryer:** I just downloaded that. I've got it. Thanks.
+
+**Tom Wilkie:** Yeah, you wouldn't download a clamp, would you? \[laughter\]
+
+**Mat Ryer:** No, I probably wouldn't... Okay, well, as I was saying - Matt Heath, thank you very much, sir, and Tom Wilkie.
+
+**Matt Heath:** Thank you very much.
+
+**Mat Ryer:** If people want to find your videos on the internet, what would they type in? Your name probably, right?
+
+**Matt Heath:** Yeah, I think so. \[laughs\] Although I'm now terrified of what comes up... \[laughs\]
+
+**Mat Ryer:** Well, you should have checked that, mate. You should definitely be checking that. I check that every day. Not really...
+
+**Tom Wilkie:** I can believe that.
+
+**Mat Ryer:** \[laughs\] No, I don't really. Okay, well, that's it. I tried to do a good ending of the show, but instead this happened. See you next time. \[laughter\]
+
+**Tom Wilkie:** Thank you.

The one with Brad Fitzpatrick_transcript.txt

@@ -0,0 +1,577 @@
+**Johnny Boursiquot:** Hello, and welcome to this episode of Go Time. Shout-out to our live listeners, those who always engage us as we record these shows; you do make it more fun. If you wanna be a part of the fun, join us in Gophers Slack; the GoTime channel is where you wanna be. If you also don't know about GoTime.fm, now you do. Go check out GoTime.fm to get access to past episodes, and all that good stuff.
+
+I'm Johnny Boursiquot, I'll be one of your hosts today. Joining me also is Mat Ryer. Say hello, Mat.
+
+**Mat Ryer:** Hello... Johnny... I don't know how to interpret that instruction, because it's too cliché to make the obvious joke... So I kind of flipped it and went meta a little bit.
+
+**Johnny Boursiquot:** Okay...
+
+**Mat Ryer:** It's not worked though, either way...
+
+**Johnny Boursiquot:** Well, we'll do what we can with it.
+
+**Mat Ryer:** Great.
+
+**Johnny Boursiquot:** \[laughs\] This is gonna be fun.
+
+**Mat Ryer:** But it's great to be here, thank you.
+
+**Johnny Boursiquot:** Good, good. We'll come back to you in a second, because I have questions about your current weather situation... But yeah, we'll talk about that. Today we have a very special guest on the show. It is Brad Fitzpatrick. Say hello, Brad.
+
+**Brad Fitzpatrick:** Hello! Happy to be here.
+
+**Johnny Boursiquot:** Glad to have you. So if the name sounds familiar, maybe you are a long-time engineer and you've come across Brad's work, maybe through a live journal, or memcached, or OpenID, or Perkeep... But a lot of us in the Go community actually know Brad from his work as part of the Go team, contributing to the Go standard library and being part of that pretty incredible project, if I may say so myself.
+
+So while we're here, we'd like to really have a fluid conversation with Brad, maybe touch on a lot of different things, Go obviously being part of that... But I'm personally interested in getting a sense of Brad's journey. If you don't know, Brad actually has his own page on Wikipedia. To me, that's a pretty high bar. That's like, "You made it!" \[laughter\]
+
+**Brad Fitzpatrick:** Yeah... Anyone can edit that thing though, so...
+
+**Johnny Boursiquot:** \[laughs\]
+
+**Mat Ryer:** So please do, guests... The most creative--
+
+**Brad Fitzpatrick:** \[04:17\] It used to say that I have 27 Boston Terrier pups, and I was like -- I don't know; there was just tons of stuff there over time.
+
+**Mat Ryer:** Well, just random people modifying it, and stuff.
+
+**Brad Fitzpatrick:** Yeah, yeah... I mean, that's my impression of what happens on Wikipedia - people come in and get bored and modify things, and other people edit it back out.
+
+**Mat Ryer:** Yeah. Suddenly, the Earth's got three moons.
+
+**Johnny Boursiquot:** \[laughs\] Alright... Again, like I said, we're gonna have a pretty fluid show. I'm interested in getting a little bit of background, perhaps the stuff that is not obvious for those of us who can go to the Wikipedia page, and those of us who are familiar with Brad's work within the Go community... Brad, what do we not know about you, that's not out there? Let's start with that.
+
+**Brad Fitzpatrick:** I don't know, I kind of tweet my stream of consciousness, so I think most of my information is our there somewhere. I haven't been too shy on the internet... But I'm currently living in a world of smoke, on the West Coast here, so it's kind of depressing looking out and seeing nothing. I haven't really left the house in a week.
+
+**Mat Ryer:** Yeah, that's really grim. It looks pretty terrible from over here, as well; it's all over the news, of course. I hope everyone out there is staying safe...
+
+**Brad Fitzpatrick:** Yeah. That's not so much about me, so much as the whole West Coast, but it's currently on my mind... But I've also been vocal about that on the internet. It turns out that the weather forecasters don't know anything about how smoke works, or how smoke affects anything else, or how smoke moves with the wind, or anything... I've been checking the weather every day and it turns out it has nothing to do with reality. In reality, the smoke just is.
+
+**Mat Ryer:** That's just because it's so different, so unprecedented events...
+
+**Brad Fitzpatrick:** Yeah, pretty much. And something that doesn't happen too often, they don't have models for it, so they just assume it fits some other model, and it turns out it doesn't.
+
+**Mat Ryer:** Yeah... I'm having a problem not quite the same, but it's very hot in London, and it was never very hot in the past, so we're not ready for it... So it's just a bit unbearable when it does spike like that. I think it's all related, but... Yeah. Changing times.
+
+**Brad Fitzpatrick:** Yeah.
+
+**Mat Ryer:** But Brad, how did you first get into computers when you first got into them? Were you quite young when you were into them, or...?
+
+**Brad Fitzpatrick:** My dad was an electrical engineer, and he was working at a company that made memory chips in Texas. And as part of this, they would test the chips after they were made, if a bit in the memory chip was stuck high, or stuck low, like it had permanent one or permanent zero somewhere, they would toss them in this discard bin... And when the Apple II came out, my dad and some of his coworkers wanted to make a bootleg Apple II, which was pretty easy to do at the time. You can get all the parts at RadioShack, and just sit around - they sat around, soldering things for a couple weeks... But the only real tricky bit was the ROM, which bootleggers were passing around, and stuff. So they had a copy of the ROM, but they just needed a ROM chip to burn it onto, and they worked at this company that made these ROM chips. So they went to the discard bin and they just grabbed a handful of them, and they just flashed the Apple II bootleg ROM onto one until the stuck bits just happened to be in the right place.
+
+**Mat Ryer:** \[laughs\] That's great.
+
+**Brad Fitzpatrick:** So we had this bootleg Apple II that I always saw my dad screwing around on, and at some point - I was like 5 or 6 - he had taught me how to write BASIC. He wrote a program on paper that was like 10 PRINT "HELLO WORLD" 20 GOTO 10 and he was like "What do you think this does?" And I was like "Well, clearly it prints something on our printer." And he's like "Well, no, actually it doesn't print on the printer, it prints on the screen." I was like "Oh, okay..."
+
+**Mat Ryer:** \[08:14\] You were sassy for a 5-year-old, weren't you?
+
+**Brad Fitzpatrick:** \[laughs\] I don't know what I said... But I do remember being confused that print didn't mean it actually went to the printer.
+
+**Mat Ryer:** Yeah, that rings a bell. I remember that same thing.
+
+**Brad Fitzpatrick:** There's tons of things in computers that are not quite right, but you just accept it and you say "Oh, sure. Whatever." \[laughter\]
+
+**Mat Ryer:** Yeah, there's lots of examples in the standard library as well. We can talk about those later...
+
+**Johnny Boursiquot:** Oh, that was like shade, if I'm not mistake... \[laughs\]
+
+**Mat Ryer:** I don't even know what that is... I could do with some today though.
+
+**Brad Fitzpatrick:** Writing a book on the standard library's history would be fun, if I had time, or books paid, but... \[laughter\] It'd be fun to have a back-story on every API of the feature, of like why we thought it was a good idea at the time, and what we were thinking... But yeah, it can't change anything.
+
+**Mat Ryer:** It's interesting though, looking at some parts of the standard library where you can see different styles of Go code being written.
+
+**Brad Fitzpatrick:** Oh, yeah.
+
+**Mat Ryer:** It's quite an archaeology to go poking around.
+
+**Brad Fitzpatrick:** There's whole different generations in there, and then when you make something new, you're like "Well, I want it to be consistent with the old stuff in the standard library, but which generation do I wanna be consistent with?"
+
+**Mat Ryer:** Yeah.
+
+**Brad Fitzpatrick:** If you saw all the talk about the file system proposal that's coming out now... People are complaining that it doesn't have context in it, and we're like "Well, yeah, but contexts don't really work with most operating system layers, and also it doesn't exist in the OS package, so it wouldn't compose well with that..." It's just so many trade-offs all the time.
+
+**Johnny Boursiquot:** We used to have this saying circling around the Go community - even I have uttered it on many occasions - "If you wanna know how Go is written, look at the standard library." But now I'm kind of hesitant on saying that, because again, if somebody doesn't know that, okay, there's multiple generations, and multiple best practices, and things you wish you could go back and redo, because we normally do things a certain way; it's kind of hard to tell somebody who's picking up Go today, like "Hey, just go and read the standard library". So are there parts of the Go standard library that you wouldn't point somebody to, but other parts of the standard library today that you'd definitely say "Okay, this is an example of how you should write Go"?
+
+**Brad Fitzpatrick:** I wouldn't read the runtime, for instance, or anything using unsafe or cgo is pretty nasty. And even Http is just full of corner cases, because the world of the internet and browsers and servers are so bizarre in so many edge cases that the code just gets unnecessarily complex... But there's little bits and pieces. When I was relearning how gif and jpeg worked, I would just -- it's easier for me to go read the Go implementation of some of these file formats, or zip files, or something... Rather than read the spec, you can just go read the Go code and it's often easier to learn how a file format or a network protocol works.
+
+**Mat Ryer:** Yeah, I've found that -- I mean, because I can read Go; I can't read whatever the specs are written in. \[laughter\]
+
+**Brad Fitzpatrick:** Yeah, all the RFCs have totally different styles too, so...
+
+**Mat Ryer:** Yeah.
+
+**Johnny Boursiquot:** One of the packages that you were pretty vocal about was the net http client, and I think at some point you even had some experimentation out there about that. What is the state of that? Where did you leave that?
+
+**Brad Fitzpatrick:** In the works... I'm not doing HTTP/3, as people always ask me... Like, if I'm gonna do QUICK. I think I will leave that battle to somebody else, with 5 or 6 years free. But yeah. QUICK just gets very invasive very quickly. It's nice, in that it combines TLS and HTTP and TCP all together, so the layering is nice, and the layering isn't fighting each other, but it's just -- they're now implementing a whole network stack in your library.
+
+**Mat Ryer:** We've got a question here in our Slack channel from Johann Brandhorst, who asks "What are some tech that you're excited about?"
+
+**Brad Fitzpatrick:** \[12:07\] Well, I was really into WireGuard lately, which is why I kind of left Google to go work on WireGuard, and making WireGuard fun and easy... I don't know, I've always been kind of so-so at networking. I've been interested in networking, but I never felt super-great of it... So this is kind of a forcing function that makes me actually learn how networks work.
+
+**Mat Ryer:** Oh, wow... Okay.
+
+**Johnny Boursiquot:** So for those of us who don't know much about networking, like what is WireGuard, and then by proxy, why you do care so much about it, in light of the work you're doing at tailscale.
+
+**Brad Fitzpatrick:** Well, I always fought networks a lot... And then I was helping my dad set up a networking port; my parents bought this RV. They are retired now and they're just kind of traveling around the U.S. - or they were traveling around the U.S, prior to pandemics and wildfires, and stuff... And my dad wanted to have monitoring on his RV. There's this CAN bus that lets him control all the lights and fans and heaters in the thing, and it also lets him see temperatures, and fuel levels, and what their black water and clean water - the levels around all that, and how long the generator has been running... So he wanted to build a dashboard to see all this, and he wanted to see it from anywhere in the world.
+
+So we had set up this whole contraption of Raspberry Pi in the RV, and an LTE modem, and all these reverse proxies, and stuff... And then I set up WireGuard for him, and that was just so cool and empowering, compared to the complexity of what we had been doing prior, that I just started playing with it more myself, like putting WireGuard tunnels between things... But it can get pretty tedious to set it all up yourself, so when I found out that David Crawshaw who I also worked with on Go was doing a company all around making WireGuard fun and easy, I was like "Oh, I wanna go do that."
+
+I had a bunch of friends that were all kind of doing startups at the same time, and I was getting some startup FOMO... Thinking back to the old days of me doing the LiveJournal startup... So I wanted to relive the startup glory days.
+
+**Johnny Boursiquot:** And how is that working out so far?
+
+**Brad Fitzpatrick:** It's fun. It'd be nice if there wasn't a pandemic going on, but... I had one good month of working in person with another tailscale person who was here in Seattle, but then after a month it was all working at home, so... Oh, well. Which means I have to listen to my kids running around and screaming, and stuff, which is not incredibly conducive towards working.
+
+**Mat Ryer:** You share sometimes some of the things that your kids say, and I find them to be hilarious. I always appreciate that.
+
+**Brad Fitzpatrick:** Oh, yeah... They're constantly exhausting, but they're also constantly hilarious, so...
+
+**Mat Ryer:** Yeah, so it's worth it.
+
+**Brad Fitzpatrick:** Yeah, well... It balances out at least. \[laughter\]
+
+**Mat Ryer:** Great. That's a great advert for kids there... If you're a young couple listening that are considering, and not sure - well, it balances out, says Brad.
+
+**Brad Fitzpatrick:** I'm told it gets better, but we'll see.
+
+**Mat Ryer:** Skeptical.
+
+**Brad Fitzpatrick:** Yeah.
+
+**Mat Ryer:** So I wanted to hear a bit of the LiveJournal story; you mentioned it... It's a really surprising thing, I think... Could you tell us a little bit about it and your involvement?
+
+**Brad Fitzpatrick:** Yeah, so in high school I did a summer internship at Tektronix, and one of my jobs was to work on their corporate intranet, and make a company directory online, and take a bunch of their tools that were kind of like not online at all - I guess they were things that didn't exist - and work on their corporate intranet... And as part of that, I was running on some commercial Unix system, I was running Motif as my desktop, and it was like a three-button mouse, and classic, old-school Unix... And I had access to run CGI scripts, and that was my first time where I had access to a server that made dynamic responses to HTTP requests. So I just went crazy, and I made their intranet do all these dynamic things...
+
+\[16:08\] And then I wanted to do that myself, so the summer before college I started making my personal website have all the dynamic crap... And I got a copy of Visual Basic, and I made the stupidest Visual Basic app, which was just like a single-line textbox that just floated up my Start bar in Windows... And I could type into that single-line textbox. There was no button to post, you'd just hit enter and you'd post. And it just made an HTTP request. I didn't call it an API for the longest time, because I didn't really think of it as an API. I was like "I don't know, it just makes an HTTP request."
+
+So everyone downloaded this Windows program, and there was no way to post on the web even... And then later, people started writing entire paragraphs in this client. They couldn't hit Enter, because Enter posts, so they would just write this whole wall of text... And I was like, "Oh yeah, I didn't really put a character limit on that." So everyone used it effectively like Twitter, because they would type until they hit the edge of their screen, and then they'd be like "Well, I'm out of space, I guess. I'll hit Enter."
+
+But then at some point people started writing big things, and wanting to do paragraphs, so I put up a web page for people to post and made it a text area instead of an input box...
+
+**Mat Ryer:** Oh, now we're talking...
+
+**Brad Fitzpatrick:** Yeah. And for a while, we all ran the CGI on our school's -- our school gave us CGI access, so we dropped the CGI file into our home directories... But that was too nerdy and complicated for people, so I made it like a shared instance, where people could create accounts.
+
+So then it was just like me and my friends from high school, my friends from the dorms in college creating accounts... But their friends told their friends, and all of a sudden we're killing my little server, and I'm begging people to send me money. In exchange for people sending me money, I put a little thumbs up next to their name on their user profile page... It was just kind of like a--
+
+**Johnny Boursiquot:** The original Verified.
+
+**Brad Fitzpatrick:** Yeah, pretty much... And then people sent more and more money, and then I raised their quota as another way to thank them. I gave them more user avatar pictures, and stuff... And at some point, my mom was like "You know you're taking money from people -- you're taking money in exchange for goods and services. You know that's a business..."
+
+**Johnny Boursiquot:** That's a business.
+
+**Brad Fitzpatrick:** Yeah. "You need to pay taxes, and stuff. People can sue you and take all your money now", and I'm like "Oh, crap..." So she hooked me up with like a family friend who was a CPA, and he introduced us to lawyer types, and we made an official company, so I was legally isolated, if people were jerks... Because you know, people are always jerks on the internet.
+
+So then it turned into a company for the whole time I was in college, and a few years afterwards... And then I sold it to a company in San Francisco, because I just got burnt out running it. It was just non-stop stress. Our site was constantly falling over. We went from one server to 150 servers or so, so we had to invent load balancers for things, and figure out sharding, which - I didn't know the name for it; the name didn't exist at the time, so we called it something else... But sharding users over database clusters, and doing HA databases, and doing HTTP load balancers, and doing a distributed file system... It was effectively S3 at the time, for photos... But the site's constantly on fire, so you're constantly writing stuff to \[unintelligible 00:19:25.00\] because we were hitting our databases too hard, so...
+
+It was good times. It was fun goofing around... And you know, there's constantly something broken to fix.
+
+**Mat Ryer:** It's amazing to hear all the different types of technologies that you had to innovate at the time. It's kind of really surprising. And now, of course, LiveJournal - it says on their website they are now at version 427... And that's great.
+
+**Brad Fitzpatrick:** \[19:54\] What?! They have version numbers? We didn't want to build all this stuff. At one point we went to buy -- we needed high-availability file storage for our images, and stuff... Because we had a bunch of machines that were net-booting/PXE-booting Linux and stuff, and we're like "Well, where do we store state?" We had the databases for posts and stuff, but we didn't wanna put all of our images in our databases, because we were spinning disks at the time, and didn't have much storage.
+
+Someone said "Oh, everyone NetApp. Go to NetApp and get one of their file servers, or whatever." So we talked to them, and we said "Okay, we need a file server. How much?" And they're like "How many users do you have?" And I was like "Um, this much." "How much do you charge them?" I'm like, "Um, this much. But not everyone pays." "What percentage pay?" And you could see them multiplying in their head... And they're like "Okay, we'll come back to you with a quote." And this was the first time that I dealt with commercial vendors... It just felt so scammy. And I was like "Oh, you're just making up a price on the fly here...", and I just got so pissed. But we had no option, because we were falling over... So we paid them all this money, which - I forget what it was, but it was like tens of thousands of dollars for something that was okay... It kind of worked, but it wasn't amazingly performant or anything. This is all pre-SSDs and stuff anyway, so even if your software is magic, there's not much you can do if you have a couple spinning disks in side this unit... Or two. I think we got a pair, for high availability reasons...
+
+So that's what led me to go build our MogileFS, which is an anagram of OmgFiles... But it was kind of our S3 object store.
+
+**Johnny Boursiquot:** So you were creating all this tech, which is now commonplace, right?
+
+**Brad Fitzpatrick:** Yeah. So that's the other reason I kind of wanted to go do a startup. When I first joined Google, they had all this secret sauce, like Borg, and BigTable... There was amazing technology, so I wanted to go to Google and see how the grown-ups did distributed systems. But now I feel like all the secret sauce is out there, in open source... But I've never had really a chance to use it. At Google I had access to it, but I wasn't necessarily using it. I could just use it if I wanted, so I used it for the Go build system, and stuff... So this is kind of like my first company that I've got to do where the cloud existed, and I can use all this distributed systems stuff for free... And it's great; we're setting up our proxy relays around the world for Tailscale for the people who don't have direct connectivity, for like 4%-5% of connections; if we can't penetrate the NAT on those sites we end up proxying the encrypted packets through our relays... We're like "Okay, yeah, let's just spin up some ones around the world", and we've just got a bunch of $5/month instances in all the regions around the world, and it doesn't take any time. That stuff used to be downright impossible, so... It's fun. I always knew that this stuff existed, but being able to use it for the first time is pretty fun.
+
+**Mat Ryer:** Yeah, it's nice to see that these things are becoming accessible, these technologies which in the past - like you - you had to just create them... Like memcached, which we should definitely talk about more. So yeah, I think it's a great way also for projects to emerge; like you say, memcached came out of the LiveJournal stuff. It's that thing of solving real problems, and having to, like you say, when it's on fire. So you have no choice; there are imperative things, you've got to solve these problems. I find some great tech comes out of those kinds of situations.
+
+**Brad Fitzpatrick:** Oh, yeah.
+
+**Mat Ryer:** Do you like working in that sort of environment, or do you prefer it more relaxed, where you can just take your time and design maybe...?
+
+**Brad Fitzpatrick:** \[23:46\] No, I need a certain amount of stress. There's probably a limit, but if there's no stress at all, the priority isn't so obvious. I don't' do well when it's not obvious what the priority is. If I have 20 things that are all kind of equal priority, but I know I'm only gonna have time to do two of them today, but it's just arbitrary which two of the 20 I pick, then I'm like "Why pick anything? 18 of them aren't getting done... So why those two?" I like a certain amount of fire, so I know what to do next.
+
+**Johnny Boursiquot:** Do you like the arbitrary stress? You know, say, product teams - you're making arbitrary deadlines, saying "Hey, we need to have a conference on this date, so we need to build this thing by that date." Or do you prefer the stuff that's being driven by customer need? Like, basically needing to scale.
+
+**Brad Fitzpatrick:** Oh, the customer needs, for sure. Seeing people react to things, on Twitter or whatnot, when you launch something - that's way more motivating than some arbitrary deadline. I mean, if it's getting things done for a conference demo or something, at least that date is real. But if someone is doing a sprint, and it's arbitrarily for N weeks from now, I'm like "Okay, why that Friday? Why that Thursday?" I don't know... But yeah.
+
+**Mat Ryer:** And then they make you complicit by going through that estimations exercise, and then it's like "Well, I'm sorry, you said it was four muffins, so it's four muffins." \[laughter\]
+
+**Brad Fitzpatrick:** I've actually never worked on a team that did any of that, so I'm kind of speaking blindly here... I've heard people talk about it, but I don't really know the process.
+
+**Mat Ryer:** Oh. It's cool. It's basically abstract. You try and just have an abstract number, a relative score for work. In some ways it's nice, because you can identify big pieces of work, usually. Someone on the team will know that this is a big piece of work, and they can talk about it... But the trouble is obviously it gets turned into dates and times, and then suddenly we're all rushing, and there's no really good reason for it.
+
+**Break:** \[25:48\]
+
+**Johnny Boursiquot:** So obviously -- well, I'm gonna assume that you're using a lot of Go in your day-to-day work today...
+
+**Brad Fitzpatrick:** Oh yeah, it's like 99% Go. We have a little smattering of other languages here and there, but...
+
+**Johnny Boursiquot:** So before Go, as Cal asked in the GoTimeFM channel, what did you use primarily?
+
+**Brad Fitzpatrick:** I started off with BASIC and C. But neither one -- you know, in BASIC you hit limits pretty quickly, and C is not very fun. It's like, it exists to do things when you need to use it, but I don't enjoy writing in C. It just feels tedious.
+
+So when I discovered Perl, in '93-'94 or something like this - it was Perl 4 at the time - I was like "Whoa, I can do things really quickly!" But this was also the language everyone at the time was doing CGI scripts in too, so it was also the language that let me do cool, dynamic web stuff.
+
+\[28:07\] So I was very in the Perl community, very involved, and all of LiveJournal was pretty much all Perl, except for little bits and pieces that needed to be C for performance reasons. Memcached was originally in Perl, but it fell over within like a second, as I suspected it would. The prototype was in Perl, but then we quickly rewrote it in C.
+
+So I love Perl, and I never really jumped on the Python or Ruby trains, because they were just not different enough from Perl for me to care. They had all the same pros and cons; they weren't much better. They were still like explode at runtime, and no type checking... Python and Ruby were even slower than Perl, so performance wasn't a reason... And maybe the syntax was a little cleaner, but at that point I didn't care, because I already knew Perl.
+
+And then I joined Google, and they use lots of Python, C++ and Java, so I wrote in all of these kind of out of necessity, but I didn't really love any of them. I worked on Android for a couple of years and wrote a lot of Java, and a lot of Google frontends were often in Java, and Google backends were often in C++... I launched a service that was a whole mx of things. It was like Java, C++, Sawzall which was Google's language for logs analysis, which Rob Pike worked on prior to Go... I kind of hated it, but that's what everyone at Google wrote their logs analysis in at the time. Nowadays they've migrated off Sawzall onto Go.
+
+Yeah, so I was never really excited about any of the Google languages. I wrote in them because I had to, because there was no choice... And then I saw Go come out. I went to a bunch of the internal talks on Go, and I was like "Yeah, yeah, whatever..." Google had a bunch of internal languages, and most of them were so-so... So I was not super-excited at the time.
+
+Somebody had a question from the audience, like "Does Go have closures? Do you have first-class functions?" And the answer was a very dismissive "No, we don't. You don't need them", or something. And I was like "Well, okay... I'm out." Perl had them, and I had this book "Higher-order Perl" that I really loved, that taught me a lot of what people would probably actually learn in Lisp in school, or whatever, but I never really had any exposure to it until that book. So I was totally in love with higher-order functions, and stuff like that. So when Go didn't have it, I was like -- I kind of wrote it off. And also was like "I don't really invest much time in another Google internal language, because I'll just leave Google at some point and I won't get to play with it anyway, and I won't get to use it for my personal projects..."
+
+But then when they said they were open sourcing it, I was like "Oh, I should look again. And then I noticed they had added closures in the meantime. I guess Russ Cox just went in one day when Ken Thompson was on vacation, and couldn't object, and Russ just added them. \[laughter\]
+
+**Johnny Boursiquot:** Russ? No..
+
+**Brad Fitzpatrick:** Yeah, that doesn't sound like Russ... One of the Russes... \[laughter\] So then after that I started just screwing around on it. I was spending like 2,5 hours a day on the Google bus, going back and forth between San Francisco and Mountain View, and I just had a lot of time to screw around on my laptop... So I decided the Perkeep, which at the time was called Camlistore, was originally kind of a set of ideas and data structures of how I would want my personal life archiving storage system to work... But I wasn't excited to write it, because I didn't wanna write it in Perl; I was sick of scripting languages... I didn't wanna write it in C++, because that just didn't sound fun for a personal project... And so when Go came out, I was like "Well, the best way to learn a language or learn anything new is to use it", so I was like "Let's try to build this storage system idea that I had. Let's try to write that in Go."
+
+\[31:55\] I've found right away that the standard library was missing tons of stuff, so that's why I started sending in lots of changes... Then one day Rob Pike was like "Hey, do you wanna do this full-time?" I was like "What?! Work on a language full-time? That sounds great! And it's open source... I don't have to feel bad about working for the mega-corp, working on advertising..." \[laughter\]
+
+**Mat Ryer:** That's amazing. So again, a lot of those pieces of the standard library came out of things you needed.
+
+**Brad Fitzpatrick:** Oh yeah, pretty much everything that I added to the standard library was because I needed it for Perkeep, or some other personal project, or... Sometimes, people were complaining about it on the mailing list, or something, but it was almost all my needs.
+
+**Mat Ryer:** Yeah. But I think that's why they resonate so well. Memcached is an interesting one. That project was originally C... But there is a Go version, isn't there?
+
+**Brad Fitzpatrick:** Well, there's clients for every language under the sun, so...
+
+**Mat Ryer:** But the original is still C.
+
+**Brad Fitzpatrick:** Yeah, the server is still C.
+
+**Mat Ryer:** Oh, okay.
+
+**Brad Fitzpatrick:** I've reimplemented the server in C++ inside Google; that's what the App Engine -- there's a version of memcached inside Google that is all based on Google's RPC system, Which is all protocol buffers, and stuff... It's kind of like gRPC, basically. And that's written in C++, but it's not -- I mean, a memcached server is kind of like a good Hello World network server. It's very simple. It's a HashMap hooked up to a Get and Set effectively.
+
+**Johnny Boursiquot:** Well, that's kind of putting it lightly... But yeah, sure. \[laughter\]
+
+**Mat Ryer:** Yeah, because you know, they synchronize, and...
+
+**Brad Fitzpatrick:** I mean, the only tricky part is you want a really good memory allocator. Originally, we just used malloc, and we worked for like a week, or sometimes like a few days, and then the CPU would go through the roof, because we had all this internal fragmentation, and the memory allocator was having a hard time finding a spot to put your memory of the right size, just searching all over... So once we switched to a slab allocator - there was that Solaris paper back in the day about slab classes, and stuff... So we wrote a memory allocator inside memcached that bypassed your system one, and just did its own, Malloc-free. It only asked the operating system for one meg at a time, and then cut up the one meg itself into size-specific classes...
+
+So that's the only real trick inside memcached... But nowadays all the normal libc memory allocators do similar things, so...
+
+**Johnny Boursiquot:** So let's talk about that home cluster project you keep posting about... \[laughter\] Why, first of all?
+
+**Brad Fitzpatrick:** Yeah, so at one point -- I've gone back and forth between... I've always had computers at home, and I basically always had servers at home, but then at one point I went like too full-cloud. All my stuff was on the cloud, and I had VMs all over the place, and bandwidth costs were going up, because the cloud charges you way too much for bandwidth, and stuff... And I woke up one day realizing I didn't understand how computers worked anymore, or what computers cost, or how fast they were, because I was so disconnected from reality and I was always using this virtual stuff... So I wanted to build my own server again.
+
+And I think about the same time I had an internet outage at my house, because -- I forget what I was using for a gateway, but I had like 2-3 gateway failures in a row, where my router failed in my house... And it was super-annoying. So I decided that my goal was gonna be to build a high-availability internet router that had no single points of failure.
+
+I was like "Well, I can't have one server. I need to have at least three, because I need to have a \[unintelligible 00:35:39.21\] quorum if one fails."
+
+First I started off with VMware, because I was using it at work, with ESXi, but I couldn't stand how corporaty and enterprisy the whole thing was, so I switched to Proxmox. So now I had just VMs that were just floating around, and I don't really know which VM is on which of the three nodes... And if one of the nodes dies, the VMs just float somewhere else. I use Ceph for storage, so the storage is sprayed over all the three machines too, and I don't have to worry about where my block devices are; they're all just highly available.
+
+\[36:11\] And my router isn't a physical piece. My router is just a Linux VM that floats around the machines, too. So each one of those is plugged into -- I have three switches, and every machine is plugged into... Well, I had two switches at first, so every machine was plugged into both switches, and they're all different power supplies, and each of the three machines has two cables, one to each switch... Two ISPs, and they just failover between the ISPs... The WiFi in the house is all PoE-powered from the switches that are on UPS... So the whole neighborhood could lose power, and I would still have working WiFi for like an hour and a half... \[laughter\]
+
+I did this to learn how servers worked again, and you know, stop being so dependent on the cloud. But of course, I always build these too complicated setups and I forget how they work, and then I have our second child and then I don't look at it for a year, and all my certs expired the other day, and all my home automation stopped working, because my etcd cluster was down... And I couldn't renew the certs, because to renew the certs it needed to talk to the cluster or something, and... I don't know. The whole thing's just screwed, so I wanna nuke it and start over.
+
+**Johnny Boursiquot:** \[laughs\] You need some observability, man...
+
+**Brad Fitzpatrick:** So now I'm looking at like "Do I wanna use Flatcar Linux, like the CoreOS continuation project, or do I wanna use K3S, or maybe I just wanna use Podman...?" So I'm kind of like debating all my options now to build a more simple thing to run containers at my house.
+
+**Mat Ryer:** It's much easier to just do it in the cloud. \[laughter\]
+
+**Brad Fitzpatrick:** I could... But I wanna have jobs that are connecting to security cameras, and doing motion detection on live video... And if I did that over the internet connection, the bandwidth cost of the cloud would be stupid.
+
+**Mat Ryer:** Right, right.
+
+**Brad Fitzpatrick:** And once I have some jobs running here, I might as well run all the jobs here.
+
+**Johnny Boursiquot:** This sounds like the typical "Oh, I needed to blog more, so I wrote a blog engine."
+
+**Brad Fitzpatrick:** No, that's how people stop blogging. They're like "I'll start my blog again once I rewrite my blog engine", and then they never rewrite their blog engine, so they never blog. That's where I'm at.
+
+I see a question on the Slack, how do I feel about Perl 6... It's not called Perl 6 anymore. What is it called? Rakuda, or something...?
+
+**Mat Ryer:** Oh, really? How come?
+
+**Brad Fitzpatrick:** They renamed it. I think the Perl 5 people were feeling hurt...
+
+**Johnny Boursiquot:** Yeah, there's some drama about that.
+
+**Brad Fitzpatrick:** Perl 5 people were feeling like they were deprecated if there was a Perl with a greater number...
+
+**Mat Ryer:** Well, it kind of makes sense, doesn't it?
+
+**Brad Fitzpatrick:** It's called Raku now.
+
+**Mat Ryer:** Is it a different team or something then?
+
+**Brad Fitzpatrick:** I don't know... There's probably some people still staying there, but... I mean, it's been going on forever now, right? I mean, it has some fun stuff, and I occasionally -- I mean, I read some of the new things that come out, the summaries of where it's at whenever I see one pop up on the internet somewhere, but... I haven't used it, but I've read some of their design docs. It has some cute stuff... I bring it up sometimes in the Go proposal review meetings, where I half-joke that we should adopt something from Perl 6.
+
+Perl 6 has the dot-dot-dot operator, which - they call it the yadda-yadda-yadda operator, or something... And it's like a to-do, which -- it's valid code; you put it in there. I said that in Go we should make dot-dot-dot as a statement; just basically be a panic. So if you're not done writing a function yet or a method, you can say dot-dot-dot, and it would compile, and it would run, but if it ever executed that line, it would panic and say "To do dot-dot-dot encountered." It would make demo code on slides really cool. It would be syntactically correct code if it's on the slide.
+
+**Mat Ryer:** \[40:05\] It'd be nice if it just ended.
+
+**Brad Fitzpatrick:** \[unintelligible 00:40:05.00\]
+
+**Mat Ryer:** Yeah. It's like "Oh, it's the end of the example." \[unintelligible 00:40:10.01\]
+
+**Brad Fitzpatrick:** Oh, I thought you meant like it would just do \[unintelligible 00:40:12.06\]
+
+**Mat Ryer:** Maybe. That'd work.
+
+**Johnny Boursiquot:** We have another question here from Lewis - where is a bug or feature written in Go by you?
+
+**Brad Fitzpatrick:** The MIME multipart package - it took us like five tries to get that right. Me and Andrew Gerrand wrote it originally when he was in San Francisco... And I don't know -- we had gone out to brunch in the morning before we made it, so we've always blamed the Mimosas for the poor quality... \[laughter\] But it must have been something else, because it was really bad for a while. We just kept finding bugs in it. But I think it's pretty solid now... But whenever I see that package, I kind of cringe, and I'm like "Oh, right, that one..."
+
+**Mat Ryer:** Yeah. It's a complicated thing, but the interface is simple. Did that evolve over time? Did that change, or...?
+
+**Brad Fitzpatrick:** No, no. We can't change the interfaces.
+
+**Mat Ryer:** Oh, I see. You mean this was for the v1 release even.
+
+**Brad Fitzpatrick:** Pretty much... I mean, it was early enough that it basically didn't change. It was just the implementation that was tricky to get right.
+
+**Mat Ryer:** Thanks for doing it though, because we can just use that now, and it works.
+
+**Brad Fitzpatrick:** Yeah... Well, you know, until people find bugs and edge cases in it, but yeah... We'll see.
+
+**Mat Ryer:** Yeah, that's alright. You can't worry too much about that, can you...?
+
+**Johnny Boursiquot:** How do you feel about -- there's always this back and forth between certain community members around adding new features to Go. The hot new thing right now is the debate going on with the generics proposal... Which kind of seems like it's on its way. I'm curious how you feel about introducing these kinds of features in Go? And do you also share the feeling that "Okay, Go is good enough as is, and introducing these things is gonna add unnecessary complexity"? Where do you stand on that?
+
+**Brad Fitzpatrick:** I'm pro generics, for sure. I know there's some community division on this, but I think it will help. I mean, people will abuse it, because people will abuse anything, and there will be bits and pieces of code here and there that are disgusting, but I think by and large it'll make most code more readable. Maybe some of the implementations of these things - like, if someone makes some generic data structures, it might be really hard to read some of this code... But most of the code that you're reading that uses these packages uses some tree-type or some HashMap-type should-be better code than exists today. There should be fewer empty interfaces, and stuff like that.
+
+So I think that'll be good. I'm still disappointed in how error handling turned out, and it seems like the whole Go team has kind of given up on trying to address error handling at this point, because all the previous efforts have kind of failed, and it was so hard to get those previous error handling proposals out, and they were so poorly received... And I feel like they just don't wanna try anything at this point, because there's no ideas... Nobody has any new ideas, and it's so contentious in the community that nobody has the energy to just deal with it...
+
+I kind of think that Go's popularity - it helps it be more popular, but I think it hurts in that it's now too hard to change the language, because there's just too many vocal opinions out there. In the early days Go could change quickly, because nobody cared... But yeah, the Go 1 compatibility thing is a blessing and a curse.
+
+**Mat Ryer:** It's a blessing for us as users of it...
+
+**Brad Fitzpatrick:** Yes and no. I mean, if we were allowed to change the standard library and make it clean and consistent, then you wouldn't see these three or four generations of styles in the standard library, right? We could put context everywhere, or we could build context into the language, and then make it implicit, or something...
+
+**Mat Ryer:** Just do a Perl 6 and make a new language with a different name... \[laughter\]
+
+**Brad Fitzpatrick:** \[44:08\] Yeah, we see how well that worked out for Python 3 and Perl 6, right? In practice, these things don't work. Ian wrote a document about all the languages, and the successful ones are the ones that don't nuke the world and start over, so... You've gotta find a way to do it iteratively.
+
+**Johnny Boursiquot:** How do you manage your modules? That's another point of contention within the Go community, right? Basically, how do we manage the introduction of Go modules and the semantic import versioning, and all that stuff? I'm curious what you feel about it, if anything. Do you just say "Well, we're gonna be at v1 forever, because we don't wanna have a v2", or how do you treat that?
+
+**Brad Fitzpatrick:** Well, it's easier to be at v0 forever. Then you don't have your rules.
+
+**Johnny Boursiquot:** Right...
+
+**Mat Ryer:** So never release, is your advice
+
+**Brad Fitzpatrick:** Yeah... It works.
+
+**Mat Ryer:** You'd be happier.
+
+**Brad Fitzpatrick:** It depends who your audience is. If you're writing stuff for yourself, it's easier to stay at v0 forever, for sure. I honestly haven't had to deal with doing any of the v2 stuff... I see a lot of people complaining about the semantic import versioning, and I could probably believe it... But I just haven't used it enough to feel the pain.
+
+**Mat Ryer:** Yeah. My packages just stay simple. I don't know for version 2... \[crickets\] Don't worry about the silence, Brad. You don't have to feel bad after I say a joke. It's okay to just be an awkward silence... Because they cut it out, you see...
+
+**Johnny Boursiquot:** You hope. \[laughs\]
+
+**Mat Ryer:** I hope.
+
+**Break:** \[45:45\]
+
+**Mat Ryer:** Well, it's that time again... I think it's time for Unpopular Opinions!
+
+**Jingle:** \[47:15\]
+
+**Brad Fitzpatrick:** I was hard-pressed to think of one, but then I just saw this Twitter discussion going by again... I feel like every year or so there's like a Twitter fight about whether HTML is a real programming language or not, with "real" in quotes. And you know, all these people fight about it, and it doesn't seem like too worth fighting about it. One camp is like "Oh, it doesn't have conditions. It's really just a markup language, not a programming language." And the other camp is like "No, don't belittle new people that are just getting started. You should encourage them."
+
+\[47:59\] But I'm pretty firmly in the camp that it's not a real programming language, but that doesn't matter; you're still a nerd typing stuff into a computer. Once you're typing nerdy, arbitrary stuff into a computer, it's not a small jump at that point. It's like a tiny step towards learning a real programming language that has if. \[laughter\] So I don't know... But I only thought of that because I just saw it again going by, and it seems like sacrilegious to say that HTML is not a real programming language... But I will say it.
+
+**Mat Ryer:** That's definitely gonna get your Wikipedia page trashed...
+
+**Brad Fitzpatrick:** Yeah, totally.
+
+**Mat Ryer:** And they'll probably do it using HTML.
+
+**Brad Fitzpatrick:** You're all my nerd friends, even if you just do HTML. You're still a nerd, it's okay.
+
+**Mat Ryer:** Yeah... I think that's it. That's the nice thing. Because that's what people mean, isn't it? They're moaning the "Yeah, we're belittling these newbs that are learning just HTML, bless them."
+
+**Johnny Boursiquot:** Yeah, "Are you a real programmer if you only do HTML?" There's so much identity associated with that, you know?
+
+**Mat Ryer:** Yeah. So any others? Johnny, you haven't given one for ages, mate.
+
+**Johnny Boursiquot:** Well, that's because I haven't been around for ages, remember? I haven't been on for like 3-4 shows, man...
+
+**Mat Ryer:** Yep... \[laughter\] So you admit it... So do one.
+
+**Johnny Boursiquot:** \[laughs\] Well, I don't know; it sounds like Brad wanted to add some flavor.
+
+**Brad Fitzpatrick:** No, I just saw a question go by - if I wanted to remove something from Go, what would I remove?
+
+**Johnny Boursiquot:** Ooh...
+
+**Brad Fitzpatrick:** I think I would remove complex numbers.
+
+**Johnny Boursiquot:** Ooh. Say more.
+
+**Brad Fitzpatrick:** I think I filed a proposal about this a while back... Because it turns out nobody uses them. \[laughter\] And if anyone does use them, I think we should have operator methods instead, so you can define complex numbers as a type, and still be able to use normal math operators on them, with the usual precedence... And then maybe some of the Gonum people would be happy, too.
+
+**Mat Ryer:** Yeah... I mean, I've never used it. I think you might be right; I don't know who uses it. I just stick to simple numbers. I stay away from the complex ones. I don't need them.
+
+**Brad Fitzpatrick:** I think the legend goes that Ken Thompson was very convinced that any modern language needed them because some other language had them, so they added them, but then no one uses them... I don't know.
+
+**Mat Ryer:** That's so interesting, that it's sort of very personal kind of opinions that end up -- you know, that lead to decisions that then, now because Go has grown so much, we sort of see at scale. So it's really interesting hearing about some of those little discussions that people have, and it turns into something, into the language that we then all are using every day.
+
+**Brad Fitzpatrick:** Or not using.
+
+**Mat Ryer:** Or not using it... \[laughter\]
+
+**Johnny Boursiquot:** It's kind of a personal preference. If someone says "Well, any sufficiently modern programming language needs to have complex numbers", one could make the same argument for generics, right? Why wasn't generics part of the v1, right? So yeah, it's preference, at the end of the day... But I have to agree, I've never used them. I think you have to be doing a particular kind of work to need them, right?
+
+**Brad Fitzpatrick:** Yeah, somebody was telling me that even the people that typically need them, there's faster ways to do it without them, or something... So a lot of people working in the math fields, where you would think that they would make sense, don't. Rendering the Mandelbrot set is always a classic example of when you need it... And then people are like "Well, actually, there's faster ways to do the Mandelbrot set without them" etc. \[laughter\]
+
+**Johnny Boursiquot:** Oh, well...
+
+**Mat Ryer:** That doesn't sound like people on the internet...
+
+**Brad Fitzpatrick:** \[unintelligible 00:51:40.26\] say gif or gif wrong...
+
+**Johnny Boursiquot:** Oh yeah, I meant to come back to that... So -- \[laughter\] I have a bone to pick with you.
+
+**Brad Fitzpatrick:** If I recall, one of the specs actually had the pronunciation in the spec, and it was gif, wasn't it?
+
+**Mat Ryer:** I don't know.
+
+**Johnny Boursiquot:** Well, I mean...
+
+**Mat Ryer:** \[52:02\] Johnny, this could be a new regular part of the show. We make a jingle, and we'll do gif versus gif... Just to find out.
+
+**Johnny Boursiquot:** Yeah... I don't know. I've gone back and forth honestly, and I don't know. I guess today I'm feeling like gif, but... \[laughs\]
+
+**Mat Ryer:** I didn't know it had it in the spec. That changes things, doesn't it?
+
+**Brad Fitzpatrick:** Yeah... The spec was called GIF89.TEXT, all-caps... And now I'm trying to find it.
+
+**Mat Ryer:** How did they spell it? \[laughter\] You just read it like that Brad. It just says it's pronounced G-I-F.
+
+**Brad Fitzpatrick:** Yeah, it's not an 89a spec...
+
+**Johnny Boursiquot:** \[laughs\] I can't believe we're doing this...
+
+**Mat Ryer:** This is what I thought a podcast with Brad would turn into... \[unintelligible 00:52:46.16\]
+
+**Johnny Boursiquot:** Oh, man... We can't help ourselves.
+
+**Brad Fitzpatrick:** Yeah. I'll get back to you. I'm sure about this... But you know, memory.
+
+**Mat Ryer:** So do you wanna tell us quickly about tailscale? Because I'm aware of this, but I'd love to hear a bit more about it.
+
+**Brad Fitzpatrick:** Yeah, so we don't have a great elevator pitch, which is maybe a problem for a company...
+
+**Mat Ryer:** Private networks made easy.
+
+**Brad Fitzpatrick:** Yeah, that works... So we're reluctant to call it a VPN, because VPN has all these different meanings to different people... Because people hate their corporate VPNs, and the personal VPNs are often very shady; people trying to hide their IP address while they're downloading torrents, or something like that, and the company is really logging all their stuff and reselling it...
+
+So yeah, VPNs - people have lots of opinions about that term. And we're kind of a VPN, but instead of sending all your stuff and encrypting it to like a hub and spoke, it's all kind of like -- basically, you put tailscale on all your devices, and then they join a network together, and all your devices can reach all of each other. They get a new IP address, that is your tailscale IP address... And anything with tailscale then can contact anything else with tailscale, or your network that you create. And all WireGuard to each other, and it does peer-to-peer, so it won't go through our servers. It'll get through all your NATs, so it will all do the NAT traversal and all the firewall punching tricks... And something like a couple percent of connections that can't penetrate the NAT end up getting proxied through relays we run... But we can't see the traffic, we're just moving UDP packets around, that are all WireGuard-encrypted already, and we don't have the keys...
+
+So it's fun. Networking gets stupid-easy at that point, because you get a connection and you know by its IP address exactly who it is. So you can do old-school host-based authentication again. You don't have to deal with cookies, or logins, or whatever. You know exactly which device and user that connects to you.
+
+Companies are replacing their corporate VPNs with this, and... I don't know. Kind of great.
+
+**Mat Ryer:** Yeah. Very cool.
+
+**Brad Fitzpatrick:** And it'll find the shortest path. So if your office is split between the West Coast and East Coast, and your employees are roaming around - back in the days, when employees could roam around and take flights, and stuff - you would just connect via the shortest path; you wouldn't bounce around the country, proxying packets around...
+
+**Mat Ryer:** Hm. Does it use the packages from the standard library then?
+
+**Brad Fitzpatrick:** Oh yeah, pretty much. I mean, there's certainly things we've had to write, but it uses a lot of the standard library, yeah. And pretty much all of our code is open source. The GUIs for Windows are not open, for instance, but you can run it in Homebrew, on Macs, and stuff, just without the GUI.
+
+**Mat Ryer:** Very cool. And we'll put a link to it in the show notes, for anyone interested in checking that out.
+
+**Brad Fitzpatrick:** Yeah, it's fun to be able to use Go, rather than just work on Go itself.
+
+**Mat Ryer:** Yeah.
+
+**Brad Fitzpatrick:** I've definitely found some parts of the standard library I'd never used before. I was like "Oh, perfect. This exists. I never even knew about it."
+
+**Mat Ryer:** \[56:05\] Well, do you miss it? Do you miss working on the Go project at Google more officially? Or has nothing changed for you?
+
+**Brad Fitzpatrick:** Well, I definitely ignore a lot more bugs that I used to, because now it's not my job to stay on top of a lot of things... But I'm still going to the proposal review meetings... So in that sense I'm still involved. I think I'm the first non-Googler to be part of the proposal review meetings now.
+
+**Mat Ryer:** Oh, congratulations.
+
+**Johnny Boursiquot:** So do you think they'll listen to you if you disagree with something or propose something strongly?
+
+**Brad Fitzpatrick:** Yeah, they're not very contentious meetings... So kind of by definition, things go in once they've had consensus. If anyone disagrees with something and has a strong opinion, everything kind of stops, while people figure out why someone has a strong opinion about something.
+
+**Mat Ryer:** Yeah. It'd be interesting to listen into one of those meetings, Brad. Would you be willing to wear a wire?
+
+**Brad Fitzpatrick:** I could ask Russ.. That'd be a good way to get kicked off of them. \[laughter\] We had discussed at times making them open, or broadcast, or recorded somehow... But it's more of a logistics issue. But I suppose we should revisit that.
+
+**Johnny Boursiquot:** Because it does feel a little bit of a behind-closed-doors kind of discussion, for something that affects the entire community. I'm not putting that sort of dark cloud on it, but if it's closed, and the decisions made during that meeting are the start of new directions, it could feel like that for some.
+
+**Brad Fitzpatrick:** Yeah, so we did change the procedure a few months back... So now there's notes/minutes kept for all of them, and no decision is made in a single meeting. So it now takes two weeks, or a month, depending on the type of issue. We announce our decisions ahead of time, and only once a whole meeting has gone by with no new data, and an assumed direction has already been picked... If this is tentatively accepted or tentatively declined, then once another week goes by with no new input, then we say "Okay, marking accepted. Marking approved."
+
+So if you're not in the meeting, you still know that the decision is coming, and it gives you a chance to object and to raise new information. We stole this from the Rust community, who had a blog post about how it works in their community. It has some name, but I'm forgetting the name now. It's called "No new information", or something like that...
+
+**Mat Ryer:** That's an interesting title for a thing...
+
+**Brad Fitzpatrick:** Yeah, I'm probably screwing it up.
+
+**Mat Ryer:** That's alright. But they aren't deaf to the -- or they aren't ignoring the community. Feedback from the community still feeds into those meetings, doesn't it?
+
+**Brad Fitzpatrick:** Oh, the meetings are entirely in response to issues filed by the community. I mean, occasionally, one of us will file our own issue that then we discuss, but...
+
+And in the past they've been pretty good about trying to bow out of the voting or discussion on things they've done themselves. I forget which proposal it was, but a while back Russ filed a big proposal, and he kind of recused himself from the decision-making on it. He's like "I don't wanna look like I'm just approving my own thing."
+
+**Mat Ryer:** That's very honorable of him.
+
+**Brad Fitzpatrick:** He's an honorable chap.
+
+**Mat Ryer:** He is.
+
+**Johnny Boursiquot:** So have you introduced your kids to programming yet? Someone from the Slack channel wants to know.
+
+**Brad Fitzpatrick:** My older one is three, and the younger one is one, so it's not quite there...
+
+**Johnny Boursiquot:** Might be a little tad bit early.
+
+**Brad Fitzpatrick:** In a couple years, yeah...
+
+**Mat Ryer:** You could start them off with Ruby... \[laughter\]
+
+**Brad Fitzpatrick:** I've been debating a lot what I'm gonna do when the time comes... It has to be engaging enough. It has to be contemporary somehow. I don't think getting him an Apple 2 terminal would be quite so exciting. Lego might be fun; the little turtle graphics is always fun.
+
+**Mat Ryer:** \[01:00:12.25\] If I had a kid, I would bring it up kind of in the same timescale that I was brought up, to have a Spectrum first... You know what I mean? Like, it can learn on that. Because I don't think kids appreciate how good technology is.
+
+**Brad Fitzpatrick:** There is a guy online - he said when his kid reached a certain age, he got to have a certain video game console, and he got him every one that had ever come out in order, and he made them use it for like six months, or something. So he could play Pong or whatever, with the little silly paddle thing, for like six months. And then he could advance to the 8-bit games, and the SEGA... So he can see progressions of decades of game consoles getting better.
+
+**Mat Ryer:** That's brilliant.
+
+**Brad Fitzpatrick:** I thought that was pretty cool.
+
+**Mat Ryer:** Yeah, I think that's great. It's basically my idea, but with some level of commitment...
+
+**Brad Fitzpatrick:** Yeah... Finding all these machines that still work on eBay and whatnot seems like a challenge.
+
+**Mat Ryer:** Yeah... But it's so fun. It is funny; I'm quite nostalgic for that old tech. I actually miss when computers were rubbish. I miss the times when the computers were just like terrible graphics, really slow... I'm quite nostalgic for those times.
+
+**Brad Fitzpatrick:** I mean, computers used to be faster, in a lot of ways. Applications used to be faster than web things are today.
+
+**Mat Ryer:** Hm.
+
+**Johnny Boursiquot:** Yeah, we don't care anymore. We just say "Hey, just throw some SSDs at it" or something.
+
+**Brad Fitzpatrick:** There was a person a while back who analyzed the keyboard input latency from old machines to modern machines, and just showed how computers nowadays - you hit a key and the time it takes to show up in Gmail or whatnot, it takes forever.
+
+**Mat Ryer:** Oh, really? Yeah, that's funny.
+
+**Brad Fitzpatrick:** You know, keyword latency, display latency, operating system latency...
+
+**Mat Ryer:** Yeah... That's funny.
+
+**Johnny Boursiquot:** Along those lines, about teaching kids to learn to appreciate tech - it would be kind of hard to say "Here, child, carry around this rotary dial phone before I can give you smartphone." It might be hard to score that, but you know...
+
+**Mat Ryer:** Yeah... But at least make him have a landline though for a bit first... And then you could be like "Right. Now, watch this!" and cut the wire. And all along it was just an iPhone \[unintelligible 01:02:34.26\] taped to an old phone, and you're like "Look at this. We blow the tiny mines." We did that, I think... \[laughter\]
+
+**Brad Fitzpatrick:** So I found a landline recently, and I was like "Oh, I wanna make this do silly things again." And I used to do silly things with Asterisk back in the day, like controlling a little PCI card that did the tone that made phones ring... And so I went to look at Asterisk again, and after not using it in almost 20 years... And I was like "Oh, it's basically the same. Some things are kind of changed..." I found some blog posts, it was like the history of how this AGI, this Asterisk Gateway Interface, which is like CGI for phones - how it changed over time; there were three generations of things - AGI, and then something else, and now they have ARI. And it was kind of like a "Oh, these are the mistakes that were made, and here's what we learned over time..." And it was kind of cool to see a project that felt like I just used yesterday, but it'd really been 20 years, and I found this retrospect of posts about like 'Yeah, we've made all these changes..." I was like "Wow!" It's kind of cool to watch a project in fast-forward like that.
+
+**Mat Ryer:** Yeah. Fast-forward. If you could fast-forward the kids as well... You could be like "Right, you've been good this week, so now suddenly... 64k. Good boy."
+
+**Brad Fitzpatrick:** "You get more memory. You get more graphics."
+
+**Mat Ryer:** Yeah. "Graphics cards now exist." Just like milestones in tech as rewards for kids. I don't know -- I mean, I just thought of it today, but I am get a domain name as soon as I can for it The first step.
+
+**Johnny Boursiquot:** Alright... So this has been pretty epic and pretty fun. It's been a great joy and pleasure having you on the show, Brad, and a great participation from our folks in the GoTimeFM channel as well. Hopefully, they've gotten all that questions answered and now we're gonna see some updates on your Wikipedia page very soon here... \[laughs\] Yeah, that gif and gif thing - I don't know, man...
+
+Anyways, it was a pleasure having you, and we'll see you next time on the GoTime podcast.
+
+**Brad Fitzpatrick:** Cool. Sounds good. Thanks.

The secret life of gophers_transcript.txt

@@ -0,0 +1,606 @@
+**Mat Ryer:** Hello, and welcome to a very special and installment of Go Time, the GopherCon Mash-up. I'm May Ryer, and welcome to everyone listening live at GopherCon, and also to people listening in the podcast in the future. Very nice for you to travel through time in that way for us.
+
+Today we're talking about the secret life of gophers. We've collected three disparate gophers from about the globe, or from across the flat Earth, if -- you know, I just wanna make sure we're inclusive... \[laughter\] And they're gonna tell us their stories and shed a little bit of light on what it's like to be them. So let's meet them!
+
+Well, Natalie Pistunovich, hello!
+
+**Natalie Pistunovich:** Hello, Mat Ryer!
+
+**Mat Ryer:** Welcome to the show.
+
+**Natalie Pistunovich:** Thank you so much. It's awesome to be here for the second time. Thank you very much for the invitation. You keep making a great host.
+
+**Mat Ryer:** Thank you very much. We've only just started; you won't probably be saying that later... Things tend to deteriorate in the second half. We'll see. So you work at Aerospike, you're a developer advocate there...
+
+**Natalie Pistunovich:** Yes.
+
+**Mat Ryer:** Cool. We'll definitely be interested to learn more about that too later on. We're also joined by Angelica Hill. Hello, Angelica!
+
+**Angelica Hill:** Hello!
+
+**Mat Ryer:** Welcome to Go Time!
+
+**Angelica Hill:** Thank you. I'm very, very happy to be here. First time, and very excited.
+
+**Mat Ryer:** Good, you're more than welcome. And you're a product manager at the New York Times, right?
+
+**Angelica Hill:** Yes, product manager at the New York Times. I manage emails, newsletters, and breaking news alerts. I work with wonderful developers!
+
+**Mat Ryer:** Okay. We'll like to hear more about those. And the New York Times is a newspaper, isn't it? In New York.
+
+**Angelica Hill:** Yeah, it's just this very small publication. I know not very many people know about it... But you know, we're really trying to build up that.
+
+**Mat Ryer:** Yeah. Like a startup. Well, best of luck to you. We wish you all the best. \[laughter\]
+
+**Angelica Hill:** Thank you.
+
+**Mat Ryer:** \[04:07\] And we also have Kris Brandow. Hello, Kris!
+
+**Kris Brandow:** Hello, Mat.
+
+**Mat Ryer:** Welcome to Go Time!
+
+**Kris Brandow:** Thanks for having me back.
+
+**Mat Ryer:** Well, it's a pleasure. You are a senior software engineer at Wunderkind, is that right?
+
+**Kris Brandow:** That is correct.
+
+**Mat Ryer:** Is it Wunderkind or Vunderkind? I feel like it should be...
+
+**Kris Brandow:** There's a lot of debate about how to pronounce it in the company... I say Wunderkind, but...
+
+**Mat Ryer:** Wunderkind.
+
+**Kris Brandow:** ...it's whatever you want it to be.
+
+**Mat Ryer:** Oh, nice. I like that. Well, let's dig in. I mean, we've met you, I think; got the introductions out the way... I wanna ask, first of all, what's the first thing you're gonna do when you can travel again? When Covid's over, what's the first thing you do? What's the first place you're gonna go to? Kris, why don't you kick us off?
+
+**Kris Brandow:** I'm a little bit boring, I'm like an introvert, so I don't travel much... But I definitely want to go back to a bar and just hang out on a Friday night with my friends and have a drink in a crowded bar. I never thought I'd wanna be in a crowded bar again, but like... I just wanna be around a bunch of people. So that's what I would do.
+
+**Mat Ryer:** So it's changing you if you're an introvert. That's a real change.
+
+**Kris Brandow:** Yeah.
+
+**Mat Ryer:** What about you, Natalie? What's the first thing you're gonna do? Any ideas?
+
+**Natalie Pistunovich:** I'm probably gonna go to some Scandinavian country and hike for a week in a row. You know, in Sweden you're allowed to put up a tent anywhere you want. You don't have to go to a dedicated camping site.
+
+**Mat Ryer:** Really? Well, not like a shop, or in a train, or something, right?
+
+**Natalie Pistunovich:** In any public area, I would say. Not in private territories.
+
+**Mat Ryer:** Huh. I'd be doing that all over the--
+
+**Natalie Pistunovich:** But you can go hike, hike, hike, in the middle of the forest set up a tent. You don't need to wait for the camping site. Just go on. Just soak in some nice nature.
+
+**Mat Ryer:** Nice. Angelica, same question to you.
+
+**Angelica Hill:** I mean, I just wanna go home. I'm on the first flight to London. I'm desperate to go home.
+
+**Mat Ryer:** Yeah, I know. It's been tough, for sure. Well, okay, so Go has lots of different fans, and it has people like it for lots of different reasons... So I'm interested, where did you first hear about go, when was that, and what was it about Go specifically that was attractive? Angelica, why don't you kick us off? How did you first hear about Go?
+
+**Angelica Hill:** So I first heard about Go about a year and a half ago... Long story short, I always loved computer science, loved technology, but never really took the time to dive into it. My background's in Shakespeare; I went to drama school, and then I went and studied English literature, and then I went to Columbia Journalism School... So nothing to do with computer science whatsoever.
+
+I started working on a very technical team at the New York Times, thought "Okay, great, this is an excuse for me to go and try to learn more. It's gonna be beneficial to my job, but also I kind of like it a little bit... Enough to invest time." I went to an after-work course and started learning JavaScript, HTML, and became obsessed. It was so fun, it gave me such happiness, such joy... So I basically started asking every single software engineer I could find what language they wrote in, asking tips and tricks... And I stumbled upon Go. One of my colleagues at the New York Times was obsessed with it, and he just went on and on how great it was... So I said "Okay, teach me." And it was the first language that I really felt I could understand, and I kind of had a bit of an epiphany moment where I was like "Oh, this is not all gobbledygook. I can definitely do this." And I just kind of went from there.
+
+**Mat Ryer:** Do you think there was anything specific about Go that made it easy to learn or easy to pick up?
+
+**Angelica Hill:** Honestly, I think it was just that I had a really good support system at the start. I was diving in self-teaching, but I had a group of people I found who loved Go, and really were willing to dedicate the time to help me understand the basics... And also explained them to me in a way that was very easy for my kind of then (I would say) non-technical brain to understand. A lot of analogies were used.
+
+**Mat Ryer:** \[08:06\] Yes, that's very helpful, isn't it? Well, some languages are kind of packed with features, and one of the nice things about Go is because it is a bit of a minimalist language, it has a sort of simplicity to its design... I've heard people say that because of that, it does actually make it quite easy to pick up and learn... And I think that's something -- you know, some other languages have very powerful things they can do, but of course, there's more concepts to learn, like object-oriented programming, for example. You can kind of skip over that in Go, and you don't really miss it, either.
+
+So that's interesting... So you don't have a computer science background; you learned because of just curiosity.
+
+**Angelica Hill:** Yeah, exactly. And I honestly would say there is no language that I don't think I now can't learn; I think it was just Go that got me over the hurdle of saying "No, I'm not technical", but now I feel very strongly that anyone can learn Go, or any software engineering language. They just need to kind of like de-spoof the idea of "I'm not technical/I'm technical." That's not a thing. Everyone is technical. Everyone can do it.
+
+**Mat Ryer:** Right. Well, that's very encouraging to hear. Okay, so Natalie, what about you? I know you've been in and around Go for a while. I've seen you at conferences, doing talks... You did a workshop this time... We've even worked together in the past... So where did you first hear about Go?
+
+**Natalie Pistunovich:** I first heard about Go when I moved to Berlin. Before that, I was thinking I will work in NVIDIA and do really cool, massive, parallel things, and all the hardware, all the things together at the same time... And then I moved to Berlin, and Berlin is all about software, not about hardware. But luckily, the company where I started working - they were using Go before it was even 1.0.
+
+**Mat Ryer:** Oh, wow.
+
+**Natalie Pistunovich:** Yeah. So in 2014, when I joined - this was like the second year of the company, I wanna say - they said "You're welcome to work here, but you have to learn Go." I said, "Alright, let's do this!"
+
+I got to do all the parallelism that I wanted still, so that was a nice turn of events. That's how I got familiar with Go. But my background is in computer; I studied computer engineering... And even in high school, my major, or however you call that, was the same field.
+
+**Mat Ryer:** Okay, cool. That's interesting then... So this was out of necessity for a job. That's unusual, that the team had started using Go before version one... But good for you, I guess.
+
+**Natalie Pistunovich:** Yes, it's indeed not a very exciting way to start... But you know what is very exciting? Taking pictures and tweeting about this podcast and about the conference with the hashtag of GopherCon. \[laughter\]
+
+**Mat Ryer:** Excellent. I just feel like we've had product placement there. Kris, do you wanna plug any products? \[laughter\]
+
+**Kris Brandow:** Not particularly...
+
+**Mat Ryer:** So where did you first encounter Go, and what was it that attracted you to it?
+
+**Kris Brandow:** I think I have a background quite similar to Angelica. Interestingly, my parents both were computer people...
+
+**Mat Ryer:** Wow. Like robots. Like androids. \[laughter\] Not quite human.
+
+**Kris Brandow:** Right. My dad does have a masters in computer engineering, specializing in robotics though, so... Close.
+
+**Mat Ryer:** Wow. That's a cool dad, ain't it?
+
+**Kris Brandow:** Yeah. But growing up, I wanted nothing to do with computers. I would build computers with my dad, but it was not a career I wanted to do. I wanted to be an author. I was like "I'm gonna be a writer, that's what I'm gonna do." At family events people would be like "What are you gonna do when you grow up? Are you gonna go into computers, like your parents?" I'd be like "No, no. I am going to be an author." And that's actually what I went to college for. I got a creative writing and broadcasting mass communications degree, and I was just like "Yeah, this is gonna be great. I'm gonna be a writer." And then senior year rolled around, or really junior year, and there were all of these websites that needed to get built, so I was like "I can sort of do this..." And I started building websites. I was like "Oh, I like this." So I kept doing it, and I learned this fun little content management system called Drupal.
+
+\[12:02\] Basically, during one of those summers I looked at Go a little bit, but I was like "I don't quite understand this, so I'm just gonna stick with Drupal, which I know", which I did for a few more years... And then one night -- I think it was actually on the night that I met Sam Boyer, who is one of my really close friends... We were just sitting in a bar, talking, and I was like "Man, there's something I don't like about the software I'm writing." And he was like "Oh, I know what that is. That's complexity. You don't like complexity. You should look at Go. Go is really a simple language, and it's elegant and it's beautiful. You have to try it out." So I did, and that was kind of the end of me writing Drupal professionally, and the beginning of my Go career. That was like 7+ years ago now. So it's been a wild ride, but that was the introduction to it. Just like "Oh, you like simplicity. Go write this language."
+
+**Mat Ryer:** That's great. I love the idea of the -- because most kids rebel from their parents, in some way... The fact that your parents were in computers - it's kind of funny that you rebel against using tech. My nephews - that would never happen. They are not gonna rebel that way. But that was really interesting to hear that.
+
+So you mentioned you wanted to be an author when you grow up... What do you want to be when you grow up now?
+
+**Kris Brandow:** Still an author... I mean, I also see myself as an author. I have this opinion about software engineering that I've been kind of circling around with my friends, that software engineering is a writing discipline, as well as an engineering discipline. The thing that we all do every day is write. It's source code, it looks a little bit different, but it's still the same sort of thing, and I use a lot of the same skills I learned in college \[unintelligible 00:13:38.10\] to write good software.
+
+**Mat Ryer:** Oh...
+
+**Kris Brandow:** So yeah, I still wanna be an author when I grow up... Just not writing novels or screenplays.
+
+**Mat Ryer:** Yeah, that's really interesting then. So tell me a bit more about that... Have you got an example of something - a principle or a technique that you can apply, that applies both to prose writing or code?
+
+**Kris Brandow:** Yeah, I think a lot of my GopherCon talk from last year talked about this too, about how similar the publishing industry is to the software engineering industry... But a good, concrete example - I think the way that we do code reviews is very similar to the way that copy editing works; the things that you do as you're copy editing are very similar to things that you do as you're code reviewing, just for a slightly different domain.
+
+**Mat Ryer:** It's a good point, actually. I think people should probably use GitHub if they're writing, because you can have pull requests and stuff... And see the changes, and the history, and things. It'd be quite nice to do a pull request to Harry Potter and change it so that Voldemort could win...
+
+**Natalie Pistunovich:** Or to the last season of Game of Thrones.
+
+**Angelica Hill:** Oh my gosh, yes...
+
+**Mat Ryer:** Yeah, redo that. Rewrite it. \[laughter\] Just do a pull request and delete it.
+
+**Natalie Pistunovich:** The whole thing. \[laughter\]
+
+**Mat Ryer:** Yeah... That'd be great. Okay, so has that changed for you, Natalie? Were you always kind of gonna be doing computers, or did you ever want it to be something different?
+
+**Natalie Pistunovich:** I wanted to be an astronaut.
+
+**Mat Ryer:** Right...
+
+**Natalie Pistunovich:** But very early on -- I got my glasses when I was in the sixth grade, so that was kind of a deal-breaker...
+
+**Mat Ryer:** Why?
+
+**Natalie Pistunovich:** Well, you can't be in space, at least these days, but I'm sure that a couple of generations into the future it will not be a big deal. Just like flying planes. And what I really want to be now is a person with enough free time to nerd about space without having to do really committing things like bootcamps, or whatever astronauts (physical training and so on) have to go through. You know, just enjoy the space without the heavy commitment.
+
+**Mat Ryer:** You've got to go in that thing that spins you around dead fast first, I know that...
+
+**Natalie Pistunovich:** \[laughs\]
+
+**Mat Ryer:** But astronauts aren't looking at things, are they? They're just in space; they don't actually drive in it... So I feel like you shouldn't need glasses, but... We'll bring that up with NASA.
+
+**Natalie Pistunovich:** Thank you.
+
+**Mat Ryer:** Yeah. So would you go to space? Because there's some commercial--
+
+**Natalie Pistunovich:** Absolutely.
+
+**Mat Ryer:** Yeah, you would.
+
+**Natalie Pistunovich:** If only I can afford this, I will absolutely do this. Now I'm watching The Expanse and that's as close as I feel to space. There's that show about Mars... There's some space content out there until I can wait it out and go to space.
+
+**Mat Ryer:** I'd probably go, but not on the first couple of flights, you know what I mean? Just let them... "Just go. I'll wait a bit, yeah."
+
+**Natalie Pistunovich:** \[16:15\] \[laughs\]
+
+**Mat Ryer:** So there's something that's interesting... Angelica, you're a product manager, and a lot of us programmers interact with product managers in some way, usually, in a team... And sometimes there's a -- let's call it a healthy tension between them... Tell me, do you hate your devs?
+
+**Angelica Hill:** I love my devs far too much. I am so mushy with my team. I think I tell at least one member that I love them every day. I am a very loving product manager.
+
+**Mat Ryer:** Nice.
+
+**Angelica Hill:** But I would agree, there's definitely a dichotomy there around having a kind of product mentality to the work we're doing, versus being ahead of the game, kind of a software engineering mentality. So I would say I have learned more and more as I go through and work more closely with my team, with other teams. The way that I see our product is very different to the way that my engineers see our product... And that's great, that's wonderful. It means that while we're having product discussions we can have all those various different viewpoints... But it has meant that as I got more technical, as I dived into Go more specifically, because that's our backend language we use, I have had to make the kind of conscious decision "Okay, now we're gonna have an engineering discussion where I'm gonna have my engineering hat on... Okay, now I need to switch to my product hat and actually talk about business value, OKRs etc."
+
+And also, I think just personally -- I was joking before we went live on the fact that I've kind of been a Gopher on the down-low, very kind of secretive about how much I love Go, just because it results in the inevitable question that I get from my manager, and have got from my manager multiple times now, of like "Oh, Angelica, you seem to really like this Go stuff. Are you sure you don't wanna be a software engineer?"
+
+**Mat Ryer:** Hm...
+
+**Angelica Hill:** So I would say there's definitely a tension... I can speak for only myself when I say I love all of my team... But whether they love me, and the fact that I am constantly switching between these two hats, I don't know.
+
+**Mat Ryer:** Yeah, that's really interesting... Because often you work with people and the perspective is so different... And of course, you don't expect product managers to understand all the technical things that are happening, but usually, you can end up in a situation where feature requests or new features and things are kind of written out, sometimes described a bit too well, and they just don't fit in the tech anywhere... So you having a knowledge of Go - does that help you see those things early? Does it help in other ways, too?
+
+**Angelica Hill:** Yeah, so I would say it works in two core ways. One is that I can actually engage in one of these great conversations my engineers are having; I can understand the issues, their complexity, and how big of an issue they are... For example, if we don't have an effective staging environment, if we don't have great unit test coverage - things that if I didn't know anything about software engineering... If an engineer came to me and said "Oh no, Angelica, we need to dedicate a month to writing unit tests", I'd be like "Okay, why is that important? We need to build this new feature."
+
+So I would say it enables me to understand what is more important in terms of tech debt etc. But it also means that when I go to leadership on the product side and they say "Why are you spending two months working on tech debt?", I have that language, I have that ability to really validate it to him and be like "No, this is why it's important. This is why we need to do this stuff. So I think it works well because of that.
+
+\[19:58\] And also honestly, for me, it means that if there are small engineering tasks - which is what I'm aiming to get to - that aren't challenging to my engineers, aren't new problems that they can dive into, aren't things that are really helping them with that growth, I want to get to a place where I can just pick up that ticket and do it. Because it's exciting to me. I'm learning, I'm very new... Why should they have to do these menial tasks, when I can very much do them and they can do the more interesting work?
+
+**Mat Ryer:** Yeah. Well, I think that's great, honestly... To me, it can only be a benefit to the team, having that knowledge, frankly... I'm sure some people don't see it that way. They see it like, you know, you have to have that distance, because you wanna be able to make all these unreasonable demands on their developers, don't you?
+
+**Angelica Hill:** Yeah, I would say there are certainly people - on both sides, both engineering and product - who feel like in order to be truly great at the product discipline you need to really be all-in, all the time, absolutely be empathetic to all the other disciplines, like design, UX etc. that you're working with... But if you get too invested, then you're not gonna be able to have that kind of unbiased, overarching mentality to prioritize what the team is gonna work on... Because I'm just gonna be like "I love my engineers", and whatever they wanna work on, I wanna help them work on that.
+
+**Mat Ryer:** Yeah. That does sound good. If you're hiring, now is the time \[unintelligible 00:21:17.06\]
+
+**Angelica Hill:** \[laughs\]
+
+**Mat Ryer:** That's a bit of a dream sometimes, because... Yeah, we've all had varying successes with product managers in the past. It is a very diplomatic way of, again, saying how much we don't like each other.
+
+**Angelica Hill:** \[laughs\] Well, maybe I can be the first one that you like.
+
+**Mat Ryer:** Well, we'll have to see...
+
+**Angelica Hill:** We'll see.
+
+**Mat Ryer:** But don't hold your breath.
+
+**Angelica Hill:** Okay. \[laughs\]
+
+**Break:** \[21:42\]
+
+**Mat Ryer:** Natalie, you're doing dev rel stuff... So what's that like? Are you just part of the engineering team, or do you have other focuses as well, similar to being a product manager?
+
+**Natalie Pistunovich:** It's interesting as a question for me, because I think different companies do this in a different way... And I joined Aerospike as the first lead developer advocate... So I get to figure out how things should be. They trust me a lot to bring in recommendations. So the person I'm reporting to is the CTO... There is an entire engineering team in parallel to me, so you know, no big deal... Me and sometimes with awesome engineers, reporting to the same person eventually.
+
+\[24:09\] I do work a lot with product, I do work a lot with marketing, and I do kind of try to balance everything altogether. It includes things like building a developer hub. You want a place that developers, once they actually heard of Aerospike, because it seems to be a database that if you know it, you like it and you use it, but if you don't know it, you're like "What is this? Is it aerospace?" That's not the reason I went to work there... \[laughter\]
+
+**Mat Ryer:** Do they make you go in a big spinning centrifugal machine to work there, or no?
+
+**Natalie Pistunovich:** The office is built like a spaceship... No, it's not. I wish it would be.
+
+**Mat Ryer:** It should be. It's a good idea.
+
+**Natalie Pistunovich:** \[laughs\]
+
+**Mat Ryer:** That could be your next initiative.
+
+**Natalie Pistunovich:** Yes... But they do bring to me a lot of questions, like "You are a developer. How would you like to get to know -- if you land on this page and you hear about this for the first time, how would you like your journey to be?" So in this sense, it's somewhere between user experience and product, but then in other things, like "Go and make a cool project now", it's like "Okay!" And then back to the engineering hat. It's a lot of everything.
+
+**Mat Ryer:** That's cool. I mean, the developer experience I think is way more important than historically people have given it credit for, in my experience... As a developer - yes, we're technical, but that doesn't mean everything can just be complicated, and ugly... So yeah, I really do appreciate people on teams that have that eye and care about that developer experience.
+
+You also do a lot of community organizing, and things... For example, do you not -- I know you do... You do the GopherCon EU, don't you?
+
+**Natalie Pistunovich:** GopherCon Europe. We're inclusive even to European non-EU country members.
+
+**Mat Ryer:** Oh, that's good, yeah. I didn't know if I was gonna be allowed or not next time... \[laughter\]
+
+**Natalie Pistunovich:** Yes, yes. GopherCon... Yeah.
+
+**Mat Ryer:** I was gonna say, how did you get into doing that stuff? The community side. And why do you do it?
+
+**Natalie Pistunovich:** Why do I do this? Because it sparks joy. For a while I said "Because I like this", but then a better answer came up; it's more than just enjoying it. It actually is fun to do all this. And how it started - you can go back quite a lot. I am doing the Go User Group in Berlin; before that, I was doing in university all the student union, and so on... I think this is something that just goes on with me wherever I go - organizing things for whatever the community of the place is. And it's nice to have discovered that this exists also in the world of engineering, so it doesn't leave you once you're out of university...
+
+And the jump from organizing a user group to organizing a conference is big, but the organizing team of GopherCon (especially Heather) are super-awesome, and really great, and supportive, and helping a lot with thinking about things, and sharing from their experience... And you make a lot of mistakes, and that's okay. \[laughs\]
+
+**Mat Ryer:** So is there anything as a community organizer - is there any sort of advice you wanna give to people that are attending events? Are there any things you wish everyone just knew and just did?
+
+**Natalie Pistunovich:** Ask a lot of questions. As a speaker, as an organizer, whatever the position I'm in on the other side, if you will, having people engage with you and showing that they want to know more, or they're curious about what you do is really great. And sometimes it's a lot of information, sometimes the talk is not very clear, or too high-level, too low-level for whatever you are, but you can always find something to ask... And that really warms the heart when people do that, so please ask questions.
+
+**Mat Ryer:** \[28:06\] That's nice advice. How do you ensure that you're gonna get good quality when you put the program together? What sort of techniques do you use to make sure -- I'm asking this because look at this conference, GopherCon, and look at GopherCon EU... The quality of the talks are kind of astonishingly good... So how do you do that? How do you keep that quality there?
+
+**Natalie Pistunovich:** Thanks for the compliment to GopherCon Europe; I absolutely agree that the talks in GopherCon here, in this GopherCon, are great. I think the secret is the same everywhere - just have a good reach of your call for papers, to make sure that you get a lot of good and mixed submissions, and that your review committee is also mixed and diverse enough, and has a good, wider image when they build the talks.
+
+**Mat Ryer:** Yeah.
+
+**Natalie Pistunovich:** I would say the reach might be even harder than actually the picking... Even though picking is a really hard job, and having been in some review committees for not just Go conferences, it's really hard to review, and it's even harder to pick what's really good... But reaching out to those people who would otherwise never hear about you when you just use your regular channels - this is the bigger challenge. Because there's awesome speakers, awesome projects out there, and it's very easy to miss them if they're outside of your network.
+
+**Mat Ryer:** Yeah. I think if people can reach those people as well -- if you know somebody that isn't part of these communities, welcoming them in I think is great. And I do think Go does a good job with that.
+
+Kris, you're talking at GopherCon... What's your talk about?
+
+**Kris Brandow:** I mean, my talk just happened, but...
+
+**Mat Ryer:** Yes.
+
+**Kris Brandow:** ...it is about diversity and inclusion, and how we can build better Go community by building a more diverse and more inclusive community.
+
+**Mat Ryer:** Okay, so I find this subject to be really quite interesting. I know that GopherCon in particular - remember the first GopherCon - had a kind of focus from the beginning on making sure that people that were otherwise not represented had a place at the conference... Have you found Go to be doing a good job with this, and is there more we can do?
+
+**Kris Brandow:** Yeah, I think in general the Go community, especially when you compare it to other software engineering communities, is pretty good at this. We have good amounts of diversity, or decent amounts of diversity, and we're a very inclusive community, I find... There haven't been many times when I've been in discussions with people where I felt like "Oh, I don't feel like I belong here. I don't feel like I can be a part of this conversation." So I think that's really good, and a really positive aspect of the Go community... And I think really the way that we get to having a more diverse community is by doing more of that stuff.
+
+I feel like when people first start thinking about diversity and inclusion, it's always like really big things, like "Oh, we should start a program at some school to get more kids to learn software engineering, so in ten years they can become Go engineers." Or we could just watch the way that we interact with each other and speak with each other, and try to be just a little bit more inclusive. So I think that's really what tends to be helpful. I think contrary to a lot of popular opinion, there are a lot of people of different backgrounds spread throughout software engineering as a whole, but I think over time people just stop wanting to participate in communities, because it can be extremely exhausting. It can be extremely exhausting to participate in this industry as a whole when it comes to work... So if you're getting exhausted at work by not feeling included, and by people disrespecting you or sliding you in small ways, you're just not gonna have the energy to go out into the community and deal with the same stuff.
+
+So as a community, we kind of have to compensate for the sad state of software engineering companies right now, and what their cultures are... And I think if we focus on that and focus on "How can we undo some of that uninclusive harm that people are feeling, that's how we grow our community."
+
+**Mat Ryer:** \[32:09\] Yeah, that is really interesting. One of the pushbacks -- and I've had this myself, because when I do a talk, I tend to call out how Go has a focus on this and it's important... And occasionally, I will get people who challenge it. And usually, it's this idea that if everything's meant to be equal, then we shouldn't be talking about this at all.
+
+For me it's about that thing, as you pointed out - you sort of inherit, we inherit these biases, just from the societies that we're in. So it does seem that active action is needed to balance that out. Otherwise, it's like object-oriented programming. You just get it from the base class. Terrible--
+
+**Kris Brandow:** And I also feel like there's a lot of nuance that comes with the words that we use to describe diversity, equity, inclusion, equality... And we haven't gotten the right words for it yet, I don't think. I think it's easy when you say "Equality", people are just like "Yeah, that's the thing." And then people try to dig into it and find ways to undermine what you're saying, when that's not what you mean...
+
+I think there's that really popular image where there's equality, and then equity, and it has a fence, and three different people... And the taller person can see over the fence and the shorter person can't; and equality has the boxes all laid out, and it's not helping the shorter person... And then equity is the one with the boxes where the shorter person can now see, because they're getting more boxes... I've always hated that image, and I didn't know why until I've read an article that was basically like "Even in this image we have problems", because what we're saying is that "Yeah, we shouldn't have the fence at all", but we're also saying the people that have \[unintelligible 00:33:55.07\] literally shorter, which isn't really the case. It's that really the ground is kind of like unlevel, and everybody's the same height. Or maybe the person that can't see is even taller than everybody else, but because they're so much further down the hill, they can't see over the fence.
+
+So what we need to do is more level the field than give people extra things to help them up. And I think that helps people start to understand what it is we're really talking about. Because I think whenever you talk about giving things to a certain group, people always get defensive, and they're like "Well, why does this group get it? Why can't we just give everybody something?" or something like that. When you say "No, the foundation - that very thing that we're all talking about, is extremely uneven", then it's less about "We've just gotta give something to people." Then it's much more understandable for people why we have to fix it.
+
+**Mat Ryer:** I think you've improved that drawing. We should have that made, because that's good. It's better than my base class thing, especially -- I mean, we're at a conference that doesn't have object-oriented programming; I don't know what I was thinking.
+
+**Kris Brandow:** Yeah. I did read that in an article, so I'll try and dig that article up and post it in the Discord channel for people, or something...
+
+**Mat Ryer:** Yeah, please. Yeah.
+
+**Break:** \[35:05\]
+
+**Mat Ryer:** So what's the benefit of having diversity within a team and within a community? Because that's another thing... I've witnessed benefits to having more diversity, but what are they? Let's talk about some of them. Again, I think some people think it's just a kind of one-sided thing, but it benefits everyone, having more diversity.
+
+**Kris Brandow:** Yeah. I haven't dug up the studies in a while, but there's a lot of study going back 30-40 years, of how even some more objective ways of measuring diversity enables you to solve problems easier... Because the more that you have similarities, the more that people get stuck on the same thing. So I think one of the biggest things about diversity, one of the base things about diversity is that it enables you to have different approaches to solving problems. And since most of our job - or literally all of our job - is solving problems, that means the more diversity we have on a team, the easier it will be for us to solve problems more effectively.
+
+And that can be from debugging something, but it can also be from just catching bugs that could be catastrophic in production earlier on, all the way to things like making sure we don't have accidental -isms. One of the most interesting things that I always find about modern day bathrooms in America is that there's a lot of these sensors all over the place, and they don't work really well for people that have darker skin, because a lot of the companies that built them just don't think about people that have darker skin when they calibrate them... So sometimes it's really hard for me to wash my hands, or get a paper towel, just because the companies themselves aren't diverse enough and weren't thinking "Oh, well, we have to augment this to a broader range of skin tones to make sure everybody can use it."
+
+So even those types of bugs we can really start solving if we just have one person that's different from everyone else in a row.
+
+**Mat Ryer:** I see what you mean about it being tiring, because... I mean, yeah, you just don't need that. I'd be totally fine; I'm so pale that all the sensors would be going off when I walk in there, no problem. My problem with American bathrooms is all the gaps in all the doors. Like, make the doors join up, do you know what I mean?
+
+Natalie, you also -- there's a big focus on diversity when you do your community organizing too, isn't there?
+
+**Natalie Pistunovich:** Yes, definitely. I try to bring as much to the table as I can from different experience... Not to bore the people who have heard the previous episode where I was, but I will say in short that I spent a year living in Kenya, and when I went back to Germany, I maintained as many connections as I can, and I use this for things like asking all the local user groups to always participate in meetups and in conferences, and submit to talks, and so on. And now that everything is virtual, this makes this very easy.
+
+This is, of course, becoming a stretch, because it's been a couple of years ago, and the last time I was there it was with Bill. We gave a workshop with Bill Kennedy; we gave a Go workshop at an amazing institute that is called Tunapanda, which is a one-year training, and then people can get jobs as developers... Not just in Go, but also in Go.
+
+And all this keeping in touch with communities that are very much outside of my network is a hard work, but it's also very much worth it. And of course, there's the easier (if you will) diversity work of also making sure to work with the local community, for example, being inclusive more towards women, which is also not a big part of the Berlin tech scene... Working with refugees as much as possible, talking to organizations and asking them to invite the people who \[unintelligible 00:40:16.17\] school kids, or if it's people in university age, and so on - so invite the relevant groups to all sorts of events that we're organizing... Yes, diversity is a hard work, and it's definitely paying off.
+
+**Mat Ryer:** \[40:32\] Yeah. Great. Well, you can see it as well in the conference... And again, I think that also speaks to why the talks and things are so interesting. There are always some great surprises at these conferences too, so... Yeah, I think that's awesome.
+
+I was gonna ask this question to each of you... If you could travel back in time, what would you say to your younger self? And you can't say "Buy Bitcoin." I mean, you can, but you'd have to tell them when to sell it as well, otherwise it's useless. \[laughter\]
+
+Angelica, what would you say to your past self, past Angelica?
+
+**Angelica Hill:** I would probably say "Don't be afraid to give in to your inner dweeb." Because I think I've always known that I was a bit of a dweeb, but I just didn't let it fully out. And had I let out the inner dweeb earlier, then maybe I would be even more dweeby now, and that's so great.
+
+**Mat Ryer:** Right. And that's a good thing, is it?
+
+**Angelica Hill:** Yes, it's a very good thing.
+
+**Mat Ryer:** Okay, that's nice. I like that. Well, I think anyone listening that's perhaps junior or just getting into Go can take that. I think that's a nice bit of advice. Yeah. What about you, Kris - if you had a time machine, what would you do?
+
+**Kris Brandow:** I think I'd tell myself to be kinder to myself. I think I was really hard on myself, or have been really hard on myself over my life. As I've gotten older, I've just realized that I can still be successful, I can still accomplish things while not being as overly critical with myself. I tend to think that everyone else around me is better than me, or smarter than me, or something like that... I'm usually just like "Man, I'm so dumb. I'm so foolish. I don't have anything to share. I probably shouldn't share anything, or I'm gonna look stupid." And that has never really been true throughout my life. I've always had something to share, I've always been able to learn from people, people have always been able to learn from me...
+
+So I think just being kinder to myself with that, and giving myself more space to breathe, or just chill out... And in a lot of ways too, just be okay with being an introvert. Like, no, you don't have to go out with your friends, or you don't have to stay up till 4 AM all the time. You can go, you can leave the party. It'll be okay.
+
+**Mat Ryer:** That's nice, yeah. And again, for anybody suffering with impostor syndrome, I think that's quite a nice thing to hear... Because yeah, we all kind of feel like that. I mean, we're here having a chat... It doesn't seem like it, but there's loads of people watching this right now... I don't wanna put any pressure on anyone, but... Don't mess up! \[laughter\]
+
+**Kris Brandow:** I also feel like as a writer I'm kind of trained to be overly-critical. As a writer, nothing I do is ever really done... Even my GopherCon talk, I'm just like "Oh my god, there's so many things that are wrong with this... Oh my god, everybody will know..." And from my channel, it's like, people really liked what I talked about.
+
+**Mat Ryer:** Yeah.
+
+**Kris Brandow:** So I think that's also in there as well.
+
+**Mat Ryer:** Does it come out of the pedantry that you have, do you think? That you're being so critical? Because when you're doing code, you really kind of need that. It's kind of an asset, pedantry, when it comes to code. Is it that?
+
+**Kris Brandow:** Yeah, and I think that's important to being a writer as well - you kind of have that drive to constantly go over everything you're writing and be like "How can I make this better? How can I make this better?"
+
+One of the things I've learned is school was like "Yeah, you can do that, you can just keep going, but there's a date at which you have to publish." There's a date at which you have to push to production. You can't just go around in circles forever. It's good enough. Put it out there, move on to the next thing.
+
+**Mat Ryer:** Yeah. And that does apply to software. We've just launched Firesearch.dev, which is a full-text search thing, and there's more things we want to do in it, but you have to release it at some point. Someone said if you're not embarrassed by your first release, then you released too late. I think there's probably some truth in that.
+
+\[44:20\] How about you, Natalie? You've gone into space, you've gone in the rocket, it's gone off, but something's gone wrong, and you've traveled through time and gone back... And now you're gonna have a little word with yourself; what are you gonna say?
+
+**Natalie Pistunovich:** Something between "Assume less" and "Ask more". In my first full-time job in that company that asked me to learn Go I was the junior member in a team full of senior backend engineers, and that was completely terrifying. And they were incredibly nice. They were nice, and they were explaining to me... And they saw that I'm scared to ask questions; they would sometimes explain to me more, and even when they'd say "Do you understand?" and I was like "I have no idea what you're talking about", I would say "Hm..." Stop doing that. \[laughs\] Don't assume that people expect you--
+
+**Mat Ryer:** Just say "No, I don't know what you're talking about."
+
+**Natalie Pistunovich:** Yes, exactly. Don't assume that people expect you to know this. Don't be afraid to ask. And another really good thing is write down the answers, because in the beginning -- anytime you go into a new company, but especially as a junior, everything is overwhelming, and a lot, and it will mess up in your head, even if it does make sense in that specific moment because somebody was patient enough to explain you for a long time... Write it all down.
+
+**Mat Ryer:** Okay, I think that's great advice. They say "When you assume, it makes an ass out of both of us." \[laughter\] So what about some technical things then? Let's see if there's any interesting technical bits of advice we can give people. What IDE do you use, for example, when you write in Go?
+
+**Natalie Pistunovich:** I'm using VS Code. I'm not using too many frameworks, too many plugins. I would say it's pretty basic, the way that I like my setup in the IDE... But also in the physical setup. Even though I've been working from home for a good few years, I only got the nicer equipment like a light ring or like a microphone now for GopherCon. This is \[unintelligible 00:46:15.26\]
+
+**Mat Ryer:** So you have a sort of understated minimalist environment.
+
+**Natalie Pistunovich:** Just work on the floor and that's it, you know... \[laughs\] No, I'm kidding.
+
+**Mat Ryer:** \[laughs\] Yeah. What about you, Kris? What IDE do you use?
+
+**Kris Brandow:** I use Vim.
+
+**Mat Ryer:** Oh, here we go...
+
+**Kris Brandow:** I learned Vim early on in my career, actually. I got tired of trying to load up some of those Eclipse-based IDEs, and I was just like "Oh, this is taking too long." I was working on servers all the time, and I'm like "Okay, Vim or Emacs. I have to pick one." I'm like "Vim is installed in more places. I'll just do that one."
+
+Now I'm just kind of stuck with it. Whenever I try to go to an IDE, I'm just like -- the proficiency loss and the productivity loss is just too much for me. I'm just like "I'll just go back to Vim. I'll just use that. I'm home there." I know how everything works, I have everything set up the way I like it.
+
+**Mat Ryer:** Yeah, you have to learn Vim when you're young, while you're brain is malleable. You can't learn it when you get older; it's impossible. But yeah, and you can do Vim -- like, keyboard shortcuts in VS Code, can't you? But...
+
+**Kris Brandow:** Yeah, I've definitely tried to move to the shinier GUI editors before, and I'd have Vim emulation modes... But I don't know, I think I just use more advanced things in Vim, so it never really quite matches us, and it just makes a giant mess... And I'm just like "I'll just keep using Vim. It's working. I'm happy." I spent the three weeks eight years ago to really get on board with this, so I'll just keep that investment rolling forward.
+
+**Mat Ryer:** And it doesn't hurt that you look like a wizard.
+
+**Natalie Pistunovich:** And you have to change that picture behind you, Kris.
+
+**Kris Brandow:** Yeah, I know. I wouldn't be able to have my Vim poster if I didn't write in Vim all the time.
+
+**Mat Ryer:** \[laughs\] Yeah, I'm surprised that's so graphical, that poster. I would expect it to just be text. \[laughter\] So Angelica, what IDE do you use when you write Go?
+
+**Angelica Hill:** I'm team VS Code.
+
+**Mat Ryer:** Yeah.
+
+**Angelica Hill:** \[48:11\] But I would love to learn Vim, and any other recommendations... I'm very much at the start of my gopherizing journey, so I would love to hear more about other text editors, other things people use, because I'm not set on VS Code. I don't know enough about all the others to really decide yet.
+
+**Mat Ryer:** Yeah, fair enough. I mean, if you use Vim, to the rest of us you look like a wizard. That's just true. So it's got that going for it. But you have to do all kinds of strange judo with your fingers for the keyboard shortcuts.
+
+**Kris Brandow:** That's Emacs, not Vim.
+
+**Mat Ryer:** Oh, okay... \[laughter\] I didn't mean to get into that, but... I don't know. I use VS Code, too. So tell me about a mistake that you've made, technical or otherwise, that you could perhaps warn people about. We kind of have talked a bit about this with the time travel thing, but... Are there any other mistakes or things that you'd just wanna share?
+
+**Angelica Hill:** Don't use your product management credential access to give yourself production access to everything... And then just try and potter around.
+
+**Mat Ryer:** \[laughs\] Okay. Yeah, that's actually a great one. Because you can... You can just go and do anything, can't you?
+
+**Angelica Hill:** \[laughs\] Don't just go and potter around the code in production as a learning exercise. Just don't. \[laughter\]
+
+**Mat Ryer:** Okay, I think that's perfectly fair. That's a good one. We'll have to hear the story about what's gone on there at some point. Maybe you can come back on Go Time at some point and you can tell us about that.
+
+**Angelica Hill:** Oh, a catastrophe...
+
+**Mat Ryer:** Yes. What about you, Kris? What's a mistake that you would warn people about?
+
+**Kris Brandow:** I've made so many mistakes I can't choose one. \[laughs\] I would definitely say like be wary of burnout, and when you're actually in the stages of burnout, and also give yourself way more time than you think you need to get out of that burnout.
+
+I think work environments are especially difficult right now, especially with companies accidentally gas-lighting people all the time... And that leads very quickly to burnout, and I think it's very hard to recognize when you're burned out, and then also just like -- it takes way longer to undo that burnout than it does to get to the place where you're burned out.
+
+So yeah, I think my advice to people would be like "Just give yourself space to recover." If you don't feel like doing anything, just don't do anything. Just relax. You don't have to do anything.
+
+**Mat Ryer:** So how does that sound from a product management point of view, Angelica?
+
+**Angelica Hill:** I'm a big advocate for taking time off and making sure that you aren't exhausted writing code. I mean, we in the lead-up to the elections were on Google Hangouts from very early in the morning until 11 PM at night, and I was not supportive of that, but I was supportive of us getting ready for the elections, elections, meaning this week is much more chill.
+
+So I definitely think that from a product point of view, absolutely, it's important. I take a lot of time to do 101s with everyone on my team to make sure that people are sleeping enough, because otherwise it's detrimental, both to the product and to them as people. I don't want them to be exhausted.
+
+**Mat Ryer:** Yeah, I think that's a good one. If I'm tired, I can do more harm than good in code, so... But yeah, it does take time to figure that out. I have to interrupt, because it's that time, believe it or not... It's time for Unpopular Opinions.
+
+**Jingle:** \[51:50\] to \[52:07\]
+
+**Mat Ryer:** So who is going to share their unpopular opinion first? Has anyone got one for us? Natalie Pistunovich...
+
+**Natalie Pistunovich:** \[laughs\] My unpopular opinion is "Commit to things that you think are way too big and make a really bad idea."
+
+**Mat Ryer:** Do you think you should commit to things that are beyond the realms of possibility?
+
+**Natalie Pistunovich:** In the right context, I should probably say... And when I say this, what I have in mind - probably the first few conference talks that I gave -- I was asked "Would you like to give a talk?" I was like "Why me? No, it's a bad idea. I don't think I should do this." "Yes, I'll do this." And then all the way until that very point, I was like "Ugh... It's gonna be terrible. I'm not prepared enough. The slides are gonna be not clear. My accent is funny... Everything's gonna be wrong." And then the minute before, it's like "Oh my god, I'm sick. Tell them I cannot come. Call off the entire thing", and then it's like "Wow, okay. Maybe it's not so bad. It was terrifying, but I am happy that I did this."
+
+So most of the decisions, the professional decisions I'm trying to make are reasonable, and I'm trying to think "Do I want to interview for this company? Do I want to take this task (when we do sprint planning)?" and things like this... But leave space for really bad ideas, that scare you. Not in the sense bad that they will harm you, but bad in the sense that you're terrified.
+
+**Mat Ryer:** I think that's a nice one. What do you think of that one, Kris?
+
+**Kris Brandow:** I definitely agree. I don't think it's that unpopular, because I definitely agree with it... Yeah, it's important to just go out there and take on big things. That's definitely what I did with my first GopherCon talk. I was like "I'm gonna submit this, but I don't think it's really worthy of getting it up on a stage and talking about..." And then I did it, and people liked it, and I was like "Oh, okay..." So aim big, even for things that you aren't quite sure now are good ideas for you to do.
+
+**Mat Ryer:** Yeah. Any other unpopular opinions?
+
+**Kris Brandow:** I've got one...
+
+**Mat Ryer:** Hmmm...?
+
+**Kris Brandow:** I feel like this actually might be an unpopular opinion... I guess it really depends on who you are, but I think a lot of the things we usually see as disadvantages, especially when it comes to the D&I space, like race, or gender, or sexual orientation - they can actually be advantages in a lot of ways... Say like "Well, you get less things. You don't get as much of a leg up because you're a black person within a white and Asian-dominated industry...", but I see that as like "Oh, well I have to work harder - yes. But then I know how to work harder, so I can just keep working harder." I have the extra stamina, I have the ability to keep going.
+
+Or as a queer person, people are like "Oh, I would never wanna be queer", and it's like "Well, I got to choose my life. I got to sit down and think about and figure out what it was that I wanted my life to be", and I see that as like a tremendous advantage.
+
+So I think in a lot of ways the things we usually see as disadvantages are more just like differences, and in some cases, as the world changes, they can become advantages.
+
+**Mat Ryer:** Oh, wow. That is very positive silver linings there... Nice one. I like that one. I really hope that's not very unpopular. Angelica, do you have an unpopular opinion?
+
+**Angelica Hill:** I mean, my unpopular opinion has already been touched upon by Kris, so I feel like it's less unpopular on this call... Because I was gonna say I think that despite what I certainly was told growing up - software engineering is not closer in my mind to the maths, and science-tied. I think it's much closer to the humanities and the linguistics side... And I would go as far as to say it might be one of the humanities; it's about interaction, it's about building stuff for the users, but it's also about a language that you are speaking to your computer. It's like French. My computer is a French person. I'm learning the language to speak to my computer, so that it knows what I want it to do, and what I'm trying to say.
+
+So I would say that's my unpopular opinion, but I think maybe shared by my fellow writer, Kris.
+
+**Kris Brandow:** \[56:16\] Yeah. I think another interesting point in what you said is that while we do use programming languages to talk to computers, I think we use them to talk to each other more than we use it to talk to the computer.
+
+**Mat Ryer:** Yeah, because the code gets compiled for the computer, doesn't it? When it's code, we're the ones reading it and writing it. Is that what you mean, Kris?
+
+**Kris Brandow:** Yeah, yeah.
+
+**Angelica Hill:** Exactly.
+
+**Kris Brandow:** Hence why it's a writing discipline.
+
+**Angelica Hill:** And it's beautifully creative. I think certainly when I was being told about what software engineering was, the creativity was left out. It's so fun and so creative, and it's a real creative outlet. It's great!
+
+**Mat Ryer:** Yeah, it is. Even like in Go, which prides itself in some ways of there not being too many different ways to do things... Still, there's so many different ways you can tackle a problem. Yeah, I think it completely is a kind of creative endeavor. I always find programmers tend to also be creative in other ways, too... So yeah, I find that to go hand-in-hand. Have you noticed that?
+
+**Kris Brandow:** Absolutely.
+
+**Angelica Hill:** A hundred percent.
+
+**Mat Ryer:** Yeah. Natalie, what do you think about those unpopular opinions that they've said?
+
+**Natalie Pistunovich:** I very much agree... And yeah, I also very much like how positive Kris said you can look on things that are -- your disadvantage is actually your advantage, and how you can go about this. I'm very happy to hear those.
+
+**Mat Ryer:** Yeah. But don't do a talk like "The silver lining to racism", or anything like that. \[laughter\]
+
+**Natalie Pistunovich:** No, no...
+
+**Angelica Hill:** I one hundred percent agree with you, Kris. But I would also say -- to slightly twist that a little bit, I also think for people who are maybe more new to Go, who are newbies to this space, that's a real advantage. How often in your adult life do you get to start from zero? Like when you went to school and learned a whole new thing from scratch...
+
+I think being new is such an advantage, and it also plays into what you would tell yourself, Natalie, when you were little - you get to ask all the questions, and people just assume you're gonna ask all of the silly questions, because you're new.
+
+**Natalie Pistunovich:** Exactly. And sometimes asking those silly questions would even shed a new light on something that people are like "Why actually do we do this this way? That doesn't make sense."
+
+**Angelica Hill:** Yeah, exactly. 
+
+**Natalie Pistunovich:**
+So you can bring a lot of contribution without even realizing, just by asking.
+
+**Mat Ryer:** That's true. I've been on a team where we had a junior member that was asking all these things, and people were saying "We didn't know that." Because sometimes you feel like you can't ask those questions, especially if you're meant to be senior in a team. So yeah, I think we should just be a bit more chilled out about things like that. We are a bit too worried; and it must come from something real, but... We are definitely a little bit worried about asking the silly questions sometimes. And you know, it's okay. That's why I think trust in a team is very important. That's why, Angelica, the way that you think about your dev team I think is a very healthy way to do it. For that reason. You can kind of foster this environment that people can make mistakes, that people can ask those questions that no one's going to ask. Yeah, I think that's a great one.
+
+**Angelica Hill:** And if they don't wanna ask it, they slack me and I ask it.
+
+**Mat Ryer:** There you go. It's like a service.
+
+**Angelica Hill:** Yeah, it's great.
+
+**Mat Ryer:** Do you charge for that?
+
+**Angelica Hill:** I'm the newbie. I'm a newbie for hire. Ask all your silly questions.
+
+**Mat Ryer:** It's a good SaaS. you could actually do that as a SaaS.
+
+**Angelica Hill:** I'm ready.
+
+**Mat Ryer:** NaaS. Noobs as a Service.
+
+**Angelica Hill:** Slack me, we can make it happen. \[laughter\]
+
+**Mat Ryer:** This is a good idea. Okay, we are running very close -- in fact, I think we are more or less out of time. That just leaves me enough time to say thank you so much to our guests... Kris, thank you very much. I hope you can come back again another time.
+
+**Kris Brandow:** Yeah, thanks for having me.
+
+**Mat Ryer:** Angelica, you too. Thanks for your contributions, especially how you treat your devs.
+
+**Angelica Hill:** I try. Thank you so much for having me.
+
+**Mat Ryer:** Yeah. And Natalie, it's always a pleasure.
+
+**Natalie Pistunovich:** Thank you very much, Mat Ryer.
+
+**Mat Ryer:** How was your workshop that you did, by the way? Did you enjoy it?
+
+**Natalie Pistunovich:** Terrifying. But yes, I enjoyed it. Like a rollercoaster.
+
+**Mat Ryer:** Well, we don't know what that means, but we want to find out another time; we've run out of time now. Thanks, everyone! See you next time.
+
+**Outro:** \[01:00:35.06\]
+
+**Mat Ryer:** I think we're off now... We may not be, I don't know; we might still be on.
+
+**Natalie Pistunovich:** I like arcade's tunes.
+
+**Mat Ryer:** Yeah, yeah. It's like Mario, ain't it?
+
+**Angelica Hill:** \[laughs\]
+
+**Mat Ryer:** Are we off-air now? Yeah, we don't know.
+
+**Natalie Pistunovich:** It shows "Live on custom live streaming services."
+
+**Mat Ryer:** Yeah, they never know what's going.
+
+**Angelica Hill:** And in Slack.
+
+**Kris Brandow:** Oh, they faded away.
+
+**Angelica Hill:** Yeah.
+
+**Jerod Santo:** Yeah, we're good now. When I came back I wasn't sure, but we got the clear, so we're all done. I'll turn this up...
+
+**Mat Ryer:** Yeah, that was great. Thank you so much. So much interesting stuff in that.
+
+**Angelica Hill:** It was so fun!
+
+**Jerod Santo:** Awesome show, guys. That was awesome.
+
+**Kris Brandow:** Yeah. I like that as like the nice like right after my talk... I felt that was like a nice little -- I can't word right now... But just a little transition.
+
+**Mat Ryer:** Yeah. We should maybe do that then at the real GopherCon next time...
+
+**Jerod Santo:** Totally.
+
+**Mat Ryer:** ...as they come off stage. Go straight into a little interview. That'd be nice.
+
+**Jerod Santo:** Yeah. Right into the recording booth for the after show, you know?
+
+**Mat Ryer:** Yeah. When you start drinking--
+
+**Jerod Santo:** Post-talk analysis, where Mat just breaks down your talk and tells you how awesome it was.
+
+**Mat Ryer:** Awesome, yeah.
+
+**Kris Brandow:** I'd very much appreciate that. How awesome my talk was. \[laughter\]
+
+**Mat Ryer:** Yeah, unless it's... Not very. But I've seen you talk, so I know you do good ones. I'm looking forward to seeing the GopherCon one. Angelica, you're speaking, too.
+
+**Angelica Hill:** Yeah, tomorrow.
+
+**Mat Ryer:** Tomorrow. Hm.
+
+**Angelica Hill:** "Anyone can be a gopher."
+
+**Jerod Santo:** I love it.
+
+**Angelica Hill:** You can hear more in-depth about my journey to gopherize.

The trouble with databases_transcript.txt

@@ -0,0 +1,349 @@
+**Mat Ryer:** Hello, and welcome to Go Time. I'm Mat Ryer, and today we're talking about databases. I'm joined by Jon Calhoun. Hello, Jon.
+
+**Jon Calhoun:** Hey, Mat.
+
+**Mat Ryer:** How's it going?
+
+**Jon Calhoun:** Pretty good.
+
+**Mat Ryer:** Have you ever used a database?
+
+**Jon Calhoun:** A couple times...
+
+**Mat Ryer:** Oh, cool! Well, we are very lucky to have an expert, I would say, in this field, which is gonna be very helpful for this show... It's Jaana B. Dogan. Hello, Jaana.
+
+**Jaana Dogan:** Hello. How are you? I wouldn't say that I'm an expert, by the way. I'm more of like a frustrated user who decided to work on databases as a revenge, type of thing... \[laughter\]
+
+**Mat Ryer:** That's even better. It's such a massive subject... Maybe we could start with some of the basics, like what are the main types of database that developers can choose.
+
+**Jaana Dogan:** Databases is a very large spectrum, if you think about the way we categorize things. There are a few dimensions. For example, one of the dimensions is we have relational databases, and then we have schemaless databases. On the other hand, we have all these niche databases such as DocumentDB, GraphDB... It's more of like the data model you have, and the way you access that data model. It's a wide spectrum to talk about how we categorize databases, and I think partially today we will be talking a lot about how difficult it is to pick, how difficult it is to verify that that's the DB that fits the problem space, and so on. Even from a very high-level overview, databases are hard to categorize and hard to pick.
+
+**Jon Calhoun:** \[03:53\] I'd even say especially now that-- I forget the name of it, but there's a couple of companies that essentially offer products that are like Google Spreadsheets as your database... So somebody who's not technical can go in and view stuff, but you can sort of use it as the database and it's slower, but it's like a new way of looking at it... And even that - you might use it for certain things, like storing stuff about customers, but it doesn't fit into any of the traditional database roles, in the sense that it's not performant in any of the good ways that you'd want for a high-load database.
+
+**Mat Ryer:** Yeah, but for small cases.
+
+**Jaana Dogan:** Definitely. Anything -- in any engine that you can store data... Not even an engine - any place that you can store data can be named as a database. I'm seeing sometimes blob storage sort of engines are also being called as a database, because you can index things interestingly, you can still query things. There are like column-based storage, blob storage, infrastructure and so on... Some people call them databases. It's hard to tell what is a database and what is not. I used to think that I need to just kind of go back and work on stateful workflows, and that's how I've found myself in databases, because everything that kind of like fits into the stateful workload kind of has a really huge dependency on the databases.
+
+**Mat Ryer:** Very interesting. I've used -- extensively, I used to just use relational databases. For people who aren't familiar, you have tables (spreadsheets), and if you want to reference data across tables, you use a foreign key. So you point to the primary key in another table, and then later at query time you can join that data together. But at scale, that gets a little bit difficult to do, because that data is physically sometimes spread around, isn't it? And is that where schemaless databases kind of came from?
+
+**Jaana Dogan:** Yeah, absolutely. The way the database works or the way it models things has a lot of things in common with the storage engine. So the way you store, the way you shard, the way you really partition the data has a lot to do with the type of capabilities it provides to query... So from a high-level perspective it's always important for a user to understand how at a lower layer things are stored, so you can estimate what is feasible, what use cases are actually a good fit for that type of database. Even though it sounds like a bit of work, I really suggest people to take a look at what type of use cases make sense, and at the end of the day at the storage level what do they do before evaluating anything.
+
+**Mat Ryer:** Yeah, it's interesting, because when you have small data, you can kind of get away with anything. If you don't have loads of data, any database is probably gonna be fine. In some cases you could even, like you say, use blob stores or things in different ways. When we built the gopherize.me website, all the artwork is essentially just files in the blob store, and then there's a process that reads them, just lists them, and then it just saves a -- I think it was a JSON blob or something, that it would then load into memory very quickly. That's because it's kind of tiny data. You can get away with that. As soon as you start hitting scale though, things get a little bit more interesting, don't they?
+
+**Jaana Dogan:** Yeah, especially if you wanna run interesting queries. The example you gave with relational databases - you wanna join multiple tables; some of them might be on one particular machine, some data might be on a different machine, and the engine should bring all this together, do the right filtering and whatever, and then join everything and then serve it to the user.
+
+\[07:55\] I think from the perspective of if there's one node, for example, in a system, in a database where you only have one node, everything is pretty simple. As soon as you start distributing it, it's becoming complicated. Even in one node you have difficulties with different partitioning schemes and so on, so it's giving you a huge advantage to learn how the storage really works. It really gives you a lot of understanding of what is feasible and what are some of the capabilities, and what type of data or schema makes sense in particular use cases.
+
+**Jon Calhoun:** Related to this - Mat, you talked about scale a couple times, and going up to scale in different size of databases... One of the things that's interesting to me about that is that -- I guess the question would more be "Do you think databases would have evolved the way they have if we had the hardware we have now when they needed to start evolving?"
+
+**Jaana Dogan:** That's a really good question. I think there's one more thing that we should consider - the world was not disconnected... So I think in the last ten years people are really looking for faster results; they're looking for more complicated queries. We just realized that this is a trend at Google especially. In the beginning I think most of our systems were much simpler.
+
+Over time, people's tolerance to latency was going low; at the same time, applications want to provide more interesting stuff, so they have to run more complicated queries, join more stuff together... And especially, think about a social network - there's all these different data models and different databases you need, because the data is becoming way too connected. You have your friends, and friends of friends, different schemas, connections and filtering requirements and so on. Everything is becoming more complicated, and at the same time, our networking for example is getting better, our computing power is getting cheaper... So there are definitely different trade-offs nowadays.
+
+**Mat Ryer:** Yeah, it's interesting, because when I think about the different types of data stores that I'm familiar with, broadly speaking schemaless data stores like MongoDB, or Cloud Firestore versus the relational databases, in a way if you know exactly what data you want, it's kind of easier in schemaless data stores. You kind of pre-prepare the data, in the sense that you often denormalize that data by essentially instead of joining data at query time, when you write the fields, you do the work to gather the data together that you need for read.
+
+A simple example is if you have a person with addresses - in a relational database that would be that you have a person table and then an address table, and there would be some foreign key that linked them together, and then at query time you can get all of the addresses for a person. And if you then create an order, it's another reference to the address, and things like this.
+
+In a denormalized world, that would be -- you would actually copy probably the address into the order, so that you don't have to go and find where that address is in order to have then the complete picture of that order. But of course, that is a trade-off, because now if that address gets updated or something changes - maybe you update your phone number on that - you then would have to actively go and find all the places where you've copied that data and update it, or do something in the application to kind of get around that.
+
+I find that with rapid development, this very fast, iterative development, relational databases in some ways are quite a lot easier, because you can cook data in different and unexpected ways after the fact. When you do with schemaless data, especially if you really want those reads to be rapid, you can't really do that as much. You have to have written it in the right way in the first place. So I find that to be sometimes a little bit more difficult.
+
+**Jaana Dogan:** \[12:11\] I think we had this conversation before at Go Time... We were talking about how it's easier to iterate on the schema, without the schemaless setting. And then once you are ready, some of the organizations are actually taking it and translating it to more of like relational and schema-based type of models. That's common, but it's also sometimes very difficult, because you're taking a database with a completely different set of trade-offs plus performance characteristics and trying to represent the data in a completely different way, in a new db; if you don't have experience with the db or migrating to, and especially if you're a slightly larger, more successful project, things are getting harder. But we see this shift too often...
+
+On the other hand, with some applications where -- financial applications, for example, they really care about consistency, they really want their transactions, and they would really want their foreign keys... So they understand that it's difficult, iteration will take a while, but they start with the relational databases, because they know that's the way to go.
+
+**Mat Ryer:** Yeah, I've seen that before as well, where you start with one type of database and then there's a particular problem that it doesn't do very well at... So them you introduce a different kind of database and have them both. That I think is probably more likely than -- we sometimes build these abstractions imagining that we can switch out our database. You're probably never gonna do that, but having different types of databases and representing data in different ways - it's kind of an optimization really after the fact, once you understand more about what you need or where your bottlenecks are.
+
+**Jon Calhoun:** The classic example of that that I've heard - I've been told that at Stripe one of the common things they've done is that they have a NoSQL database that they're using for all the really high-speed transactions, but then on the backend when they wanna run analytics and do all these other things, it's really hard to do that, and a lot of times people want SQL, they wanna be able to use some tools that use SQL for that... So they actually take a lot of that data and translate it into a SQL database. And while it's delayed, it's only used internally, so that's okay... So they're taking that trade-off and deciding "It's useful to have this data in both formats..." And like you said, they didn't switch from one to the other; it's more of a "This makes sense for this use case, and we ported over to this for another use case."
+
+**Jaana Dogan:** In my experience I'm seeing always two or three databases in a system. You can't really fight the trade-offs; you get benefit from them differently. There's usually a relational database, another database for warehousing reasons, like analytics and so on, and then there's usually a database or something like Elastic for search reasons. So you can at least list three core data resources.
+
+**Mat Ryer:** Yeah, and then of course backup could even be a different one, where you're taking a backup and putting it in some kind of cold storage, or less active place. It's common, I think, for developers to want to get the perfect solution from the beginning and just build that, but probably a better strategy is to just start with something, one thing, simple, do what you're gonna do with it. And then as it starts to become a problem - hopefully you are keeping an eye on it - then you can start to think about these things and see if there's perhaps a different technology that would work.
+
+**Jaana Dogan:** \[15:40\] I've seen a trend of -- people are starting from the database they know. For example, we're talking about a lot of limitations around MySQL, but if you consider the number of people who are using MySQL, it's kind of contradicting. But each time I talk to people who are very large users of MySQL, their number one argument is this - databases is a huge world; we really know MySQL, and that's a really safe path, because we know how it scales, how it fails... Even though there are difficulties or even though we will have to do some more work ourselves, at least we know what to do... And then they pivot from there. That's a really common thing - start at MySQL for your relational database, and then start evaluating things over time. Probably that's not going to cover whatever you need from a database.
+
+**Mat Ryer:** Yeah, yeah. I think also if there's a technology that you'd prefer, as well, in the early stages of a project that goes quite a long way if you enjoy that sort of technology. I'm building a project at the moment using Cloud Firestore, which is this schemaless database, and it's extremely fast. Reads are extremely quick. But it's quite limited in what you can do, and you certainly can't join data in the way that you would with MySQL. In fact, if you want to even query or filter on multiple fields, you have to actively create an index for that. So that is quite a change, too. If you've only used relational databases in the past, that's quite a strange thing to then be faced with... And since you don't do joins in it, it means you get quite good at denormalizing and then working around the trade-offs that come with that denormalization.
+
+It is very interesting... And in that case, with that technology, the limits actually really force you to understand the underlying technology and how it works, and that means -- I feel like we are making better decisions. It's not just a case of "The data is just there and I can get whatever data I like." We really have to think about "How do we want to read this data? How are we gonna present it to the users? What do they care about?" You really have to think about that at write time, versus being able to just do whatever you want later.
+
+**Jaana Dogan:** Yeah, and even adopt new databases if your requirements change. You're iterating on your product and at some point you realize that you actually need this type of querying, and this is your latency requirements etc. and then you just bring in another database. You duplicate some data, but that's how it works.
+
+**Mat Ryer:** Yeah, and that's what we've done. We actually have BigQuery as well, as a part of this too. And there's other database technologies, and another one is Spanner, the Google product. And Jaana, you work on Spanner, don't you?
+
+**Jaana Dogan:** Yeah, so I actually switched to the Spanner team almost a month ago, or maybe it's been more than a month... Time is flying to me nowadays. One reason I actually wanted to work on Spanner is I wanted to go back to databases. Before coming to Google I was mainly working on databases... And the reason I came to Google is actually I was table-flipping every day, because we were losing data, or we had performance problems and so on, and I just wanted to go to a company where databases are sorted out, so I can delegate my state work, state problems to something else, and I can focus on my app, my business logic, my higher-level stuff.
+
+It's funny now, because I came to this company not to work on databases, but I eventually ended up working on databases... And I was talking about the Spanner team a couple of months ago, maybe around February, and I was wondering "Hey, what's going on?" I used to think that Spanner has a very different approach to some of the problems we're discussing right here, and I was wondering "What's going on with their --" they have a cloud product also, "What is the usage?" and so on. I thought that it's a huge missed opportunity; nobody's actually talking about Spanner, nobody's actually filling the gaps for Spanner, because its approach - which I can explain briefly what it does - is very different.
+
+\[20:20\] It's a really interesting mental exercise for me to go back to databases, because it's on the other side of the spectrum, and from that perspective I learn more about the traditional databases. I constantly am comparing the design decisions and different fail modes and so on.
+
+**Mat Ryer:** Yeah... When you say "table-flipping", that's not a database term, is it? You mean you were getting frustrated and turning tables over, right?
+
+**Jaana Dogan:** \[laughs\] Yeah. Sorry, yes...
+
+**Mat Ryer:** Table-flipping sounds like that could be a thing in databases, doesn't it?
+
+**Jaana Dogan:** \[laughs\] It is going to be my new tool, if I ever build a tool, like some table, whatever...
+
+**Mat Ryer:** Yeah, table-flipping. You can do it, it's cool.
+
+**Jon Calhoun:** Just waiting for data to get thrown all over the place, just from random spots...
+
+**Jaana Dogan:** Yeah.
+
+**Break:** \[21:11\]
+
+**Mat Ryer:** I remember when Spanner was first announced, and it did get quite a lot of press. It seemed like it was doing the impossible. What does it do differently, and what problems does it address?
+
+**Jaana Dogan:** We've been talking about a lot of trade-offs, but we've never mentioned the cap theorem. Eric Brewer a while ago came up with this idea that there are three things you can have... In distributed systems you can't have three of these things; you have to pick two. And those three things are represented by CAP. C means consistency, A means Availability, and P means Partition tolerance, like network partitioning tolerance.
+
+What he says is if you want 100% consistency and 100% network failure partitioning tolerance, you can't have 100% availability. You always have to make that sacrifice. We're talking about relational databases, as well as like schemaless NoSQL type of databases... Actually, relational databases are more like CP systems. They have higher consistency, they are more tolerant to partitioning... On the other hand, NoSQL databases are compromising from consistency; they're eventually consistent... But they provide higher availability. So they're AP systems.
+
+So if you have this mental model, I think databases are becoming easier to understand, because you know that there's a limit, there's physical limits to the world, and you can't have it all.
+
+**Mat Ryer:** I've worked for some project managers that really just don't agree with this. They want all three.
+
+**Jaana Dogan:** Yes. Because it's just hard to explain. It's almost like -- if you don't know about the fundamentals of the limitations about distributed systems, it's hard to convince people, even if you are bringing all these papers and concepts and whatever.
+
+**Mat Ryer:** \[23:59\] Yeah, it is counter-intuitive. You've mentioned eventual consistency. Developers should be familiar with this concept, because it is quite important, especially when you're working with data at scale. And essentially, it's the idea that if you're gonna put some data into the database, if you've got that data spread across physical locations, which you might do for some sensible reasons, it's gonna take some time for those changes to propagate. And it does seem counterintuitive to people, because you think "Well, I've put this in the database, and then I did a query, and that result didn't show up." And if you think about the user experience of that, it is quite bad. The users just created something, then they go back to look for it and it's not there... And that's just because maybe they hit a different node when they did the query, or you're waiting for the indexes to update, or whatever it is. How do we solve that? ...let me ask that question. How do we solve problems like that?
+
+**Jon Calhoun:** I think there are some approaches... It kind of depends on what you're doing, but there are approaches depending on the problem. Let's say you've got this really popular website with comments, for some reason, and you need to go to a NoSQL database. Well, when a user posts a comment, you could basically say "Okay, this user needs to read from the spot where it was written to." We need to make sure that syncs up, so the user sees their own comment... To have that sort of consistency in their head.
+
+But for other users who are browsing that website, if it takes a couple of seconds or a minute for that comment to show up, it really doesn't matter to them, because they don't really know any better. So it's one of those things where you can take trade-offs like that, where as long as the user who posted it gets that real-time feeling, the rest of the users can kind of have the "Don't see it for a minute" type approach. But it really just depends on what you're doing.
+
+**Mat Ryer:** Yeah, it's funny you mentioned the feeling, because I've got around this problem in the past by in the browser just basically caching the comment. So it's gone to the API, it immediately puts it into the DOM... So it feels very snappy; it's done it immediately. But in the background, it's still waiting for that request to finish. And it is kind of smoke and mirrors sometimes, isn't it?
+
+**Jaana Dogan:** It is. What you said was so true; it really depends on the use case. Users have to go through this gymnastic themselves to pick the right thing... And it has a lot to do, again, with the CAP; you can't have three of them, you have to pick two. What Spanner does, interestingly, is they claim that they actually are beating the CAP theorem... Which was something controversial, because what Eric Brewer says in theory - it's a model to think about very extremes. So Spanner says that we have them all. You don't have to make any compromises.
+
+But in Eric Brewer's mental model about his theorem - if you think about the very extreme cases, like 100% availability, 100% consistency, 100% partitioning - that type of extremes can't exist because of the physical limitations of the world... You will have some sort of network partitioning of some sort... And Spanner is actually a typical CP system; it has 100% consistency, and it's very tolerant to partitioning, but its availability is significantly higher than any other relational database. It provides five nines of availability, which means like 5 minutes downtime a year. That's amazing. Most other relational databases require 10 minutes or whatever a month for maintenance and so on... Or if you wanna upgrade the schema, it requires downtime. Or the failover requires downtime. So how did this happen...?
+
+The Spanner team says they're beating the CAP theorem, because they provide this high availability... And it has a lot to do with the way the internals of this distributed system are working, plus our good networking infrastructure. We're just kind of like improving the availability -- not to 100%; we're still talking about five nines, but five nines is actually a lot in practice.
+
+\[28:09\] So our goal is maybe you shouldn't make as many compromises. We will try to provide you a higher availability, but you will still have the transactional relational database. But at the same time we have a lot of limitations around the type of the schema limitations, for example, some SQL limitations... Because it's hard to deliver really complicated queries on a very highly distributed system.
+
+Latency-wise, for example, the way we handle writes are completely different from traditional databases... But we are trying to pick the best. For example, unlike other traditional databases, when a write comes to Spanner, we go and write it to multiple replicas. It arrives at the leader, but we synchronously sync it with other replicas.
+
+But we use Paxos \[unintelligible 00:29:19.24\] so if a replica goes down, it doesn't really stop the write. And in traditional databases they don't have this concept, so it's just kind of like if something goes down - that write fails, or there's going to be huge latency until something comes back up again, and so on... So they are trying to pick up those different flavors of things... Because you know, the world has changed a lot; we have better networking now, we have better computers, we have specialized hardware, and so on... Everything is going distributed; we need larger-scale, more resilience... Why not think about a completely new database in this new world, with the new rules...?
+
+That's why I like the project, because it looks things from a different perspective, and then you are internalizing all the other hard problems in other databases by looking from that perspective.
+
+**Mat Ryer:** Yeah, that is so cool... It's funny, because that's kind of how I see Go as a language as well. Someone commented that Go is kind of just like C. But C was designed back when computers had a single core, and not much memory... You know, things were different, and Go was designed for more modern architectures, so it definitely does make sense that you would do that for databases as well.
+
+**Jaana Dogan:** And Spanner has the same level of opinionation. It has an opinionated sharding scheme, for example... But it works for the people, and you can tweak it. You have to learn how to tweak it. But it just works for a lot of workloads. We use Spanner at Google for lots of things. There are databases actually built on top of Spanner; they are trying to migrate everyone to Spanner databases. It just works for a lot of cases. It gives you some sort of opinionation, but as a result of that, you don't have to think about these hard problems of sharding, all the failovers... You have this global consistency which is great, because you can start a transaction in one microservice, and then you can update the same table on another microservice without making them collaborate. You don't have to acquire an exclusive lock, or whatever. The database is just going to handle it. It will see if they are serializing, if there's any conflict or whatever, and it's going to abort.
+
+There are different trade-offs, and this database is really designed for distributed systems... And with these new rules, in the new world, this is what I think a database should look like.
+
+**Mat Ryer:** That's amazing. I love that you're building databases on top of Spanner. That is quite a testament to how good Spanner is, isn't it?
+
+**Jaana Dogan:** \[32:07\] Yeah, I mean storage systems are always bootstrapping from each other, because it's just hard to come up with something from scratch, for example. Spanner had metadata in Bigtable when it was first bootstrapping, and it's using our earlier storage, blob storage engine and file system Colossus. So they're all sort of dependent on each other... And it's funny, because I think Colossus nowadays is using Spanner for metadata, or something like that. So there's always these bootstrapping and interdependencies, but that's how we evolve. You can't build everything from scratch. You have to pick the right things that are available to you. It's almost like they wrote the compiler in C, and then migrated everything to Go at a later time, when everything is just more -- you know, Go is more sufficient enough to support the compiler itself. So you see all this bootstrapping and common usages and so on in storage.
+
+**Mat Ryer:** Yeah, it's so meta. I love it.
+
+**Jaana Dogan:** Yeah. \[laughs\]
+
+**Mat Ryer:** Another thing that the Cloudfire store thing does that's quite interesting is with IDs. IDs is another thing that you sort of -- in the early days I took for granted that the database could just create the IDs, and often they would just be an incrementing number. So the first entry that goes into the database is 1, and then the next one is 2, and so on. I don't have to teach people how to count... Are there any problems with that when you're talking about data at scale, or in a distributed--
+
+**Jaana Dogan:** So in the storage layer, if you are -- there are some databases that takes that ID and then stores things in that order. For example, if you have ID 1, 2 and 3 will be in the same partition, and so on.
+
+**Mat Ryer:** Next to it, right?
+
+**Jaana Dogan:** Yeah. And that sometimes creates these hot spots. For example, if your ID is -- for example in Twitter's case, if you have your ID which is something gradually incrementing, you will always have the same shard, that all the recent tweets are coming to -- you know, you have to query from. So in order to deal with this, databases use hash functions; they take the key and then hash things, and so on... But some databases, rather than going that way, they are like "Okay, we're not going to. We will tell the user that we're going to use the ID when we're storing things, so they can tweak things maybe." They suggest you not to use order increment, because it's going to create all these hot spots, and so on. Again, it's a design choice.
+
+Spanner also doesn't support order incrementing. We're actually doing some work around to support it, but Spanner's sharding mechanism has a lot to do with the primary key. The primary key is very important in Spanner, and we're trying to actually now build things on top of that to make sure that we are not creating those hot spots.
+
+**Mat Ryer:** Yeah, that is really interesting. What we're doing is actually random IDs, and having an element of randomness to it... Which when you think about it, it does seem crazy, when you think "We're gonna just create a random ID, and then --" Of course, you have to think "Well, what happens if on that tiny change -- is it possible that you could have two of these random IDs created at the same time?" It's technically possible, but just within the laws of physics it's so unlikely. Is that it?
+
+**Jaana Dogan:** \[35:46\] But it's still possible to reject the insert, right? Let's say that you picked 8 bits of some random generator \[unintelligible 00:35:55.03\] and there's a lot of collisions. Then you can actually retry by generating another ID in the app level, and then insert with that number. So it's very important for the database to have some constraints. In Spanner we still have the constraints; the primary key cannot be duplicated. It has to be unique. So it provides to us the type of constraints which help at the application level a lot.
+
+**Mat Ryer:** Yeah, that's interesting, because I think understanding the database often does drive changes in the design of the application, like you talk about. But what happens in that case though if, say, two IDs just happen to create the same ID. If it's an eventually consistent thing, wouldn't they both feel like they've created the record? At what point do they find out that that's failed?
+
+**Jaana Dogan:** This is why you need transactions. In databases there's this concept called transactions. Transactions have some properties in relational databases. This is summarized by ACID. ACID means Atomicity, Consistency, Isolation and Durability. It means that whatever you commit needs to be consistent at the end. You shouldn't have two IDs at the same time, and whatever. So the database, if there's something conflicting going on, can reject the second insert, if you generated a duplicate key. And then you at the application level can write logic to retry with a newly-generated thing. So you will have consistency all across, without eventually waiting for things to happen... Because your transaction is going to be aborted by the time you are trying to commit, because you generated a duplicate key.
+
+**Mat Ryer:** Yeah.
+
+**Jon Calhoun:** Even if you didn't have the transactions, I imagine you could come up with some sort of strategy, but it just might be a little different... One example is if you have five servers, each one might have a unique ID, and they might take your unique identifier and at the very end throw their unique ID. So even if it does succeed, you know each one is gonna be separate. You just have to have whichever one you wrote to first tell you what the actual real ID is at this point.
+
+**Mat Ryer:** So it puts the Node ID on it, and mixes it together.
+
+**Jon Calhoun:** Yeah, something like that. You know it's unique, so even if both worked, they're both gonna be unique; they're just gonna both have a different suffix of some sort, or something like that.
+
+**Jaana Dogan:** The other strategy is to have a generator, so the generator generates and makes sure that it's actually unique. It kind of like stores the fact that it's actually generated. This is a complicated problem and I don't wanna talk about all the fail modes, but your IDs can come from a trusted source, let's say. You can provision maybe exclusive locks, and whatever... So you can make sure that you can trust that ID.
+
+**Mat Ryer:** Yeah. I've used a system before, I worked in a place and there was an ID service, and you could say "Give me ten IDs", and then it would give you the guarantee, I guess, that they were unique.
+
+**Jaana Dogan:** Yeah.
+
+**Mat Ryer:** It was weird, as someone that was only used to just the database, just incrementing the number itself, to then have to do different things and write different behaviors to solve that. It was really an interesting thing.
+
+**Jaana Dogan:** Yeah. And one of the other alternatives is to do optimistic locking. What you can do is add a version number or something to the record, and you can say "Only insert this if the version is 1." So you can make sure that no one has ever written to that record before you. That also gives you -- you can implement that constraint yourself at the application level.
+
+**Mat Ryer:** Yeah, so if somebody did get in before you, of course the version number would have changed, and then you know that; you notice that it's changed, and then you could either take those changes and reapply yours, and then apply them back. Yeah, that is interesting... It's nice when the database does that for you.
+
+**Jaana Dogan:** \[40:08\] Exactly. That's why I work for Spanner, seriously.
+
+**Mat Ryer:** \[laughs\] Yeah.
+
+**Jaana Dogan:** One of the things that I wanted to discuss is I'm so tired and frustrated about the fact that I need to do what my database is supposed to do. All that logic shouldn't be leaking to the application level... But that's sort of what the reality is right now. So I'm just trying to understand "Hey, what is next in databases? How can we make people more productive?" If it's just giving them a slightly more opinionated thing, that does some of these things, maybe more like consistencying better, then why not?
+
+**Jon Calhoun:** I think that's probably why Rails was so popular. It took that even a step farther, and was like "You don't even need to know SQL. You just have to have an idea of how it works, and you can be productive." I think a lot of people really liked that. But there is the downside of, like you said, some things still do leak, and you can have really slow queries and things like that as a result... But it did try to make that leap, which I think is why so many people just felt productive in Rails.
+
+**Break:** \[41:18\]
+
+**Jaana Dogan:** I realized that there was no way for me to explain what Spanner does, so I wrote that article about things I wish more developers knew about databases, to sort of like give people a catalog of different things that you have to worry about. It's definitely not a full, comprehensive list, but it kind of gives you -- there's a lot to worry about... And you as a random person kind of like discovering those problems along the way as you experience problems, experience data loss, and so on... And it's such a challenging thing for people who are working in databases, because they can't explain what the database is good for, because nobody understands what is the large list of the problems around.
+
+So I find it equally challenging to be on the other side, as a person who is working on databases, because it's so hard to explain the whole spectrum. Why would anyone care? That's the database's one and only job at the end of the day... But if you give them an opinionated set of things that will make them productive enough, they wanna give it a try and they like it.
+
+**Mat Ryer:** Yeah... You know, there's some other design principles and properties of a system that are important, especially when we're working, like you say, in modern architectures with message queues and databases that have eventual consistency and these kinds of things...
+
+One of them that I think is very important, that I don't know that loads of developers are familiar with it, but it's idempotency. Operations that are idempotent essentially no matter how many times you apply the operation, the end result is the same as just applying it once. For example, if I were to have a counter and I apply the operation +1, if I ran that three times, then the number is gonna go up three times. That's not idempotent.
+
+\[44:13\] An idempotent version might be for me to set the number to three, and then if I set to three three times, the end result is the same; it's set to three. Little things like that, and designing systems with that sort of knowledge I think really helps... And you can kind of design for failure. You know that "Yeah, this message system has this at-once guarantee", but that means it's possible that it may deliver the same message multiple times. Well, if that operation is idempotent, for example, that's okay that that happens. But it doesn't mean that you can't be useful without knowing that, but of course, it does help. Are there any other things that you wish developers knew about when it came to databases?
+
+**Jaana Dogan:** I have a really good book to suggest. Martin Kleppmann's book on Designing Data-intensive Applications, I think that's the name. It is a really huge, comprehensive list of all the things you have to know. If you need a catalogue or a reference, what are some of the fail modes, or what is the spectrum of the problems in this particular area? You can take a look at that book, you don't have to read it page by page... But if you need to understand a certain topic, you can just go and -- he does a really good categorization plus explaining what is out there, giving really good references. I really suggest that book to everyone who wants to learn more.
+
+**Mat Ryer:** And I will recommend Jaana's blog post, because it's phenomenal. We'll post it in the show notes.
+
+**Jaana Dogan:** Do you need a fun fact?
+
+**Mat Ryer:** Yes, please. I think we do. \[laughs\] This is an intense episode; I think a fun fact is just what we need right now.
+
+**Jaana Dogan:** \[laughs\] I actually was reading Martin Kleppmann's book, maybe two months ago... And then in my dream, I saw myself writing that blog post; as soon as I woke up, I took notes... I think I drafted ten items on the cover of that book. That blog post actually probably came from some of the ideas that I got from his book, and so on... So that's really funny -- I saw that the article was being very useful in my dream, and it turned out to be true. It's so funny.
+
+**Mat Ryer:** Well, if people didn't feel stupid before, the fact that Jaana is coming up with this stuff in her sleep is certainly gonna do that. That's how Paul McCartney wrote Yesterday, by the way. He just woke up and had the song.
+
+**Jon Calhoun:** My dreams are way less productive.
+
+**Mat Ryer:** Yeah. I just dreamed my legs were made of jelly. That's not helping anyone. \[laughter\] I can't turn that into a blog post... I probably could. I might.
+
+**Jaana Dogan:** I actually am an impostor. All of these good ideas are coming to me when I'm dreaming. I'm serious.
+
+**Mat Ryer:** Yeah, wow...
+
+**Jaana Dogan:** That's why I'm sleeping. I should just go back to sleep.
+
+**Jon Calhoun:** I don't think that makes you an impostor. I think that means you know this stuff so much that it can't slow down.
+
+**Mat Ryer:** That's amazing though... So in your dream, were you actually thinking through these -- do you remember the dream at all, or did you just wake up and had the point? Were you trying to solve this in your dream?
+
+**Jaana Dogan:** I remember the dream; I saw that I wrote the article. The article had ten items, and I remember that the first -- it was something about network partitioning tolerance which became the first item. There was something about ACID, there was something about consistency, there was something about optimistic locking... I think ten of the items or seven of the items in the actual article came from my dream. And I can send you maybe the screenshot of the frontpage; so I'm waking up from the dream, taking notes, and then going back to sleep. \[laughter\]
+
+**Mat Ryer:** \[48:02\] Great. I love that. Wow... If you're gonna do a talk and you don't quite have enough, can you just have a quick nap and top it up?
+
+**Jaana Dogan:** \[laughs\] I sleep when I need to debug. There is one funny thing - I used to see myself as a package in the systems that I designed before coming to Google. That was the way for me to debug the things that I was working on. Then I came to Google and we have distributed tracing, so it became slightly easier... But I remember myself seen as a message in the RabbitMQ on a topic, or something; that was a horror movie, it was a very overwhelming nightmare... But it really helps me to internalize some of the mechanics of the things and debug.
+
+This is a known phenomena, by the way. Most of the people see some of the problems they're dealing with when they're sleeping, and then they're like "Oh, gotcha. I got it!"
+
+**Mat Ryer:** Yeah, it has happened to me. It is very satisfying. It is whatever's going on in the sleep, that organizing process, putting the memories in and all that stuff is kind of awesome. That is awesome, I can't believe that.
+
+**Jon Calhoun:** I'm just gonna put this up to something your brains do differently than mine, because otherwise I'm gonna feel like mine's broken.
+
+**Mat Ryer:** What are you dreaming about? You've probably got some great dreams.
+
+**Jon Calhoun:** I've even seen things -- well, somebody talked about when people think. whether or not they think thoughts and words in their head, or if it's just something else... And if you're in one camp or the other, it's really hard to imagine the other, because I don't know how that would work.
+
+**Mat Ryer:** What do you mean? Like, people think in words...
+
+**Jon Calhoun:** In words.
+
+**Mat Ryer:** ...rather than abstract--
+
+**Jon Calhoun:** Rather than something abstract, or I don't know... I don't know, because I think I'm more of a words person. But even then, thinking about how you think is a hard thing to put down...
+
+**Mat Ryer:** Yeah, it's a bit too meta.
+
+**Jaana Dogan:** Yeah, I heard that people don't talk to themselves when they're thinking. There are people who only can use visuals or abstractions when they're thinking about anything, anything random. Their thinking process doesn't have that inner voice, which is very hard for me to relate to, because that's such a hard concept for me to have that inner voice with me all the time.
+
+**Jon Calhoun:** I imagine those people don't talk to themselves, because that's usually my inner voice just coming through my lips... And if somebody looks at me, I'm like "Oh, don't worry. I'm not crazy. I swear." \[laughter\]
+
+**Mat Ryer:** Oh, you're just saying it out loud.
+
+**Jon Calhoun:** Yeah, sometimes. If I get really caught up in what I'm thinking about, I have caught myself saying things out loud that I'm sort of thinking in my head... And it makes you look really weird at the time.
+
+**Mat Ryer:** I bet that has got you into some trouble.
+
+**Jon Calhoun:** Nah...
+
+**Mat Ryer:** \[laughs\] It depends what you're thinking.
+
+**Jon Calhoun:** It's always programming, it's never anything that would get me in trouble.
+
+**Mat Ryer:** Well, that's because you're a nice bloke.
+
+**Jon Calhoun:** Yeah, sure. \[laughs\] No, I don't know... I'm gonna say it's because those are the things I think hardest about. They require the most thought.
+
+**Mat Ryer:** Do you ever have that feeling where you know that an idea is about to happen? It hasn't happened yet, but you know that something's gonna happen somehow... And it can be a few seconds where you think "Oh, hang on... I'm about to have an idea", and then it does happen. But it's just a feeling before that, and I can't figure out how that could happen. Did you ever have that?
+
+**Jon Calhoun:** I think I've just had the opposite, where I'm like "I had a good idea, and now I don't know what happened to it." \[laughter\]
+
+**Mat Ryer:** Jon, I bet you just can't remember your dreams; you're probably solving all kinds of things in there... You just can't remember it.
+
+**Jon Calhoun:** I mean, that could be it. I'm told that if you write down and think about your dreams consistently, you'll start to remember more of them, but I've never done that.
+
+**Mat Ryer:** Well, it's amazing. Well, it's that time again... It's time for Unpopular Opinions!
+
+**Jingle:** \[52:00\] to \[52:16\]
+
+**Mat Ryer:** So... Has anybody got an unpopular opinion?
+
+**Jaana Dogan:** To be honest, I haven't practiced for this particular question... But I can think about one. I have too many unpopular opinions. The only criticism I get is most of my unpopular opinions are actually popular opinions, but I never feel like I can verbalize them.
+
+**Jon Calhoun:** You haven't found the crowd that agrees with you to hang out with...
+
+**Jaana Dogan:** \[laughs\]
+
+**Jon Calhoun:** I mean, I can tell you the same thing that I said on Twitter earlier, Mat - I definitely think that at some point in most developers' careers they should do debugging through printing out code, and walking through it step by step. Now, it definitely doesn't work for very large applications or anything crazy, but we had to do this in school for this one programming team I was on... And I've found when you're thinking about something complex, like an algorithm or something that's a little bit more substance than just print out 'foobar', or whatever... That by printing it out it forces you to take a step back, and instead of being like "Why is this one test case--" or in your mind you're thinking it's an edge case; "the whole thing's right, I've just gotta figure out why this one edge case is wrong." But sometimes you need to take a step back and rethink the entire algorithm, or everything that you're basing your assumptions on... And when you print it out, you don't have the ability to just change one thing and rerun it, because you'll catch yourself doing that with a debugger, or something like that. You'll just change one line, rerun it real quick; change another line, rerun it, and be like "Alright, why isn't this working?" But if you actually have to sit there and trace it with your head, with paper, you don't have the opportunity to do that, so you're forced to be like "Could this be wrong? What fundamentally might be wrong with this?"
+
+**Mat Ryer:** Yeah, that really resonates. I find just thinking is the best way to debug something. Obviously, that requires you to really kind of hold a good chunk of the system in your brain somehow... And obviously, the bigger and more complex the system, the more difficult that is. It's another argument for keeping things simple.
+
+But yeah, I think some people are more hands-on, and they want to go and use a debugger and look at the variables and just watch things changing... And I find that to be quite a difficult thing to do. So for me it's easier to just -- I mean, you probably just have a siesta Jaana?
+
+**Jaana Dogan:** My opinion on this is, you know, I start with `printf`debugging all the time, because it's much easier, convenient, whatever... But if I need to break certain conditions -- sometimes things are extraordinarily hard to debug, because they happen very rarely... And if you are really targeting that particular thing, maybe it makes sense to use a debugger. But it also depends on how good the overall workflow is. Some languages or some platforms have a really good debugging experience; that makes it so much easier. But in Go, for example, I always use `printf`. I am actually very sad that with the new module thing you need to explicitly overwrite the stuff that you are depending on, because it's sitting in a read-only type of -- there's the file modes for the modules are different now. It's not like you have the GOPATH and you can change every line whenever...
+
+I'm a little bit sad, because it's not a big friction for me, but it's still kind of like -- I need to tell my editor "Yes, you can overwrite this type of thing" and you can see that your text editor is asking whether you wanna overwrite, even though there's a \[unintelligible 00:55:46.28\]or whatever... I like to, in short -- to summarize, I really like `printf` debugging. That's the number one thing, and I think it helps you to internalize the source code much better, just because you're just switching and reading things more. In the debug mode you just only are caring about the execution path.
+
+**Jon Calhoun:** \[56:07\] I should say that I don't hate debuggers -- I think I'm similar to Jaana, where I don't hate debuggers or anything like that; I just think that learning to debug in these other ways helps you train your brain to think about the right things...
+
+Mat, when you said it's important to sit there and think about the problem, I think that's the reason -- in college especially, I think people should be printing out some of their code, because that's when you're learning to think and learning to actually be a programmer... And it's a useful tool for that. I haven't printed out code in the last ten years, so it's not like something I'm doing daily or something like that... It's just doing it a few times sort of forced me to actually avoid those bad habits that otherwise might just leave you floundering and tweaking one line at a time.
+
+**Mat Ryer:** Jon, you're talking about literally printing a hard copy of the code...
+
+**Jon Calhoun:** I've printed hard copies of code, yes. So what I mean by that is -- a lot of this stems from the fact that I did ACM ICPC is what it's called; Intercollegiate Programming Competition. And the way they do it... If you've ever done Google Code Jam, or Top Coder, or... There's different competitions like that. They're mostly algorithm-based, and they're like problem solving. So most solutions are less than 200 lines of code, generally speaking... But the ACM one specifically is for college students, and you're on teams of three, with one computer. So because of that limitation, anytime you have a bug, your first thought is supposed to be "Print out your code and let somebody else do something else", because you don't wanna hold somebody else up, sitting there, when you don't even have a clue what's wrong with your stuff.
+
+So it sort of forced you to sit down and be like "Alright, where in this code could it be wrong?" and you come up with a couple "This is what I think is wrong. Here's how I can test it" and then when you're finally ready, you say "Okay, I'd like to be back on the computer if somebody else is done with whatever they're doing."
+
+So that's where I first started doing it... But this was also working with much more complex algorithms. It would be like implementing a minimum-cost maximum-flow type things, that are a little bit more complex than a bubble sort, or something like that. And when you're doing that, it's easy to have some bugs in your code. So as a result, it's something where you just had to sit and think about it.
+
+**Jaana Dogan:** There was a really quote from I think Ken Thompson or Dennis Ritchie... Their approach to debugging is `printf` debugging. They were talking about how reading the code and looking things from a different perspective actually helps you to debug things, and `printf` debugging is sort of like helping that process. And I kind of am relating to that. I really like to go and dig in and just put some printlines here and there.
+
+I don't wanna necessarily focus only on the execution path, because you just wanna have a better sense of the codebase in general when you're debugging things, to see how things fit together, and so on. So I love `printf` debugging.
+
+At school, by the way, just because you asked, they ask us to design a debugger. And I created this trampoline thing, which basically takes every symbol, puts a `printf` before calling the function... It just basically rewrites the program to put a `printf`, and then calls into the actual function. So when you run it in the debug mode, it just prints out all the function calls. To me, that approach is really nice, and is a really good starting point. Everything that is interactive I think is secondary and is more for those advanced cases where it's hard to reproduce some of the cases and you need to evaluate and break the breakpoints with some conditions, and so on.
+
+**Mat Ryer:** Yeah, that's it. It's very specific when you're doing that. You're right; you're kind of watching the execution of things, and that might be very appropriate. One of the nice things about just thinking is you can go anywhere very quickly. It's a very fast way of navigating this virtual model of the system... And you can consider more things more quickly, it seems, by doing that.
+
+\[01:00:08.00\] I don't know if I've always done it like that, but... I mean, I did a tweet about this; the best tweet I've ever done, and it's also inspired a lot of trolls as well... For the first time I've had actual trolls, which is very exciting for me. It was like being a woman on Twitter.
+
+**Jaana Dogan:** \[laughs\] One more thing about this topic is I was kind of like a bigger fan of debuggers, and then I spent a couple of days like 10-15 years ago, because our debugger had some bugs with concurrency... So I was never able to reproduce some of the bugs, and that was the moment that I actually stopped trusting debuggers. I always do `printf`, because debuggers do all these crazy rewrites, and so on... I just want minimal impact on my code; I just wanna be able to see what's going on first, and then yes, if I couldn't really figure it out, I could always take another look with the debugger.
+
+**Mat Ryer:** Yeah, that's actually a great example, because when you're stepping through code, you're not running code concurrently like you would normally. So if you're just thinking, you can actually consider those kinds of more abstract things and more complicated things. It's quite a good example of a place where, you know - yes, it might be too narrow in some cases, but it's really whatever works for people. Don't troll me, please.
+
+**Jaana Dogan:** \[laughs\]
+
+**Mat Ryer:** Well, I think that's all the time we have today. Jaana, thank you so much. I really feel like I've learned so much, and I'm sure our listeners have also... So thank you very much. Jon, I didn't learn anything from you today, mate...
+
+**Jon Calhoun:** It's okay.
+
+**Mat Ryer:** \[laughs\] I'm just kidding.
+
+**Jon Calhoun:** I was supposed to tell Jaana today... Jerod, one of the producers behind Changelog, told me that I was supposed to ask "What's the best database?" and then I was supposed to ask "And why is Postgres the best database?"
+
+**Jaana Dogan:** I have some opinions even for Postgres... But you know, Postgres is one of these -- at least it's doing the basics right, and it's not super-crazily surprising... It's like actually doing the bare minimum. There are some gotchas that you have to learn about every database, and I think -- yes, I agree with the fact that Postgres is one of the best databases ever. And this is a strong opinion, and it's an outrageous to probably say out loud, but I am a big fan of Postgres.
+
+**Jon Calhoun:** I mean, I'll agree with it, because it's my go-to tool for most things, because it works.
+
+**Jaana Dogan:** It works.
+
+**Mat Ryer:** Yeah. Great. Well, thank you very much. Thanks for listening, everyone. We'll see you next time.
+
+**Outro:** \[01:02:59.27\]
+
+**Jaana Dogan:** This topic really needs like five hours. I could really talk about like five hours, and we would be able to still capture only 5%. And that's my daily challenge right now, like "How are we supposed to give people an accessible entry point to these problems, and what we are doing with these problems" and so on. So this was very useful for me, because I got a lot of opinions from the way that you see things, and that really helped me to see your mental model about these problems, and so on. Thank you.

There's a lot to learn about teaching Go_transcript.txt

@@ -0,0 +1,599 @@
+**Mat Ryer:** Hello, and welcome to GoTime. Today we're talking about teaching. We're gonna be exploring this subject with some great and experienced teachers, including Jon Calhoun. Hello, Jon.
+
+**Jon Calhoun:** Hey, Mat.
+
+**Mat Ryer:** How's it going?
+
+**Jon Calhoun:** Pretty good, how about you?
+
+**Mat Ryer:** Not bad, thanks. We've also got Johnny Boursiquot. Hello, Johnny.
+
+**Johnny Boursiquot:** Hello! It's good to be back.
+
+**Mat Ryer:** It's always good to have you here, Johnny.
+
+**Johnny Boursiquot:** Thank you.
+
+**Mat Ryer:** And also, you know, Mark Bates is here... Hello, Mark.
+
+**Mark Bates:** Hello, Matthew.
+
+**Mat Ryer:** How are you, mate?
+
+**Mark Bates:** Yeah, I'm doing alright. How are you doing?
+
+**Mat Ryer:** Yeah, not bad, not bad. I'm excited about this episode, because I feel like there's a lot to this subject.
+
+**Mark Bates:** I think what you're looking for is there's a lot to learn about teaching.
+
+**Mat Ryer:** There we go... Beautifully put. Thank you very much.
+
+**Mark Bates:** Thank you. I think I stole if off a Bazooka Joe gum wrapper... \[laughter\]
+
+**Mat Ryer:** That is a reference I get.
+
+**Jon Calhoun:** I haven't had one of those in forever.
+
+**Johnny Boursiquot:** I know. Are they still around?
+
+**Jon Calhoun:** Halloween coming around, I'm like "I need to go somewhere and find some Bazooka Joe."
+
+**Mark Bates:** They are still around. My kids had some recently, and they haven't gotten any softer. \[laughter\] I'll tell you that much...
+
+**Mat Ryer:** Okay, so maybe we could start off by a little bit of an intro, because I don't know that our listeners really know how much experience you all have in teaching... So how did you get into teaching and what sort of teaching do you each do? Bates, we'll start with you, mate.
+
+**Mark Bates:** Yeah, I have a feeling that the three of us do very different styles of teaching, which is gonna be fun obviously to talk about today. For me, I first started getting into it with writing books, about ten years or so ago... And then about three or four years ago I met Cory, and Brian Ketelsen... And we were talking about training and thinking about doing something together, and we started Gopher Guides.
+
+\[04:22\] For me, my experience with training has been a lot of in-person, in a classroom, three days, four days, teaching. Not a traditional teaching style, but the in-classroom teaching. And then even last year we started to do a lot of virtual training, so of course, this year now we're doing nothing but virtual trainings...
+
+So that's my experience - it comes from doing workshops and in-person teaching, virtual teaching, with the big groups and dynamics, and somewhat to the book stuff early on that led me into it.
+
+**Mat Ryer:** Yeah. And Johnny, you do classroom-based teaching sometimes too, don't you?
+
+**Johnny Boursiquot:** Yeah, my journey is an interesting one. A lot of the teaching I was exposed to came from being part of workshops that were aimed at teaching programming to beginners, right? So think RailsBridge, and GoBridge, and these kind of things... So basically at first I'd go and I'd participate as a TA, and help students and whatnot. Then eventually you start picking up enough of how people learn, the different styles of learning... There are folks who can just hear something, and internalize it, and know what to do. There are folks who hear it, but it doesn't really stick until they actually try something and do it... You see all these kinds of very different learning styles: auditory learners, visual learners, all these kinds of things... And that's when it dawned on me that "Hey, I seem to have the ability to tailor things for the different learners."
+
+So I wasn't trying to learn how to teach, I just wanted to help... But okay, you're able to convey certain information differently to different folks in the audience, so that everybody walks out satisfied with their journey, with their learning process. And the biggest thing is knowing how to leave enough of a spark there, so that when they leave the workshop, they can continue the learning on their own.
+
+So it became this thing that I eventually also started doing professionally, not just as a hobby or a way of giving back. It became a professional kind of thing. I do other things now in that realm, but it just so happens that it wasn't something I really wanted to get into.
+
+**Mat Ryer:** That's interesting you talk about the different people learning in different ways, because -- I mean, how do you do that? How can you create a course or something that is tailored to these different--
+
+**Mark Bates:** Could I just say, before we get into the one-size-fits-all metaphor and the deep topics - can we hear from Jon, see how Jon got into learning?
+
+**Johnny Boursiquot:** I know, right? Jon.
+
+**Jon Calhoun:** Mat doesn't wanna hear from me. It's okay.
+
+**Johnny Boursiquot:** Mat's like "Forget Jon." \[laughter\]
+
+**Mat Ryer:** I was gonna get to that in a minute.
+
+**Jon Calhoun:** I have faith in Mat.
+
+**Mat Ryer:** Well, Jon, you don't do classroom-based teaching as much, do you? You have video courses. So that is quite different.
+
+**Jon Calhoun:** Yeah, mine was -- when I started learning Go, I realized that there weren't any resources like... Like, Rails tutorials is how I learned Ruby on Rails, and I really liked that project-based book, or video course, or whatever; it ends up being both if you want, but... I really liked that approach, and when I was learning Go, they didn't have it, so I basically just started taking notes with my journey, and I got to a point where I was like "I should probably turn this into something and share it with people and try to create something similar."
+
+And then it just happened to be that the startup I was working at was in some rough times, and it made more sense for me to just branch out and see how that went. So at first, I started doing consulting and teaching, with the two sort of augmenting each other, but it didn't take very long until I've actually found a place where I was living off of just the teaching income, so that's what I kind of focused on.
+
+\[08:14\] It's been a lot more fun; it's really rewarding to see people get stuff and to take a different approach. It's kind of interesting to me, because I see the benefit in having all these different styles put together, and I really feel like - and I'm hoping as a result of Covid we start to see all of these sort of get merged together, and find ways to find the best of both worlds... Because up until now, I just don't feel like that's happened. But we can go into the one-size-fits-all discussion first, that's completely fine. Or, Mat, do you wanna talk about your teaching experience?
+
+**Johnny Boursiquot:** See, Mat's making that face like he doesn't think he's a teacher, but we all know he is.
+
+**Mark Bates:** Even better, he just farted. Go on. \[laughter\] It's a very similar face with that.
+
+**Mat Ryer:** I'm gonna have to ban you for that, I'm afraid, mate... Well, yeah, I've done it at conferences and stuff. Sometimes at conferences you get asked to do a workshop or something, but I haven't done that very often. One thing I'll say is it's absolutely exhausting... And that shocked me, of how draining it actually is to do teaching. Did you find the same thing?
+
+**Mark Bates:** It's exhausting.
+
+**Johnny Boursiquot:** Yeah, absolutely.
+
+**Mark Bates:** It's funny, because like both Jon and Johnny, I used to do the TA, and then the RailsBridge... In fact, Johnny, I think you and I might have even taught a couple together over the years \[unintelligible 00:09:36.19\]
+
+**Johnny Boursiquot:** Yeah, I'm sure...
+
+**Mark Bates:** And I did obviously the video stuff, which I still do, and I had Metacast, which was those kind of bite-size snippets for a long time... So there's a lot of different ways you can learn, but overall, regardless of how you're doing it, it's still very daunting and still challenging. And when you've got a classroom full of people... I know for me if I'm at a corporate client for 3-4 days and I'm standing there for eight hours a day, training and being energetic and motivating and accessible, and all those things you need to be to be a good teacher, it's exhausting. I go back to the hotel room and I lay on the bed and I pass out. That's it, that's my night. My night is I go back to the hotel room, order room service and I just watch TV until I fall asleep, because I'm so drained from that day.
+
+**Mat Ryer:** Yeah. Why does this sound like an alibi? \[laughter\]
+
+**Johnny Boursiquot:** It very well could be. No, here's one thing I'll add on top of that - the funny thing is most folks think that to be a teacher you kind of have to be naturally outgoing, have the personality that brings out people from their self-contained personality types, and that kind of thing... But it's really not. I know for me - I wouldn't say I'm an extrovert... I have little spikes; so when I'm teaching, I have to literally -- almost like flipping a switch in my mind, that says "Okay, now you need to be someone else."
+
+**Mark Bates:** You need to be energetic, and motivating, and exciting.
+
+**Johnny Boursiquot:** Exactly. Literally, you have to flip that switch that says "Okay, now it's no longer about you. You'd normally like to be introverted, and have your own space, and have quiet, and all these things... But this is not what you're about to do now." So you literally have to change and become someone else in order to do that.
+
+Obviously, the more time you spend doing it, the more experienced you are, the more naturally that comes... But it is not an easy thing at all to put yourself in the right mindset to do it.
+
+**Mark Bates:** It's not, and I have the exact same problem. I'm famously what I call an extroverted introvert, where I'm almost always introverted, but I have spikes, like Johnny said, when I'm doing something like this, or I'm teaching, or I'm at a conference. You see me at a much higher level of activity; I'm much more outgoing and energetic than when you see me offline, when I'm just kind of quiet, sitting in the room and just relaxing. But you do have to put yourself in that mindset, and that's the daunting and exhausting part, at least for me anyway.
+
+**Mat Ryer:** Is it like a persona then? Do you feel like you have a different persona when you -- do you feel like you become someone else to do it?
+
+**Mark Bates:** \[12:18\] A little bit... I don't know, I think we all have personas when we hit the stage... And whether that stage is a classroom, or in front of a thousand people, or at GopherCon, or something. When we're in front of an audience, we all have a persona, whether you admit to having a persona or not. We're all just a little bit different, and we all tweak ourselves slightly different just to be up on that stage... And that goes for just speakers, too; we're also educators. We shouldn't discount those who give 30-minute talks, 45-minute talks at conferences. They're also educators, and we could talk about what it means to be that sort of an educator at some point too, which would be kind of fun.
+
+**Johnny Boursiquot:** In some regards I'm also curious to hear from Jon about this, because I've been fortunate enough to have done the face-to-face, I have done the online, and I think the worst possible combination of all these things is when you have to do virtual training, where there are no faces, there's no cameras. You can't see, you can't tell that somebody's getting something or not the way you can in a live classroom. These are the absolute worst for me. And I power through them, but they are my least favorite, when there's no cameras and you can't see anybody. You're just a voice on the other end.
+
+But it's a slightly different thing perhaps with Jon. Maybe you're doing a screencast, and you're putting together a video... I've done those; you can give yourself a bit more -- you can forgive yourself a little more, because you know that if you have a bad take, you can record it again... And I know you've done this with Metacasts as well.
+
+So I'm curious, Jon, do you feel like you have to become someone else, a different personality when you're doing the screencasts and the recordings?
+
+**Jon Calhoun:** It's definitely there some. I remember one piece of feedback... I don't even think it was that long ago, but somebody basically made the comment that my writing felt way more energetic than this one video that they watched... And it was like the very first video I'd released for one of the older courses. And those older ones - I really did not quite get that transition; I didn't know how to do it. Even now it's something I struggle with, because I'm not interacting with people as much, so a lot of it feels -- I feel like I'm a random person on YouTube that's crazy high-energy. And I know I'm nowhere near that level, but it just feels so weird to me, because I'm like "That's not me at all." So it's kind of challenging.
+
+And you're definitely right that it is hard, because you don't get that immediate feedback... So I can't go do a workshop and then come back and be like "Okay, that worked. That was the right energy level." It's literally just release some videos and hope that it's right. And I do occasionally get that feedback, but it's probably not as immediate. Maybe people are more forgiving, because it's a recorded thing and they just kind of realize that's the way it works, I don't really know. Maybe that's why I've got the really bright colors in the background, that'll brighten me up and make me seem energetic...
+
+**Mark Bates:** Yeah, I think a lot of people just don't know how hard it is to do video, like Jon does... I used to do it with Metacast. There's a lot that goes into that... You know, that idea of a) just having to come up with fresh content constantly is very difficult. But then that energy level, and that "How do I get this concept across to nobody?" It's not even like a virtual training, where at least you know there are people on the other end to hear your words. You're recording and you don't have that, and you really have to pretend, like "What is this audience that I'm talking to? Who are these people? What do they wanna hear? What are they reacting to in this video?" It's really hard. That's why I stopped doing it... \[laughter\]
+
+**Jon Calhoun:** \[15:52\] I think people also forget for every ten minutes you see, there was probably like three hours of all sorts of other stuff, including bad takes... I've literally sat there one morning and spent three hours trying to record the same 15-minute video... And I'm just like "This isn't happening. I don't know why, it's just not happening today." And it's so frustrating.
+
+Meanwhile, you've got people like "When is that gonna be out?" and you're like "I'd love to give it to you, but it's not there."
+
+**Mark Bates:** It's frustrating when you're doing a really, really good take too, and something doesn't work. Like there's a bug... And if I refresh, nothing happens; everything's broken. Okay...! Hit stop, let me back up... What the hell just happened...? And then you've gotta get everything back in the same place in terms of your screen... Because you know, if I move to the browser, I have to get back to my editor, make sure the cursors are in the right place and everything lines up again... It's a real struggle when you hit those things. At least live you can make a joke about it and move on...
+
+**Johnny Boursiquot:** Keep it moving, yeah.
+
+**Mark Bates:** Yeah. There's that, too. I'm curious, has everybody else done virtual conference talks this year?
+
+**Johnny Boursiquot:** Yeah.
+
+**Jon Calhoun:** Yeah, I haven't given one, but...
+
+**Mark Bates:** Jon, you haven't. But Johnny, you have, right?
+
+**Johnny Boursiquot:** Yup.
+
+**Mark Bates:** It's been interesting... You know, we're all having to do these virtual talks... And Johnny, you were talking about the energy. I've had a couple conferences this year request that we pre-record our talks, as opposed to giving them live... Because you know, it's a virtual world, they can do whatever the hell they want. And I'll tell you, I hated it. I did it once, and I really, really hated prerecording my talk.
+
+**Mat Ryer:** Why?
+
+**Mark Bates:** The energy. I'm standing there in an empty room, talking to nobody, and I can hit record as many times as I need to. There's no spontaneity there, there's no energy, there's no nervous energy that kicks you up a bit... So even knowing that people are virtually listening on the other side, even if I can't see their faces, that nervous energy of just knowing that my words are being heard, and my face is being seen, I have to kind of be on my game. Yeah, that's one of the biggest problems I've been finding with the virtual, is that energy.
+
+And then what Johnny said, the feeding back. I scan the crowd; you've gotta scan when you're teaching. If you're not constantly scanning your audience and looking for people in trouble, then that's -- you're not doing it right. You're not doing your job right.
+
+**Johnny Boursiquot:** Right. One of the worst things is -- normally, in regular conversations, there's opportunities for impromptu jokes, and a little banter, that kind of thing... And I find that these things work their way into my teaching as well. Something will happen, it's unscripted and it's natural, it feels authentic, there'll be opportunities for a joke, or something like that... You can kind of play with these things.
+
+If you're inexperienced, these things feel like "Oh my God, something bad just happened." But the experienced speakers, they actually turn these around and use them to their advantage. That sort of skill -- you can't use any of that skill or any of these moments when you're doing a prerecorded thing. Imagine having a joke on a prerecorded talk - how weird is that? So you have to imagine in your head--
+
+**Mark Bates:** I've done it. \[laughter\]
+
+**Jon Calhoun:** Did you add a laugh track as well?
+
+**Mark Bates:** No, but here's how sad I am - I paused for the laugh.
+
+**Mat Ryer:** Mm-hm... Yup.
+
+**Mark Bates:** I just took a breath, let the laugh happen, and then you continue.
+
+**Mat Ryer:** Yeah, you can pause for laughter...
+
+**Johnny Boursiquot:** Amazing.
+
+**Mat Ryer:** Yeah, but it shouldn't be like two minutes, Mark. You're not getting a standing ovation after every joke.
+
+**Mark Bates:** That's true. Admittedly, the last conference talk, I did. I did wait for the standing ovation, and I did keep telling people to calm down, and just relax... It was mayhem.
+
+**Mat Ryer:** \[19:57\] Oh, that's great.
+
+**Mark Bates:** But like Johnny said, those jokes -- I'll tell you, one of the biggest things for us at Gopher Guides... Like, when we're doing the in-person stuff we don't have this opportunity with the virtual nearly as much, because of what Jon said, and Johnny (and everybody's kind of saying that), that lack of interaction. When we're in classrooms, or we're at companies and we've got 20-30 people in a room or whatever it is, you're talking, you're having lunch, you're learning about them as people, as individuals, as a company... And there's a lot of stuff you -- I like to take some of those in-jokes from them, and incorporate it back in; like some of what they're working on.
+
+If I'm trying to come up with an off-the-cuff kind of an example to describe something, if somebody had a question or whatever, I try to use their experiences. If it's Uber, which we've done a ton of training for...
+
+**Mat Ryer:** Other car shares are available... Sorry, carry on. \[laughter\]
+
+**Mark Bates:** Yes... But you know, I'll use self-driving cars a lot as examples, because that's the group I was doing a lot of training for... And using those examples -- because they understand them, and they can relate to those things.
+
+But the other big thing that you don't get virtually for us was the in-class demos. I love doing live demos in class, and Johnny, I'm sure you have the exact same thing, right?
+
+**Johnny Boursiquot:** Yup.
+
+**Mark Bates:** Someone asks that question, and you're like "Oh, let me just open up Vim. Let's code it, let's see what happens." And what's interesting that we've found is that every single class we've done over the last 3+ years, we've taken some of what we've done in class as demos or discussions, and it's been incorporated back into the training material, usually that night or the following week.
+
+But you need that face-to-face, you need those questions, that spontaneity of someone just raising their hand - not even raising their hand, just shouting it out - that you don't get over virtual. It's just been tough.
+
+**Break:** \[22:03\]
+
+**Jon Calhoun:** So we've talked a little bit about what's bad about the virtual... Have there been any positives to it?
+
+**Johnny Boursiquot:** No travel...
+
+**Mark Bates:** No travel is huge.
+
+**Jon Calhoun:** And maybe it's a little bit different, depending on what you're doing... But I know from my end, one of the things I really like about doing videos, despite that there are definitely some drawbacks - it's the fact that you have this massive reach, in comparison...
+
+\[23:52\] So rather than spending six months traveling, I can spend six months like "How do I really refine this challenging topic and make a much more involved project?" And then the second part for me has always been that the amount -- like you said, it's exhausting to teach in-person workshops, so realistically, you couldn't go someplace and teach for a month. That would just kill you.
+
+**Mark Bates:** It'd be too much, yeah. I mean, teachers do it every day. Let's not discount what -- all of our teachers - and God help them - they're all doing virtual right now. So kudos to every teacher in the world right now who's having to deal with all this... But yeah, go on. I couldn't do it for a month, let's put it that way. \[laughs\]
+
+**Jon Calhoun:** But basically, you can make bigger projects and teach them in the online venue, because you kind of know that people have that chance to take breaks themselves when they need them, you don't have to produce it all at once... Are there other advantages you've noticed to doing the virtual stuff?
+
+**Mark Bates:** I've found that instead of doing big 8-hour days when we're doing in-person, we've scaled it back so we do more days at less hours... Because as we've all come to learn this year, some more than others, sitting in front of Zoom for eight hours a day is a real drag, and it's really energy-sapping, and it could be so hard on you.
+
+So we've kind of moved into these four-hour blocks, with nice breaks... And people need that. They can't sit in front of a computer. But we've found that's working very well; people are engaged for that time, and they're active, and then they have the rest of their day to do their regular work, and stuff like that. So for us, the virtual has actually been quite nice. Our platform was kind of built for it anyway...
+
+But what I miss personally, like I said, is that interaction, that face-to-face, that one-on-one... Which is odd, because I hate people. \[laughter\]
+
+**Jon Calhoun:** I tried to simulate that with a Slack community, but it's still -- I know it's not the same, but it's one of those things where you do get the feedback from people, like "Oh, I'm on this video, and this is the challenging thing I've ran into", and you can incorporate that into future lessons, or you can give examples there... So there are ways.
+
+I imagine with corporate training that's trickier, in the sense that you can't bring up a Slack and be like "This is your corporate Slack for just this thing." Maybe you can make a channel, but even then that's kind of hard.
+
+**Mark Bates:** Yeah, you usually have to use it. Usually they have Slack, we've found, or stuff like that, as a back-channel...
+
+**Mat Ryer:** Sorry, I need to just say that other annoying apps are also available.
+
+**Mark Bates:** That's true.
+
+**Johnny Boursiquot:** Not just Slack.
+
+**Mat Ryer:** Not just Slack.
+
+**Mark Bates:** Not just Slack... You know, Zoom. We've also sometimes just used the chat in Zoom, or whatever platform we're using, too. When you're doing the live training, that kind of offline community isn't as important... Certainly when you're in the room. But virtually, you definitely need a way for people to funnel... And Zoom's got the webinar features, with the raising your hands, and asking questions, and polling... So there's some decent stuff built-in for reaching, and that can be quite helpful there. But like I said, having an interaction is really hard over Zoom.
+
+**Mat Ryer:** Is there anything specific about teaching Go? Because I know, Bates, you did Ruby training, and things. Does Go's minimalist design make teaching easier? Is it easier because Go has this smaller API, generally, in the language?
+
+**Mark Bates:** Go is simple, that's what's so nice about. I think that's why we all like Go. I often joke in class that Go is kind of a WYSIWYG language... What you see is what you get. You look at a Go file and there's no magic there; the imports are at the top, there's all your stuff... There's not a lot going on. There's not that crazy meta-programming that you get with dynamic languages.
+
+\[27:53\] When I was doing Ruby - and Johnny definitely can attest to this - you're trying to mentor junior developers at work, and stuff like that... And there's a million ways to do anything, and none of them are necessarily easy or make total sense. But a language like Go, where it's very clean, and easy to read, and it's optimized for reading - I think that makes it a lot easier to teach. I think the fundamentals are so much easier to teach in Go than a lot of other languages, to be perfectly honest. That's my opinion.
+
+**Jon Calhoun:** Johnny, I think you said you taught Ruby stuff, too. I could be wrong though...
+
+**Johnny Boursiquot:** I have.
+
+**Jon Calhoun:** Did either of you run into cases where you're teaching and somebody's like "That goes against our style guide, how we do these things"?
+
+**Mark Bates:** All the time.
+
+**Johnny Boursiquot:** Oh, boy, I've got stories for you.
+
+**Mark Bates:** All the time.
+
+**Johnny Boursiquot:** And here's the thing. Currently, I do trainings on platforms like \[unintelligible 00:28:47.28\] but also I do the private corporate stuff... And the private, corporate stuff, I tell you, it's the one that has folks coming to it with the strongest held views and beliefs about style guide, and approaches, and this and that... So I feel like I spend about a third of the class - if I look at the amount of time I spend sort of saying "Well, this is like C, but different in that way. This is like C++, but different in THAT way. This is like Java, but different in that way. Don't name your interfaces with an i in front, because we don't do that." All these kinds of things... I'm basically spending my time getting them to leave the baggage at the door, so that they can appreciate Go for what Go is, rather than saying "Well, I have this way of doing OO in my head. How do I do that in Go?" No. You don't just overlay that knowledge on top of Go... Which is why I think as a teacher you have to know how to relate things that folks coming from other languages and other technologies might already know... Because they're gonna come in the door with that knowledge. You have to somehow reframe it.
+
+At the same time, as a learner - and I can't say this often enough - you have to leave that baggage about all other languages that you know and love at the door when you come to learn Go. You have to really learn Go for what it is, and then have your preferences and your judgment, and "Oh, it's better at this/worst at that", or whatever it is. Leave that stuff at the door.
+
+**Mark Bates:** Yeah, that's a really tricky thing to balance... Because I know exactly what you're saying; I'll often say "This is kind of akin to this in Java", or Ruby, or whatever I can try to make some comparisons, hopefully to languages they know in the room... And that hopefully I also know, that I can use to relate... \[laughs\] But at the same point, a thing I always tell people is don't code Go like Ruby, don't code Go like Java... And also, don't code Ruby like Go, don't code Java like C. These things carry across all languages, but you do have to keep reiterating that in class.
+
+A lot of people come at you and say "I would do this in C\# like this", and it's like "Well, that's C\#. We don't do that in Go. That's not a thing in Go." "But how do I do it?" You don't. You don't need to do it in Go; it's not a concept in Go. "Yeah, but I need to do it in C." But you don't need to do it in Go! It's like, how do you get people sometimes past those things, like "I know how to do it in this language"? It's hard.
+
+**Johnny Boursiquot:** You're not swapping syntax. This is not just a syntax swap, right? It's a different way of thinking, it's not a syntax swap.
+
+**Mark Bates:** Yeah, exactly. I once saw somebody code -- and this still blows my mind a little bit, because I would never contemplate doing it... But I saw it and it just blew my mind that somebody did this. They wrote a try-catch mechanism in Go. \[laughter\]
+
+**Mat Ryer:** With panics.
+
+**Mark Bates:** Yeah, with panics, and...
+
+**Johnny Boursiquot:** Recover...
+
+**Mark Bates:** \[32:01\] ...recovers, and stuff like that... And it was both remarkably beautiful and terrible at the same time. \[laughter\] It was in a package called Structs, which was even better... \[laughter\]
+
+**Johnny Boursiquot:** Because that's where you put exception handling, clearly... \[laughter\]
+
+**Mark Bates:** That's where you put exception handling! And so it's a clear case of somebody saying "I need this. This is how I know how to code. I only know how to code with try-catch and finally", and that's what they built. They built a system that allowed them to do try-catch and finally, because that's how they know how to code... And the idea of error handling in Go just made no sense to them.
+
+**Johnny Boursiquot:** Great.
+
+**Mark Bates:** Yeah. Breaking people of previous habits is really hard...
+
+**Johnny Boursiquot:** It's hard, it's hard.
+
+**Jon Calhoun:** One of the things I've found there is that when people ask "This is how I do it in C", sometimes it's more about trying to figure out what's the underlying problem that you're trying to solve, so we can look at what is the correct approach to take in this other mindset... And that can be very hard, because sometimes people come at you with just "No, I just wanna do the same thing." And other times they're like "No, I'm actually trying to solve a problem", but they don't ask that question. They ask "How do I do this other thing?" and you're like "Well, let's take a step back."
+
+I know that can be challenging, because we see this on the Gophers Slack all the time, where beginners ask that and you try to say "Well, what problem are you solving?" and they don't know why you're asking that. And then when they finally get it it's awesome, but up until then you almost sound like that jerk who's just like "No, I won't answer that question. You need to give me a different question."
+
+**Mark Bates:** You've gotta pull it out of them sometimes. You've gotta pull up that problem, definitely. And there's some people who just want that answer... And it's funny, there's always those people - and they're the people hung up on that syntax; they're the people hung up on their way of knowing... But when you keep telling them that's not a thing you do in Go, that's not an issue in Go...
+
+I'll often say "I've been programming Go for seven years now, and I've never had to do that. That has never come up", whatever the thing is. And they're like "But how would you do it?" I'm like, "I don't know, it's never come up. This is truly a non-issue." \[laughter\] But you've gotta try to answer that as best you can, though it's not always easy.
+
+**Mat Ryer:** Yeah. So if we could turn that then into a piece of advice for someone that's gonna learn Go, and maybe they do come from a different language, I guess that advice is that you have to be open-minded... And you're never gonna do away with everything you've learned, especially because people invest years and get really good.
+
+I used to do C\#... I've built some amazing, abstract class hierarchies and all sorts of things with abstract classes, with generics... In C\# with the generics you can put constraints on the types that you're allowed... They can be interfaces, or they can be base classes, and things... And I felt like I'd become really good at that. So in a way, we're asking people to sort of -- you say "Leave the baggage", but that baggage is valuable to them. They've invested a lot in that, haven't they?
+
+**Johnny Boursiquot:** Context though... I mean, you have to have context. Certain skills are gonna be more valuable in certain contexts than in others... Again, going back to what kinds of problems are you solving, and what problem domain are you working in... Is this tool better suited for this kind of problem? All these virtues that we as software engineers aspire to - did I use the right tool for the job? Craftspersonship, and stuff like that. We aspire to all these things, except when it comes to our fair programming language. We feel the need to defend, "Well, Java does things the right way. I like the way it does things. Why can't I do the same as in Java?" Well, this is not Gava, this is not Guby, this is not Gython... This is Go. So learn that tool.
+
+\[35:58\] But I will admit, I'm gonna get off the soapbox for a little bit... I will admit that it's not unrealistic to say that the language you've been trained to use or learned to use for years and years and years has shaped the way to think about problems.
+
+**Mark Bates:** Absolutely.
+
+**Johnny Boursiquot:** It's not realistic to say. So naturally, you're presented with a problem. You want to solve the problem, but the tools you have, the facilities you have in your mind are still contextualized for the previous language you were using. So the people who jump from language to language to language - I'm sure we've all done this, because we all work in multiple languages these days... You'll swap your editor, and you'll start to type Go inside of your TypeScript, or your JavaScript, and then you're like "What am I doing?" Or you'll switch to VS Code where you have your Go code and start typing TypeScript. You're kind of jumping around.
+
+But I think those who have to work in these multiple environments, multiple languages, are in a much better position and have the ability to change their thinking at will... To say "Okay, for this particular problem, this language solves it this way." Whereby if the only thing you'll ever -- sort of the one hammer you have, it's Java, or it's C\#, or it's C++ or whatever it is, you're gonna try to hammer every nail with it, right? So again, one of those things where we basically tell ourselves in the software engineering community "Hey, use the right tool for the job." Well, if we're gonna say that, let's apply it across the board.
+
+**Mat Ryer:** It's funny on this, because I do write JavaScript and TypeScript as part of my day job at the moment, and I actually do feel like I've taken lessons from Go, and I do apply them actually... Because it's simpler. Because Go is simpler, I feel like you can almost do it that way. For example, there's loads of language features in JavaScript and TypeScript that I just don't use, and I'll just write very simple, little functions... It has actually changed the way I write JavaScript, too.
+
+**Johnny Boursiquot:** You could be seen as being not proficient enough though... Again, it's one of those things where somebody who's an expert JavaScript developer, or they really know TypeScript inside and out... And they see your code, they think "Oh, you're just a novice. You're not using these super-duper elegant mechanisms, this special, esoteric knowledge about the language that I happen to know because I have dozens of years of experience in that thing." They're gonna see you as a novice... Whereby you're trying to keep things simple. But they see it as a lack of knowledge.
+
+**Mat Ryer:** That's interesting, yeah... I even do it in Go, because I don't really use channels very often these days in Go, for example...
+
+**Mark Bates:** Neither do I. I very rarely have ever used channels, to be perfectly honest... As a matter of fact, one of the things when we teach channels - we have a slide at the beginning, and there's a couple quotes; Mat, one of them is from you...
+
+**Mat Ryer:** Oh, dear...
+
+**Mark Bates:** ...which was...
+
+**Mat Ryer:** Cut out Bates.
+
+**Mark Bates:** I can't remember what your quote was, but it was about not using channels. I can't remember what it is now. But Cory said "I find that once I -- you know, I'll start using channels, and about at the end I ended up refactoring them out anyway."
+
+**Mat Ryer:** Right...
+
+**Mark Bates:** That's part of the learning experience though... Sometimes overusing mechanisms when you first start learning a language.
+
+**Mat Ryer:** Yeah... What concepts in Go are in your experience then particularly troublesome for people? You mentioned errors earlier...
+
+**Mark Bates:** Interfaces, I've found, have been one of the biggest struggles in class. People really struggle with how interfaces work in Go. Everything else -- we can talk about channels and people would get it; I can talk about goroutines and people would get it... All this sort of stuff. But when I start talking about interfaces, people are like "Wait a minute... I don't get it."
+
+**Mat Ryer:** Right... Is that because they have already this previous knowledge, and they know how interfaces work?
+
+**Johnny Boursiquot:** \[40:08\] Yeah, where's the *implements IPersonal* or something, right?
+
+**Jon Calhoun:** I think people are just so used to that--
+
+**Mark Bates:** \[unintelligible 00:40:13.11\]
+
+**Johnny Boursiquot:** Right... \[laughter\]
+
+**Jon Calhoun:** Interfaces are definitely weird, in the sense that you're expecting them to follow a certain paradigm that usually comes from like a Java, or something like that, where you return the interface... It's not the same as Go, where it's like, when you write the implementation you don't mention the interface at all. And they're like "How is this related? You didn't even mention the thing."
+
+**Johnny Boursiquot:** How are you connecting those dots? Yeah, if it's not in the code, how do you know? It's magic. I don't like magic.
+
+**Jon Calhoun:** And it is confusing.
+
+**Mark Bates:** It is. Well, I always say it's implicit versus explicit. We don't have to say "foo implements bar", it just does. And then people are always like "But how do you know it does?" I'm like, "It just does." They're like, "But how do I know that my thing implements it?" It's like, well, you look at the function definition you're trying to call, and it says "I need an ioWriter." You look at the doc for an io.Writer, and "Hey, my thing happens to implement that. Fantastic." But people don't get those two things.
+
+And the big question I get a lot, too - as I'm sure you guys - is "Well, how do I know if I'm not implementing some other interface?"
+
+**Mat Ryer:** Yes.
+
+**Mark Bates:** Well, who cares?! Who cares if you are?
+
+**Mat Ryer:** Right, yes.
+
+**Mark Bates:** It could be, it doesn't really matter.
+
+**Mat Ryer:** Yeah, that's true. I remember that exact thought, like "What happens if by accident I happen to just match the method on signature for something else, and then it gets abused, or it gets used in the wrong way?" But it just doesn't happen. And because of the simplicity... For example, you can't have overloaded methods. In some languages you can have the same method name, but with different arguments. And Go doesn't let you do that. So yeah, it feels at the beginning like that's constraining you, and it feels like "Well, they need to add that feature." But actually, it pays dividends when it comes to maintainability, readability and clarity, doesn't it?
+
+**Mark Bates:** Yeah.
+
+**Johnny Boursiquot:** But those are abstract concepts you're talking about here. These are things that are subjective. Readability - what you find readable may be not so readable for me. So it's hard to get someone who's very used to seeing a certain style of code and attribute their own idioms for what readable code in that other technology that they know and love is... Basically applying that and saying "Well, that's what readable code is supposed to look like." It's very hard to carry all that stuff over to Go. It's almost like you have to relearn "Okay, what does readable-- at its core, what is readability all about?" You have to get to that underlying level.
+
+I'll tell you, in a class of eight hours, you're not gonna have time to talk about the theory of the readability, what's readable, what's not - you don't have time for that. You're trying to cram as much information in people's heads as possible, in as efficient a manner as possible, so that they can retain all of that knowledge you're dumping on them in such a short amount of time. You don't have time to dive into the theory, you have to get to the practical. But the thing is, without the theory to understand how to approach the language, then it becomes hard to say "Well, trust me - I've been doing Go for a long time. You're not gonna run into this particular problem you keep telling me about right now."
+
+**Break:** \[43:37\]
+
+**Mark Bates:** Changing topics -- Mat, sorry; allow me to take over hosting here for a second...
+
+**Mat Ryer:** Please do, mate... Please do.
+
+**Mark Bates:** A lot of people have been asking, what would we recommend for people who want to learn how to write Go? What ways of teaching, or what ways of self-learning I think is really the question that we've been getting on Twitter and Slack. "How do I best self-learn Go?" And I'm curious to hear -- because again, I think we've all taught Go in very different ways... Some of us in multiple ways, whether it be books, blogs, videos, workshops, in-person, virtual, podcasts, conference talks... There's a lot of ways we have taught Go, but I'm curious to hear how everybody on the panel here would -- people would suggest to people who are learning Go, what are some great ways for approaching it.
+
+**Jon Calhoun:** I can say the number one piece of advice I have for people is to not just follow along... Whether you're in a workshop, or you're doing a course, or whatever else, if you just follow along with the material there and that's it, it doesn't stick. But if you take the material there and you try to slightly change it... Let's say Johnny's teaching something and he's like "This is how we create a channel, and we send some values through it", and they try to maybe create a buffered channel, or they create a channel of a different type, or they mess around with it a little bit and experiment and try to get a feel for what these different things do - those people tend to have a much better understanding of how it works... And sometimes you'll write code that just doesn't work, and you don't quite get it, and then all of a sudden you have a question to ask. You can say "Why doesn't this work? I apparently didn't understand it as well as I thought."
+
+And even for courses, it's really hard, because some people just wanna follow along with the tutorial, or a course, or whatever, and at the end they've built a thing. But really, the value doesn't come from building a thing, it comes from "Okay, can I use this tutorial to build something sort of similar, but not the same. Far enough away that I'm gonna learn some stuff along the way." And those people tend to be the ones who get the most value out of reading tutorials, or courses, or workshops, or whatever else. So that's the biggest piece of advice I'd give to people - you need to be doing more than just following along.
+
+**Mark Bates:** Yeah. And to add to that, don't copy and paste the code from the browser into your editor.
+
+**Johnny Boursiquot:** No. Type it in. Type, it, in.
+
+**Mark Bates:** Hand-type it. Hand-type that code. It's muscle memory, and you will learn the code so much easier. If you just copy-paste it, you're not learning anything; you're just moving on. Spend the time, hand-type it. It's painful, but it's awful. But no, those are really good points.
+
+**Jon Calhoun:** The hand-typing is one of the things I loved about physical books - there was no copy-paste. It's like, "Good luck, buddy."
+
+**Johnny Boursiquot:** \[laughs\] Forcing you to do it... Yeah, there's truth to that. But interestingly enough, the other day I ran a quick Twitter poll, obviously limited to my audience and my reach and whatnot... But of the folks who follow me, pretty much mostly, 90% of them are technologists, and engineers in some way... I wanted to find out how they learned technical content. And the options I had on there were like video, blogs, documentation or something like that... But basically not the kind of content, but the medium; what medium do they use to learn.
+
+\[48:07\] I was expecting video to be the runaway winner. It actually ended up being a nice distribution across all the different mediums, including reading (the long way), watching videos, reading tutorials and how-to's, and having a live training, that kind of thing... Again, reinforcing the idea that different people learn in different ways. But in the comments, what I was getting - there was like an underlying thread there.
+
+What I was getting was that many folks need a combination of mediums to reinforce their learning. And I saw myself in that group, because literally, if I'm trying to learn a topic, I will get the video on Udemy, I will buy the book on Amazon, I will find a live training workshop to attend... Basically, I'm approaching it in almost like a 360 style, because I'm gonna get something from one of these instructors or teachers that I'm not gonna get from the other one. So I'll try to get myself like a complete 360 degree sort of knowledge even before I jump in to start actually writing code.
+
+I'm not sure what you call these kinds of learners, but I spend a lot of time gathering knowledge before I start to apply anything... And again, that might be slightly different from the typical recommendation to "Hey, learn it, apply it, type it" and whatnot... Some learners, like myself - we like to gather a lot of knowledge before we start applying it... But it's a different mechanism of knowledge acquisition.
+
+**Mat Ryer:** Yeah. See, I'm the opposite of that, because what I'll do is I'll start to build something. And I use that almost to guide my learning.
+
+**Mark Bates:** That's what I do.
+
+**Mat Ryer:** Yeah. Also, Mat, you mentioned Amazon - I should just say, other tax-dodging megastores are also available.
+
+**Johnny Boursiquot:** Right. \[laughs\]
+
+**Mark Bates:** I am a bit more like Mat. I like to start with a practical problem. I struggle reading thick textbooks... That's just never been my forte. When I first started programming, that's all we had. In the mid' or late '90s there wasn't a lot of online video courses and tutorials. You got a book at the store and you hoped that it was up to date. But I like the practical approach. I need to have a problem that needs some solving. Then I start working with it and investigating it.
+
+Often, if I have a big project I'm working on for a client or something, often there's a piece of it - "Oh, I've never done this before." Or maybe there's a new API I need to work with, or a library that I'm gonna have to work with... So I'll just open up a simple, main, basic project, and I'll start just a rudimentary playing with it in there; I'll pull up docs, and I'll watch some videos, and I'll read some blog posts... Until I get something that is actually working, and that's simplistic, almost linear, serialized programming top to bottom, just to get it work. That's how I learn. I need to physically be in the problem to get the education out of it.
+
+**Johnny Boursiquot:** There's an interesting dynamic going here I'm wondering if teachers feel the need, or if some teachers feel the need to get enough of a complete, broad set of knowledge, but then they advise their students to not learn in that way. They advise their learners to actually get the hands on, rather than try to acquire all the knowledge first... Because that's part of my reasoning. That's why I try to acquire as much knowledge as possible... Because only then do I feel comfortable telling somebody "This is what you should learn and how you should learn it." Because I feel the need to know enough about a topic. I feel the need to develop expertise on a topic before I can then pretend to be a teacher for that thing.
+
+**Mark Bates:** \[52:00\] Oh, if we're talking about teaching, versus learning - absolutely. I'm not gonna go and teach anybody how to write MUMPS. I've never written MUMPS in my life, and I'm not gonna pretend to sit there and teach people how to write MUMPS code. But learning I think is a very different process, at least for me anyway. For me, learning is that hands-on "I've got to be practical. I can't get it into my head any other way."
+
+**Mat Ryer:** I don't wanna self-promote, but I do have a book still available... You can get it on Amazon and other tax-dodging superstores. \[laughter\]
+
+**Jon Calhoun:** I just wanna let people know there are other books available. \[laughter\]
+
+**Johnny Boursiquot:** Not just Mat's.
+
+**Mark Bates:** There are other non-Mat-Ryer books available on Amazon.
+
+**Mat Ryer:** There are. Actually, there's loads. Most of them aren't written by me. But my book is a blueprints book, so it is about building real things. So the way I learn is kind of the way I teach... And I also extend that to trying not to over-teach or trying not to include everything. I try and pick the important bits, enough for people to be useful, so they can do something, and then start that reward process that you get when you code something and it works. That to me is what then inspires the next bit of learning, and it's kind of that tight feedback loop. So the sooner you can build something real - for me, that's a very exciting thing. It may not be the case for everybody, but yeah.
+
+I think the criticism in some cases might be "You didn't tell the whole story there. You've only touched the surface, or something..." And there's loads of other things that people could know about this particular thing. And I try and focus on what's pragmatic, what's practical, and what can be useful, because really, that's what we need to start with. And then it guides your learning, right? If you need something more, you know, because you're blocked.
+
+**Mark Bates:** Yeah. You also have to be sure you don't over-stimulate, too. You don't wanna throw too much information at people at any given time, because even if it's practical and incredibly relevant information, it's gotta be handed out in small doses, so that people get it.
+
+One of the big things we focused on at GopherCons in our content is when you see a slide, that slide's about one topic. One thing. And it could just be a very simple thing - one way of variable declaration, and that's all we talk about on that one slide. And the code example -- maybe it's a bigger code example, because we need more code, but we only show the 3-4 relevant lines to students, because we only want you to focus on those 3-4 lines that are important to this slide, for the topic we're talking about.
+
+I've found if you show somebody a huge, complex code example, you know what they're gonna do? They're gonna spend the whole time just staring and trying to figure out the huge code example; they're not gonna follow what you're saying. If you show them 3-4 lines and you just move through those slides, you're giving them the whole picture just a little bit at a time. And then they'll all catch up by the time you get to the exercise. Now it's like "Okay, take all of those little bits and you can build something a little bigger."
+
+But it's hard, you can't overload people either... And I see it all the time; that's one of the worst things you can do as a teacher.
+
+**Mat Ryer:** Another thing that happens sometimes - if you've just made up an example, just because it doesn't matter, I've done it where my example is not a good example, and that becomes then the focus... And that obviously gets in the way then. So that's a fail on my part. And usually, breaking it down and only showing the relevant bits is a good way to avoid that sort of noise that otherwise can get in the way.
+
+**Mark Bates:** \[55:57\] It is. It's also really hard -- you just touched on something very important, Mat, which is writing examples. It's probably the hardest thing you have to do as a programming teacher. Because we keep talking about Go, but we're just talking about programming languages. All this applies to every other programming language on the planet, right? And honestly, learning anything, or teaching anything in general... But yeah, it's -- I've lost my train of thought... \[laughs\]
+
+**Mat Ryer:** Well, writing examples is hard. You use the Beatles a lot, don't you?
+
+**Mark Bates:** I do use the Beatles a lot, just because I'm a big Beatles fan. Well, they offer just enough variety, and those forenames - they're not hard names, they're not long names; they play instruments... It's just little bit that you can use, and they're good enough to give you more than 2 or 3. Because 2 or 3 things in an example is usually too small. But anything more than 5 or 6 is way too big; it's just too complex. So 3 or 4 is usually a good example... But finding small, tight, concrete examples of the thing that you wanna show that's also kind of real worldy too, because the smaller and tighter you try to get an example, often the less practical that example becomes. It becomes a toy, and I'll tell you, we've spent so much time, Cory and I, sitting in Zooms together, talking about examples, going "This one just doesn't work in class. It's too small/It's too complex/It's not showing the problem tight enough..." And it's hard. The words - that stuff's the easy part. It's those descriptions - that's what people see. And the developers wanna see code; if you show them code, it's gotta be good code, that illustrates it exactly, and that's really hard.
+
+**Jon Calhoun:** I think this is one of the reasons why so many people struggle with structure... Aside from the fact that there's not a one single solution, when you try to show examples it's really hard to show an example that's small enough to actually take in at once, while also appreciating the benefits of like "This is why we structured it this way, and this is why it works." Because you almost need a large codebase for that to make sense, especially when you get into the more complex structures... Like you're using domain-driven design; really, if you're building a Hello World app, domain-driven design is just complete overkill and it's not helping you.
+
+So it's like "How do we give an example that illustrates these points and is complex enough to feel the pains that this might solve, while also being simple enough that you can actually take it in in a short period of time.
+
+**Johnny Boursiquot:** My trick to that is to offer layered examples, as we get through the curriculum. Reintroducing and saying "Hey, remember in the last module we talked about this. Now we're gonna add an additional layer of complexity to that." So it's building up that mental model and introducing more complicated things. But basically, you have the same one or two set of exercises that are getting incrementally more complex. That way, it's not a completely new problem you have to solve with each new example of code I'm showing you, or with each new module; it's just a continuation... Which is why I really enjoy projects that take that approach, to say "Hey, we're gonna build a Kubernetes cluster together." And we start from the ground up and we keep adding layers. But by the end, you know you're gonna get a fully working Kubernetes cluster or something out of it. But we're introducing the topics and adding the necessary bits as we go. But you have something you're working towards.
+
+**Mat Ryer:** That's great, yeah. Would you have any advice for people that want to get into training? And I'm also gonna ask a cheeky question...
+
+**Mark Bates:** Don't.
+
+**Mat Ryer:** \[whispering\] Is it good money? Is it good money? \[laughter\]
+
+**Mark Bates:** It's a good job if you can get it. Here's the thing - we've really just been talking a lot about the actual presentation of the material... And just at the end there we were starting to touch a little bit on code examples, and stuff like that... But what I think people really don't understand is that when you go to teach something, if you wanna teach three days, you have to have three days of content. And three days of content doesn't grow on trees. \[laughter\] It takes you a lot longer than three days to write three days' worth of content. It takes months.
+
+\[01:00:24.00\] So if you tomorrow say "I'm just gonna go and start teaching", well good luck to you, because you're not gonna get very far unless you're really ready to sit down and spend hours and hours and hours fretting over your content.
+
+**Mat Ryer:** You could just buy Jon's course and play it on an iPod in your ear... \[laughter\]
+
+**Jon Calhoun:** That would be pretty interesting...
+
+**Mark Bates:** You could, yeah. That would be the same. You'd have to have the same tone...
+
+**Jon Calhoun:** I think it'd be hard, because my stuff isn't designed for in-the-classroom really... But maybe. But Johnny mentioned this earlier, where as educators we like to learn everything; like, really understand all of it, and then we sort of figure out "Okay, this is how I'd recommend learning it, now that I've actually spent a large amount of time learning this all." And I think that's where a lot of the value in these trainings and workshops and all this stuff comes from, is not just in the content, it's the fact that I've spent hundreds of hours figuring out the best way for you to spend 40 hours and learn the most amount of content.
+
+And a lot of people are like "Why does it cost so much?" It's like, look, you can do this on your own, but given that you as a developer you probably make at least $25/hour, probably much more than that, it's really cheap in comparison to how much time you're gonna waste.
+
+**Johnny Boursiquot:** Right. And this is where we slightly deviated, so talking about the business value. I don't know if I sound like a suit but really, this is where you have to look at "Okay, what am I paying, but what am I getting in return here?" You're not just getting basically 50 bucks, 100 bucks, 300 bucks, $1,000 worth of content. What you're getting is -- and let's be honest, anything that I teach in my classroom, and I'll go as far as to say anything that all of us on the panel here teach, you can google around for it if you want. There's nothing that we're teaching that is proprietary, or magical, there's nothing you're gonna learn in our classrooms that with sufficient time and persistence you're not gonna find out there and teach yourself.
+
+**Mark Bates:** Well, how do you think we learned it all?
+
+**Johnny Boursiquot:** Exactly, how do you think we've found it?! The value is in having somebody who's spent the time distilling all of that stuff down to the key nuggets to help you pick it up in 40 hours, rather than 400 hours. So what is your time worth? We find joy and satisfaction - and hopefully monetary gain - from doing this work, because we enjoy doing it... But really, at the end of the day, what is your time worth to you? Do you wanna spend $200 googling around until you learn and master something?
+
+**Mark Bates:** Or your team's time.
+
+**Johnny Boursiquot:** Right, yeah.
+
+**Mark Bates:** When you're talking 40 people, that's a lot of time.
+
+**Johnny Boursiquot:** That's a lot of money. That's a lot of time and money.
+
+**Mark Bates:** Yeah, absolutely.
+
+**Jon Calhoun:** To even back Johnny's point, I've talked to people who make courses where literally everything taught in the course is available for free on their blog. It's just all put together in one big project, in a really nice order... And they're like "Look, people pay just for that. It doesn't matter that it's all there free." They'll read one little bit free, as a blog article, and be like "I want the whole course. I don't wanna go searching for the rest of this."
+
+**Mark Bates:** Well, that's kind of the Michael Hartl Rails tutorial model, where he gives away the -- or used to; I don't think he does it anymore. He used to give away the entire Rails tutorial book online for free, as HTML. You can read the HTML version. Or you can spend hundreds on the videos, and the PDFs, and ePUBs... And he does great with it. I can't tell you how many people I've seen go up to him and thank him 100 times over for his courses, and stuff. He will be out at a conference, or hanging out, and people just come up to him and thank him all the time.
+
+But he didn't get there by accident either. Again, he puts a lot of effort and time into all of that. So if you're thinking about getting into teaching, consider all that we've just said about that time, and that effort.
+
+**Jon Calhoun:** \[01:04:08.12\] I would just say start small.
+
+**Mark Bates:** And start small, yeah.
+
+**Jon Calhoun:** Start small and see if you like it. Because it's good to have more teachers.
+
+**Mark Bates:** One of the things I always recommend -- when I used to run development teams, I used to force brown bag lunches every two weeks, and force all of the engineers, one at a time, to do a presentation during a brown bag lunch. And it could be on the library, a plugin, something that they're working on at work, it doesn't really matter. It's a safe, small environment, with half a dozen of us in the room, or whatever... But it gives everybody a chance to stand up and start to learn a little bit of how to teach and how to mentor. Because at the end of the day, as we grow in our careers, whether we make our livings as educators or we're working in dev shops as senior developers or directors, or VPs, or whatever - as you move up through your career, mentoring and teaching is a big part of your job. And as you move up, that's why you get paid so much. You get paid to mentor the people below you, and you've gotta start learning how to do that now.
+
+So if you are thinking of that, or you're just looking at your future prospects, I think you need to start -- try doing some of those things in-line, at your company; a brown bag, teach some stuff. Have other people teach some stuff. I think you'll do great.
+
+**Mat Ryer:** Yeah. Well, it's that time again... Bates, hold this space guitar, mate. Jon, get on the drums. Johnny, piano... It's time for Unpopular Opinions!
+
+**Jingle:** \[01:05:52.19\] to \[01:06:07.05\]
+
+**Mat Ryer:** Do we have any unpopular opinions about teaching? \[laughter\] Mark's really confused by that... But they're gonna put the music in, aren't they?
+
+**Jon Calhoun:** Has Mark been off the show that long?
+
+**Mat Ryer:** \[unintelligible 01:06:04.26\]
+
+**Jon Calhoun:** I don't wanna offend anybody, but I guess without a better way of putting it, I think the smartest people often make the worst teachers.
+
+**Mat Ryer:** Hm... Thank you. \[laughter\] Bates is brilliant, by the way.
+
+**Mark Bates:** I'm trying to decide which way I should be offended.
+
+**Johnny Boursiquot:** I know, right? \[laughter\]
+
+**Mark Bates:** Either I'm a great teacher and I'm really stupid, or I'm a terrible teacher and I'm really smart. So I'm trying to figure out where I sit in there.
+
+**Mat Ryer:** Tell us what you mean though, Jon. That's interesting.
+
+**Jon Calhoun:** So what I mean by this is - in my opinion, the smartest people tend to be people who get things very quickly. They don't struggle with them, so they don't really start to relate with how the average person is going to learn the material. So as a result, the smartest people are the people who can't produce a learning path that is going to appeal to the average person, and they're just gonna try to go too quickly.
+
+The best examples I can give is I had a friend in college who had -- it was like a Calc 2 class, or something, and every time somebody would ask a question, they had this genius math teacher who would just be like "Oh yeah, it's real simple" and he'd just throw something up on the chalkboard real quick... And the whole class would sit there like "We don't know what you just said. We do not understand anything you wrote on the board."
+
+I think part of it was because it all just made sense to him and it all clicked, so he didn't quite understand where they were struggling. He didn't have the empathy to really relate to him.
+
+**Mat Ryer:** Maybe he just needs to be a bit smarter. \[laughter\]
+
+**Jon Calhoun:** So what I mean by that - smart people can become good teachers, but I think naturally if they take what they learned in the way they learned it, it's not going to do well. Another way to phrase this is I feel like you have to be dumb enough to struggle with the material, so you understand how the average person is going to struggle... But you have to be smart enough to figure out how to teach the stuff that you struggled with. And I just feel like the smartest people don't necessarily have that struggle.
+
+**Mat Ryer:** I think you've managed to find an unpopular opinion.
+
+**Mark Bates:** \[laughs\] I think it's a--
+
+**Jon Calhoun:** I mean, I'm calling myself dumb as well, so...
+
+**Mark Bates:** \[01:08:10.22\] It's an interesting opinion. I don't know if I fully agree with it, but I can kind of see what you're saying. I think there are definitely a class of really smart people who struggle with interpersonal skills and who struggle connecting with people and primarily live in their heads... And they definitely make terrible teachers, because they're not willing to slow their brains down, or don't know how to slow their brains down. But again, I don't think it has anything to do really with smart, but more in terms of the personality and those interpersonal skills. Because if you're really smart and you have interpersonal skills, you can make those cognitive leads, and you can say "Hey, how am I gonna get this person from A to B?" And I think you have to be smart to do that sort of stuff... But you have to have the personal skills. I think that's the difference.
+
+**Jon Calhoun:** Smart is really not the right word, but I don't know the correct word for it. It's almost like the people who just innately get things are not necessarily the best teachers. And I realize it's a generalization and there are exceptions; I don't expect that to be true of everybody.
+
+**Johnny Boursiquot:** Jon, I think I'm gonna connect the dots for you.
+
+**Jon Calhoun:** Okay.
+
+**Johnny Boursiquot:** And I don't know if it's an unpopular opinion or not, but... Those who are smart and know it, and bring that to the teaching with them - I think those make the poor teachers. In teaching, in training other people, your ego does not belong there... Because it's not about you. You are trying to convey information, you're trying to relay information, you're trying to teach another person how to fish, to paraphrase the old proverb. So it has nothing to do with you; you're just a conduit for information.
+
+So if you go on stage, whether you're giving a conference talk, or whether you're on a podcast of professional teachers as a panelist, or something...
+
+**Mat Ryer:** Insulting them all... \[laughter\]
+
+**Mark Bates:** Or you're Mat.
+
+**Johnny Boursiquot:** Or you're Mat, yeah... Or you do this professionally or not, on occasion, whatever it may be - your ego should not be part of that equation. It's about how you convey information to people in different ways, meeting them halfway, meeting them where they're at, and sort of conveying that information. You're nothing but a conduit. Leave your ego out of it.
+
+**Mark Bates:** Can I add to that and say not only you're a conduit for teaching information, you're a two-way conduit, because you should be learning while you're teaching. And if you're not learning while you're teaching, you're not gonna be a good teacher... Because you need to learn your audience, you need to learn who they are, you need to learn how to talk to them, learn how to work with them... But also - I think I said this earlier - I'm not afraid in class when I get asked a question to say "Honestly, I have no idea. I've never thought of that before. I've never seen that before. That would never cross my mind..." And I'm okay saying that, because then the next thing I say is "You know what - let's open up Vim. Let's see what happens."
+
+**Johnny Boursiquot:** \[01:11:14.12\] Exactly.
+
+**Mat Ryer:** Right.
+
+**Mark Bates:** "Let's learn together." And great, now I know. And I have learned something from that class that I didn't know before I went through it. Like I said, Cory and I try to make an effort of putting that right back into the material, so we have it for the next time. So it's like "Yeah, someone asked a great question, we didn't have an answer, we didn't have a slide... We're gonna learn that answer, we're gonna learn how to present that, and we're gonna give it back to the next class that comes along." And if you're not willing to do that, you're gonna be a pretty \*blip\* teacher.
+
+**Johnny Boursiquot:** Yup.
+
+**Jon Calhoun:** I've struggled with how to convey that opinion just because--
+
+**Mark Bates:** What, with me swearing?
+
+**Jon Calhoun:** Because I completely agree with you guys not this swearing...\[laughter\] It's a struggle to convey that opinion, because it's almost like I'm saying somebody who learns how to present material and has the smarts to understand their audience and relate with them and figure out the best way to teach them isn't smart, and that's not what I mean... I think it's more like Johnny said, there are some people who are -- it's almost like they're just so focused on their own intelligence and showing people how intelligent they are...
+
+**Mark Bates:** Yeah, they live up here... They all live up here.
+
+**Jon Calhoun:** Yeah. They have to be able to step down and relate to the audience. And I think, in my experience - at least with university, and stuff - the teachers who tended to be like that were the ones who were there strictly for research, and they were forced to teach a class, and it showed. They were not teachers, they were very smart people who did not wanna be teachers.
+
+**Mark Bates:** They wrote some \*blip\* thing on the whiteboard and expected you to know it, and that was it. Yeah, I'm with you...
+
+**Mat Ryer:** That's your second strike, Bates... \[laughter\]
+
+**Mark Bates:** Oh, no way! I'm banned!
+
+**Mat Ryer:** So do you have any funny stories? Because I did teach the Gopher Guides--
+
+**Mark Bates:** Okay, two guys walk into a bar...
+
+**Mat Ryer:** No, no, hang on... \[laughter\] I taught the Gopher Guides material in London, Bates, if you remember...
+
+**Mark Bates:** Yeah. Years ago.
+
+**Mat Ryer:** And at the end of the three days, this young man came up to me and said "I think I'm your cousin", and I'd never met him before... And it turns out he was my cousin.
+
+**Johnny Boursiquot:** And he learned that as a result of you teaching the Gopher Guides material?
+
+**Mat Ryer:** Yeah, I was teaching a class at a company...
+
+**Johnny Boursiquot:** Wow...!
+
+**Mat Ryer:** ...and I've found a cousin. So you can get extra cousins if you do teaching. That's one of the perks. \[laughter\]
+
+**Mark Bates:** It probably could have been worse. He could have maybe introduced himself as your son. \[laughter\] "I think you're my dad... Because of this channels example!" I'm not quite sure how a buffered channel would get you to that conclusion, but... \[laughter\]
+
+**Jon Calhoun:** He got into the unsafe package.
+
+**Mark Bates:** \[unintelligible01:13:47.04\]
+
+**Mat Ryer:** Just in time, and to save Mark's career, we are gonna end here. So thank you so much to all our--
+
+**Mark Bates:** I think it's too late to save my career... \[laughter\]
+
+**Mat Ryer:** Thank you so much for listening, everyone... We'll see you next time.

Unusual uses for Go GUIs_transcript.txt

@@ -0,0 +1,283 @@
+**Jon Calhoun:** Hello, everybody! Welcome to Go Time. Today we are joined with guest Andrew Williams. Wanna say hi, Andy?
+
+**Andy Williams:** Hi, everybody. Thanks very much for inviting me along. It's really cool to be part of the chat.
+
+**Jon Calhoun:** And then we have one of our hosts, Johnny Boursiquot...
+
+**Johnny Boursiquot:** Hello, everybody. Good to be back.
+
+**Jon Calhoun:** And myself, Jon Calhoun. Today we're gonna be talking about unusual uses for Go; just the weird ways developers are using Go. Maybe not weird, but just not quite the normal ones.
+
+First, we just wanna establish what are some of the more common use cases of Go, and then we're gonna jump into what we'd consider anything that sort of falls outside of that, is what we'd say is slightly  unusual. That isn't to say that all of these are unusual. Some of them are more common than others, so you'll see that going in all different directions... And we're not trying to encourage anybody to not use go for any specific reasons; this is more of just exploring some different ways you might use it, that you might not have traditionally heard about.
+
+Who wants to take the common use cases? What do you think those are?
+
+**Johnny Boursiquot:** I can take a stab... Do you mind if I take a first stab, Andy?
+
+**Andy Williams:** Yeah, on you go.
+
+**Johnny Boursiquot:** Okay. So my own personal experience is that when I first heard of Go, it was all about back-end, systems-level kind of work, queueing technologies, databases, that kind of thing. High throughput, networked applications and services; what was being called at the time lower-level kind of back-end things, as opposed to your traditional web developer framework stacks we were familiar with. And really, that's been fairly consistent, up until about, I'd say, three(ish) years ago. Then I saw an explosion of all kinds of different uses of Go. But yeah, we're gonna get into that... But that's my personal experience of what the common uses of Go have been for me until a few years ago.
+
+**Andy Williams:** Yeah, I think that sounds about right for me as well. I've only really been in the community for not even two years actually, so after all of these new and exciting areas started opening up... But the common use cases still seem to be focused around web servers, back-end systems, like you say; that's where the examples are. And often, if you ask somebody "What's this language for?", they're pretty much gonna say "What it was designed for." I think, like you said, that's solidly where it started.
+
+**Jon Calhoun:** \[04:00\] Yeah, I'd say my experience has been pretty similar. I've definitely seen -- even on the web side of it; you see it a lot more in APIs and things that are returning JSON than with the -- like, you don't see the templating library being used quite as often to generate HTML. A lot of times when people think about web frameworks, they think about Rails, or something else... And that stuff has started to exist in Go, but it just wasn't nearly as common until more recently, as tools have started to emerge.
+
+Okay, let me jump to the next question... Before we deep-dive into any specific areas, what is the strangest thing you've seen Go used for?
+
+**Andy Williams:** Wow... [laughs]
+
+**Jon Calhoun:** Am I gonna throw you both off with that one?
+
+**Andy Williams:** I mean, I'm biased working so much in the GUI space; I've seen some pretty crazy things done there, and there are some ideas around running user interfaces using native, that's then rendered through the web browser instead of using the web technologies. That feels pretty whacky to me. Interesting area, some cool applications, but not my first choice for how to build something.
+
+**Johnny Boursiquot:** To me JavaScript is  a weird one to have in Go. Specifically, I'm thinking of projects like GopherJS, Vecty... There's some we've got listed here, goplay.space - I'm not sure what that's about... That's kind of odd. I don't wanna sound like I'm anti JavaScript in Go, but typically the way I see it is that if I need to do JavaScript, I'll just go do JavaScript. I wouldn't try to bring it in into my Go world. I have this odd desire to use the right tool for the job, and to me, trying to force that in -- I mean, I'm sure it has its use cases, but for me personally, I haven't come across one where I really am like "Well, you know what - yeah, let me do some JavaScript through Go." It just never felt right to me.
+
+**Jon Calhoun:** I think the JavaScript one is the same for me, but mostly just because every time I've looked at it and been like "You know, I wanna try this thing. I wanna see what it's like", and then I look at it and I just think "I almost feel like I need to know JavaScript", and then  I need to know the Go version on top of it. So I'm like, if I've already learned JavaScript at this point, I'm not -- like, there's probably some benefits in different things... And that's not to say that it's not a cool project. I definitely like that people are experimenting and trying different stuff. I love that. It's just, I see it and I'm like "I cannot see me using this in a production environment, or trying to sell it to a manager." I don't know how I'd make that sale.
+
+So if somebody has actually made that sale, please let me know how you did it, because that's definitely an interesting one to pull off.
+
+**Johnny Boursiquot:** It's a super-power. [laughs]
+
+**Jon Calhoun:** Yeah... When you see those -- goplay.space is an example of something built with Vecty and GopherJS, and it's really cool, but I just sometimes wonder if you might be better off just finding somebody who knows the JavaScript stuff really well, and just doing it that way... But it's hard to say. It is cool that people are doing it. But on the other side, the WebAssembly stuff I think has a lot of potential to be really cool in the future... And I'm guessing things like GopherJS sort of helped pave the way for that, so I'm like "Alright, if it's helped me do that, that's pretty awesome, too."
+
+**Johnny Boursiquot:** I'm curious what the impetus is for doing these other things through Go. I imagine if you're really passionate about Go and you wanna use Go for all the things, maybe you can generate your JavaScript from Go, and that's fine... Again, I see these things as more of being sort of nice, fun thought experiments, and nice projects that are pushing the boundaries of what's possible, and I think you do need these kinds of projects in any ecosystem, to sort of show "Hey, let's think outside of the box that was defined for the language from day one" kind of thing. I think those definitely have a place in the community, and they will always play a role...
+
+\[07:57\] But to your point, if I'm trying to build production-grade applications -- not to say that those projects can't produce production-grade applications, it's just I'd have a hard time selling not doing JavaScript in JavaScript, or in a framework that is tailored for doing, say, graphical user interfaces or web interfaces, or even mobile interfaces with JavaScript tooling. Trying to do that through Go - I'd have a hard time selling that.
+
+**Jon Calhoun:** Yeah. So it sounds like what we need to do is get somebody who's an expert in these areas and have them come on and change our mind, which would be pretty awesome.
+
+**Johnny Boursiquot:** Yeah, I don't know. Do you think we can find one of those?
+
+**Jon Calhoun:** We can look, we'll see. So we have Andrew here with us, and Andrew has a little bit more experience in what I'd consider more the native graphical user interface area. For those of you who are not familiar with Andrew, he created - or helped create; I'm not sure how that started - [Fyne](https://github.com/fyne-io/fyne), which is something that allows you to sort of build native graphical user interfaces (GUIs). So I guess the first question I wanna ask is "Why don't we see more people doing graphical user interface type stuff in Go or in some of these languages?" What makes it challenging? Because truly, people want to build these applications that work natively on any OS.
+
+**Andy Williams:** Yeah, it's actually a bit of a mystery to me, honestly. When I was first thinking about how might you reimagine building graphical applications, which is kind of where Fyne came from, I looked at Go as a language and compared it with a few others and it just seemed like such a great fit, with the concurrency, memory management, and just the language semantics seemed to fit really well.
+
+Partly, I think maybe there's not much going on in this space because that's not what people originally expected that the language would be useful for, so it's kind of a few years behind that opportunity, because it was particularly not worked on for a long time... But like Johnny said, it's been expanding over the last few years into lots of different areas, and people are starting to think outside of that area, so the graphical toolkits along with other things are coming along now... And you probably look at the Awesome Go list and see like 25 different toolkits, all trying to do something with graphical user interfaces in Go, and maybe half of those are on the embedded website, and the others would be to some flavor of native. That said, it is a really hard thing to do well, and if I tell somebody "Oh yeah, we're building a new graphical interface toolkit from scratch", they just look at you and go "Why? That's so much work... Surely, the ones that exist are good enough." So it is, I think, one of those challenges that people just go "What's the point...?"
+
+**Jon Calhoun:** I know one of the big issues I see -- because people have been trying to make this cross-platform GUI-type thing for a while... You see React Native, and you see a bunch of others out there, and even on mobile, there were people who would be like "Okay, come learn our thing and you can develop for both iOS and Android at the same time", which was always a big challenge... So I guess the first question is "Do you think it's really possible to make good user interfaces that are cross-platform one time, or are we stuck just building it for each different operating system separately, because they all expect different stuff?"
+
+**Andy Williams:** Well, I absolutely think that it can be done. That is the purpose behind the project, to show that one of these considered impossible tasks actually just hadn't been tackled with the latest toolset, a language that supports these ideas. And if I didn't think it was possible, I'd be doing something else with my time, for sure... So I think it's a question of trying to figure out "How do you apply current technologies on top of all of the learning that we've had over the last 20-30 years of building graphical interfaces and the toolkits that support them, and bring all of that together with a bit of fresh thinking?"
+
+\[11:59\] One of the things that really drove me initially was when the smartphone apps were really taking off and we saw what good usability could really look like... And then people looked at desktop and just went, "Oh, these two are incompatible." Well, you know, maybe that's because actually we needed to take the opportunity to reimagine it and see how this could work across all platforms with a fresh look, and can we apply the design learnings, the usability, and take those concepts onto both desktop, but also the cross-platform with one codebase concept. I think there's a lot of space there to really come at this with a fresh angle, and that's what we're trying to do.
+
+**Johnny Boursiquot:** There's always a trade-off. If you're trying to create something that's cross-platform, in some environments you might say "Well, for the Mac environment - we can take advantage of certain things here, but for a Windows environment we can't take advantage of the same things; we have to sort of give up some things here in order to get that..." So where are the trade-offs that you're making? Are you hiding those from the developer, from having to create sort of specific OS-based APIs and things for allowing a developer to do what they wanna do with the project?
+
+**Andy Williams:** Yeah, absolutely. I guess there's definitely trade-offs in anything that you're trying to do cross-platform, although as the Go team have showed us, you can actually find clever ways to work around most of these and still have an elegant API at the end of the day. It probably would be worth looking at what native means, because there's a lot of different toolkits that are trying to be native and they can take different approaches. The [andlabs UI project](https://github.com/andlabs/ui), which is doing a really fantastic job of abstracting a standard API across system standard components so that you build an application with one codebase, and when you run it, it looks exactly like any other application on the system it's running on...
+
+So the trade-off that they're gonna be taking on board is, I guess, the lowest common denominator to element, although they're managing to build more complex components off the standard items available... So Fyne's in a place where we thought "Actually, let's have a standard user interface across all of these systems." So the trade-off there is probably the immediate recognition that a user might expect when they're loading a new application. We're presenting them something that is a little bit different. We're going for consistency across the platforms, as opposed to specifically consistency with the current system, which is a potentially courageous, but design choice that we made... So there's a bit of trade-off there with user familiarity.
+
+The one that  we're looking at at the moment is around system dialogs. If you're running on, for example, the Apple desktop, there's a lot of functionality there (the iCloud document store etc.) that if you're saving a file, you would expect to have it presented to you... So we're needing to look at a system-by-system basis, how exactly that integration might work, so that people can get access to the files that they would expect on their system without a huge variance in the capability of the software running on different platforms. There's all sorts of ways that it has to vary across systems, but that's the one that's certainly on our mind at the moment.
+
+**Jon Calhoun:** When you said "The Go team showed us that we can actually use some clever things to get around the differences between the operating systems", just for anybody who's not familiar, I'm assuming you're referring to having build tags and having specific Go files that compile depending on the language you're building for?
+
+**Andy Williams:** Absolutely, that's a big part of it, but I guess I was more thinking that from the language and a standard library level you really don't need to worry about it at all, for the most part. If you're reaching for a build tag, you're probably wanting to do something specific for a certain platform when you're making that choice to break away from the guaranteed consistency across systems... And that is certainly a challenge when you start working in the graphical world, where those guarantees don't necessarily go away completely, but they certainly introduce a lot more challenges... And how you manage to have an API that is as easy to use as the standard Go library is, whilst dealing with these more system-specific concepts is a challenge, as well.
+
+**Jon Calhoun:** \[16:17\] Yeah, so what I meant initially was that you can have multiple Go files, and  each one, depending on the language you're building for, is the one that's used... But the actual API that people are calling is the same functions, and they're generally expected to do the same thing, it's just how they do it might be a little bit different.
+
+**Andy Williams:** Oh yeah, absolutely.
+
+**Jon Calhoun:** But then the flip side to that is you kind of get lured into this almost wanting to have two things act in different ways, and you don't want developers to call a function and have to mentally think "If I call this function and I'm running on Mac, it's gonna do this thing. But if it's on Windows, it's gonna do something similar, but not quite the same... My code needs to adapt for both of those", which would be very challenging, especially -- like you said, in the GUI world, everything from alerts and notifications and just permissions... There's just so many different things that are really challenging to pull off, because that's where the differences in the operating systems really start to stick out.
+
+**Andy Williams:** For sure, yeah. I think if anybody's ever trying to design an API that's gonna be consumed outside of your team at work, it's important to consider the path of least surprise for any developer that's gonna be using the API... And if you do want to put platform specifics in there, the outcome should really be entirely consistent, irrespective of the specifics that are happening behind the scenes... So there may be a significant difference to how things are functioning, but really the end result should be consistent. Notifications is a really interesting example there, but there's probably a hugely long list of those sorts of system items that are going to be challenged to do consistently across systems... But from the developer's point of view - yeah, they call a function, and a thing happens; the documentation, whether it's GoDoc or something more elaborate, is going to describe the functionality, not the platform-specific item, and I think that's important.
+
+**Break:** \[18:09\]
+
+**Jon Calhoun:** So you're working on Fyne, which does (I believe) mobile and desktop... Is that correct?
+
+**Andy Williams:** It was desktop initially, and just in December we added mobile, so that's iOS, Android, and also a Raspberry Pi fitted in there, because it's running the same chipsets for the graphics output as the mobile devices are, so... It's not really mobile, but it was a nice added bonus at the same time.
+
+**Jon Calhoun:** So when you're thinking about that, do you think about those -- like, when you're exposing an API, do you guys feel that that API should be the same for mobile and desktop, or is this something where you've actually drawn a line and said "One's different enough from the other that we can't just make it universal"?
+
+**Andy Williams:** \[20:10\] No, absolutely... These APIs that we're building have to be consistent completely. We're following Go's design principles on that, and everything that we do is idiomatic, or aiming to be idiomatic to the language, and consistent APIs is really important there. So if somebody's rating an application with Fyne, then they know that it's going to work across all of these different devices in the same way.
+
+That said, there are sometimes differences between devices that you want to enable that aren't generic. The one that springs to mind right now is a virtual keyboard. That's not typically available on a desktop platform, so there are some APIs that are device-specific, and you can use appropriate calls to say "You know what, if I'm running in this environment, then take this action." So that's available if people want to customize their systems to the device it's running on, but it's not really encouraged, because we want to make as much of this completely transparent as possible.
+
+**Jon Calhoun:** I know historically one of the big downsides to using something that does the GUIs across platforms tends to be performance, but a lot of times I think that's showing up in JavaScript worlds, where everything's running through JavaScript, rather than running in something that was actually compiled... So do you think using Go actually helps prevent that issue and keeps it snappy?
+
+**Andy Williams:** Yes, absolutely. The performance that we've experienced has been really phenomenal, actually. You're quite right, a lot of the technologies that try to tackle the cross-platform do suffer in some of their choices, and performances  can be challenging to keep up there... But when the Go code is compiled down to the machine, apart from some implementation details that we might have, it's gonna be running at the same speed as the native code as if you'd been building with the toolkits that the platform was designed with... And partly because the graphics drivers that we have implemented are going straight down to the same hardware acceleration that the Swift or Java codes would be using as well.
+
+**Jon Calhoun:** So I guess I find that interesting, because if you're getting down to that level, I know one of the common concerns that I hear people talk about - and this isn't specific to GUIs, but I talk to people in the system administration space and stuff like that, and they always make the argument that like -- say I'm a Mac sys admin; they say that eventually I need to get into Swift to touch the lower-level things that I need to touch. I guess, do you ever run into cases where that's the case, where you really need to touch OS-specific APIs? Because that's one of the issues...
+
+Or I guess a better way of putting this is that it seems like to write for Mac, you need to know Swift; to write for Windows, you need to know one of those .NET languages that actually can interact with the things they want you to sort of restrict... And it seems like they don't care as much about supporting other languages with those really specifics... So has that been a challenge, or is that something -- you're just not doing those things?
+
+**Andy Williams:** It really is a challenge, yeah. Absolutely. There's certain things that there's just no way to address using a language that's not what was intended... And it's something that occupies my mind when we're working to do new capabilities on systems. But our best efforts are to hide all of that complexity from anybody that would be building their application with our APIs... So the project exposes a pure Go API, which is great for everybody who's working on it, and it makes a lot of sense in all the standard tools... But if you were to go and look at our source code, then you would find Java, and Objective-C, and some C... It's all brought together under the hood, depending on the target build system. And there's a couple of tricks in there to make sure that you don't have to have all of those variants installed all of the time just for a build to work. This is largely gonna work out of the box if you have Go and a C compiler.
+
+\[24:15\] So there's a lot of complexity under the hood... But it means that we're able to hook into the platform-specific APIs, the types of things that are only available on an Android device if you're accessing the Java APIs, or that might only be exposed through Objective-C on an Apple computer... But we want to make sure that that is never anything that you would need to think about if you were building on top of our system.
+
+There's obviously going to be areas where we haven't completely added all the support that we need to, but we're working on it over time, and if people find an area that they're having to reach out to some other language for, then we'd encourage them to open a ticket and help us work that support into the main project.
+
+**Jon Calhoun:** So you say that you build with Java and seeing these things... I guess can you talk a little bit more about that? Are you using just C bindings and connecting to things that are gonna be on every system? What does that look like? Are you talking like OpenGL, or is it something else?
+
+**Andy Williams:** Yeah... So it is pretty much through [cgo](https://golang.org/cmd/cgo/) like you say and the main dependency issue is talking to OpenGL, which gives us access to the graphic subsystems. In fact, that is the only dependency for this on most systems. On desktop we don't need to do any clever things, for the most part... And some Apple APIs that might be required are actually accessible by C, or Objective-C that it compiles down to using the standard toolchain, so it's not really too much of a problem. When you're building for mobile, this becomes a lot more challenging, but the Go mobile team have done a fantastic job of actually solving a lot of the challenges there for us, and through extending that project we've managed to add that support.
+
+I think the craziest thing that I saw in terms of making that work was that the Android target has some Java code in it that's pre-compiled into a text binary that's then bundled into the Go source code as a data asset, and then that's extracted as part of the build process to give you your bootstrap into the Go runtime. Now, I would rather hope that nobody using this would ever know that that existed under there... But if you wanted to delve into "How can I improve this for Android specifically?", you're gonna find some really weird stuff.
+
+**Jon Calhoun:** Yeah, I could imagine that being confusing, at the very least, when you're getting started... So the Go mobile project is what was helping you bootstrap into that... Was there anything that you've found - open source or otherwise - that helped you get started with the overall project, or getting into OpenGL and that sort of space?
+
+**Andy Williams:** Actually, the graphics is a complicated area, I guess... And it's very low-level, so the code reuse between projects is not exactly shining in that area. Really, I guess I was just basing on experience I had with previous projects as much as anything. I can't think of anything in particular that we really called on... Although I suppose initially the project did use the render pipeline from the EFL toolchain (that's the Enlightenment Project). That gave us some abstraction for the graphics driver. But in the end, we realized that to really build an idiomatic API top to bottom, we couldn't depend on an abstraction built in another language. It just didn't really speak the right language for us, and we were working a lot to work around the way it functioned, and we had duplicated code in there, and thought "Actually, this doesn't make sense", so we took it out and implemented it from scratch, right the way down to the hardware.
+
+**Jon Calhoun:** I mean, it's probably even more pronounced in your space, but I think in a lot of spaces you'll see that sort of difference, as to whether or not they started with the end user -- like, they're designing the API for end users, versus starting with something they had to work with to touch the back-end, and then working their way towards the end user... Because an API that's designed for end users will look very clearly like they designed this for me to use it.
+
+\[28:22\] And then there's other cases where you get an API where it's really confusing, or everything's just scattered all over... I'm trying to think of an example, but I can't off the top of my head. But I know I've definitely seen even just Web APIs, where you can tell the Web API is based off of the data models that they're storing things in, rather than what end users are actually gonna wanna use, and as a result, it just ends up looking not very user-friendly, because they didn't think "What is a user actually going to want here, versus what's easiest for us to give to them?"
+
+**Andy Williams:** Absolutely, yeah. And sometimes you want a design that's  close to the hardware, or close to the data model, and there are projects out there that would use that design on purpose... But I think really for our project we're trying to make this as easy to use as possible, even for a first-time graphical app developer; so it needs to be built with them in mind, with as few lines of code as possible, very clear intent in every single line of code, so that nothing is excess, nothing is confusing.
+
+But there are different approaches for a lot of graphic libraries that expose exactly how  a graphics pipeline works, and you're feeding instructions into that pipeline. If you were building a game engine, that would be really important for you. But we're pretty sure that really enabling folks to quickly build applications that are user-friendly is what's most significantly lacking at the moment, and especially cross-platform. There's just not really anything there if you don't want to worry about the fun of web technologies inside your seemingly native applications.
+
+**Jon Calhoun:** Yeah, that makes sense. And I guess one way I view it is -- you mentioned game engines, and I think there are some people who wanna definitely access the low-level stuff, but for the most part, at least when I'm watching people make games, it seems like a lot of them want to use engines of some sort; they wanna use Unreal Engine, or some engine of some sort that abstracts some of that away for them, so that they can kind of speak in a not quite system-level language; they can actually start talking about things in a way that makes more sense to them.
+
+So I get what you mean, there's definitely some people who need that low-level, but I think there's a lot more people that tend to talk at that higher-level, a little bit easier to understand language.
+
+**Andy Williams:** Yeah, absolutely. I mean, every API is still built with a design in mind, and I think you're right, it's important to know who you're designing for when you set out, because changing that after the fact is gonna be really difficult, and probably make a rather confusing product at the end of the day.
+
+**Jon Calhoun:** Okay, so we talked a little bit about gaming stuff... Do you think a lot of this is why the gaming space is just not that common in Go as well? Is it basically the same type of challenges getting that stuff rendering, or do you think there's different challenges there?
+
+**Andy Williams:** Well, there's a lot of projects out there that can actually get something rendering to the screen; I'm not entirely sure that that's a challenge now. Maybe people are still cautious because it's new or not strictly part of the language... If you'll excuse the phrase, but I've heard it a fair amount. I think that it's more around putting together the libraries and the support, the type of stuff that you described. "I don't want to write a game engine, I just want to pull together a game", and that is another huge amount of work. I wouldn't be surprised if sooner or later we do see something emerge that is a pretty compelling approach. Maybe close to a cool typed API that would look really great for Go.
+
+I just imagine that maybe it's gonna take a while before folks are happy to put the effort into building that sort of an engine binding to an existing engine in Go. It wouldn't really make any sense in my mind -- the differences between the languages in this space are so vast that I think you could spend a lot of time and realize that actually it's just too hard to read.
+
+**Jon Calhoun:** \[32:16\] Yeah. I think one of the things that might also help there is that there's a lot of tools out there that are also very good at cross-platform game stuff. Steam has done a lot of work to make sure a lot of their games run on every OS, and as a result, you can kind of use their tools... So we're not really lacking. Whereas, like we talked about earlier, the rendering of graphical user interfaces across  systems - at least every approach I've seen so far has had issues of some sort that people typically run into.
+
+Basically, the really big UI that would start to lag if you had too much data or too much of anything in there was a big one that I saw a lot of companies complaining about. Because if you're doing an Airbnb and you have a  bunch of listings, it would start to not render correctly when you're swiping up, or it just wouldn't feel natural. So Go is  a good fit for that. But if you've got games that are kind of working, I get that maybe it's not as big of an issue to get something in that space when there's already somebody who's solved that problem in another language, another way.
+
+**Andy Williams:** Yeah, I guess that's a good point. I'm not entirely sure how solved it is. Just from my experience with the games Christmas sale, I saw this huge list of great bargains and thought "Oh, fantastic! I'll get a load of them." And then of course, only one in ten actually worked on the platform I was running on. So it may be technically solved, but there's clearly challenges there that mean it's not absolutely consistent.
+
+This is a real problem actually with any API or platform that tries to do cross-platform; if there's really any roadblock to just rolling out across all of them, then you're gonna find people drop off really quite substantially. And if we're seeing new games getting released only for one platform out of three or four on a major system like steam, then I have to think -- I don't know the APIs, but I have to feel that there's probably something in there that makes it a real challenge. And if you're gonna claim cross-platform, it really needs to truly be cross-platform, without having to jump through hoops or do special things to make your code work.
+
+**Jon Calhoun:** So given that that's the case, do  you think that anybody who's making a GUI library in Go needs to actually keep track of what languages things are being released in? If somebody's using Fyne and you find that a lot of your users are only actually releasing to MacOS, they're not actually releasing to Windows, Linux, Mac, or maybe they're choosing two of them - do you keep track of things like that, or try to keep an eye on that to sort of see if that's a sign that you're not doing as well as you could be in one operating system?
+
+**Andy Williams:** That is probably something that we should do, actually, to have a better idea about where this is working for people and where it's not working so well. I imagined that we would get a lot of feedback from the community, which -- we have an awesome community, very supportive, very active, but the challenge here is that actually we're finding more people than I expected are using this in their workplace, to add a user interface to something that was maybe  a command line tool, or an ugly web form that they wanted to get rid of. So these types of projects are not really going out in the open, to be able to give us that visibility... So it's kind of lacking there a little bit.
+
+We're trying to figure this out by publicizing the applications that are available. Anything that is open source, we're putting together a list of, and recommending that people check it out. But when it comes to "What are you distributing for?", we're really trying to emphasize the cross-platform approach and say "Look, just run the build for different target platforms." That's the only step. It doesn't make sense with the way that we're designing it to target only one platform. It may be that you only want to perform one upload or submit it to one store, but if the links are available on a website or on an open source repository, it's there for the taking; there's no additional complications to just rebuild it for the additional platforms... In theory.
+
+**Johnny Boursiquot:** \[36:08\] I'm curious about the design process, and the thinking that goes into deciding basically which direction the project should go in terms of this API footprint, what should it do, how should it do it, when should feature X, Y and Z come... I mean, you're at version 1.2.1 right now... Along those lines, are there design decisions that you made up to this point that you wish you could take back, or are there some things that you've found out later on that basically you think could have been caught during the design phase?
+
+I'm asking because I don't know this world, and I don't know how different it is from the traditional software-engineering, practice and design, and all of the thinking and best practices that goes along with that.
+
+**Andy Williams:** Sure, yeah. There's absolutely things that I would change with a year-and-a-half of hindsight, but not as much as maybe I'd imagined we would. We're very careful about our design process, both from an interface design point of view, but also from an API design point of view. And the design of the API came before really a single line of implementation. The project started with an ambition to rethink how you could build graphical user interfaces across platform, and then came a broad strokes design outline as to how that API might function. Then we started implementing, initially with this other back-end component; so we didn't have to write the graphics drivers, but then eventually coached it all the way down.
+
+So it is a very considerate approach... This is the way that I would consider any software engineering project if I was in a workplace. So when I started to think about this open source project, I wanted to make sure that we didn't compromise at all in that way. Obviously, there's other ways to build projects - get the code running, share it with some people and start building from there, but I felt that without a real design backing to this, it would struggle to keep its consistency over time. So when people ask for new features to be added, not only do we think "How would that look as an API?" and maybe even ask them how they'd like to interact with it, but we also have to consider "Well, is this something that makes sense for the majority of our users, and is it something that can make sense across the different target platforms that we want to support?" So that we're not just dropping in a small feature for one platform, that then doesn't do what you'd expect on other systems.
+
+That process - I think it has served us very well. It does mean that sometimes features take a long time to develop... And you know, others are dropping quickly, but not always. So we have a roadmap that I've put together probably initially, two years ago, and it evolves all the time; first of all, we wanted to get desktop apps working, and then we thought "Okay, that's solid. Let's add some new widgets to it." Then it came time to look at mobile, so we targeted that for the 1.2 release last year.
+
+Actually, as part of that release, we wanted to get data binding in there as well, because it's really simple to build a simple application, but then if you want to back a big data model into it, or connect more complex systems, display lots of items - even though that's pretty slick, there's still a lot of code to be written, and we thought "If we're really gonna do this properly, we need a good data binding system." And we started designing it.
+
+There was a lot of discussion, there was a bit of experimentation to see what could work, and it came close to release time and the mobile stuff was polishing out quite nicely. We had to say "Look, actually, if we put this in right now, we don't think that we could commit to this being the API going forward forever." So we took that out of the release and invented a new 1.3.
+
+\[40:04\] We were initially going to go directly to a big 2.0 drumroll, but we thought, "No. Actually, to do this properly, we need to take more time. We need to engage with more external developers, so we're not just building as a development team what we think is right, but actually what makes sense for everybody else."
+
+I'm sure that the guys who are working on that could think it's been a lengthy process, because we've been building that API now for over three months... That's quite a long time in any engineer's lifetime, I suppose. We're confident that we're gonna get it right, and actually the demos that are coming together now - they're blowing me away, actually, what a bit of time and consideration has created. It's really cool. So we're gonna continue to think really hard about all of these design elements.
+
+The items that we would change if we could - they're very small details in the grand scheme of things, and maybe when 2.0 comes along, we can change some deprecated stuff and do a walkthrough about how people might update their code, but it shouldn't be a big deal... Long answer.
+
+**Johnny Boursiquot:** No, it's great insight into the process. Yeah, it sounds like there's just as much agonizing over what the API should look like, what the developer experience should be that goes into any great API. So along those lines, you've written a book, Hands on GUI Application Development in Go, and I'm curious basically what the process, having worked with a lot of different GUI toolkits, from GTK to Qt, to any number of the ones you talk about in your book - I'm curious what makes Go uniquely suited for Fyne, for the current project you're working on? What makes Go a good fit, or even what are some areas where it struggles compared to what you're familiar with in other frameworks, in other languages? Where does Go shine in this project?
+
+**Andy Williams:** Yeah, wow... Thinking back to when we picked this as a language - I think part of it was how well-thought-out the language is, and the documentation that goes with it. If you're going to learn a new language and you just want to learn a new language, Go would be a fantastic option. It's so well put together. The documentation is there, the community support is there, and also the broader open source community and the way that all of this complex functionality is readily available for any developer, without having to know additional tooling. That was very compelling from a language design point of view.
+
+But really, the process of writing the book, thinking about the challenges that have existed in graphical user interfaces through the ages, and looking at the challenges that the existing toolkits push on to their end users, I just couldn't help but notice that concurrency, memory management and building across multiple platforms - they were just like the three standout issues that really there was an opportunity to solve. And if you put a bullet list together like that and you say "Which language is this a good fit for?", you don't have a very long list, and Go was really clearly at the top of it.
+
+**Break:** \[43:17\]
+
+**Jon Calhoun:** JavaScript really shines at some of these aspects. The async model that JavaScript tends to be, and the fact that it can react to events makes it great for graphical stuff... But I think that there's other areas where it doesn't quite shine the same way as Go does, which could make it a little bit different... And there's also the fact that people have done this in JavaScript enough times that clearly -- I don't think that would have necessarily solved the problem to just do another JavaScript library to do it.
+
+**Andy Williams:** Absolutely. Honestly, when it came to pulling together the content for the book, it was commissioned, and one of the key items in what the book should contain was that actually, as well as being easy to build, it was important that these applications would be performant, as well as being easily built, that they would run really well... So for that reason, we thought "Well, actually, let's not go with embedded browser engines and a JavaScript stack." And that's not to say that it can't be done with JavaScript, but if you were looking to compare a lot of technologies, I don't imagine there would be a huge number that really thought that performance was really top of the list, if that was the technology stack you were using. So I took the opportunity to say "Well, in that case, let's not look at web technologies, let's just look at the way that people are doing this natively." I think it felt like a much more clean story about the history and future of graphical user interfaces... Of course, with Go.
+
+**Jon Calhoun:** So if we're looking at different ways to build GUIs in Go, I guess first off, can we talk about just what options are there? You created Fyne, you mentioned andlabs UI, which is the one that compiles down to something that looks native to the OS... What other ones are out there, and can you talk a little bit about how they're going about rendering the UI? Does that make sense?
+
+**Andy Williams:** Yeah, I can try. Honestly, the list is too long; I'm going to miss people out if I try to pick through them. I hope this list doesn't show any of my particular prejudices, but it's somewhat inevitable. See, I mentioned andlabs; there are absolutely great projects that will bind to existing technologies as well. There's a couple of different ones for GDK and for Qt, so if that's something that you would like to play with, there's some really great stuff there. You might need a bigger amount of free space on your hard drive to be able to set those up, but they are established projects, with really substantial APIs.
+
+If we're looking at -- things are a little bit different, I guess... A few years ago there was a couple of famous projects: Gx-ui and the Shiny project. Gx-ui, actually, I don't know a huge amount about, because that kind of went dormant a number of years ago... And although it was interesting maybe when I started researching the book, it seemed like it wasn't really going to pick up... So I looked at Shiny instead. This was a project that was put together by some people on the Go team, although I honestly can't remember their names right now... Apologies.
+
+\[48:19\] This was, I guess, a really powerful technology demo of what could be done with Go. They implemented in very similar ways the OpenGL drivers in an abstraction on how you might paint using Go primitives... And in fact, that is used (I believe) for certain runtime configurations in the Go Mobile project to this day. It's a really solid project, it just doesn't really have much widget toolkit built on top of it. It sort of stopped where it was.
+
+Then there is the [unintelligible 00:48:48.01] It's actually kind of contemporary with Fyne; it's been going for a couple of years or thereabout. But the approach that that project has taken is - although rendering a similar low-level API behind the scenes, they're using an immediate mode API, as opposed to the retain mode that Fyne has put together.
+
+Basically, what that means is that each time the user interface wants to render a refresh, then the developer's code is going to  describe how the system should look at that point in time. That's really powerful for games development embedded systems... And actually, it's a  manner of putting together a graphics API that's gaining popularity in many areas. But the approach that I wanted to take with Fyne was to say "Actually, this is gonna be minimal code for the end developer. We're gonna make a lot of assumptions on their behalf", and in that regard, Fyne is  a very opinionated toolkit. It looks a certain way, it behaves a certain way, and if you like it - brilliant; if no, it's maybe not gonna be for you... Whereas a toolkit that's got more flexibility could be very tempting to folks who wanna control every single aspect of how their application is going to work. I am sure that I have missed a couple off that list...
+
+There's also bindings to other systems... The Nuclear project is quite interesting, and that contains Go bindings that are pretty easy to use. Then you've got Wails and Walk for Windows-specific APIs. There are other platforms out there for solving particular problems, but I've really only looked into the ones that we're aiming to make cross-platform graphical apps.
+
+I hope I've not missed anybody out there that thinks they're really happening in the space right now...
+
+**Jon Calhoun:** I think everybody understands that it's hard to list everybody... So I've only looked at Wails very briefly, but does it use Vue.js, or does it use something -- I wasn't sure what it was using to actually render stuff... Because it seemed like Fyne had this model of -- like you said, we kind of have an idea of what each component looks like, and you kind of use our predefined component type design... Andlabs was, like you said, trying to get everything native-looking... And it looked like Wails was more of like a wrapper around a JavaScript-type view, that you had a little more customization around, but it wasn't necessarily native... Is that correct?
+
+**Andy Williams:** That's my understanding. Actually, that's probably about the limit of my understanding of the project as well, because it falls in the category of hybrid or using web technologies. I haven't explored it anywhere near as much as the other ones that I spoke about there.
+
+**Jon Calhoun:** Yeah. I know that this is a space that it would be fun to go build projects... Try to build a small project in each one of these... Because even like you'd said, some of these are a little bit more involved, and I've never dove into those... You said -- is it Q T? I don't know how you pronounce that one.
+
+**Andy Williams:** Well, I think it's pronounced Qt.
+
+**Jon Calhoun:** Qt? Okay. I was gonna say, I don't even know how to pronounce it, that's how limited my exposure is there... So I'm just like "Alright, I see people doing this, but I've just never gotten there."
+
+**Andy Williams:** \[52:13\] People tend to pick a camp, one that works for them for the reasons that they need at that time, and they just get passionate about it because it's solving the problems  that they want. I think that people who are picking up graphical app development now are looking to solve different problems.
+
+One of the challenges I realized when exploring this is that very few of them have any web services integration; they don't really help you with persisting state between user sessions, or even different devices... And I'm thinking if we can solve all of those things, you don't have to be on the web to take advantage of modern cloud-based technologies... So yeah, the right tool for building something is not necessarily the same tool as you want for communicating through the back-end systems.
+
+So yeah, native apps I think have got an opportunity to perhaps gain a little bit more popularity if we really could make the cross-platform work well.
+
+**Jon Calhoun:** I also liked -- you had mentioned that a  lot of people are using Fyne to do things like take something that's command line and make it a little bit more user-accessible... Which I find useful, because I find myself doing random, one-off tasks for my wife... My wife's a photographer, and there have been times where she's imported all the photos twice in a folder, so she's got a bunch that have the " 2" at the end... And as a programmer, you're like "It takes no time at all to write something that just goes and deletes those all." She doesn't wanna go through it manually and delete them all; that's just really tedious for thousands of photos... So I see that and I'm like "Alright, this would be cool, to build a UI for her to work with, that she's not gonna mess things up. She can choose the right folder and do it."
+
+I think there's a lot of small problems that could be solved that way. So if somebody wants to get into this, building something with a graphical user interface, do you have suggestions for them? Where would you suggest they start, what types of projects should they keep an eye out for?
+
+**Johnny Boursiquot:** Before you answer, Andy... Jon, there is nothing wrong with a command line interface.
+
+**Jon Calhoun:** No, there's nothing wrong with it. We're using it. But if I'm trying to show my wife how to use it, there's a problem...
+
+**Johnny Boursiquot:** [laughs] It sounds like you need to teach your wife how to use the command line, my friend.
+
+**Jon Calhoun:** What happens if she says "Honey, please come do this for me again", so I just do it every time?
+
+**Johnny Boursiquot:** [laughs]
+
+**Andy Williams:** I think how you get started depends on what you want to do. There's a lot to explore, and somebody who's curious should definitely check out the [Awesome List](https://awesome-go.com/). Go to the GUI section in Awesome Go and see what's there, and just try them all out, really.
+
+Personally, I would think if somebody wants to get up and running with something graphical in Go, then they should absolutely head to the Fyne.io homepage and have a little read. We've put together a tour that kind of follows the format that the Go Tour used, because it's just so easy to pick up really quickly... So we step people through what it means to put together a graphical application - how applications are linked to Windows, and how content is handled, and callbacks, that kind of thing.
+
+So I'm gonna absolutely just say that that's the way you should go... But given the graphical nature of it, I would also say head to YouTube, see what you can find. There's such a good selection of demos out there... And if you put a couple of key search terms in there, you're gonna see a huge variety of different things... And the code tutorials often come with them. That's sort of my biased angle on it...
+
+Also, the GopherCon talks - I know this has come up a couple of times at GoLab, and hopefully you can discover more upcoming conferences as well... Although as an unusual use for Go, it's not something that gets a lot of air time on the bigger conferences, because it's not what people go there for.
+
+\[56:04\] Yeah, so a bit of Go code up and running in your favorite editor and start typing out some keywords, and it's gonna discover these for you and suggest how you pull something together. A little application is really only gonna take a couple more lines of code than a command line application. Not everybody's really thinking quite so big as the Fyne team, who are currently reinventing the desktop by building the app from scratch in Go as well. There's a very big space for innovative and exciting applications in between those two, and I think it doesn't have to be difficult anymore. This is something that really is so much simpler in a modern, higher-level language.
+
+**Johnny Boursiquot:** Don't sell yourself short, don't sell this project short... At Go conferences currently - and I can speak very confidently, at least for GopherCon - there are way more folks that are new to Go coming into the community than there are experienced developers. That's currently the state of affairs. We have way more newbies than any number of experienced developers... So a lot of folks are gonna be coming to the language and to the community through different avenues. The bigger the population, the more variation you're gonna have, and basically for people to take up the language.
+
+So if somebody who's familiar with other GUI frameworks and they want to also learn Go, maybe Fyne is the gateway; that's the gateway by which they come into the Go community. So I think you're gonna see a lot more of this type of adoption through non-traditional avenues for new Go beginners.
+
+**Andy Williams:** That's really good to hear, actually. I think you're right, there's so many more people coming into it now that we do need to think how you're bringing this to brand new folk, not to established Go developers... And I think that we've been bearing that in mind, "How do you make this obvious for the first-time developer?", of which a lot do fine and then apologize for the questions, and we say "No, no, we want to know what you're struggling with, so we can make it better."
+
+I think actually one of the unexpected challenges is that sometimes people come across the Fyne project and they want to learn Go and they're familiar with GUIs, and they're confused about why a thing that's really difficult for them isn't even present in the language or the toolkit... And there are a  couple of times where folks say "No, no, you're missing this capability", and actually we have to say "Well, we've designed it slightly differently. That thing that you're familiar with and struggling with - just completely forget it."
+
+So the preconceptions that people can have is actually harder work to help with than somebody coming in and going "Oh man, I completely don't know this. What do I do?", because in that regard you point them at some documentation or a video... But if somebody knows an old system and they want to try and break free of it, that's a lot of built-in learning, and you don't necessarily know where it came from or why they picked up a previous system... So perhaps some reeducation to do, but also perhaps we do need to make something that looks more familiar, or adapt to familiar use cases, if that seems like  the right thing to do.
+
+**Jon Calhoun:** Mat likes to do this segment, where we ask our guests what is their unpopular opinion... And it can be a tech opinion - I think he kind of leans towards that - but it doesn't have to be.
+
+**Jingle:** \[59:32\] to \[59:48\]
+
+**Jon Calhoun:** So Andy, what is your unpopular opinion that you'd like to share?
+
+**Andy Williams:** Well, so I thought about this, and I don't know how unpopular it is... I think that people might agree with it, but often don't... So I thought I would say that to me, a quality engineered approach is more important than the speed of development. This may be completely obvious to some folk, or it may be pretty challenging... And if you're building a project to a deadline, there's obviously gonna be a time pressure, but I'd far more like to be involved in a project where they took their time, thought it through, and built something that was pretty solid... And maybe it was late or maybe it didn't have the functionality that was expected, but it's something to build on.
+
+Looking project to project, it's perhaps a little bit difficult to think which is more important, but I look at the overall ecosystem of applications that we build our lives on now, and I think "Goodness, if actually somebody had taken a little bit longer and thought this through, maybe it would work a little bit better, or maybe it could be easier maintained."
+
+So although the Fyne project is about helping people get up and running with graphical apps really quickly, actually what's more important to us is helping people do it well, build a codebase that is easy to understand three, four, ten years down the road, rather than having to replatform or reinvent things.
+
+I don't know if it's too out there or not, but it's something that I'm surprised that not everybody agrees with, so... I thought I'd just say it.
+
+**Johnny Boursiquot:** I will see your unpopular opinion and raise you that level of quality of a project should match its urgency to get to market. [laughter]
+
+**Andy Williams:** Yeah, I like that. It's a challenging thing sometimes though... It's obviously important to get somewhere fast, but if you get there fast and then have to back-track, or need a whole bunch more time to fix the things that got you there, it's difficult to see that it was worth the effort... [laughs] Anyway...
+
+**Jon Calhoun:** A lot of this comes down to -- we've talked about this a million times, but the size of your team, and everything else... Because if I'm working on a project completely by myself, what I can get away with is drastically different, especially if I'm not open sourcing something... Because it's like, I'm the only person that sees this, and if I need to rip this all out, nobody's gonna give me any grief for it, and I already understand how it all works, so it's not like I have to figure out "What does this hodgepodge of code do?" But I definitely agree that if you're working with teams on projects, especially like with you guys, releasing a graphical user interface, and having other people use it - that is a very different beast to tackle... And when you're doing that, you can't have people using it and then all of a sudden say "All these APIs we had - we're ripping them all out. Good luck. If you wanna migrate to version two, you basically just throw all your code out." That's not a realistic migration strategy.
+
+**Andy Williams:** No, that's painful. There's been a couple of famous situations  with that over the last few years... It's difficult to say that it could have been done better, but you just have to think "Well, maybe a little bit less speed and a lot more thought in designing this could have been beneficial."
+
+[01:03:14.15] It just occurred to me that the thing about team size is really interesting - it may be easy to not think about it so carefully when it's just you, but I find it to be helpful to think or actually do any new project out in the open, and think "Would anybody who saw my code think I was doing a good job?", and imagine that the rest of the community is your code review peers. They're probably never gonna look at it, but I think it helps keep me honest, especially early days.
+
+A project like this one - it had huge ambitions, and I had to start somewhere, and without the community as it started... It was helpful to think "Well, if I was looking at this and had no idea, would it make sense?" So yeah, I just pretended that the rest of the internet was reviewing my code.
+
+**Johnny Boursiquot:** That's a lot of pressure to put on yourself... [laughter]
+
+**Jon Calhoun:** I was gonna say, I think you could write perfect code, and the internet would still be brutal... But maybe I'm too pessimistic, I don't know...
+
+**Johnny Boursiquot:** It's true... [laughs] Somebody, somewhere is always gonna have an issue.
+
+**Jon Calhoun:** Alright, Andrew, thank you for joining us. Everybody, thanks for joining us for Go Time. Hopefully, we'll get to cover the subject again in the future, and we'll cover some other unusual use cases for Go. If you have any ideas or suggestions for weird ways that people are using Go, definitely reach out and get in touch with us. We'd love to hear about them.

WFH_transcript.txt

@@ -0,0 +1,851 @@
+**Mat Ryer:** Hello, and welcome to Go Time! I'm Mat Ryer. Today we're talking about working from home, or remote working. We're gonna have some (hopefully) reassurances for people who are new to working at home; there'll be some tips and tricks for people that do it already, some interesting tidbits that you might be able to apply in your own lives, and hopefully it'll just be a lovely, entertaining chat. That's the goal.
+
+Joining me today, we have -- you can hear him snickering already, it's Mark Bates. Hello, mate!
+
+**Mark Bates:** Hello there, mate. How are you doing?
+
+**Mat Ryer:** Good. How's it going?
+
+**Mark Bates:** You know, everybody's hunkering down over here. The family's doing alright, and trying to stave off cabin fever. How about you, Mat?
+
+**Mat Ryer:** Yeah, similar... We're also joined by Carmen Andoh. Hello, Carmen!
+
+**Carmen Andoh:** Hello!
+
+**Mat Ryer:** Welcome back! How's it going?
+
+**Carmen Andoh:** Thank you. It's going well, I am in the same boat as everyone else. I am in day one of that homeschool life, that remote work life, that remote gym life, and everything else...
+
+**Mat Ryer:** Yeah, we'll definitely like to hear more about that on today's episode... And we're also joined by -- it's only Jon Calhoun. Hello, Jon.
+
+**Jon Calhoun:** Hey, Mat. How are you?
+
+**Mat Ryer:** I'm good, sir. And yourself.
+
+**Jon Calhoun:** Good. So far it's work as normal for me, it seems like...
+
+**Mat Ryer:** Because you already work at home.
+
+**Jon Calhoun:** Already working at home, and I'm in a small town, so it's pretty easy to get out and exercise and that sort of stuff without running into people.
+
+**Mat Ryer:** That's interesting... I think what we're gonna find is we're all kind of in slightly different situations, and in some cases very different, probably... But certainly, of all the people that are now working at home, our job, I feel like - we ought to be able to do this quite well. How do you feel about that? I'll open that out to everyone.
+
+**Mark Bates:** \[04:01\] I think as an industry we tend to do better at this. We've been doing this a lot longer, I think, for the majority of us... But I think having everybody else at home now makes a handful of other challenges. We can definitely talk about that later, I think...
+
+**Mat Ryer:** So you've gone from being a home worker who's at your home office, on your own all day, to now lots of other people being around. That's the change for you, right?
+
+**Mark Bates:** Yeah.
+
+**Mat Ryer:** That's interesting...
+
+**Mark Bates:** For me, I had a quiet, empty house all day, I had my schedule, I did lunch on my time, I took breaks that were convenient to me, and now it's everything from -- you know, I've gotta make sure to feed the kids, to I can't turn my music up, because that might upset my wife because she's on a conference call... You know, lots going on now. It's a very different world at home... For me anyway.
+
+**Mat Ryer:** And you're not a natural sharer, are you?
+
+**Mark Bates:** No, I'm not. I'm not very giving... \[laughs\]
+
+**Mat Ryer:** Carmen, what about you? Do you normally go in an office?
+
+**Carmen Andoh:** I go into New York City every third week... So I do two weeks at home, I'm in Upstate New York, and then I try to get into the Go team New York City office for one week, Monday through Friday. So it's like an interesting hybrid, where I'm a commuter and an office goer one week, and then home every day the next two weeks.
+
+But before that, when I was with Travis CI, I was 100% remote for four years, and then at the startup before that, 100% remote... So I've been a remote worker for about 5+ years now, or hybrid.
+
+**Mat Ryer:** Yeah... See, I've been home-working for about 5-6 years.
+
+**Mark Bates:** It's been ten for me.
+
+**Mat Ryer:** Wow. What about you, Jon?
+
+**Jon Calhoun:** I think it's seven or eight for me.
+
+**Mat Ryer:** Okay, so we've been doing it quite a while then, and obviously, we've been successful at doing it...
+
+**Mark Bates:** To varying stages... \[laughter\] You know, I'm not gonna say it was ten years of wonderful working from home lifestyle...
+
+**Mat Ryer:** Right.
+
+**Jon Calhoun:** Oh boy, no... There was a couple of years that are just hard to get everything right.
+
+**Carmen Andoh:** Trial and error...
+
+**Mark Bates:** It's also easy to slip into depressions, but we can talk about that too a little bit later. Let's talk about Jon's thing, too - just the getting into the schedule.
+
+**Mat Ryer:** Yeah... So what were the challenges, Jon?
+
+**Jon Calhoun:** I think some of the things that make it harder just learning -- like, when you're in an office, you have all these things that sort of set the tone of "You're going to work". People talk about getting dressed and not wearing PJs, or they talk about a million other things of routines that set that tone that you're starting to work... And when you start working from home, you just don't realize that you're losing all that. It's really easy to not realize that you're not training your brain to go into work mode, it's easy to let random tasks eat up your time...
+
+One example is you'll be asked to take the kids to the doctor, and then that'll lead to "Well, I need to go to the pharmacy now to get some medicine", and then all of a sudden while you're at the pharmacy you're gonna pick up some groceries. Then after you get the groceries, it's like "Well, I might as well whip something up for dinner, because they can't take these meds on an empty stomach", and before you know it, your entire day is gone, and you're like "What happened?"
+
+**Mark Bates:** Yeah, that's 100% accurate. I know, Mat, you don't have kids so much, so those sorts of daily templates...
+
+**Mat Ryer:** Yes... So much...
+
+**Mark Bates:** Well, yes, that you know of... \[laughter\] I'm assuming Carmen definitely relates to that, because I 100% do... Whenever a kid is sick, I get called. Whenever there's a dentist appointment, a doctor's appointment, an after-school event, orchestra every Tuesday... I haven't been on the podcast this much this school year because it directly conflicts with me having to drive my son to orchestra once a week.
+
+**Mat Ryer:** Right. To work on Kubernetes, or...
+
+**Mark Bates:** Yes, yes...
+
+**Carmen Andoh:** Not orchestrator. Orchestra.
+
+**Mat Ryer:** Oh...!
+
+**Mark Bates:** Yeah, you can only imagine my surprise when I've found out...
+
+**Mat Ryer:** Because of your flexibility, you become the default doer of random tasks.
+
+**Mark Bates:** Yeah, I think we can all agree with that... Jon, Carmen, wouldn't you agree?
+
+**Carmen Andoh:** Yeah.
+
+**Mark Bates:** \[08:05\] And my wife - I'm sure your spouses as well - are very supportive and totally understand... But my wife works downtown Boston, where I always joke she has a "real job". She goes into an office daily, she has a staff... She can't be coming home to pick up a sick kid from school if I'm a five-minute drive away, right?
+
+**Carmen Andoh:** Right.
+
+**Mark Bates:** That's an unrealistic expectation to ask of her... So it's just stuff that you end up having to internalize in your work-from-home lifestyle, and figure out those blocks, and just make it work.
+
+**Jon Calhoun:** It's also hard because other people who don't work from home I feel like just assume that you are free, even though you're not...
+
+**Mark Bates:** \[unintelligible 00:08:46.16\]
+
+**Jon Calhoun:** There have been so many times where people are like "Oh, it's nice because you can do these things..." and I'm like "Well, I still have to work." I'm still working, I'm just at home while I'm doing it. I can't just run around mowing the lawn, and planting a garden, and doing all these things. I still have to get work done.
+
+**Carmen Andoh:** Yeah.
+
+**Mark Bates:** I've been doing this for ten years. My mother called me two weeks ago on a Friday, in the morning, "Can you come over and help dad change out all the plugs and switches in the kitchen today?" No, I can't come over and do electrical work for several hours on a Friday. I've got stuff to do.
+
+**Mat Ryer:** Why do they want to change all of the switches?
+
+**Mark Bates:** Because they're retired and have nothing else to do... But that's getting off the subject entirely. They didn't like the color of the switches. It's a whole thing... \[laughter\]
+
+**Mat Ryer:** That's fair enough.
+
+**Carmen Andoh:** Yeah... Well, now with COVID-19 I don't think that anyone is going to have these assumptions. Since we're all in the same boat and many of us are working from home, we won't be saying "You're not just doing anything" or "Just come on over", so that will help, at least temporarily...
+
+**Mark Bates:** Yeah, for those working from home now... Because there's no school, there's no sports... Somebody asked me "When are you free this week to have a call?" and I was like "Literally, any time between now and mid-April. Just throw something on my calendar. I am so \[unintelligible 00:10:04.18\]" So those I don't think will affect us. But we've still gotta make lunch for kids, we still... You know, we all have stuff to do, it's all gonna be different; we all have to add it into our days. I think that's Jon's point...
+
+**Carmen Andoh:** Yeah. I think one of the things that I'm thinking about for this episode is thinking about working from home, but also thinking about working from home in this new Coronavirus context... Because they're two separate things.
+
+Me working from home while having my kids go full-time to a school, and my husband going to his job, and activities, shuttling different people around - it's very different from everyone sheltering in place, which is as of March 17th, which is the air date of this show, is the reality for many people in California and Europe, and increasingly on the Eastern \[unintelligible 00:10:53.20\]
+
+**Mark Bates:** And Massachusetts has been shut down for a few days now...
+
+**Carmen Andoh:** Oh, okay...
+
+**Jon Calhoun:** Can I ask - Carmen, are you also in the situation where when you work from home you're generally the only person there?
+
+**Carmen Andoh:** Yes. So I get wonderful, deep worktime. Once the kids would go to school, and my husband would go to job, which is around 8:30 - 8:45 in the morning, if I didn't have meetings, I had real silence and an opportunity to do some deep work... Which is essential for productivity.
+
+**Jon Calhoun:** Okay. So I think one of the things that's interesting for me is that my wife and my daughter have pretty much always been around... And I know that for a while, one of the things that we struggled with was sort of social cue -- basically, communicating that I'm working, or when I am working, versus when I'm not.
+
+One simple example of this is if I would walk -- I have an office, so I have a space that I work in, and she knows "Okay, he's in there. He's working." But for the longest time I didn't, and it'd be really hard to communicate "Are you just going to the bathroom, are you just making lunch, or are you able to sit down with me and talk while you're having lunch? Or do you bring it back to your desk, you're still in work mode type thing?" And there were a lot of different things like that, that took a while to sort of get down or figure out some system that worked for the two of us... And I think that a lot of people are gonna struggle with that now that they're both around the house.
+
+\[12:09\] In an office, it's really easy to be like "Okay, he's clearly at work. I don't wanna ask him this simple question." But when he's just in the other room, it's easy to walk in and be like "Can you do this, or can you answer this question?" and that can be really distracting when trying to work... And I think that's something that I've been lucky enough to figure out already, but it's gonna be hard now that everybody else has everybody in the house.
+
+**Mat Ryer:** Yeah, because it's important that -- blocks of uninterrupted time are kind of really vital for productivity. I mean, for me, that's definitely the case. If I've got 30 minutes time, I can't really usually start anything. So 30 minutes isn't enough time to even do anything.
+
+**Mark Bates:** Yeah.
+
+**Mat Ryer:** So when I started writing blog posts, it was to actually fill the little gaps when I found I had little bits of time that I could try and use... But yeah, it's so important having uninterrupted time.
+
+I used to think that working at home was gonna be impossible or difficult because of all the distractions at home... But like you, Jon, I kind of got into a situation where I've figured out a routine that works, and a situation that works... And actually now, being in an office is way more distracting for me.
+
+**Mark Bates:** Oh, yeah. If I go on-site with the client, I can't believe how noisy and distracting it is.
+
+**Mat Ryer:** Yeah... You must get used to it, but it is strange going back into it.
+
+**Jon Calhoun:** You find ways to tune stuff out.
+
+**Carmen Andoh:** Yeah.
+
+**Mark Bates:** Yeah... So to Jon's point, I can tell you some things that we've done here that have helped us. One is we were fortunate enough that we were able to carve out two spaces. I already had my own office in the house, and we had a guest room, so we were able to make my wife her own office, which is super-awesome, because she's on the phone all day. She's on calls all day, and whatever... So she keeps her door shut. When her door is shut, that means "Do not interrupt. I am working."
+
+For me, I generally keep my door open because my dog likes to come and go... And the kids are pretty good and they know if I'm in my office, I'm working. And if the door is shut, that means I'm on a call, don't come in.
+
+Then the other thing we do is we have headphones. Headphones are another cue. We both have earpods, or a variant of them... And if you see one of us walking around the house - usually her, because again, she's usually on calls - if they're in, it's like "Okay, don't bother them. They're working, or they're clearly not in an open-eared environment. They're not ready to listen."
+
+And then texting - we still text, just like we used to... Even though we're one room over, it's still like "Hey, what time are you finishing up tonight? Any thoughts on dinner yet?" Just so we don't interrupt those flows and keep that async, just like she was in Boston and I was at home.
+
+**Jon Calhoun:** It's funny, because we do the same thing. She'll be upstairs, texting me, and people are liek "Really?" and I'm like "This is the easiest way to communicate asynchronously." It's like I'm in an office, and it works.
+
+**Mark Bates:** Yeah, yeah.
+
+**Carmen Andoh:** When I had children that were smaller, so we didn't have a space that I could carve out of my own - and we can talk a little bit about making sure that you have a psychic space that is just for work, and try your best not to make it on your bed, because there is that psychological benefit...
+
+For me, it was a corner of a shared space. They were little, and what I used was a silly headband; it was for Halloween, like a ladybug headband... And I remember "If mommy is wearing that, you can't--" Because this is tips and tricks for people who have maybe small children in the house.
+
+**Mark Bates:** That's great.
+
+**Carmen Andoh:** \[15:45\] Amazon or other online merchants have a red "Do Not Disturb", and when that's on... Or green. I know some people that have done that for the office space. As they got older, headphones is the clear statement. When my kids come into this room and they see that I have headphones on, they'll always leave me alone... And they only come if it's really like I have to go, or there's an emergency. So that was always negotiated over the last 5-6 years. They're older now and it's better, but...
+
+**Mat Ryer:** Yeah, headphones is a good one. Headphones works in an office, too. That used to be the way you would tell people you're focusing.
+
+**Carmen Andoh:** Right.
+
+**Mat Ryer:** It really worked... And then of course, to get around it, someone invented Slack, and now it doesn't matter if you've got headphones on or not; they can get you. But yeah, headphones and those sorts of clues are good... And I suppose it's just that sort of communication with the people around you. Set up the rules, think about it, talk about it, and agree it. That probably is quite important.
+
+**Carmen Andoh:** Right.
+
+**Jon Calhoun:** I think it's also worth, like Carmen said, picking a space that is your workspace... Because part of it is that psychological telling yourself you're going into work mode... And for me, it was even to the point that -- like, I don't like using the same desk for playing video games as I do for working. It's one of the reasons why I don't run a Windows operating system for anything work-related; it's because if I'm playing games, I'm on Windows, and that kind of transitions my brain... And it's just an easy way of me being mentally aware of what I'm doing.
+
+Even when I was in a space where I couldn't actually set up an office, instead of having something on my head, I set up a little drop cloth curtain that went around my desk. Anything you can do to isolate yourself, or make it clear "This is where I'm closed off at" helps a lot... And it also just sort of helps your brain transition into that. But I think it's also useful to have routines that help you move into that work mode.
+
+We talked about this maybe a little bit before we went on air, but a lot of people are no longer gonna have to get up in the morning, shower and dress and go to work. That's not part of their routine... So I think you're gonna need to come up with similar routines that help you, again, move into that mental transition of getting ready for work.
+
+**Carmen Andoh:** In terms of the space, the spacial ritual of "This is my workspace, this is the time for me to think about work", I remember having a colleague at Travis who they had their dining room table that sat only four, and when they worked, they would sit in one chair. That chair was only for work, and when they ate, they sat in a different chair. That was kind of the mental ritual that they had, to be able to (as Jon said) separate work from non-work when they had to use the very same exact space.
+
+So it's things like that that you should think about... That way you don't get burnt out and you don't start blurring the days, and you have a clear separation of work versus home stuff.
+
+**Mark Bates:** Yeah, don't sit on your sofa with your laptop, or on your bed... Don't sit in your favorite chair, or anything like that. Find a different place. At the other end of the dining room table is another suggestion that I was gonna say.
+
+**Carmen Andoh:** Right.
+
+**Mark Bates:** I've found that just trying to make a point of being at my "desk", wherever that space may be, by somewhere between 9 and 10. I usually try to get my toast, and my coffee, and I'll sit down at my desk somewhere between 9 and 10, depending on what's going on that morning... And sit down and then start going through my emails, and stuff... Even if I hadn't showered yet; at least it helps get me in there. I make it a point, I sit there, I read those emails... I don't do it on my phone, I don't respond to Slack on my phone; I do it all at my desk. I mean, I might do later in the day, but that initial step is just "Get into work, Mark."
+
+Then those emails trigger the work thoughts, and then the rest of my day is work -- like, I'm in work then.
+
+**Mat Ryer:** Yeah... So I read something that was talking about one of the things you lose when you work from home is the commute... Normally, your day is kind of bookended by getting to the office, and then getting home again... Which often can involve some kind of - even if it's just sort of like a walking exercise, or something... You know, you do get some different kind of -- it just feels different, doesn't it? And it bookends... And you can lose that. So I think that's quite important, even if they seem -- because that sounds quite funny, Carmen, when you say they'd just sit in a different chair. That sounds quite funny when you think about that. But if that works, then absolutely.
+
+\[20:11\] One thing I have to say, when we talk about finding the space as well, there are lots of people that - especially London, New York, San Francisco and other cities where you just don't have that at all... I mean, you're saying dining room table -- I just got a dining room table.
+
+**Mark Bates:** You have a lovely flat. You have a nice dining room table AND an office. You're doing fine. \[laughs\]
+
+**Mat Ryer:** Yeah, most people though... Yeah.
+
+**Mark Bates:** No, it's true.
+
+**Carmen Andoh:** For people in urban spaces, who are living in a studio...
+
+**Mark Bates:** New York City...
+
+**Carmen Andoh:** Yeah, I have colleagues who are trying to maximize that space in a studio.
+
+**Mark Bates:** I feel really bad for people -- and in San Francisco, too. People are living in these small little places, and they're stuck in there now... It's gotta be hard if you can't find that space, but just do whatever you can. Anything you can. Even if you just take the chair and move it to the other side of the room for the day and slide it back again... Just do something. Change your angle. Face it to the window. Just do something to change it.
+
+When the weather gets a bit nicer, if you can open the window, even better. If you have a balcony or something you can sit out on - that's great. The weather is still a bit -- meh...
+
+**Jon Calhoun:** I think some of the things that can help too are just making sure that you set those boundaries and stick to them both ways... Because one of the things that always killed me was I could get into work mode, but then I wouldn't switch out of work mode. It's really easy to justify too, because you'd be sitting on the couch and you'd be like "Oh, it's a work email. Let me check it real quick." And then you're like "Oh, it'll take five minutes to respond to", and again, that leads to all those different things... But it's really not fair to ask your family to leave you alone when you're working, and then whenever you're supposed to be spending family time, you're like "Okay, now I can just drop everything and go do work, instead of hanging out with you guys."
+
+**Carmen Andoh:** Yeah...
+
+**Jon Calhoun:** So going both ways is a big part of it too, because otherwise you'll just work forever.
+
+**Mark Bates:** Yeah, it's important that just as you start your day, to end it. That could be closing down your computer, walking away and not touching it till tomorrow, it could be signing off of Slack and then just going away, but... Yeah, you've gotta end your day.
+
+**Carmen Andoh:** Yeah.
+
+**Mat Ryer:** A friend of mine has two accounts on their laptop... So they sign into their work account and have all the work apps and everything set up there, and then they have a different account for the personal stuff. Something like that is quite nice, little virtual boundaries.
+
+**Mark Bates:** Yeah, you can do that on one computer too, which is quite nice.
+
+**Mat Ryer:** Yeah.
+
+**Carmen Andoh:** One thing I'll say from experience, as we all adjust to staying at home indefinitely for the coming weeks, is that the end of the day - stick to that boundary, because you may have only gotten four good hours of work, but then you had kid interruptions, or you had XYZ... And if you only had four hours, and you resolved to end at (say) 6 PM, just do it; be nice to yourself. If you're a team lead, or a manager, or working, just know that we are going to probably be at a reduced capacity for some time now... Because there are problems when you don't set that boundary and you say "Well, I only got four hours done. It's six o'clock... I'll make some up now from 6 to 10 PM." But that is not good self-care, and that's not setting good boundaries.
+
+Things happen, and we just need to be okay with the fact that we have other responsibilities, we're living with other people, society is shutting down, and that's okay for the short-term.
+
+**Break:** \[23:43\]
+
+**Mark Bates:** One of the things that when I first started it took me ages to figure it out, and then I started talking to people more and more about, is when you go to an office you don't get eight hours of uninterrupted work. You don't. You get four, maybe. That's on a good meeting day... No, I'm serious. Between people coming to your desk in interruption, you go to get a cup of coffee and you're there for ten minutes, somebody in the hall, that hallway meeting... You'd be surprised how when you add those up over the course of the day, those add up to hours in a day.
+
+So if you're at home, don't beat yourself up because you decided you're gonna take the dog for a walk in the afternoon for 20 minutes to stretch your leg. You would have done that to go to the coffee machine, and talked to Joe in accounting about his stupid March Madness that isn't happening, and you're like "I don't care, Joe. I just want a cup of coffee", and you're there for 20 minutes... \[laughter\]
+
+Those things are there, and especially now, where we do have to make time for family, they're no different than the time we'd have to make for our co-workers, socially, in the office, and the other things that just kind of come along with being in the office.
+
+So I don't worry anymore about taking breaks, I don't worry anymore about making bread in the afternoon, or something like that. I'm thinking, I'm doing work in my head, I'm just away from my desk, and that's okay. I try to do those things when I need that shift of thinking \[unintelligible 00:26:27.16\] I'll make the bread, I'll take the dog for the walk, I'll go to the grocery store, whatever. So I try to work those tasks into those spaces, too.
+
+**Mat Ryer:** Yeah, that's a really good point. I think it's a very naive view, and people think that you have to just be working solid amounts; like, you have to work eight hours a day, solid, uninterrupted. I've worked in situations where it's office-based, and sometimes the people around you don't necessarily understand the job in the same way, and they might not be as technical, or just have a different perspective completely...
+
+I've had some experiences where they really value the amount of time you sat looking at the screen. And if you measure that as a way of deciding how productive you're being or how good an employee somebody is, it's really a mistake, I think. Focusing on actually what gets delivered - that's the important thing. That shift - and especially if you work in a trusted team - can make all the difference.
+
+So things like what time you got started, or how long you spent on your little break, or whatever - that becomes kind of background noise really, because it isn't that important.
+
+**Jon Calhoun:** I think there's also probably a correlation between -- like, we've all worked in a place where people check things like Reddit, or do other random things as little breaks... But I think the amount of time you spend doing stuff like that changes drastically, depending on whether you're in a job where they count the amount of time you're in front of a screen, versus just be as productive as you can be.
+
+\[28:00\] Because when I sit down, if I have three hours just to get something done, I don't check Reddit or do all these other things; I sit down and I do what I need to do. But if I'm gonna be in an office for eight hours and I know I'm stuck there for eight hours, then it's like "Okay, I can just throw this in, check this thing, check Twitter", do whatever. It's a lot easier just to slip all that stuff in and allow yourself to get distracted.
+
+**Mat Ryer:** Yeah. So speaking about the commute then... Carmen, you do both - you go to an office every third week, and you also work at home, so you can compare really those two things... What do you do with your commute? How long is your commute? When you work from home, do you think of ring-fencing that commute time to put to some dedicated use? Or is it just blending into the rest of your day?
+
+**Carmen Andoh:** It's for exercise. My commute is kind of considered my exercise during my New York weeks. Depending on the subway schedule, it could be - if I catch all the trains right - about 35 minutes; if I don't, it'll be closer to an hour, or if I'm off-peak... So when I'm home, I use that time to schedule -- it's just a completely different schedule. I have a different work schedule when I'm home versus when I'm in the office.
+
+Someone in the channel asked if anyone plans to offset their work schedule to accommodate for the family being home, and my answer is 100% yes. I love to get up very, very early, and start my deep work at about five...
+
+**Mark Bates:** In the morning?
+
+**Carmen Andoh:** In the morning... Because I'm already an early bird.
+
+**Mark Bates:** Which timezone at that?
+
+**Carmen Andoh:** Right... Well, this started when my team was based in Berlin, and I was the only one in the New York timezone, and I agreed that in order to collaborate more we needed to have more chunks of hours when we all were together... And I already really liked -- so nobody in my house likes to wake up early; I'm the only one, and I already do it to get some space.
+
+This doesn't work for everyone, if you're not an early bird... But I would work from five, and then end my workday at one. That was perfect in my old job.
+
+I also manage expectations with my team now, and say "Listen, my hours are gonna be a bit blotchy", but I want some of that deep work. Now that none of the kids are going to school, they don't wanna wake up till eight... So for me, five to eight is great, deep worktime; I just hit it out of the park. I don't check emails, I don't go to social media... Whatever I had set up as my big rock thing to do from the day previous gets done during that time, and it really sets the tone for the rest of my day.
+
+So yeah, to answer, Mat, that is exactly how I ring around that... And I just have internalized very differently what a workday looks like when I'm here at the house, versus what a workday looks like when I'm in the Google offices in New York City.
+
+**Jon Calhoun:** My brother has three children, and he would walk two of them to the bus... He did something very similar to what you did, where he would get up every morning at 5 or so, he would get basically one big thing he wanted to get done for that day, he'd start working on it then. Then as soon as the kids were getting ready for school, he'd walk them down to the bus and do that... Which would interrupt his day, but -- the way he communicated it to me was basically that it allowed him to make sure that he got the most important thing done for the day, and then the rest of the day, if there were distractions, it didn't matter quite as much.
+
+**Carmen Andoh:** Right.
+
+**Mark Bates:** My wife would take and drop one of our kids off, and then go into Boston... Then in the afternoon we'd go take a dog for a huge walk, like an hour-long walk, even though school is like six minutes away... But just get a nice, big walk in in the afternoon. That was nicely scheduled around picking him up. Those things are gone now, for a lot of people.
+
+For me, I'm still trying to keep that timeframe, if I can, and still keep trying to take him out in the afternoon... Although I've found that now that my wife's home, we're trying to see if we can coordinate a little bit more time in the afternoon for her and I to maybe go for a walk. You know, just a little break in the afternoon, which is quite nice. So if you have a spouse, or a partner, or somebody like that, you can schedule a nice walk in the afternoon, take a break, with kids, whatever...
+
+**Mat Ryer:** \[32:07\] Hm, lovely.
+
+**Mark Bates:** ...walk the dog... Those are all good things. I know she's turned her morning commute into running. She runs half-marathons, so she gets up ever day, and instead of getting up at 5:30 in the morning to go running, she can get up at 6:30 in the morning and go running instead. That's her big like "This is wonderful! I can get a nice, big run in every morning."
+
+**Mat Ryer:** Mark, let me ask you this, mate...
+
+**Mark Bates:** Yeah.
+
+**Mat Ryer:** What do you wear for bed? \[laughter\]
+
+**Carmen Andoh:** Well, you're gonna have to qualify that for why you're asking in the context of a free show...
+
+**Mark Bates:** Yes...! I'm not sure I'm gonna answer that. \[laughter\] I'm sure there's a code of conduct violation just waiting on the other side of that question...
+
+**Mat Ryer:** Just answer the question...
+
+**Mark Bates:** It's entrapment, is what it is...
+
+**Mat Ryer:** \[laughs\] Okay, fine.
+
+**Mark Bates:** It's entrapment.
+
+**Mat Ryer:** Well, one of the bits of advice you hear a lot is that you should have different clothes to work in. Have you heard this? We didn't talk about this already, did we?
+
+**Carmen Andoh:** No. No, we didn't.
+
+**Mark Bates:** By different clothes you mean like I don't have to get up, put on a suit, and work at my desk all day, and then change into jeans in the evening, right?
+
+**Mat Ryer:** Well, that depends. Some people do actually do that.
+
+**Mark Bates:** My brother does that, actually.
+
+**Mat Ryer:** Right. So he presumably doesn't sleep in a suit...
+
+**Mark Bates:** He might.
+
+**Mat Ryer:** Well, then he's not doing that, is he? \[laughs\]
+
+**Mark Bates:** He's very conservative...
+
+**Mat Ryer:** Then he's working in his pajamas, which is what they say you shouldn't do... \[laughter\] And I don't know how I feel about this one. Jon, what do you think?
+
+**Jon Calhoun:** I think it depends on the person. Figure out what works for you... I mean, I said this before we got on air, but -- I don't sleep in pajamas. But I own a bunch of pajamas that when I wake up in the morning, I put them on, because it's cold in my house... And I will go out and I'll make my coffee, I'll go downstairs, I'll get on my computer and I'll answer work emails and I'll do a couple things like that... And I don't shower for the first couple hours of the day, because I tend to go out and exercise in the afternoon. Somewhere around lunchtime is when I like to go out and exercise.
+
+So a lot of people think that's weird, because if I hop on a video call or something, they're like "You clearly woke up and didn't shower..." \[laughter\] And I try to limit those calls to just people that are okay with that.
+
+**Mark Bates:** I got that this morning, as a matter of fact...
+
+**Jon Calhoun:** But to me, that's just part of my routine that works. It's kind of like Mark said, I like to get straight to emails and straight to doing that stuff... I like to do these things, and then eventually when I work out -- well, I don't wanna shower twice in a day; I don't wanna shower in the morning, and then work out, get sweaty, and then not shower, or shower a second time... So it just makes more sense to do this.
+
+Then after I do that, I can put on whatever clothes I want for the rest of the day, and do whatever. I can even link the rest of my day around that, where like if I'm gonna go to the grocery store or do that sort of things, it tends to happen after I've showered and put on some normal clothes, not when I'm sitting around in my pajamas.
+
+But that also allows me to sort of -- it's almost like my pajamas are my deep thinking work clothes, because... You know, I'm clearly not leaving the house at this point, I'm not doing that stuff as much... I say "clearly", but I've definitely left the house in my pajamas and my wife yelled at me, but... \[laughter\] But most of the time.
+
+**Carmen Andoh:** Yeah... Really, it's about that mental preparation, whatever habit that you need to get into. It could be that you wear your blue pajamas to bed, and the green pajamas to work... It's just the point meaning that you're preparing your mind for a work mindset. And it's the same trick as I sit in this spot of the small dining table for work, and this spot to eat meals.
+
+Again, it's boundaries, separating the psychic space of "This is the home, where I try to relax and unwind from work, but also where I now have to work."
+
+**Mark Bates:** The one thing I will say is slippers are 100% allowed.
+
+**Carmen Andoh:** Oh, yes...! I'm so happy.
+
+**Mark Bates:** And highly encouraged.
+
+**Carmen Andoh:** Yeah. I totally love that these are what I wear all the time. I'm showing it...
+
+**Mark Bates:** I've got some UGG slippers that are just -- I got them a couple Christmases ago and they're just amazing.
+
+**Carmen Andoh:** Yeah, cozy house slippers are definitely probably top five perks of working from home. \[laughs\]
+
+**Mark Bates:** Invest in some nice, quality slippers.
+
+**Jon Calhoun:** That will be like all the links for this show, is just referral links for slippers... \[laughter\]
+
+**Carmen Andoh:** "What brand do you use...?" \[laughs\]
+
+**Mark Bates:** You know what - let's actually talk about comfort, because that is important. Office chair.
+
+**Carmen Andoh:** Oh, yeah...
+
+**Mark Bates:** \[36:12\] If you can, if you have the space and you have the ability to get a nice office chair, you absolutely should. I've gone through a couple different chairs now, and I have a Steelcase chair now, which I absolutely love. I got a Herman Miller maybe 6-7 years ago...
+
+**Carmen Andoh:** Ooh, fancy...
+
+**Mark Bates:** It was like a low-end Herman Miller... But it was the first nice chair I had ever actually splurged on, and didn't just go to staples and buy the $100 "executive model." This is a nice chair... And I remember saying to my wife, I was like "Oh, I can't believe how much I spent on that Herman Miller..." and she's like "Mark, you sit in it 40 hours a week. That's a justifiable purchase. Of all the random junk you buy, a nice chair and a good desk are okay things to purchase."
+
+**Mat Ryer:** Yeah, all the Easter Island heads that you bought... \[laughter\]
+
+**Mark Bates:** I have the full collection, by the way. One was just broken. But my heart goes out to them...
+
+**Carmen Andoh:** This is one of those things people are probably asking "Should I make that kind of investments with an unclear amount of time that I'm gonna be at home before returning to the office?" I don't know if that's the answer, but this is gonna be more than 5-6 weeks. Your back will thank you if you have the means and the space to invest in a chair.
+
+**Mark Bates:** Yeah.
+
+**Carmen Andoh:** I use this GT racing chair - I'm showing people in the Zoom chat, but... Gamer chairs are made for this kind of stuff; they're made for people who are in it looking at the screen for a very long time... I got a racing chair for Christmas as a gift, and it was the best.
+
+**Mat Ryer:** You code with one of the steering wheels as well, which I think is amazing.
+
+**Carmen Andoh:** \[laughs\]
+
+**Mark Bates:** It also has multiple cupholders, which is amazing.
+
+**Carmen Andoh:** \[laughs\] Yeah.
+
+**Jon Calhoun:** So I have a Herman Miller at my desk, for work... So I've definitely spent money on a more expensive one, because I use it so much... But upstairs at our dining room table we actually have -- Costco sells an office chair that's like $150 (in that ballpark), and it's not quite as nice as the Herman Miller, but it probably gets you 90% of the way there with ergonomics, and everything... And I would highly suggest, if you're looking for a cheaper option, to go look there. Where all the Herman Millers and those ones can be $600+, that one will be like $150 or something... And it's a decent option to check.
+
+And even as far as space goes - my wife and I keep an office chair at our dining room table. It's just one of the chairs there. And while we both don't like to work at the dining room table all the time, there are times where I need to go up and watch my daughter while my wife goes and does something, and I can sit there and work... And again, that separation is still slightly there, even though I'm slightly in dad mode.
+
+**Carmen Andoh:** Nice tips.
+
+**Mat Ryer:** Yeah, one of the things about being in an office is you tend to get up and walk around a lot more than when you're working at home... So I think that has to be a bit of a conscious thing you do as well, even if you do have a Herman chair. So excuses to kind of get up and walk about I think are also probably quite important.
+
+**Mark Bates:** Get a glass of water every hour.
+
+**Carmen Andoh:** Yeah...
+
+**Jon Calhoun:** Or get a dog.
+
+**Mark Bates:** A dog will definitely help.
+
+**Jon Calhoun:** Dogs are awesome for this.
+
+**Mark Bates:** For the first time I got my very first dog just over a year ago, and it has changed my life. I can't believe I've worked from home all this time without a dog.
+
+**Jon Calhoun:** My dog will literally -- if I'm in my office too long, he'll come over and start nudging me with his nose...
+
+**Carmen Andoh:** Really?!
+
+**Jon Calhoun:** Even if he just has to go to the bathroom or something, it makes me get up. But then on top of that, every day I'm like "I can't skip the walk today, because he needs a walk, too."
+
+**Carmen Andoh:** Yeah, he needs to go outside. That's great.
+
+**Mark Bates:** They're great excuses for walks, for entertainment... I talk to my dog; he's my rubber duck. I'll talk to him about code...
+
+**Carmen Andoh:** That's great...!
+
+**Mat Ryer:** Is that another one of your pets, Mark?
+
+**Carmen Andoh:** \[laughs\]
+
+**Mark Bates:** Yes, yes...
+
+**Mat Ryer:** Take him for his little swim, in the back.
+
+**Carmen Andoh:** So I have this watch, that -- you know, if you can't have pets in your building or your house, if it's not allowable, I just have a watch and I just set it to... If it detects no motion for 60 minutes, it beeps. And I hate it sometimes, especially when you're in deep work.
+
+\[40:13\] And it doesn't have a setting where you can say "Only notify me between these hours of the day"; just any motions... So I sometimes turn it off. But it's a nice way to -- or maybe you use a browser, a Chrome add-on that can say "Get up now." That's also an option to get you up and out and walking around.
+
+**Mark Bates:** If you have an Apple watch, they do that. They ping you, to try to get your stand goals...
+
+**Carmen Andoh:** Oh, yeah...
+
+**Mark Bates:** I know that's one of the things my wife and I -- we now always see who gets their rings first. So the rings on our watches help us make sure we get enough activity, movement and standing during the day. That's really nice, and you can see that, and it encourages you, it pings you during the day...
+
+**Carmen Andoh:** That's nice.
+
+**Mark Bates:** I'm sure Android watches do similar things. If you don't have those things enabled, definitely enable them. They'll ping you, they encourage you, they want you to do things. It's quite nice.
+
+**Carmen Andoh:** That's nice.
+
+**Jon Calhoun:** You can also do the pomodoro technique, and all those things. I know lots of other people use those too, and use them effectively. It's just not what I personally use.
+
+**Mark Bates:** Yeah, I need something a little less manual... And that's one of the things I like about the Apple watch. It's all automatic; it just yells at you, and you're like "Fine, I'll stand up. I'll go get a glass of water now." Like, "I've finished this water, I'll go get another one."
+
+**Jon Calhoun:** Oh, and related to that - don't get a really big glass of water. I used to get this really big mug to bring down, so I wouldn't have to go refill it... And I've basically found that having a smaller glass that forces me to go refill it is useful.
+
+**Mark Bates:** Yeah.
+
+**Mat Ryer:** This is a brilliant episode. Don't have a big glass of water.
+
+**Jon Calhoun:** It sounds crazy, but...
+
+**Mat Ryer:** I mean it. I love it.
+
+**Mark Bates:** But you're supposed to drink a lot of water, generally; it gets you up, and then it also gets you to the bathroom, which is another thing that gets you out of your desk. These things sound silly, but when you've been doing this for a long time, these are tricks that you do learn. Jon is speaking the truth, I know what he is talking about. I switched to a smaller glass years ago too, for the same reason.
+
+**Mat Ryer:** I just can't wait for them to take that clip and put it on the Twitter as the promo for this show, though...
+
+**Mark Bates:** \[laughs\] Drink a glass of water.
+
+**Mat Ryer:** Don't have it too big.
+
+**Jon Calhoun:** People are gonna imagine this with like these tiny, Winn-Dixie plastic cups that you get just to rinse your mouth out, or something...
+
+**Mark Bates:** \[laughs\] The little dentist rinse-and-spit, yeah...
+
+**Mat Ryer:** Well, actually, standing up for calls and walking around while you're on calls, if you can... You know, if you're not demo-ing, or anything like that...
+
+**Mark Bates:** Yeah.
+
+**Mat Ryer:** Little excuses like that to be active I think is important. It's definitely something that I try and do consciously.
+
+**Mark Bates:** Yeah, pace around your room.
+
+**Mat Ryer:** Yeah, I personally find also that's quite a good way to think as well, if you've got a particular problem that you wanna work on in your brain.
+
+**Mark Bates:** I'm gonna stand up right now...
+
+**Jon Calhoun:** So now I know why I pace every time I'm on the phone, and my wife yells at me for it... \[laughter\]
+
+**Mat Ryer:** Yells at you for what? Tiny glasses of water?
+
+**Jon Calhoun:** I walk, or I pace, anytime I'm on the phone... And I think it comes from sitting around, so when I get a chance to do it, I do it... And whenever I'm on my phone, my wife will be like "You can sit down." I'm like, "No, I'm fine. I sit all day." I also have to tell people this all the time when I go places, when they're like "Sit down", and I'm like "You know, I sit a lot through the day. I'm fine standing."
+
+**Mat Ryer:** Yeah. It's rude almost, isn't it? I was at an elderly relative's house, and they were like "Oh, sit down", because it's rude for them to not invite you to sit down... But then it got flipped on its head and just became I was the rude one for not wanting to sit down... So I just sat.
+
+**Mark Bates:** Yeah, but that doesn't surprise anybody...
+
+**Mat Ryer:** What, that I'm the rude one?
+
+**Mark Bates:** Yeah.
+
+**Mat Ryer:** Oh...
+
+**Carmen Andoh:** \[laughs\]
+
+**Jon Calhoun:** Okay...
+
+**Mark Bates:** Well, you did ask me what I wore to bed, so... I think I'm justified in my criticism today, Matthew.
+
+**Carmen Andoh:** You two are great.
+
+**Jon Calhoun:** \[44:00\] So I guess that can lead to another subject here... How do you communicate differently when you're remote? How do you not ask people "What do you wear to bed?" in a weird way? \[laughter\]
+
+**Mat Ryer:** Yeah, don't ask in a weird way. Ask in a good way.
+
+**Mark Bates:** Yeah, you can ask it in a totally appropriate workplace way, absolutely.
+
+**Mat Ryer:** Yeah, yeah.
+
+**Jon Calhoun:** You know, when we see you in the video chat walking through Victoria's Secret, it's probably not an appropriate time to ask.
+
+**Mat Ryer:** \[laughs\]
+
+**Carmen Andoh:** One of the things that's happening is that we are no longer having in-person communication... So on reflecting what makes in-person communication superior to (what would be the next level down?) video chat, then superior to text, is things like being able to see each other and read the room... Right? Look for emotions...
+
+Right now I'm in Zoom with you, and I see Mark nodding his head, Mat nodding his head, Jon nodding his head. We all have a shared understanding because of the facial expressions and body language that you've given me that you've understood what I've said.
+
+Same thing for emotions - we communicate far more than just our words. We communicate our tone, our intent, our values, how we feel about things... And that is what creates bonding and shared understanding in the team.
+
+One of the things that \[unintelligible 00:45:13.27\] in the chat said - "If we go from in-person meetings to chat rooms, people have a tendency to turn off their video and just do audio..." And I think now is the time when we wanna say that's not the best thing if we wanna continue connecting. So put your video on and keep it on.
+
+There are different communication platforms that allow for larger groups, for us to see everyone at a time. Unfortunately, Google Hangouts is not a good one; they have a limit of five, where you can see people in a grid, and then maybe eight in a sidebar. Zoom is a little bit better, but you have to pay for that. So there's just trade-offs to all the different platforms, but if you can, try to see each other's faces, try to look for emotions and body language and shared understanding... And also just to see other people and stay connected in uncertain times.
+
+**Jon Calhoun:** Yeah, we mention this to everybody who comes on Go Time - we use Zoom just so we can see each other when we're recording, and there's a Gallery view, where instead of just saying whoever is currently talking, it shows you everybody in a grid. And that does so much to help with communication... Because if somebody wants to talk, they can raise their hand, or you see that they're trying to talk... There's a million different ways that you can non-verbally communicate that you wanna say something. And you can tell when a guest doesn't have that on, because they won't notice some of this stuff happening.
+
+So it's definitely something to keep in mind and to check out. I think that's what you were talking about, Carmen - Google Hangouts caps it at like five(ish).
+
+**Carmen Andoh:** Yeah... And you know, I work at Google, and I know that this is a thing they're trying to figure out, but... When we think about the coming months, and keeping teams from breaking down in communication, but also connection, I think that more than ever we need more and more video chats, and they need to be intentional. So talking about ground rules...
+
+Also, one of the things that we've set up is a snack room. A room that's open 24/7. And just like you would go to the watercooler down the hall, or whatever (we have cafes and micro-kitchens at Google), this is a thing that's open 24/7. If you feel lonely, if you feel isolated, go to this room.
+
+I will typically go in and there's one person, two people, three people, and the agreement is as long as you're logged into that space, you can social chat. Sometimes some of us mute their video and audio, but they want maybe this background, so that they don't feel so alone.
+
+So that's a thing that you can do, whether you wanna do it free with Hangouts, or if you have paid Zoom, or have another video-chatting platform. That's one step that you can take to continue to build rapport and connection with your team.
+
+**Mark Bates:** And don't skip your daily stand-ups either.
+
+**Jon Calhoun:** Related to that, one of the best remote atmospheres I had was one where it was completely normal to just message somebody and say -- my dog wants me to get up right now...
+
+**Mark Bates:** \[48:10\] What a weird thing to message somebody...
+
+**Carmen Andoh:** Oh, I see his tail wagging... It's so cute! Anyway, continue...
+
+**Jon Calhoun:** Anyway... It's very normal to just message somebody and say "Do you have a minute to jump on a video chat?" This wasn't a "Put it on my calendar"... As long as they could, obviously. And oftentimes these video chats were maybe three minutes. It could be like "Hey, can you look at this thing, cover it?" But it really helps simulate that walking up to somebody's desk and asking a question type thing.
+
+I mean, you wanna try to not interrupt everybody's workday and let them get productive stuff done, but having these quick video chats is complete okay. Video chat doesn't have to be something that's scheduled on a calendar, and takes 15 minutes, and all this other stuff. It can just be a "Look at this real quick with me. Walk through it, make sure it makes sense." Because sometimes that can save you 20 minutes of trying to write up an email explaining what the heck you're trying to think, and everything else. But on top of that, just that communication and just having a little bit of banter or something can help brighten up your whole day
+
+**Carmen Andoh:** Does anyone have a rule of thumb for that? Like, how much time should you put in a thing as you start to write it before you say "You know what - this would probably just be a lot easier on video"?
+
+**Mark Bates:** For me it's more -- like, if I'm in Slack and I'm already going back and forth with somebody about something, or trying to ask them questions (a client, or whatever), I'll just be like "Do you have five minutes so we just jump on a chat right now?" You'd be surprised how quick those really complicated conversations in Slack go to like "Oh, sorry. So we were in violent agreement then? Okay, fantastic", and the call is over. \[laughter\]
+
+Just feel free to -- tools like Slack have video capability already built in, so you can start that conversation then and there.
+
+**Carmen Andoh:** Yeah.
+
+**Jon Calhoun:** This is also a big part of why I think being willing and ready to jump onto an actual video chat rather than just audio is still useful when you're working from home. It's one of those things where I feel like people who work from home more frequently are more willing to jump on video chats and actually have video on... Whereas somebody who does that once or twice a week or something like that is less likely to do it, because it's kind of the exception for then. But when you do it a lot, you kind of realize that being able to jump on a video chat is very useful.
+
+I know sometimes, like I said, I wouldn't shower until the afternoon, but I just kind of got over that, and I'm like "I've worked with people that were okay with it, so it's fine." And I know that I'm also a white guy, so I probably have some privilege there... But hopefully, it's not too bad for everybody.
+
+**Mark Bates:** More importantly, I can't smell you from here, so I really don't care if you've showered or not. \[laughter\]
+
+**Mat Ryer:** It's another advantage to remote work.
+
+**Carmen Andoh:** Yeah, smelling.
+
+**Mat Ryer:** Actually, the flipside of this is -- one of the things that I've seen this used for before is actually like surveillance of the team, and things... And mandating that everyone has to have the video on all day, and things like this... It can get a little bit -- it's not great. And actually, I think if you feel like you have to have that kind of supervision or surveillance on the team, that's when there's probably deeper problems there in that team...
+
+**Mark Bates:** That's what it feels like...
+
+**Mat Ryer:** But the flipside of what you were all talking about is actually exploring and enhancing our asynchronous communication skills as well. In open source, Mark, you probably never ask someone to just jump on a quick call.
+
+**Mark Bates:** Can't say I do... \[laughs\]
+
+**Mat Ryer:** Open source projects tend to be asynchronous... And there's real value if you can do things asynchronously, because suddenly you actually remove the need for you to even be online at the same time. So yeah, it's in some situations that those skills are gonna be good to have... And I think generally that's a good idea anyway, because it falls into that whole not wanting to interrupt people. If you can resolve that thing, wherever it is, asynchronously, that's probably great.
+
+\[52:03\] You can lose out, of course, if you only do that; you lose out then on the social stuff. Having the random channels, or being able to socialize and have a bit of rapport, as you've said, I think is important. But yeah, async communication is kind of a perk, in some ways, of remote work.
+
+**Carmen Andoh:** What are some of the async communication tips that you - I mean, I'm just asking around - have found invaluable over the years?
+
+**Mat Ryer:** Well, I think what happens is you learn little things of etiquette, for starts... I've heard this thing said on Twitter - don't just reach out to someone and say "Oh, hi Mark." Don't just say that...
+
+**Mark Bates:** Hi, Mat.
+
+**Mat Ryer:** ...and then write your -- it's from a film; I'm not gonna bother getting into it.
+
+**Mark Bates:** I'm sorry, the phrase "Hi, Mark" is from a film?
+
+**Mat Ryer:** "Hi, Mark", yeah.
+
+**Mark Bates:** I hear it's from several. \[laughs\]
+
+**Mat Ryer:** Okay, could be... I don't know, I'm not IMDb. Yeah, so... Actually that. It's like your first message is kind of an interruption payload, and you're gonna drop it on somebody. So make a good one; make it contain all the bits you need. If it's a question you're asking, you know, "Oh, hi Mark", and then ask the question. And then press Enter.
+
+Mark's not then on the other end waiting for you to write that first thing. That actually turns out to be very useful. But there's other communication skills... And I think it is a skill of being able to write and communicate through that way. Write something and leave it, and it stands on its own... And usually, that involves preempting questions sometimes, but...
+
+**Mark Bates:** When you're talking with async, first of all, patience is a virtue... Because that response might not come back immediately. It might be hours, for now...
+
+**Mat Ryer:** Well, if it does come back immediately then it's not async, is it?
+
+**Mark Bates:** Well...
+
+**Jon Calhoun:** Could be.
+
+**Mat Ryer:** Could be. Fair enough.
+
+**Mark Bates:** Yeah... Anyhow. I mean, I think that's from a film, Mat...
+
+**Mat Ryer:** That's from loads of films.
+
+**Mark Bates:** Yeah, absolutely.
+
+**Mat Ryer:** And I should know...
+
+**Mark Bates:** The other thing I would say is don't wait till you're blocked to try to reach out to somebody, because that's the worst time for both you and them. Try to get those questions out earlier if you can, to give them time to get back to you before you get a super-block.
+
+I also try to \[unintelligible 00:54:17.03\] I'm like "Okay, I have a question...", whether it's a super-block, or just kind of "I just need some more guidance", or whatever... I'll put it out there and then I'll try to find another task to move on to. So that has now become async to me. I'm waiting for the callback to come back now, so I'm gonna go and run this goroutine over here, which is now doing my taxes, or whatever... Or fixing another bug, whatever that thing is, while I wait for that routine to send a message back down the channel and it's ready to unblock me there. Sorry, I had to pull in a little Go...
+
+**Mat Ryer:** Yeah... \[laughs\] Thank you.
+
+**Mark Bates:** Legally, we had to talk about Go...
+
+**Jon Calhoun:** We already had orchestration pop up briefly...
+
+**Mark Bates:** That was in the pre-show, I think... Wasn't it?
+
+**Mat Ryer:** I hope so, yeah.
+
+**Jon Calhoun:** I think it was in the show.
+
+**Mark Bates:** Oh... Either way.
+
+**Jon Calhoun:** Somebody - I forget who - mentioned on Twitter that another thing with remote work and doing the async stuff is that because you don't really get the tone as much, just assume that everything that's written to you has a smiley emoji afterwards; it's a great way to make remote work better.
+
+I'd love to see a study done on text written by people, remote versus not, because I feel like they use emojis a lot more, just as a by-product...
+
+**Carmen Andoh:** I think I have a link, I'll have to find it. But I did read an article about that.
+
+**Mark Bates:** I use emojis all the time.
+
+**Carmen Andoh:** Yeah...
+
+**Jon Calhoun:** As I said, when you work remote, you have to... Because it's like "I don't want them to think I'm writing one sentence to be mean. It's because I'm busy... So here's my one sentence :)"
+
+**Mark Bates:** I write these emails or Slack stuff and I feel like I'm putting a happy face at the end of every sentence most of the time...
+
+**Carmen Andoh:** That's great.
+
+**Mark Bates:** ..because I don't want people to feel like I'm trying to be mean. I'm just letting you know, I'm okay with all of this; this is just an explanation, and here's a smiley face to prove it... I'm not upset, here's a smiley... I'm just being terse in my communications." \[laughs\]
+
+**Carmen Andoh:** \[56:09\] Russ Cox actually said that in a 2015 keynote, like "I tend to get terse when I'm pressed for time, and people interpret that terseness as rudeness, or impatience." They wanna interpret something, because it's all done in text. So emojis typically help mitigate that.
+
+But I had a request from a colleague about 3-4 years ago... He asked something - it was my manager - and I just put "Okay." And he said "I have a request. When you're writing 'okay' to me in text, I need you to add either an exclamation mark, or a question mark, or something... Because I can interpret it in my head as 'Okay...?', or "Ok", or "OKAY!!", or "okay..." Exactly.
+
+**Mat Ryer:** Like, suspicious... "Okay......."
+
+**Carmen Andoh:** Exactly. And I just love that, when he felt it enough to be able to ask that. So as team members, be okay to ask for these little quirks of yours.
+
+The other thing that was a masterclass in async/text communication - I said something, and the person wrote back and said "I can interpret what you've just said in three different ways", and they took the time to say how those were interpreted. Then I realized "Oh, my goodness..." I was not clear at all, and I wanted it to be the second way. Now, that takes more time, of course... You can always jump on a call... But I thought that was great, for the benefit of everybody else who's going to see that in the room. And then it helped me really think about it.
+
+So in terms of interpreting in good faith, not just emojis, if you're unclear, try to broaden the perspective and say "I could take it this way, or I could take it this way." Or just say "I don't know how to interpret the comment." Just be able to do that. Because behind this piece of communication is both what is said, but also maybe if there's frustration behind what's being said, or if there's agreement behind what's being said... And that gets lost.
+
+**Mark Bates:** An old speaker trick works really well here, which is to repeat the question.
+
+**Carmen Andoh:** Ooh...
+
+**Mark Bates:** Speakers know this - you repeat the question from the audience, because they don't have a microphone half the time, so that people hear it... But it's also to clarify the question.
+
+**Carmen Andoh:** Do you paraphrase when you re-ask?
+
+**Mark Bates:** Yeah. At a conference, or even in Slack, whatever it is - if there's a question that's ambiguous, or I don't fully understand it, I'll just say "Okay, so just to be clear, I think what you're asking me is..."m blah-blah-blah-blah. By doing that, it clarifies to everybody. Because they all of a sudden go "No, no, no. That's not at all what I was trying to say", or "Exactly. That's 100% right. You got it", and then everybody's on the same page. It's an old trick, but it works really well.
+
+**Carmen Andoh:** That's great.
+
+**Mat Ryer:** Yeah, and it's that sort of stuff that really are the skills that help you be effective at asynchronous communication. It is about that, thinking about clarity the first time in things. And it's a kind of difficult thing. Of course, you need patience, but it leads me to think about another kind of demographic that we work with, which is kind of new developers, or junior developers that were currently maybe in teams, getting mentorship, or learning by osmosis with the more senior people. They potentially stand to lose out quite a lot from remote working, don't they?
+
+**Mark Bates:** Yeah.
+
+**Jon Calhoun:** I'm very openly against junior developers working remotely, if they can avoid it... And not to be mean, or anything; obviously, if you have to, that's fine... But I say that because I feel like working remotely is just a whole unique skillset to learn, plus there's the developer skillset to learn, and just trying to get all of that down at once, plus not having just the learning by osmosis type thing... It just makes it harder.
+
+**Mark Bates:** I spent the first half of my career working in an office, under mentorship of some amazing developers, and I can't imagine trying to become a developer without that mentorship, and without those experiences.
+
+**Carmen Andoh:** \[01:00:00.14\] Yeah... I will say how I was able to rise up as a junior remotely - it was because we got really creative with terminal sharing. Think about things like Tmux, Byobu... I just had the senior devs come in and drive. tmate has a great one where everyone can see my terminal, and they can help set up things. They'll see what I type in the command line... I think VS Code, in Golang, everybody has remote editing things. That's one way.
+
+**Mark Bates:** Pair programming.
+
+**Carmen Andoh:** Pair programming, but not just pair programming. It's seeing exactly what is being typed into that terminal by that junior. And say, "Okay, this is what you're doing wrong" or you maybe set up your workspace wrong, or "Oh, you didn't set your path." They can help troubleshoot. It's a lot more intense and work-heavy, but that's exactly what happens when you're in an office; they can look at your screen and ask you to type x command, and what do you see... We use tmate.io. I heavily recommend it. It's a fork of Tmux, and it allows for read-only, if you don't wanna be putting in commands into the person's... But also write to help drive. That's very high-touch for the mentor/mentee, but that's how I learned.
+
+**Mark Bates:** I had a client a few years ago who did all of their work on a shared EC2 instance... And everybody paired, and everybody would log into the same EC2 instance. Then you'd get over Zoom, or whatever, but everybody's in the EC2 instance, SSH-ing in... And one of the users is driving. And you're using Tmux, so you can see the other person who's driving right there on your terminal, and stuff like that. So virtual pair-programming is not only doable, I highly encourage it.
+
+**Mat Ryer:** Yeah, it's my preferred way. David and I - we worked like that; we work remotely entirely. We screen-share all day, whenever we're working... Which isn't all day. And I don't think should be, by the way; that's a whole different episode... But yeah, it's that -- and you learn little tricks just by seeing things. Sometimes I'll see something happening and I don't know how he's done that, and I just ask him, and it's usually some keyboard shortcut or something that I then get to learn... So it is a great way of having that osmosis of the information, isn't it?
+
+**Mark Bates:** I think Carmen has to go...
+
+**Carmen Andoh:** Yeah, I'm saying goodbye to everyone... It's the top of the hour for me, and I have another meeting... Speaking of increased frequency of meetings for remote... \[laughter\] I have to go live it now.
+
+**Mark Bates:** Have fun with that.
+
+**Carmen Andoh:** Bye, friends.
+
+**Mark Bates:** Bye, Carmen.
+
+**Mat Ryer:** Bye, Carmen. Oh, so that's what happens sometimes... People have to go.
+
+**Mark Bates:** I feel like with Carmen gone, that might be the end of the show.
+
+**Mat Ryer:** Oh, that's really nice of you to say, mate.
+
+**Jon Calhoun:** I suppose we're getting there... Can I get one more tip in here that we--
+
+**Mark Bates:** Oh, absolutely... As long as Mat doesn't. Go on.
+
+**Mat Ryer:** \[laughs\]
+
+**Jon Calhoun:** So when we were talking about communication, another thing that I've noticed that works well for me is when you're writing a lengthier email, it's easy for people to miss questions that you really need answered, or something...
+
+You were talking about paraphrasing or rephrasing the question... Even when I'm writing my own emails I'll find that I'll -- if you take the questions that are really important and separate them from a paragraph on their own line, and sometimes even make them bold, that really tends to help sending out these asynchronous communications, so then somebody can glance at this and quickly see "Okay, there's two bold questions. That's the meat of this email." The rest of it is all just qualifying.
+
+**Mark Bates:** To that point, I also do similar things, too. I don't do the bold, but I break the questions up, because it also gives people an opportunity and the space to answer the question in-line.
+
+**Jon Calhoun:** Yes. So it just make all that easy.
+
+**Mark Bates:** They see the question, and there's the space, and they can just jump in there and they can answer it... And that works for everybody.
+
+**Jon Calhoun:** I feel like when you write emails this way, you kind of get used to this "I don't want them to miss these questions. Here they are." You make them very clear, and that allows asynchronous communication to happen a little bit more effectively, because you don't have to actually "Oh, well you didn't actually answer my question", and now this email is gonna bounce back 2-3 more times and take even longer.
+
+**Mark Bates:** \[01:03:55.29\] I would also say -- for even full-timers, I'd probably say go out and read a book on consulting... Because I think there's a lot of things that you can take from the consulting world into the remote/working from home world. We've been talking about a lot of those things, and a lot of them come down to communication and clarity... Because as a consultant, you don't wanna be on calls 24 hours a day. You just wanna get the work done and do it well, so you tend to come up with these things that allow you to do the work really well, quickly and efficiently, and move on... So I think there's a lot of stuff that people can learn just from that lifestyle.
+
+**Mat Ryer:** Yeah, same for open source, actually... Open source is remote working, and look what things have been built with open source; it really is possible. I personally love remote working, and I don't think I will ever go back to work in an office. I never know what's gonna happen, but... Yeah, I feel like for me it's the preferred way to work; it's how I can be most productive.
+
+Sometimes if I've had to go to an office, I tend to do it on days when I don't have much work to do, which is kind of bonkers really, but... It's true. I stay home if I've got some serious work to do, which is kind of crazy...
+
+**Mark Bates:** It's the only place to get work done.
+
+**Mat Ryer:** Yeah, for me... But everyone's different, I suppose. Well, I think that indeed is our time. We've got some very interesting shows coming up, so keep subscribing, or... I don't know what you do.
+
+**Mark Bates:** Click on the link below.
+
+**Mat Ryer:** Yeah, that's it. They always say that.
+
+**Mark Bates:** So CLICK ON THE LINK BELOW! \[unintelligible 01:05:23.06\]
+
+**Mat Ryer:** \[laughs\]
+
+**Jon Calhoun:** Oh, boy...
+
+**Mat Ryer:** Number five. Number five is gonna blow your socks off.
+
+**Mark Bates:** Exactly. The number five remote working chip is going to blow your mind...!
+
+**Mat Ryer:** If that's so good, make it number one, I think, but... What do I know.
+
+**Jon Calhoun:** Well, save the best for last...
+
+**Mark Bates:** Number five is "Wear slippers."
+
+**Mat Ryer:** Oh, yeah... \[laughs\] That is a good one though, to be fair. I didn't expect that to come out. Nor the tiny glasses of water.
+
+**Mark Bates:** The tiny glasses of water...
+
+**Mat Ryer:** And the micro-kitchens at Google.
+
+**Mark Bates:** What do you expect from the insane dentist...?
+
+**Jon Calhoun:** Since I started working from home, I've bought way more slippers than I ever thought I would. \[laughter\] I have these ones that look like sleeping bag material, and they're just really nice and warm for my feet, but they wear out in about a year when you wear them every day.
+
+**Mat Ryer:** Right.
+
+**Jon Calhoun:** So my wife is like "Why are you ordering more of these?" I'm like, "I love these things. I wear them every day. I'm gonna buy more."
+
+**Mark Bates:** I've got a really nice pair of UGGs, and they even have their lovely outdoor tread on them; so if I need to run outside to the car, or whatever...
+
+**Mat Ryer:** If you need to embarrass yourself in public...
+
+**Mark Bates:** If I need to embarrass myself... At least you know you can. And trust me, about a month ago I was running through, sloshing them, chasing the dog down the street, because he got out...
+
+**Mat Ryer:** Oh, yes...!
+
+**Jon Calhoun:** Let me just say that if I wear my slippers outside, they're not longer slippers for inside. My wife is very strict about these things.
+
+**Mark Bates:** Oh, wow... Okay. Well, there we go.
+
+**Mat Ryer:** We've got some amazing shows coming up, by the way... Next week we're talking about CHAOS conferenceering... We've got Natalie Pistunovich and Ronna Steinberg to talk about GoBridge...
+
+**Mark Bates:** Oh, Ronna. Lovely.
+
+**Mat Ryer:** Yes. And also, later we've got Matt Heath and Tom Wilkie to talk about monoliths versus microservices, so that'll be an interesting one... And real-time communication as well, which is another part of this, I suppose, that's quite important... But it's more the tech side of this. WebRTC, those kinds of technologies, and stuff... So that's cool. But if anyone's got any other ideas, tweet us. Mark, what's your Twitter?
+
+**Mark Bates:** @MarkBates.
+
+**Mat Ryer:** Jon, what's your Twitter?
+
+**Jon Calhoun:** @JonCalhoun. No "h" in the Jon.
+
+**Mat Ryer:** No "h" in the Jon. And mine's @MatRyer.
+
+**Mark Bates:** No "h" in that either.
+
+**Mat Ryer:** No "h" in mine. No Jon.
+
+**Mark Bates:** And none in mine either. We're all h-free.
+
+**Jon Calhoun:** I mean, there's an "h" in Calhoun is my problem, so that makes it really...
+
+**Mat Ryer:** Oh, you can't just say "No h."
+
+**Jon Calhoun:** I can't...
+
+**Mark Bates:** Yeah, that's awful. It doesn't work.
+
+**Mat Ryer:** Arrgh! It must be so difficult being you sometimes, Jon.
+
+**Jon Calhoun:** It's terrible...
+
+**Mat Ryer:** \[laughs\] Yes. Well, that's it. That's our show. Thank you very much. We'll see you next week!
+
+**Break:** \[01:08:14.15\]
+
+**Mat Ryer:** Hello, and welcome to Go Time! Today we're talking about working from home/remote working. We're gonna have hopefully some reassurances for those of you that are new to working from home... There might be also some tips and trick that you can pick up, that we're gonna discuss, and hopefully we'll just have a nice, lovely chat as well, which is always the goal. So hello to everybody else that's on the podcast today, which is -- I'll tell you what, let me start again, because that was the worst one I've ever done... \[laughter\] Sorry. That's the worst one I've ever done.
+
+Okay. Alright, I'll do it again. Don't worry, everyone. That's a special DVD extra that you get for free for listening live.
+
+**Mark Bates:** The gag reel.
+
+**Mat Ryer:** Yeah. What?
+
+**Mark Bates:** The gag reel.
+
+**Mat Ryer:** Oh. I thought you said something else. Hello, and welcome to -- you know, you can't do "What?!" halfway through the intro, Mark... \[laughter\]
+
+**Mark Bates:** Well, what else could it have been?
+
+**Mat Ryer:** Okay, never mind. I'll email you later. \[laughter\]
+
+**Mark Bates:** Okay...
+
+**Mat Ryer:** Hello, and welcome to Go Time! I'm Mat Ryer. Today we're talking about working from home, or remote working.

We have regrets_transcript.txt

@@ -0,0 +1,367 @@
+**Jon Calhoun:** So today we're gonna be talking to you about regrets. Specifically, we're gonna be talking about technical regrets; no need to go into how I broke my nose two weeks in a row... We'll talk about what caused these mistakes, how we learned from them how teams and companies can maximize the learning potential from whenever somebody on your team does make a mistake... And really just trying to show everybody that it's okay to make mistakes, that we all make them... Because I think that's something that we all need to understand going into tech, is that you are eventually going to mess up, and that's okay. What's more important is how you move forward.
+
+To start with, does anybody wanna talk about a time they made a mistake?
+
+**Aaron Schlesinger:** Well, since it's crickets, I'll give it a shot... When I came to Go, I was really psyched about the concurrency. I came from Scala at the time, so I was doing actors and thread pools and all that... And for me at least, it was super-complicated, and I had to re-read docs every other day to figure out what was going on under the hood. So I came to Go, and I was like "Yes! Goroutines, channels... I've got this." And then I immediately wrote actors in Go, which just felt in retrospect like fitting a... What is it- a square peg in a round hole? Or the other way around? So definitely a mistake. I regret it, because it cost me like a month, after it was all said and done... But looking back, I think I learned a lot about Go concurrency basics, without at the time having really read through anything deep about concurrency in Go.
+
+**Ben Johnson:** I feel like everybody has the channels phase of Go, and whenever new people come with learning Go and how excited they are about that, I always almost forget those channels. I just don't use them that often. It's like "Oh yeah, there's that thing. It's really cool, too." But yeah, I agree, there's definitely a lot of ways to go around that issue of concurrency, and figuring out which one's appropriate at the right time.
+
+**Jon Calhoun:** \[04:06\] I've seen that a lot -- so people go to learn how to build web servers, and the first thing they'll ask is "Well, how do I use this concurrency?" because they're really excited about it... And I think -- I don't know if they're disappointed, but when they realize that it's already there and happening for them, and they don't have to worry about it, I think some of them are just like "Oh. Well, I thought I was gonna actually get to do something, and use these things." It's like, "No, you just write regular code, and just make sure it's thread-safe, and you're fine."
+
+**Kris Brandow:** I guess over the time of going into new jobs, whenever I'd look at a codebase I can kind of gauge how familiar the person who wrote it is with Go, based on the number of channels and goroutines I see sprinkled throughout the code. I tend to see a ton of new people. And I definitely did it when I was first learning Go, that's just like "Oh, I should just put goroutines everywhere. It'll be fine. There won't be any problems", and then you try to debug the code three months later and it's like "Who wrote this, and why did they do it this way?" And you look and it's you, and you're like "Oh. Oops..."
+
+**Aaron Schlesinger:** Break out the old Git blame, and then you shut it down real quick when you see your name on there... \[laughter\]
+
+**Jon Calhoun:** I think that's okay though, making those mistakes, I should say... Because -- one of the mistakes I made when I first was jumping into Go was that I just felt like I over-planned, or I tried to over-optimize for getting things perfect... So you'd read about how you shouldn't use MVC, and you shouldn't do all these different things... And I sat down, I'm like "Alright, I'm gonna write this side project." It was a side project, it wasn't my main work stuff...
+
+So I was like "I'm gonna build this thing, and it's gonna be good Go code...", and I think I spent so much time rewriting some stuff, because I was like "Oh, this is a bad way", and I realized why, and then I'd go back and rewrite it... And in the end, I'm pretty sure what I ended up doing was just using a simple MVC model, and just got it done... And then later I was able to come back and tweak things and adjust. But I wasted so much time trying to meet the expectations of everybody, I guess it's how I'd put it. To me, that was a big mistake, because I feel like you learn more by just kind of jumping in and doing stuff, rather than trying to find the optimal path.
+
+**Ben Johnson:** Yeah, I totally agree. I think that definitely early on it's less important. I feel like once you start sharing code, and libraries, and stuff like that, there are a lot of those idiomatic things where if you didn't do it at first, it might require an API change in the future, you might have to change your versions to a new major version... I think it's more complicated there, but yeah, first, for sure, just jump in. All my code was basically Ruby inside Go, the first couple projects... So that was rough. But you figure out as you go, for sure.
+
+**Aaron Schlesinger:** And I find over and over I make a lot of the same mistakes in terms of over-designing. Most of the time I like to do it because it's fun... And then I turn around, and I say "Well, this could have been way simpler. I could have deleted this function" or whatever. But I keep finding over and over that Go is a really forgiving language when you have to uncomplicatedify... I hesitate to say "simplify", because it's really not making the code semantically simpler. It's literally deleting syntax. It's just really forgiving, and allowing you to do that without changing large swathes of code outside of that area you're working with.
+
+**Jon Calhoun:** I also like to think about -- I don't know if it was Cory LaNou... It might have been him that I was talking to, but he'd basically mentioned that when you're writing code, getting something working is more challenging than refactoring something. So just keeping that in mind, and knowing that it's okay to make mistakes, and that's part of the reason we have code reviews, and peer reviews, and all these other things in place, is so that we can learn from each other and improve things... Not because we expect people to have the perfect first version.
+
+\[07:58\] I think that's a mistake a lot of people new to programming make, is that they see code that somebody like Kris or Ben would have wrote, and they're like "Oh, I need to write this the first time", and I'm like "They probably didn't even write it like that the first time, so I don't know why you're expecting yourself to do it."
+
+**Kris Brandow:** I think that's an important point about the difficulty of writing new code. I'm definitely experiencing that myself right now. I have this new project at work, and it's like "Oh, write from scratch. Don't use anything else anybody has. Just write a brand new thing", and I'm just struggling to remember how to build all this stuff I'm used to just being there already... And that's a really odd experience for me. But it makes me appreciate a lot the coming into a codebase and just being the person that gets to "fix it", or "make it better and improve it."
+
+**Ben Johnson:** Yeah, I'm working on a database project right now, and it's just in its infancy, but it's all just some for loops over some basic data structures, and it would just make any database engineer just cry, just how slow it'd be. But I'll get some tests around it, and it's so much easier once you have all that in place.
+
+**Aaron Schlesinger:** Yeah, and I feel like if you go on GitHub and you look at some popular open source projects, I feel like that's sort of the Instagram of programming; you see the perfect result, after everyone's put on the make-up, and rolled the camera, and everything. We put that on ourselves... Whereas, like you just said, Ben, you could just go in and write a for loop and it'll be good for a while. You've gotta start somewhere.
+
+**Jon Calhoun:** And there's always the tell-tale sign of initial commit has 20 source files and 4,000 lines of code, and you're like "That was your initial commit?" It's like, "That clearly was not your initial commit."
+
+**Ben Johnson:** I like to just kill the folder, and just restart after probably a couple hundred commits in.
+
+**Jon Calhoun:** Oh, I'm the same way. I've got the initial version, and then that eventually gets to something, and I'm like "Okay, I'm okay with sharing this." And then I delete and just go from scratch.
+
+On one hand, I get why you do it, because you don't necessarily want that bad history there, but on the other hand it is kind of hard for somebody jumping in who actually thinks that might be the initial commit.
+
+**Kris Brandow:** Yeah, seeing code evolve over time is definitely helpful when you're trying to figure out how some feature in a codebase works. I've definitely done that with things I've been curious about, like "How does this feature of this big open source project work?" Kind of just like dig through the path... So it's really nice when the whole history is there, but I think there's kind of like an engineer ego a little bit, that's like "I don't want people to see my very early on work, where I made all the stupid mistakes that I made. I spent five commits trying to figure out why this for loop wasn't working correctly", or something.
+
+**Jon Calhoun:** That's a hard one too, because I think naturally everybody doesn't like to admit that they make mistakes or that they're wrong. You almost don't wanna be seen as human... I don't know, I just feel like that's something that's hard to get over, is that fact that it's okay to be seen as somebody doing that stuff.
+
+I suspect that some people who livestream and do that sort of thing more often are better at it. I've always been bad at livestreaming, because I'm like "What if I make a big mistake on there and look like an idiot?" For whatever reason, that's just a mental barrier. But then I see other people who stream, and I've even seen people who learn to code a new language, livestreaming the whole process...
+
+**Kris Brandow:** That is bold.
+
+**Jon Calhoun:** I don't know how you do that, because I would just sit there and be like "I'm gonna get made fun of a lot here, or something..."
+
+**Kris Brandow:** It's certainly been my experience in the past... You just kind of get used to it over time. I remember working on the Go driver for MongoDb; getting the first big release out was so agonizing... So I'm like "No, it has to be perfect, and everything has to be golden." And then I remember there was a really big minor release we were pushing out; we released it, and either a test had failed, or something didn't work, and we had to immediately release a patch for it. I was like "Oh, okay. Yeah, whatever. Mistakes happen, we can't get it perfect all the time." That was such a large growth for me from the beginning, to that point, to be able to just be okay with that. I think a lot of people would just be like "Oh my god, no. I messed up. I'm done forever. I'm gonna go run away and hide now."
+
+**Jon Calhoun:** \[12:14\] I think the other thing with mistakes is depending on how an organization or a company handles it can kind of shape how people address them in the future, I guess is a good way to put it. One example that comes to mind - we've all probably heard a story about some junior developer or somebody deleting something, or breaking something... Somehow breaking production, whether they delete a database file, or whatever they end up deleting... And I think some organizations handle it well as a teaching experience and a learning experience, of like "How can we prevent this from happening again?" and others are like "He's fired. It's not gonna happen again." I'm like "That doesn't prevent it from happening again. The problem's still there." So beyond the obvious "Don't fire people for making a simple mistake", are there other things you've seen that work well in organizations to sort of foster that growth through mistakes?
+
+**Aaron Schlesinger:** One thing I've seen at least is encouraging the folks who did make the mistake to build tooling to prevent it in the future, whether it's production tooling, or like a CLI to help you automate the thing you messed up, or whatever it may be; they get a deeper understanding of what they did and how the right way to do it is. Also, they just transmit that knowledge via code to everybody else.
+
+**Kris Brandow:** I think too deeply analyzing the system and processes in place that led to that failure or mistake is super-important. I think a lot of the time when we go through and we do a blameless retrospective or something, we might not pay as critical attention to the system overall as we probably should, and figuring out "Well, what things do we not have in place, that would have prevented this not just on this team or for this project, but across all of our projects?"
+
+**Jon Calhoun:** I think a lot of orgs tend to call them post-mortems... Is that what you had in mind, or is that a little bit different to you?
+
+**Kris Brandow:** Yeah, I think I meant post-mortem, not retrospective. It's definitely the word I was looking for.
+
+**Jon Calhoun:** Either one's fine; that wasn't meant to be a terminology -- I wasn't trying to fix or correct you there, but that's the term I've heard... And I've seen -- I think GitLab has done some, where they've lost data and had a public post-mortem. I think some companies do like just private ones internally, depending on the mistake and the severity of it... But I was just trying to make sure that's what you mean and that's the kind of thing that you're referring to...
+
+**Kris Brandow:** Yeah.
+
+**Jon Calhoun:** Okay. Yeah, I definitely agree that sitting down and having a discussion about it is useful, especially -- one piece of advice I've seen is to sometimes do that without computers too, just to sit down, and like it's more about the human interactions, and openly as a group... Like, not trying to Git blame and look through the code and see who messed this up. It's more of a "How can we as humans make sure these types of mistakes don't happen again, and how can we work through this trying to be empathetic?"
+
+**Kris Brandow:** And I think writing things down afterward is important too, so that when new people come along and they wonder "Why do we do this thing like this?", you can have something to go reference. It's like "Oh, we had this really big outage, and here, you can go read about it. That's why we do this in a certain way." So that's the first thing that I tend to do, but I think a lot of people do when they go to a new place, is start questioning the processes and being like "Well, why are we doing it this way? Is this because we've just always done it this way, or is there some deeper underlying thing that caused us to do it this way?"
+
+**Aaron Schlesinger:** Yeah, and I think it's all the better if the company's able to publish them publicly, because the person can get a feel for how a company deals with it before they even join.
+
+**Jon Calhoun:** \[15:51\] I'll definitely say companies that do public post-mortems like that, I'm far more inclined -- like, if I was looking for a job, I'd be far more inclined to consider them than the ones that privately don't tell you what happened and you're always sitting there wondering "Did the person get fired for this? What happened?" Especially when they're mistakes that you're like "I could have easily made that mistake." And sometimes I don't even think it matters how much experience you have.
+
+**Ben Johnson:** To touch off on the outage piece - I feel like that's another side of it. A lot of these companies - I've talked about this before - where like GitHub will go down and everyone kind of stops work for a while, they're all freaking out... But no one's leaving GitHub. The cost of that outage - it's frustrating and whatnot, but if they're down for an hour, we all find ways to get around it... And sometimes those things that seem so big, those big outages aren't usually as damaging as you might think; I mean, personally.
+
+Sometimes, depending on what you do, you really need to have that super-high uptime, but other companies just aren't the same... I think it's important to get some perspective on that side, too.
+
+**Aaron Schlesinger:** I think the public aspect of it builds trust in the company, too... Because if they're willing to not only admit they made a mistake, but also give you a look inside of their infrastructure in order to tell you what happened - at least for me, that gives me a really good feeling that the company is gonna be around for a while, they've got a good internal culture, they're gonna be able to track good people to make it better in the future, the features are gonna stick around... There's a long list of things that I immediately go to when I see something like that, and almost all of the things on that list are good.
+
+**Jon Calhoun:** I think it's also nice to see what they're doing, so it doesn't happen again. The best post-mortems are like "Here are the steps we're taking so this problem does not happen again." Then there's some that -- the biggest one that comes to mind is when we all had the credit leaks from... I don't know which company it was, so I'm not gonna name any of them...
+
+**Ben Johnson:** All of them, probably...
+
+**Jon Calhoun:** All of them, probably...? So whichever ones it was... But even now, I don't feel like I know what they're doing so it doesn't happen again. As far as I know, they were like "We're gonna give you free credit checks for a year", and I'm like "Well, that doesn't prevent this from happening, it just gives me a credit check for a year." So it'd be nice to see a list of like "Here are the steps we're taking so that this won't happen to you again." Without that, it's really hard to have confidence that they're not going to have the same issue.
+
+**Break:** \[18:24\]
+
+**Jon Calhoun:** Alright, so let's take a step back and look at another mistake that somebody has made, and dive into that. I don't wanna call anybody out, so if anybody wants to jump in...
+
+**Aaron Schlesinger:** I already bared my soul, so...
+
+**Ben Johnson:** I guess kind of a long ongoing mistake in my career is just not understanding the underlying technologies that I use... Especially early on, using an HTTP framework, web framework. I'd just assume that that's what you do. But understanding that framework, and maybe even layers on top of that framework, is a lot of times more complicated than just understanding HTTP, or whatever that underlying technology is. So I feel like I've done that for a long time in my career... But more recently, just trying to step back and just understand what do those frameworks actually give me, what do they add... And a lot of times *net/http* is enough with a router.
+
+**Jon Calhoun:** So I guess first off, are you talking about all the way down to actually understanding how TCP works, and going even below that, or...?
+
+**Ben Johnson:** I mean, I think it's a trade-off of like "What do you get for the abstraction that you're working with?" I'm not pushing bits across an Ethernet port, but if I can understand just basic headers, and what they do, rather than having some other library on top of that to actually set certain parameters... I feel like just understanding that underlying HTTP/RFC or some aspects about it helps me to write more direct code, instead of the candy-coding on top that you don't really need... Syntactic sugar. That's the word.
+
+**Jon Calhoun:** Okay. So when you're looking at that, I guess my follow-up question would be like how do you balance that with the "taking on too much" aspect of it?
+
+**Ben Johnson:** Sure, I think it's definitely a trade-off. I guess I don't know when -- there's not an easy way to say "This is when you should do it. This is when you shouldn't." So I guess it's probably more trying both... And a lot of times I can work faster not using a web framework, when I can using one. So I think for that aspect it's been much better.
+
+I feel like another example would be like ORMs. I used those for many years in my career, and honestly, just going back to straight SQL - there's a bit of a learning curve there, but once you get past that, you just have a much more direct idea of what's going on in your code, how data is moving, and you can optimize it and you can change it around... Whereas the ORM has just this whole, crazy -- all these little tentacles going out, and trying to figure out what it's doing.
+
+**Jon Calhoun:** I've definitely seen both sides of that. Back when I was working on a Rails project - it was a startup, and one of the people I was working with wasn't technical, but they knew enough to get the code running... And at one point they started writing what was -- basically, they were trying to use the Rails ORM to be like "You can call user.comments and look up all users' comments." So they had this query written out, and I looked at it and it made absolutely no sense to me. And then I realized that they just assumed all this stuff was magically there; like, you're just looking for stuff that wasn't there. And it kind of clicked at that point, and I'm like "Oh, I could definitely see somebody - especially somebody relatively new - not really understanding how this maps to SQL queries and how this maps to a database table, and just being confused by what's going on.
+
+So I guess I see both sides of it, because I will definitely say that I've never written web applications as fast as I did with Rails in Go, but at the same time, I've never understood my application in Rails as much as I do with ones that I write in Go. And there's definitely trade-offs to that, in the sense that when you really understand things and things go wrong, I think that's where it sticks out the most.
+
+**Kris Brandow:** I think it definitely comes down to a trade-off with like "Where do you want to spend your time?" If you just wanna get something done really quick, I think then yeah, go with a framework... But that's gonna come back and bite you later, when you've had some bug that's in that underlying thing, and if you don't understand how that works, it's gonna be a lot more frustrating to figure out "Oh, where is that weird thing that's happening?" I've definitely encountered that a number of times in my career, where it's just like "Oh, we didn't set a timeout properly on a TCP connection, so we have a process that's trying to write to a TCP connection for like 40 hours." And it's like "Well, how do you figure that out? How do you find that out if you don't know how the underlying stuff works?"
+
+**Jon Calhoun:** \[24:17\] I think that's part of why I'm so hesitant to try a new technology. I have not used Kubernetes pretty much at all, and part of that is because I'm like "I'm gonna push something into production, it's gonna break, and I'm not gonna have a clue what's going wrong." And that's slightly terrifying to me, especially when we're talking about something in production, where you really don't want that happening.
+
+**Ben Johnson:** Everything seems to be trickling down from big companies. I've worked in moderately large companies, I've worked in tiny companies, and just the approaches you need for both are so different that -- I mean, everytime I've used Kubernetes, it just seems like one crazy layer of complexity on top of another, and I don't ever feel like I'm writing software; I just feel like I'm debugging software. It's frustrating.
+
+**Aaron Schlesinger:** I feel like you can kind of feel it when there's such a big barrier to getting something up - whether it's Kubernetes or something else; there's just this point where you feel like "Okay, I wrote this app, it's X number of lines long, and it took me a month. And now (I'm taking Kubernetes as an example) I'm writing 500 lines of YAML, and for each line I have to go back to docs, and look something up." If you're an indie developer or you're a small team, you kind of get that feeling that "Hey, this isn't something we're ready for yet. It's a signal that, like you said, Jon, something's gonna break in production, we're gonna have no idea how to fix it, and we're gonna be under the gun." And that's a feeling I think that people develop over time as developers, especially when they push stuff to production and get those battle scars.
+
+**Kris Brandow:** Yeah, I think the point around different size companies needing different things - I think that's really important, and I think a lot of small companies dive into Kubernetes because they're like "Oh, Google needs this thing, so it occurred to me I need it." It's like "Well, you probably need a VM that you know is gonna keep running and you'll be fine", and if you have 40 minutes of downtime, it's gonna be okay. You can make that trade-off, instead of like this huge complexity that comes with something like Kubernetes.
+
+**Jon Calhoun:** I think that's a general mistake that so many people make - wanting to use these complex technologies... And some of it is just genuine interest, which is fine. I completely understand that. If you really just wanna learn something, then by all means, you should go learn it. But I think sometimes trying to push it into a company or an organization is not the best way to learn it, because it does present those problems. And I've just seen countless examples where you get that added complexity that you don't really need to be there, and then you're just left with something that's really hard to maintain.
+
+**Ben Johnson:** I find it crazy - with Moore's Law we've just expanded computing power at a crazy rate, but somehow I feel like our applications are slower, because we just keep adding all these little bits of cruft here and there... It just makes me sad.
+
+**Jon Calhoun:** Yeah, I mean - GraphQL is one that I see a lot of people pushing towards... And for me, I completely understand why Facebook came up with it; there are times where I just don't understand it for -- like, if you're writing an API and you have literally one web client that's utilizing it, I'm like "Just make it return the data you need." You don't need a query language for that; just return what you need. Whereas the Facebook use cases - there's tons of different versions of mobile apps, and people are in different places where bandwidth is limited, and there's all these different factors that come into play... And I'm like "Yeah, it completely makes sense there." But I do think it's a mistake to try to let others dictate what technology we should be using, or to just mimic what other people are doing... And I see that all across tech, whether it's databases, or all these different things that they use, and I think that's a big mistake. It's almost like it doesn't matter if you tell somebody not to do it; you almost have to let them make that mistake.
+
+**Aaron Schlesinger:** \[28:17\] Yeah, we love shiny things as engineers. It's so hard to really strip down to what you need in the practical sense, and separate that from "What do I wanna learn? What's the newest thing that I wanna put on my resume?" or something like that.
+
+**Jon Calhoun:** It's especially hard with small teams, because you're usually so busy that the only way to learn something is to do it on a work project. So then it gets there, and if you're not doing it for your resume, that means you're probably looking to leave, and that makes it even worse. \[laughter\]
+
+**Kris Brandow:** I think too there's this -- sometimes we do things as like a hack, to make things look better. I think a good example of this is clean URLs, which came around in the PHP time, and you're like "Oh, I just want it to be like /user/123, and then make URLs hackable. That's really cool and really fun." And I think that just got translated into our APIs, so we started building these complex web frameworks that could pull wildcards out of the URL... But HTTP comes with query parameters, which essentially do the same thing, and actually have a nice name. But it's super-popular across most of the industry to just use clean URLs with APIs. And I always find that so confusing. It's like "Why do you care if your APIs are a little bit uglier?" ...if you have ?user=1234, instead of just having it in the main path itself? And I think we do that with a lot of things, where it's just like "Oh, well someone did that because there was this specific use case where it made sense", and it was a good user experience enhancement, but it might not apply to other places where you just kind of pick it up and try and put it everywhere else.
+
+**Jon Calhoun:** One example I've seen of that where it's useful is I was actually looking at the source code for Pace with Mat Ryer, and they have URLs where they try to stick -- basically, the name of the task is in the URL, so if you copy-paste it somewhere, somebody can actually see... Just glancing at the URL, they can tell what it pertains to. But they don't actually use that part of the URL. It's just like a slug that you can change to anything. And there's another part of the URL that actually has the ID, and that's what matters. And doing it for cases like that completely makes sense if you're just trying to make more visibility for something... And I think Medium's a good example of this too, where most blog posts have a slash - the title of the blog post, and then at the end f that tile there's a slug that they actually use. In those cases, I'm like "Okay, cool, you're making this more readable." But then there's others, like you said, that are /user/3/comments/456 and you're like "That doesn't really tell me much, aside from I'm looking at maybe this user's comment... But it's not really helpful."
+
+**Kris Brandow:** Yeah. I think too there's that key difference of like "Is this a URL that someone's gonna copy and paste, or is this a URL that's just like two machines are gonna send between each other?" I think in the latter case it doesn't really matter what your URLs look like, if they're just machines sending them between each other...
+
+**Jon Calhoun:** Have you guys seen -- it's like a JSON API spec where you have a self-documenting API that sort of defines the URLs for additional resources... I forget what it's called.
+
+**Kris Brandow:** Oh, like hypermedia APIs, or...?
+
+**Ben Johnson:** I know what you're talking about, but I can't think of the name.
+
+**Jon Calhoun:** There was something for APIs where if you looked up a user, it might give you a bunch of links to like "Here's the path to look up this user's comments", or something. It was supposed to be like a fully self-documenting way to design your API... But what never made sense to me was I was like "Who uses an API and doesn't understand what the end points are, that they're gonna start scouring at this way?" I'm like, I'm not writing code that's like "Tell me what I can get and I'm gonna magically decide." So some of the use cases there just didn't fully make sense. What you said made me think of that, for some reason.
+
+**Kris Brandow:** \[32:05\] Yeah. I think that's the general idea behind hypermedia APIs. I've spent a lot of years studying hypermedia APIs. I think I'm a pretty big proponent of them as well. I think yeah, that first confusion of like "Well, I should know what all my URLs are." I think the main reason behind wanting to differentiate those two things comes down to semantic versus runtime, I guess. We often tie those two things together, and that makes APIs pretty brittle. If you wanna have a new version of your API, now you're shoving v2 and stuff all throughout your URLs, because you can't use the same name to mean a different thing.
+
+So I think that's partly where that stuff comes from, but I also think it's very poorly explained to people, especially when you first come into a hypermedia API or something like that... Like a new concept, and people just know that. That curse of mastery, where you just assume that everybody has acquired this knowledge that you have... So it's like "Oh, I understand the difference between what a target URL is and what a semantic name for that URL is, so everybody must understand these things."
+
+And I just think too, in some ways we haven't pushed it enough as like a sub-community of the industry. Hypermedia is a thing that we openly accept when it comes to browsers, but with our APIs not as much. We'd go probably pretty mad if Facebook decided to update some URL somewhere and Chrome was just like "Oh no, that's not what I have recorded in my source code, so you can't use Facebook anymore." We'd all be like "No, that's not how it works." But I think making that translation from how something works in the browser to how something works in our APIs is difficult.
+
+**Ben Johnson:** It's also something that, Jon, you were saying - you should know what the API does. This is a little bit of a tangent, but just documentation in general I feel like is a regret I've had in the past, of just not doing enough, and that projects that have actually done really good documentation... like Bolt I did a lot of documentation on it; I feel like that was the main selling point for a lot of people... It's just like "Oh, well, this looks like a legit project, not just three lines in a readme." So I think that's another regret to put on there.
+
+**Jon Calhoun:** I think that's what a lot of companies should have... I see so many API companies, then you go read their docs and you're like "I don't know how this works." It's frustrating. And then you look at something like Stripe, and - generally speaking, Stripe is one of the better ones. And not to bash on them, but there are definitely cases where I'm like "This could have been better." And they're still the gold standard, because so many other APIs it's just like "Here's a list of endpoints, and that's about all you're getting." And I'm like "That doesn't really tell me what the general process looks like at all.
+
+**Aaron Schlesinger:** I think you can tell at what stage of the development process they wrote the docs, just based on reading them... Because you know, when you see incomplete stuff, I would guess they were trying to rush it out the door to meet a deadline. And they may understand everything in their head - they being the development team, or the PMs, or whoever... But sometimes it's hard to get such a complex topic down to size into that paragraph, or whatever it may be... And I would go with "Write the docs first", and that just feels like it takes a lot longer, so I think that may be why you see some of those docs that are just "What does this mean?" That's everyone's reaction to them.
+
+**Jon Calhoun:** I guess the companies also make the mistake of -- like, they make the API or something like that purely for the publicity, or just to claim they have one... And you can kind of tell whether it's a first-class citizen and all the things they offer, versus "Oh, it's here, but we don't really care about it."
+
+**Aaron Schlesinger:** \[35:57\] Yeah, I think that's a good skill for engineers to have too, to know "Is this a real thing that's gonna last?" or like you said, "Is this just off to the side and they're gonna ditch it the minute they can?"
+
+**Ben Johnson:** I think it requires empathy too, just to put yourself in someone else's shoes that doesn't understand this stuff. I think it's a good exercise.
+
+**Break:** \[36:13\]
+
+**Jon Calhoun:** Somebody in the chat is asking about mistakes in testing; specifically, theirs was using time.Sleep in tests, so they said "Thanks for the clock package, Ben."
+
+**Ben Johnson:** You're welcome.
+
+**Jon Calhoun:** Do you have any other examples of times that you've made mistakes with tests, or what you've learned from them?
+
+**Ben Johnson:** I can start with table tests, I used to do those a lot... And I hate them now. I find they are way more complex than they should be, especially if you do a sub-test now; I think that clears up a lot of the needs of table tests. Table tests could just be a function, and just do a bunch of them. So that's my rant on table tests.
+
+**Jon Calhoun:** I will have to see what you mean, I think... Because I've definitely seen table tests where just having a couple types and that's all that's in the table is very hard, especially because once you set a type up, you almost need to dynamically get something from it for what you expect in the result, and that makes that type of table test tricky, I guess. Does that make sense?
+
+**Ben Johnson:** No, but...
+
+**Jon Calhoun:** \[laughs\] Let's say I'm testing a database implementation, or some sort of store implementation. If I create a user, a part of my verification that it worked might be to check that the ID is set to something, and then we update that later, and make sure some other fields are checked... But I don't know what the user's ID is going to be early on. I kind of almost need to be like "I need results from some steps to move on to the next steps and figure out what makes sense." And sometimes that's hard with a table test. I still feel like it's a terrible example...
+
+**Ben Johnson:** I mean, sometimes for very simple pieces of code I'll use table tests. I could see using a table test if you have multiple implementations... But I would say those are probably the two use cases where I'd use it.
+
+**Jon Calhoun:** Are you writing more unit tests, or are you writing more integration, or what would you say the classification is there?
+
+**Ben Johnson:** Yeah, it's a good question. I definitely write a lot of unit tests... I try not to write tests for things that are too simple, I guess... But I guess it's so much easier to iterate when you have something very low-level. But yeah, integration tests are definitely important, so I'd say -- that's a good question; I have to think about it. I would say I have transitioned more to integration... If you can test a bunch of things at once... I guess it depends on the complexity of your app, too. I don't have a great answer for you, honestly. It really depends. Just that more tests are good.
+
+**Jon Calhoun:** \[39:36\] I had asked because a mistake I've made is when you're writing unit tests for a relatively small piece of code, that's when I feel like table tests are very easy, because it's just like an input/output is all you need... And when you write a bunch of table tests, it feels like you're being really productive, but you really might not be. Because you're writing code, you feel like you're doing something productive, but you might not actually be testing things that are that important. If you have a package that splits strings on spaces, or something, do you really need 50 different test cases, when a couple might give you enough confidence just to move on and do other work?
+
+**Ben Johnson:** Yeah.
+
+**Jon Calhoun:** So I guess that's why I was wondering if it was a case where when you moved more to integration testing maybe you just didn't see it as much, or what it was...
+
+**Ben Johnson:** Yeah, I think integration tests definitely give you a more real-world kind of test. Actually, one thing I will say I do a lot more recently is more kind of a quick-check test. There was testing/quick in the standard library. So - quick-check tests, you know what they are. It's just basically randomized testing. And there's really fancy quick-check stuff in other languages, but on Go you're basically just handing that off a rant, like a math rant to a test, and then generating tests from that. I find that to be one of the best ways, especially for libraries, where they're more self-contained, to actually go through all kinds of different iterations you wouldn't expect.
+
+I had an implementation of immutable sets, immutable collections, where I'd actually make a -- I had the immutable collection, which... So immutable collections, if you don't know what they are, is whenever you make a change to the collection, you make a new copy of that collection. And it doesn't make a whole giant copy, it just gives you a new copy with a little bit of a change.
+
+So I did the tests... It was a quick-check kind of test, where I would basically randomly insert/delete/change around the collection, but I'd also do it on another kind of just in-memory basic collection. It was very inefficient, the in-memory one, but I knew that it would work... Just like a Go map, basically. So then I could test that both implementations did the same thing, with random different inputs. I find that to be super-helpful.
+
+**Kris Brandow:** I think it depends a lot on the type of thing you're building, what types of tests you wanna have as well. I noticed that when I'm building more library-like things, I tend to try and test more of the granular components, because I know I'm gonna mess around with them a lot, and change them a lot... But I think if you try and do that too much with an application or a binary, you get this kind of code paralysis point, where you wanna change a function, but now since that function is so well-tested, you've gotta change a whole bunch of unit tests because of that.
+
+I think that's definitely one of my regrets, is the way I used to write unit tests, where I'd just try and test and cover everything, and it was always about implementation details, instead of testing the interface. And in my more recent libraries/applications I've built, I've started to move more toward black box testing, and really getting my interfaces correct, and making sure that I can test the different code paths throughout my application from the outside, and not needing to finagle with internal state, or anything like that.
+
+**Aaron Schlesinger:** That one hits me really hard, because even these days I find myself writing unit tests, because I wanna be able to tell exactly what broke if the test suite fails... And then yeah, I end up adding something to a function or whatever, and then all the things broke. It's like eight tests that failed, or ten tests that failed underneath the function... When in reality, like you said, I could have just had a black box, if I needed to test the mock version of the function... Throw an interface in there and then just call it a day, and probably get the same amount of coverage off of that one test, instead of those eight.
+
+**Jon Calhoun:** I think that's a mistake I've seen, where we see how other people are writing tests, or somebody will say "Oh, 80% unit tests is the way we go", and we'll think that's what we should be doing, when in reality it depends so much on what you're building that you can't just go based off what others are doing; you have to sort of decide what makes the most sense for your project... And one example I saw of this -- like I said, I was looking at the code for Pace when I was talking to Mat, and it's not maybe all of their tests, but a good chunk of their tests are integration tests that literally spin up a Firestore database, and they spin up some other stuff...
+
+\[44:01\] I think it actually spins up their server, and then has a Go client that just makes web requests to it, and that's the whole test. So the entire thing is running, and they're doing all this stuff... And I think for some people that would seem a weird way to test it, but for them, it gives them the coverage and it allows them to pinpoint what's working and what's not... So I'm like, "If that's what works, don't let others dictate what you should be doing."
+
+**Ben Johnson:** Yeah, one thing I'll say on the black box testing - I don't see a lot of developers doing this, but when you write tests, you can actually make your test package different than your regular package. You could just do \_test. I do that a lot, but I don't see a lot of other developers doing that... But that kind of forces you to actually stay black box testing.
+
+**Jon Calhoun:** I've seen that, and then I've also seen occasionally the export -- they used to make an export file, and then they'll assign variables to unexported functions, so they can access them inside of the black box testing, almost defeating the purpose of it... And I'm like "I don't think you understood why we did this..." But it's okay, I guess they're trying... It's just funny, because I think when they do that, they're doing it because they see other people doing it, and they don't fully understand the reasoning behind it.
+
+**Ben Johnson:** Yeah, it makes sense.
+
+**Kris Brandow:** Yeah, I feel like the word "unit" and "unit test" has also been something that's kind of gotten a little bit out of control. I think about a month ago I was on this spree of watching YouTube videos of Kent Beck and other people that came up with test-driven development, and all these other ideas... And it's quite apparent that what they were thinking about when they talked about unit tests or test-driven development - it's not what I think a lot of us have come to associate with those terms... And I think that leads to confusion. For me, it's definitely one of those "How did we get here?" moments. Are we doing these things just because everybody else was doing them, not because they're things that actually make sense for us to do. I think it's real tough to get away from that herd mentality of using technologies or using test methods just because everybody else is using them.
+
+**Jon Calhoun:** Yeah, I think that's a hard one... I see so many people that just assume you need to learn programming and do test-driven development because it's a hot term... And I don't know, for me that was a mistake, when I tried to just do purely test-driven development. It just never worked for me. So I don't know if it's just me, or what - it's just I could never write tests first for everything before I wrote code. I'd sometimes have to get a basic implementation, then I'd go write some tests that I wanted, and some of those would fail then... But I kind of needed a basic thing to wrap my head around before I could write tests.
+
+**Aaron Schlesinger:** Just real quick, I'll share a dirty secret that I have... I've probably written the tests before I wrote the code like ten times in my career... It all seems to work out for me. But I think the reason I have to do it that way is I only know the user experience that I want. I don't know yet the ins and outs of exactly what the implementation needs to be. I keep the user experience in mind while I build, but after I'm done building and I've found all those warts in the edge cases and everything, that's when I have to go back and really think about how I'm gonna test this thing.
+
+**Kris Brandow:** Yeah, I feel like TDD is really a stand-in for a certain phase of the software engineering lifecycle. \[unintelligible 00:47:19.17\] when you're actually writing code and trying to figure out how you should build something. I think TDD works really well for some people, but I think there's a bunch of us that are just like "No, we really like to think about it in our head, and then just write the code that does that, and then confirm what we thought would work with tests afterward. I think that's a completely legitimate use case. Usually, you'd get yelled at if you try to push for that, but I think both ways can lead to really high-quality and good code. I think it's more about the process of actually thinking through what you're going to build before you build it, or as you're building it, more than the specific practice that you're doing.
+
+**Aaron Schlesinger:** \[48:04\] I see that kind of a guideline - I see that elsewhere in development, too... One that popped in my mind is REST API, the concept of REST APIs. I know it's not actually -- for a long time it wasn't actually specified anywhere, but it seemed like there was this tribal knowledge, just like there is with testing, of what's "right" and what's "wrong"... And I definitely feel that sort of pressure in both scenarios, where it's like... For TDD, for example, "Oh, I'm a bad person for not writing tests before I write code." I think it's hard to really realize your style of writing code before you really start following the trends, or even evaluating the trends that you hear.
+
+**Jon Calhoun:** Along with your style, I think everybody has a different level of tolerance for mistakes and bugs... Earlier we talked about deploying and needing no downtime deploy... But for a lot of us, that doesn't matter. I could take many of the apps I'm running and I could restart them at midnight, and even if it took ten minutes to restart, nobody would know except for me. And we have this idea that we have to do these things, like that there's one way to do it almost... So I guess what I'm saying is with testing I've found that there's a lot of stuff that I just don't bother testing, at least well, because I'm just not too worried about it breaking, and my tolerance for that is pretty high... But a lot of that is because I'm the person who fixes it if it's broken. It's just me working on it, and most of these are projects that just aren't that important. If they go down -- I've got a fly in my office, apparently... That's what I get for opening windows in the hot summer.
+
+But you know, if you've got that tolerance, that's okay. But if you're running a bank, maybe that's not so much the case. I think people need to acknowledge that sometimes it's a mistake to assume you need to test everything thoroughly, and sometimes it's a mistake to over-test.
+
+**Kris Brandow:** On that too, I think sometimes we don't really define what we mean when we throw out numbers. People say "Oh, we want four nines of uptime", or something like that. But that's just part of the metric. You need more qualifiers than that, and I think that also makes it really hard to determine when you're actually in compliance with something, or where you can make mistakes, or where you can ease off of these high restrictions. I think a lot of the time, no matter how many nines someone usually says -- or business people tend to think of it as like "Oh, 100%! It should always be up." And I think by not defining those tolerances more concretely, we do ourselves a disservice.
+
+So instead of saying "Oh, we have to have four nines of uptime within this window of time, at the minimum", so you can actually properly track it and understand that that's what you're actually doing.
+
+**Aaron Schlesinger:** I had seen some pushback on that concept of four nines. It was like, someone said you're allowed to be down for some number of seconds per year if you've got four nines of uptime, and that illustrated to me that with any complex system you can claim uptime pretty much whenever. You can say "Oh, the user service was up, even though the website was down." Or "The database was working, even though whatever else was down." So I'm really with you there, that you have to really qualify what you mean in detail by four nines of uptime, or whatever else it might be.
+
+**Jon Calhoun:** I've seen some that claim they're up like you said, but for a subset of users it's down, and it's like "Well, what does that classify as?" Or for a subset of users they're just timing out, and it's like "Well, we're technically up; we're just taking forever to respond..." I'm like "Well, that's not useful." So there is a lot there that's tricky to figure out.
+
+\[52:05\] And when you were saying the pushback, I think one of the ones that always was interesting to me is that if you're relying on other services that also have four nines of uptime, if you just have a couple of those and if any one of those goes down it counts as downtime for you, well then all of a sudden you pretty much have to be perfect. It's like, "Well, that's not possible."
+
+**Ben Johnson:** Yeah, I think performance is a bigger metric that I don't think people focus on as much. It's just one of those grey ones, where it probably helps you a lot more than actual uptime. Being down for an hour a month or something like that - that's not great. But if your website takes two seconds to load, or your whatever app call, I feel like your users are just gonna get tired of that. That's their actual real experience for most of them.
+
+**Jon Calhoun:** It's also probably important to pay attention to when these things happen. I imagine if you're Stripe and you're down during Black Friday, even if it's only for ten seconds, that ten seconds is way more important than being down for an hour on some day at 1 AM when very few people are shopping.
+
+**Ben Johnson:** Yeah, I bank at Chase, and they're down all the time... I'm still with them. \[laughter\]
+
+**Aaron Schlesinger:** You're captive audience for it... \[laughter\]
+
+**Jon Calhoun:** They can claim 100% uptime as long as somebody is in the bank during business hours. Is that how it works?
+
+**Ben Johnson:** Something like that.
+
+**Kris Brandow:** I think that in certain parts of the industry we think that we need to have this ridiculous uptime, or these ridiculously high experience metrics for our customers, and then you go look at things that everybody in the world depends on, and it's like "Oh no, we're just gonna take our banking site offline for four hours while we do an upgrade." If something that important can go do that, do we all really need to be able to do seamless upgrades all the time? Is it really worth the trade-offs of having people wake up at weird hours of the night to accomplish some of these goals that we have? We don't really need it.
+
+**Aaron Schlesinger:** But those technologies that enable it are certainly shiny...
+
+**Kris Brandow:** Yes, they are...
+
+**Aaron Schlesinger:** Kubernetes... That's all I had; they're shiny and they get a lot of attention, and folks pick them up.
+
+**Jon Calhoun:** It's because it's kind of boring to talk about SQL at this point... And even writing about it, despite the fact that you can help people, it's hard to justify writing, because you're like "There's a million articles out there that already teach these things." At the same time, we've got people who look at an article and they're like "Oh, that was published a year ago. It's no longer good." And it's like "That's not how all technology works..."
+
+**Kris Brandow:** I've definitely noticed that with myself, where I look for books or something and I'm like "Oh, this book was published four years ago. It can't have any good information in it." It's like "No, no... Actually, knowledge does not change that fast for everything." It's just for some specific things... Kubernetes, where it changes very rapidly and old information can be not as good.
+
+**Aaron Schlesinger:** I just stepped into the JavaScript world two weeks ago, and man - if we thought it was fast-moving here... I feel like I have to know the JavaScript calendar before I can figure out what framework to use.
+
+**Jon Calhoun:** One of the things that's interesting to me is when I look at a Go post, usually all you need is the minimum version of Go, because that's when the feature is introduced. But with JavaScript, you almost need an exact range of like -- you've gotta be between 16.12 and 16.16, otherwise this doesn't work.
+
+**Ben Johnson:** I wrote a React Native app two years, and I can't even get it to run. I can't get it to light up, I can't do anything anymore. It's gone.
+
+**Jon Calhoun:** That's a hard thing to explain though... Because I think people associate dates so much with that, which - I write on a blog enough that I see that, where they're like "You should put dates", and I'm like "What I really should do is go back and write what version of Go I used for this, and which ones it's good for", because that's really all you care about. But sadly, that's a lot of work.
+
+**Ben Johnson:** \[56:07\] One other topic I'd be curious to hear other people's opinions on - I've been debating about this... So I have a new database project I'm working on, and I'm debating about open contributions on open source projects. One other database project, like SQLite \[Sequel Light\], for example or SQLite \[SQ Light\]... I think it's actually SQLite \[SQL Ite\], as I learned from a previous Changelog podcast... They actually don't allow people to contribute code. You can file issues, you can talk to them, they have a community, but they don't actually take contributions, in part because it's a lot of overhead and just maintaining other people's code, which may be coded differently than yours, and you start mixing styles in your codebase... So I'm debating whether I open up contributions or not. I'm curious other people's thoughts on that.
+
+**Jon Calhoun:** I will definitely say that one of the biggest mistakes I made in the first popular open source project I had was that I just accepted anything and everything, without fully thinking it through. And even if something has tests, that doesn't necessarily mean you should accept it. So I definitely see the restrictive side of it.
+
+I think if it were me, I'd probably lean towards the - it's okay to tell people to follow your styleguide, it's okay to maybe say "I'll accept this, but I only wanna include these things, because that's all I'm comfortable maintaining..." Like, to basically restrict what you're willing to accept. But I would probably still accept code if somebody was willing to jump through those hoops, for a lack of a better way to put it. But I completely disagree with -- if somebody has the idea that you should accept any PR, and "I'm gonna submit these things, and you should take my feedback", I think that's kind of a bonkers viewpoint.
+
+**Kris Brandow:** I definitely agree with that as well. There's a dance you have to do, or I guess a dance you probably should, around which contributions you accept, and the level of effort people will put in... Because there's lots of times where people just do a drive-by PR, where it's like "Hey, I have this problem, and I fixed it so it worked for me, so here's a PR", and it's like "Well, no. There needs to be equal effort on both sides if we're gonna actually accept something."
+
+But I really think having other contributors can be useful, especially for people that are really invested in the project and want to help you maintain it and want to help you develop it. So I think if you do eventually (or ever) want to have help building the project, I think it can be good to have contributions... But I think there probably should be a bar, especially if it's anything complex, like what you're talking about, before you can actually get your code in... Show that you actually understand how things are working, and actually want to be a part of the contributor list, and just someone that's just like "Yeah, well there was this tiny bug that I fixed... And I don't care about your styleguides, I don't care about actually understanding anything at a deep level."
+
+But I also understand the cost of having to sit down and analyze other people's code to see if it does fit, to see if it has fixed everything that they're purporting it to fix, and it doesn't break other things... So yeah, I think it's probably a case-by-case basis.
+
+**Aaron Schlesinger:** I've seen some repos that just set an expectation upfront that they'll code, but they might close PRs without acknowledging it, or they might take forever, and then ask for fixes a month later on a PR... And just kind of set the stage that you can try, but it's not guaranteed that you'll get on their review right away.
+
+**Jon Calhoun:** \[59:32\] Generally speaking, I would say that I think as contributors, people should try to reach out before they write code. Kris, when you mentioned the drive-by PRs, I think part of that is junior developer -- maybe not junior, but people who don't really know any better, who are just like "I'm just trying to help." Because I know when I was really early in my career I had a library that did that, and I submitted a PR, and they basically said "We're not gonna take this for these reasons." And they were really nice about it, but I just didn't really get it at the time... But now I completely understand that it probably wasn't gonna help them in the long run. It would have just caused more issues. So I definitely think there's a lot to balance there.
+
+And I do agree with what you were saying, Aaron, that reading that code is definitely hard, even though -- like, doesn't GitHub have a feature now that you can allow the person you're submitting a PR to to modify it?
+
+**Ben Johnson:** Yeah, you can do that.
+
+**Jon Calhoun:** I was gonna say, even that would help, but you still have to read through the code and understand it, and I think that's a balance of -- you almost have to communicate that if it's easier for me just to implement the fix myself, that's what I'm probably gonna do... Versus reading and understanding your code, and trying to help fix it to the point that it needs to be. But I think that's hard, because there are also open source projects where -- there's one in Go, I forget what it was; it was like an HTTP framework, that essentially people claim that they just took submissions, and they copied the code and submitted it and didn't give anybody credit for it, and people got upset with that... And I get why they got upset, but I don't know how you balance that, I guess is the best way to put it.
+
+**Kris Brandow:** Yeah. There were a number of times when I was working on the Go driver, where someone would be like "I want this feature" and they'd submit the code, and be like "Oh, actually there's this way we can add this interface over here, put in this extra flexibility that allows you to do what you wanna do, and also a whole bunch of other people to do what they wanna do, so we're gonna do that instead." That's definitely a good way to go about those types of things, but that also, once again, takes a lot of time and energy and effort... And especially when something is a side project, I think it becomes difficult to weigh those costs well.
+
+**Jon Calhoun:** So let me ask you this, Ben - are you planning on getting a community of active developers around the project?
+
+**Ben Johnson:** Using it, or contributing to it?
+
+**Jon Calhoun:** Contributing, I guess.
+
+**Ben Johnson:** Not necessarily. I guess my experience with writing Bolt was that there were a lot of contributions that were usually small fixes; there weren't a lot of complex contributions. So it was almost kind of a trade-off. And even if you'd get a complex contribution, you've gotta wrap your head around how they implemented it, and there's some of those where, looking back, I probably would have done it a little differently, to keep the same code style, and whatnot.
+
+So yeah, I mean, as far as contributions, I'm still on the fence about it. I'm just kind of leaning towards closed contributions, but open issues.
+
+**Jon Calhoun:** I guess the way I'd kind of view it is if you're building something like Go, where you want more than just your core team, you want other people actively involved in the project and continuously writing code for it, I think the upfront effort of reading their code, understanding their code, teaching them styleguides and that sort of stuff makes sense... But if it's a project where that's not going to be the case, it's gonna be one or two contributions and that's about it, then I think you're completely justified in saying "Just give me some issues and PRs and I'll prioritize and fix them myself."
+
+**Ben Johnson:** Yeah. I'm just curious how that lands in the community, as far as closed contributions. I haven't seen any other Go projects do that so far.
+
+**Aaron Schlesinger:** There was one, it was an implementation of WebPack. It did the same thing as WebPack, bundling up JavaScript... I wish I could remember what it is. I can try and dig it up later. Anyway, at the bottom of that readme it said "This is alpha software. I'm working on it, but I don't plan to take any contributions."
+
+**Jon Calhoun:** That's probably a challenge with GitHub in general, is that in some ways you wanna share code, and in other ways -- like what people view as open source is kind of different, and it can't be the same thing to everybody if they all have different expectations. I think we even saw this in the Go community recently, with them changing language in the internal stuff, and people flipping out... And it's like, "If you aren't involved in the project, I don't really see (at least me) how your opinion is relevant."
+
+**Aaron Schlesinger:** \[01:04:03.05\] Yeah, I think there's nice ways to say that you're not interested in other people to contribute, but you are interested in people to use it. It's just a matter of...
+
+**Ben Johnson:** Finesse.
+
+**Aaron Schlesinger:** Yeah, that's the right word. \[laughter\]
+
+**Jon Calhoun:** It's especially hard too, because I think people view open source as "Just submit code", and in every successful open source project I've seen that's not the way it actually works. You don't just submit code. You talk about an issue first. And then eventually, once you've sort of talked about how you're gonna implement it, you can do it... But it's very rare that just submitting code is a good way to go about it.
+
+Alright, so we're a little bit late, and I forgot to do the Unpopular Opinion segment. I don't know if any of you guys have prepared for this, but usually we'll ask our guests for an unpopular opinion...
+
+**Jingle:** \[01:05:00.03\] to \[01:05:15.14\]
+
+**Jon Calhoun:** So does anybody have one they'd like to share?
+
+**Ben Johnson:** The closed contribution one...
+
+**Jon Calhoun:** I think for any maintainer it's not that unpopular...
+
+**Ben Johnson:** That's fair, yeah. That's fair.
+
+**Jon Calhoun:** ...but for anybody who's looking at open source, it might be.
+
+**Ben Johnson:** Yeah. I would say -- I don't know. I feel like most new technologies are just not necessary, I guess. That would be my unpopular opinion. Docker, honestly, or Kubernetes, and a lot of those - I feel like you can run a stack... Not the same stack, but you can run basically what we used to run back in the '90s or early 2000's, where it's just like "Here's a web server running." You can run a business off of that and it probably would be fine. I mean, obviously, back up your data, but... That'd be my unpopular opinion.
+
+**Kris Brandow:** I think for me an unpopular opinion I have is... I guess once your organization, or I guess the engineers in the organization reach a certain level, that you shouldn't really just take software from other companies. I think gRPC is a big one for me, where it's just like once you have a group of people that understands how to build things with TCP and HTTP, you could probably just build it yourself. And you should do that, because your organization's needs are gonna be very different than what Google needed for when they built gRPC.
+
+So I fall closer to that -- what used to be called "Not invented here" syndrome, and I think that's probably where we should be edging back to, but I realize that that is wildly unpopular with a lot of people, and they usually say "Just use whatever is out there, because that's better."
+
+**Aaron Schlesinger:** Well, I mentioned REST APIs earlier... I don't think they're good. That's my unpopular opinion. I think they cause more confusion than problems they solve. I do think URLs and HTTP methods are good, but the pattern of having specific rules for specific types of requests, and we also mentioned putting things in the right place in URL paths - it's all well and good, but documentation can solve those problems really well, and a good SDK on top of the REST API... Well, the API; maybe not REST API. \[laughs\] Documentation and good SDKs can really solve those problems a lot better than some magical document or tribal knowledge floating around somewhere.
+
+**Jon Calhoun:** I would definitely agree that the best APIs are the ones that you don't know if they're running JSON -- by JSON I mean like a REST API, which is JSON responses... GraphQL, XML, or whatever - if their library is good enough that you don't need to concern yourself with that stuff, that's almost always the best API experience.
+
+**Kris Brandow:** You mostly don't have an SDK for your language then...
+
+**Jon Calhoun:** Well, then you're just not gonna have a good time. But the best ones have all those things down. \[laughter\]
+
+**Aaron Schlesinger:** \[01:08:01.18\] I think even then, if it's all Post and Get with query strings and post bodies, every language can do that pretty easily... And I think the hardest part in most SDKs is figuring out OAuth anyways... So you're probably wrestling with that more than...
+
+**Ben Johnson:** I agree with that, yeah.
+
+**Kris Brandow:** I guess on that REST API thing, too - it feels like we lost the spirit of hypermedian REST, and just kind of started looking at these strict rules that everybody laid out... I think that's what causes a lot of the pain.
+
+**Jon Calhoun:** I think what makes it hard for me is that I don't truly know exactly what a strict, by the definition, REST API is... But I think the idea of separating things by resources can be useful, but I definitely think there's times where it's not. One example is when I was building the backend for -- so I have Go courses, and every course has sections and sections, basically chapters, and chapters have individual sub-chapters or lessons or whatever you wanna call them.
+
+If you're looking at them as strict resources, it's be like -- you know, a course would just return the course, and then you have to call an API end point to get all the sections, and another one to get all the lessons... And separating resources that way didn't make sense from a REST API standpoint... But at the end of the day, what really made more sense is four end points that just return the data I need for different pages... And to me, that made more sense, because it's what I needed. But I could see if you get caught up in this, you have to follow REST, do exactly what it says; that would be kind of rough.
+
+**Kris Brandow:** We do that a lot in our industry, whether it's Agile, or TDD, or whatever it is... There's really good ethos, that's just like "Hey, we're just gonna help you make something", and then people just put a bunch of rules around it, and then if you don't follow those rules, it's like "You're doing it bad!"
+
+**Jon Calhoun:** I feel like a lot of that stems from this desire to make things generic enough that it can work for anything... And that's usually not useful. I think we do that with even like databases, where the interface we interact with is generic enough that it works for everything. At the end of the day you're like "Why are these queries so slow?" It's like, "Well, it's because of the way you wrote this." It's not possible to actually realize that you need to do a bunch of joins and do them properly.
+
+**Kris Brandow:** Yeah, I think -- to go back to something Ben said earlier about learning the underlying technologies... I think sometimes we try to create all of these interfaces and these nice facades that really mix a bunch of layers together and create these very complex things we have to deal with, when it'd be much simpler if we just properly layered everything and just had very simple layers that we interacted with.
+
+**Jon Calhoun:** Alright. Thank you, Kris, Aaron, Ben, for joining me today. This has been a good episode of Go Time. If anybody else has comments or anything like that, feel free to reach out in the GoTimeFM Slack.
+
+**Ben Johnson:** Thanks for having us.
+
+**Kris Brandow:** Thanks for having me.
+
+**Aaron Schlesinger:** Thanks a lot.

WebRTC in Go_transcript.txt

@@ -0,0 +1,441 @@
+**Mat Ryer:** Hello, and welcome to Go Time. I'm Mat Ryer. Today we're talking about real-time communication, specifically WebRTC and how we can get peer-to-peer secure communication going in Go.
+
+As usual, I'm joined by a cohort of wonderful and strange people. Firstly - calm down, everyone, it's only Jaana B. Dogan. Hello, Jaana!
+
+**Jaana Dogan:** Hello! Hi!
+
+**Mat Ryer:** Welcome back.
+
+**Jaana Dogan:** Yeah, good to see you again.
+
+**Mat Ryer:** You too, welcome back. Don't worry, you don't just have to tolerate me, you also have to tolerate Jon Calhoun. Hello, Jon.
+
+**Jon Calhoun:** Wow. What an intro.
+
+**Mat Ryer:** It's a nice one, isn't it? Well, it's different.
+
+**Jon Calhoun:** You have to tolerate me, that's what I'm here for.
+
+**Mat Ryer:** \[laughs\] And we're also joined by a special guest today, Sean DuBois. Sean, you founded the Pion project, is that right?
+
+**Sean DuBois:** Yeah, and thank you for having me. So it's grown organically. I wouldn't really say "found". I committed 50 lines of Go code and I had no idea where it would go, but... Yeah, that's what my life has been for the past 2,5 years now.
+
+**Mat Ryer:** We will get more onto that later, talking about what that project is, but maybe we could start with -- for anybody not familiar with WebRTC, what we're really talking about and what you can do with these kinds of technologies... Maybe you could give us a bit of an overview.
+
+**Sean DuBois:** Yeah. So if you are a user of WebRTC, what you're probably used to is "Hey, I have two web browsers and I wanna send video between them. You wanna build a conferencing room application. That's the common use case, and that's what people first start with... But really, the technology underneath is so incredibly flexible and does so many things that you can build all these different use cases. I don't know if you've seen that Google Stadia is over WebRTC, people are doing web torrent over WebRTC, so you can download torrents in your browser... Another one is co-streaming; you'll see a lot of people that will stream videos together, like they'll perfectly sync up and distribute through WebRTC.
+
+\[04:15\] So really what the magic behind that is -- the first is the peer-to-peerness. WebRTC allows two people to distribute a list of their known addresses. So I give you what my local IP address is, or what my public IP address is, and then we exchange this list and we find the best way to talk to each other. And that's the really cool magic behind it.
+
+Most people may think like "Oh, I wanna exchange a file. I have to upload it to S3 and then I download it." But with WebRTC we exchange our list of peers, and then we connect directly to each other via this really cool thing called NAT Traversal, where you both punch temporary holes in your public IP so you can talk to each other.
+
+So if anyone gets a chance and you're googling and following along, the first thing you wanna look up is NAT Traversal, to get you down this rabbit hole of RTC.
+
+**Mat Ryer:** That's awesome. And is it just one-to-one communication, or can you do one-to-many, or many-to-many?
+
+**Sean DuBois:** Yeah, so it's just one-to-one, and that's a super-interesting rabbit hole to go down. Let's say me and you establish our one-to-one communication, but most of the time what we have is like a conference call; you wanna talk to multiple people. So then you end up doing a one-to-one with every person in your conference room. Everyone does all these one-to-ones. So if you have four people, you're uploading your video three times, because you're doing this mesh topology.
+
+That's the value-add of a lot of services like Zoom, and Hangouts, where you upload your video once to them, and then they fan out. But that also adds -- now, because you upload once to them, they have to decrypt your video, and then they fan it out. Essentially, you're in a WebRTC call, but with a WebRTC peer that's running up in the cloud. Your browser just thinks that it's talking to someone, but really it's like this heavyweight server running the cloud that's simulating WebRTC connections.
+
+**Mat Ryer:** I see. So how does WebRTC relate to protocols that we are more familiar with, like TCP and HTTP?
+
+**Sean DuBois:** WebRTC - everything runs over UDP, and it's super-important that it happens, for two reasons. One is the NAT Traversal attributes of it. So if me and you both wanna connect to each other, we can't really do NAT Traversal via TCP. With TCP it's like this one stateful connection. If me and you are both inside our private networks, how do we establish a hole to go through?
+
+With UDP what you do is you send one packet out to what they call a STUN server, and then that STUN server responds and that establishes a temporary hole. So if anyone sends in on port 53000, there's a temporary hole where 53000 maps to your computer. If you ever set up your Xbox back in the day, and you remember having to do port forwarding, and stuff like that; this is basically automatic port forwarding.
+
+**Mat Ryer:** Right.
+
+**Sean DuBois:** So first you get that automatic port forwarding with UDP, and the other thing is an attribute of live video. If I'm trying to watch a live video - and you've probably seen this with Netflix, how it will flip the quality on the fly. That's called HLS, where they will upload a video in four different qualities, and you download it via HTTP. Instead, what WebRTC does is me and you are communicating real-time on the quality of the video. So I will flip between resolutions and bitrates on the fly, because we have this duplex communication. And that's super-powerful, because now I can pick what's the best quality for us to have this call in. For example, if someone else gets on your network and the network quality goes down, I can fluctuate on the fly.
+
+**Mat Ryer:** So would it tell me to start sending it at a lower resolution?
+
+**Sean DuBois:** Yeah, exactly. I send you statistics, like "This is the amount of package you sent me", and UDP, since it does the drops automatically, it says "Hey, Mat sent me 500 packets, but I only got 470." You say "Okay, I'm gonna lower my bit rate a little bit until I can get to a good quality of loss." So it has this really cool congestion control.
+
+\[08:13\] The other cool thing is since the congestion control isn't built into the protocol, like TCP, you can decide on what you want. I can say "Okay, I'll do five seconds of latency, because I want perfect picture." Or "I wanna do 200 milliseconds, because I'm having a live call." There's just all these cool things you can do with WebRTC, because you get this flexibility.
+
+**Jaana Dogan:** So is WebRTC just a protocol, or does it also contain some of the utilities, like some encoders for video and so on? Because you mentioned that you can switch back and forth between different qualities... I wonder if there's any utilities out there.
+
+**Sean DuBois:** Yeah, so WebRTC itself is basically a bundling of all these existing protocols that have been around since the '90s. If you go look up ICE, and RTP, and SCTP - these are things that have been around, and WebRTC bundled it up. You can bridge with existing things. But then it's kind of left up to the user to implement it.
+
+You can build a WebRTC implementation if you can parse RTP and you can send H 264. It's kind of like a soup; there's a couple implementations of WebRTC out there. There's these RFCs that define how it works, but then you go and you can grab -- Python has an implementation, Go has an implementation, there's a C++ implementation by Google... And the cool thing is they all work interchangeably and people are building stuff with them.
+
+**Mat Ryer:** And Pion has a Go implementation of WebRTC, right?
+
+**Sean DuBois:** Yeah. And the thing that excites me about the fact that Go is -- when I started working on it, I just loved Go for how easy it was to do things. I didn't appreciate everything that would come from it along the lines, but by writing it in Go, it's become so incredibly easy for people to contribute and learn, which I think is the hardest problem with WebRTC right now. I'm sure a lot of people are listening right now and they thought "Oh, I thought WebRTC was just so I can do video between browsers." You don't appreciate everything that's happening. But I'm hoping, because it's in Go, you can go and click through the symbols, and look at things. And then the ease of deploy...
+
+And I think the big one that's gonna happen in the future is the security aspects. Everyone is running C and C++. I think that's really scary, to be running that stuff on the public internet. I still do C stuff as my day job, but running AddressSanitizer, and MemorySanitizer, and all this stuff. I constantly make mistakes still, but Go - I can sleep well at night, knowing that I won't find out that every service has been exploited and everyone can see the calls flowing in clear text. You can't convince me that someone out there doesn't have a zero-day we don't know about with existing RTC stuff.
+
+**Jaana Dogan:** One thing that captured my attention - you said "I started with 50 lines of Go code", and I was really impressed... You know, for a new product only implemented in 50 lines. But then I was thinking, "Oh, with 50 lines actually I was able to bootstrap a lot of projects." Isn't that interesting - it's not a verbose thing, but you can get a lot of things done, very quickly sometimes.
+
+**Sean DuBois:** Yeah, so the 50 lines was totally cheating. The 50 lines -- basically, how Pion started is the magic of cgo. And I know no one wants to use cgo, but what I did is all of these hard protocols, like SCTP, and SRTP - I just used cgo to make it happen, and then I slowly rewrote each implementation into pure Go.
+
+**Jaana Dogan:** Yeah, it makes sense. That's such a common case, actually. Lots of people migrate over from C++ or C; if you're migrating from C++, you end up writing some trampoline functions... But you know, you just gradually do it. That's how I started to write Go...
+
+**Jon Calhoun:** It's also so much nicer than "Let's wait till we've refactored or rewritten the entire thing to try it." It's like, "Okay, we can see how this is working, and make sure things are moving smoothly, and actually test it all." Versus the whole rewrite, which is almost always awful.
+
+**Mat Ryer:** \[12:02\] Yeah, and you learn so much by that process. Sean, you just said you did it cheating, but actually, I think it totally makes sense to do whatever is necessary, hack away at something, because the understanding you get is really valuable, but you may not keep any of the code... And actually, sometimes knowing you're gonna throw code away really helps with that process, because you sort of don't worry about any of the practices or things that we do when we're properly software-engineering. You get to not worry about that too much. Does that spirit come easy to you, Sean? We were chatting a bit earlier about your background; I'm interested if that plays a part there.
+
+**Sean DuBois:** Yeah, I think it does. My background was that I don't have any computer science education. I left high school and I worked at a VoIP company. I was lucky enough that they gave me a minimum wage job writing PHP scripts and just kind of hacking things up and learning stuff along the way.
+
+And then also I learned so much via IRC. I would sit on freenode in the PHP and MySQL channels and I would just beg for answers. So now I feel for people who are on \[unintelligible 00:13:08.25\] I will never get upset at stupid questions, because I know I have pained so many people with my stupid questions... But it does come naturally. I've been very lucky to grow up with that mindset. I was never given "Hey, this is what the problem is, and this is how you're gonna do it. Go solve it."
+
+I think Go's community is also really great about that, as well. People definitely don't fall into that "Perfect is the enemy of good" kind of problem. I think with a lot of other languages people are like "Unless it solves every single problem and it's perfect in every way, I refuse to use it." But a lot of Go developers and companies are like "Let's solve the problem we have, and then we'll go from there."
+
+**Mat Ryer:** Yeah, I wonder if it's something to do with the fact that it's relatively new, isn't it? The community is relatively new, so it doesn't have a lot of baggage, or history, or legacy that other communities had. So it probably does allow us. I think also the way that the Go community does issues of diversity, and things like that, making it a really inclusive environment, I think maybe for the same reasons it does the same thing; it's able to take a fresh look and have fresh attitudes of things. It doesn't have all these legacies. That's really great to hear that, whenever people that come from other communities say that about the Go community; I always like to hear it.
+
+**Jon Calhoun:** I wonder if, on that front, some of that comes from how the people use the languages... Because I'm thinking of like Ruby or JavaScript, where people tend to build these big frameworks that literally do everything for you... And you're like "I'm not switching to a framework unless it's doing all this for me", whereas Go has very much been a "Don't use a framework. Here are some building blocks. Put them together." Each one does something small, but it's not gonna give you the whole thing. It's kind of in that mindset of like "Use these building blocks", but you don't actually have to have something that does everything that you might want.
+
+**Sean DuBois:** Yeah. And I love that, because it encourages intellectual curiosity. It's not like "Okay, drop this thing and it just does everything." Everyone's like "Oh, I'm using this little piece, but maybe I can do it better", so they go and rewrite it, and rip it into it. And that's what's been so exciting with Pion. We're getting close to 100 contributors. I think that's my favorite part of the whole thing, is the fact that these 100 people - and many of them are first-time open source contributors. They decided "This is a welcoming environment. This is a problem that I can dig into", and they just dove right in. And the Go tooling, and everything... Yeah, I don't have enough good things to say about how Go encourages people to be curious and get involved.
+
+**Mat Ryer:** It's great. Was it difficult writing a WebRTC implementation in Go? Or did it sort of naturally fit in some way?
+
+**Sean DuBois:** There was some stuff that is/was super-painful. I think the lack of libraries is definitely a hard one. So the big one - Go doesn't have a DTLS implementation, so there's not TLS over UDP, so we had to write one. There was no SCTP implementation, there was no ICE implementation... So it took us probably a year and a half to get to that point. But I don't think that's really a fault of Go. I really love how opinionated Go is.
+
+\[16:06\] I feel as you add more and more people to a project, they all bring their opinions, and it was nice that I could just run GolangCI and be like "No, there's no discussion. This is how it reads, this is how it flows", and that's the end of it. Then we go argue about things that are actually important. I think humans love to argue, they love to have their things, but--
+
+**Mat Ryer:** No, we don't. No, we don't.
+
+**Sean DuBois:** \[laughs\] Yeah... I guess I just picked the most dysfunctional group of people then. I don't know what happened. \[laughter\]
+
+**Mat Ryer:** No, I suspect not. It definitely makes sense.
+
+**Sean DuBois:** Those are the only things that I have to say, but man, the upsides of Go are just so wonderful. The safety of the language, the portability... What's so exciting is that I can do go build, and site architecture, and all of a sudden I can throw my binaries on these little cameras, and the WASM stuff... So that's the other cool thing about PION, is that I can write one binary and I can deploy it to the web, and I can deploy it to my servers, and I can deploy it to my set-top boxes.
+
+There's this one company, Strive, and they do a decentralized peer-to-peer CDN. So if you're sitting in an office building, why are you downloading the same video file multiple times? You should download that video file once and then you should distribute it to all your co-workers.
+
+They have one codebase and it targets Android, iOS, web, servers... It's amazing the amount of things you can target. And it feels natural to me. I'm not frustrated using it not just on my servers. I guess I'm probably preaching to the choir. If you're listening to this show, you probably already love Go, so I don't need to sell it.
+
+**Mat Ryer:** I like to think that we accidentally get onto some people's playlists, and they're too lazy to get up and change it... So yeah, for those people - welcome. Relax. Enjoy yourself. You're in for quite a ride.
+
+Yeah, so it's really interesting then... Let's talk a bit more about the use cases, because -- WebRTC do think of video and audio chat, but are you limited to the kind of data that you can transfer in this way?
+
+**Sean DuBois:** Not at all. WebRTC has two distinct things it can transport. It can transport media, which is audio and video, but it can also transport data. And the data - you can do all these interesting things. You can do out-of-order transfer... So if you're doing a video game, you don't wanna do that head of line blocking. You don't care where the player was 3 seconds ago. You only want the latest packets. And since it's over UDP, you can conditionally say "Okay, deliver my packets out of order to make them lossy", so you get the highest performance possible. So you can just send this binary data and have all these interesting attributes.
+
+And because of these data channels thing that you can do, I see all these interesting use cases. The one that I really wanna see take off, in people doing ops and stuff like that, is getting rid of the bastion server. If you're a small company, you probably run an SSH server sitting on the outside of your network, you jump into that, and then you jump into your internal things. But with WebRTC, what you can do is you can do that temporary hole, and then you could just have all of your servers inside running WebRTC, and then you can hook to them directly without running a VPN, without punching a temporary hole... Because it has all these attributes of NAT Traversal, which I think is super-powerful.
+
+So I'm excited to see what this can do. If I want two computers to talk to each other, they don't need to have public IPs. They can talk to each other without any kind of direct communication.
+
+**Mat Ryer:** Yeah, I really can't wait to play with this. It does have that hackability sort of feel to it, doesn't it?
+
+**Sean DuBois:** Yeah.
+
+**Mat Ryer:** How does it compare with technologies like WebSockets, for example, where you've got the browser making a long-running connection, essentially.
+
+**Sean DuBois:** \[19:58\] The WebSockets has two problems. One is the head of line blocking. Since it's over TCP, it's lossless. So if you're sending something that's large, you're gonna incur that penalty, that like as you do the retransmissions, it's gonna slow things down. And then the WebSockets also require TCP, so you have to be communicating with something that's directly addressable. If I'm using WebSockets, I have to connect to a public IP address, I have to be in the same networks.
+
+But beyond that, using in-the-browser, using data channels and WebSockets feels exactly the same. You just call `.send` and send some stuff and it just magically shows up on the other end. So as an end user, you don't really care, it doesn't really matter. But as a hacker, and doing stuff underneath, there's a crazy amount of power. The one that I shared in the GoTime channel, that's one of my favorites, is the Cloud Games. So someone went and built something where you can play NES games, and all these Game Boy games and all this stuff, and basically they run an emulator on a server in DigitalOcean. And they send all the video frames to you, and then you send all your key press events via data channels back to them. So you can play NES games with someone else, you can do multiplayer, you can share your screen, you can persist your game state... It's basically the open source version of all these game streaming services coming out.
+
+**Jaana Dogan:** There's one thing I wonder, since you've mentioned there's actually a lot of things to consider in terms of trade-offs and so on... Is there any sort of visibility, are there any debugging tools, or what's the good way to start maybe debugging? What do you have in terms of tools at least, at this point?
+
+**Sean DuBois:** That's a tough one. I think that's probably the biggest problem with WebRTC right now - the education and the debuggability. For me, what I do most of the time is I sit with Wireshark, I grab packet captures, and then I export the actual keys from Pion. So if Pion if connecting to the browser, you have your TLS connection. I export those and I decrypt the packets. But for a lot of people, Pion is the way the way they debug. Because you can't really debug what's happening browser to browser. But if you do Pion to browser, I can just print the stuff that comes out.
+
+With Pion, the way it works is you just get access to the video bytes directly. So instead of getting this on-track event and putting in a DOM element, you just get a stream of raw H.264 or VP8 or something like that. In my opinion, Pion is your best debugging tool, because it decouples everything and gives you the ability to peak, and poke, and play with things. And then it's great also for the load testing, and attacking your own servers, and figuring out where your rough points are, because you can spin up -- before, people would spin up 10,000 browsers to simulate what a load for a WebRTC service would look like. But now they can spin up 10,000 little Go processes that send precanned video.
+
+**Jaana Dogan:** Yeah. Even from the perspective of learning, that would be useful for lots of people to just begin looking at, so I'll try to check them out. Thanks.
+
+**Sean DuBois:** I think that's the thing I wanna encourage most out of this... Pion is an idea; it's not like a single software project, but it's like "I wanna get people into it, and I want to teach them that peer-to-peer is possible, decentralization is cool, both from the resiliency it brings to the internet, and for what people can build." And I want people to learn, I want people to get excited and bring a little joy to their lives.
+
+**Mat Ryer:** I think that's great. I love these kinds of projects as well, because they're sort of borne out of either a love for something, or a real problem that you have that you wanna solve. And they're always the best projects, I think. When you see open source projects that are addressing something real, or built with that sort of passion, you can always tell, compared to where we've kind of set out to try and deliver a package, and imagine something.
+
+\[23:57\] So I always think that's always worth pointing out, when any these projects are always like -- some people might consider the WebRTC API to be quite a boring read... I did read through it, and it's not an easy thing to read... So having this abstraction as someone that likes to play with different technologies and build little things is great. So thank you very much for starting the project and continuing your support and involvement, Sean.
+
+**Sean DuBois:** Yeah. For me, Pion was born out of frustration with -- at the time, there was a C++ implementation... So when WebRTC came around, I wasn't involved, so this is all anecdotes or whatever I heard from other people... It's that Google wanted to bring WebRTC to the browser without plugins. So they went and bought this company called GIPS, and they went and bought On2... So they acquired all of these very great, very battle-tested software stacks and they've put them in the browser.
+
+The hard part about -- if you've built Chromium, you know it's a very painful process. There's all these proprietary build systems... It's not what people are used to. And WebRTC kind of suffers from the same thing. So with Pion, my goals was also like "It's one go get away." Which is another fantastic thing about Go - if you wanna play with something, it's one go get. It doesn't matter what platform you're running, there's not "Install these libraries/Play with all this." It's like "How can you get stuff into people's hands?" So that's where the passion originally came from.
+
+Now the passion is like "How can I get as many people using WebRTC? How can I get as many people into this system and have them owning their projects and getting them excited?"
+
+**Jon Calhoun:** The go get thing is -- I feel like it's something people don't emphasize enough. I see people all the time who are like "If you write blogs, you need to put dates on them, because it goes out of date really quickly." And I think a lot of them come from backgrounds like JavaScript or something, where just trying to install something all of a sudden won't work anymore, and you're like "I don't know why, it's just not working." Whereas, like you said, with go get and everything else - they all tend to work, to the point that it's like "I really don't need to date most of these things, because they shouldn't break."
+
+Unless you're using a third-party library and they make breaking changes, or release a major version, or something. But that stuff tends to be pretty rare, compared to other languages where that is more problematic. Or like you said, they have really big build processes, where so many different things can go wrong.
+
+I think I talked to -- maybe I told you about it once, Mat, where I was working on a project where basically something in Brew got removed (Homebrew, for installing), and as a result, nobody on the team could rebuild on a fresh laptop to get everything up to spec, because we just couldn't get some certain library, or something. And it got to the point where we ended up having to do a bunch of work to move a bunch of things to upgrade stuff because of that... And that's a really painful process. So having those complicated build processes is a challenge.
+
+**Sean DuBois:** And also it's like that broken Windows thing, where even -- a lot of people will say... There's this kind of arrogance of like "I don't care about new developers. I only want seasoned developers to come use my project, because I don't really care. This is like a full-time job etc." I think that's what a lot of these projects suffer from.
+
+But even if a seasoned developer comes and uses your product, if it's hard to use and hard to set up, it just gets people on an awful mindset. You just start off using my thing upset, which I never want. I want people to be like "I had a wonderful Tuesday afternoon playing with WebRTC."
+
+**Jon Calhoun:** A good way to put that is they wanna spend their time learning WebRTC, not figuring out how to install software, which they've probably already figured out a million times before... Even if they're seasoned, they've figured it out, and nobody enjoys that.
+
+**Mat Ryer:** And just to be clear, you can do this on any day of the week, can't you, Sean?
+
+**Sean DuBois:** \[laughs\] No, Tuesdays only.
+
+**Mat Ryer:** Tuesday afternoon only, okay.
+
+**Sean DuBois:** Tuesday afternoon, yeah. \[laughter\]
+
+**Mat Ryer:** Yeah, that's quite an interesting way of rate-limiting your code.
+
+**Jon Calhoun:** It's got a whole build process that tears down the whole repo. Everything just disappears, then it comes back up again...
+
+**Break:** \[28:06\]
+
+**Mat Ryer:** That's interesting, I think other projects could learn that lesson about the developer experience. There's a few little tricks you can do in Go. I like doing them like putting the test code in a separate package, so that you are accessing your package from the outside it seems, in your test code. When you do that, you get a kind of feel for your API. You're kind of experiencing as your users are gonna really experience it. And having that focus I think helps every project. Go, because of its minimalism, kind of encourages that... Because if minimalism is given, then you have to be really selective about what goes in, so then you really have to think about it.
+
+So just having that sort of restrictive minimalist mindset I think does help us, and it's kind of trained in Go to think like this. Obviously, people -- you know, the languages do the same kinds of things, and none of the ideas are really necessarily even original, but it did get an early focus, and I think you can see signs of that.
+
+**Sean DuBois:** That's my biggest fear with Go ten years down the line, that it's so easy to add things, but it's impossible to remove things. So I hope that Go keeps that minimalism. I think above the Go... I love that the Go community -- you write codebases, you throw them away, you rewrite, you do new things, but the line where Go starts... You can't change that forever. It's like an unbreakable contract. So I hope we keep that minimalism in the language and then allow the explosion of crazy ideas outside of it.
+
+**Mat Ryer:** Right, yes.
+
+**Jon Calhoun:** Earlier you talked about NAT Traversal and connecting to each other... So that very first thing, where they tell each other IP addresses - or I assume it's IPs; I'm not sure.
+
+**Sean DuBois:** Yeah.
+
+**Jon Calhoun:** How do they communicate that sort of thing? How does this whole process get kicked off? Can we get into the details of that?
+
+**Sean DuBois:** Yeah, yeah. So what most people do - they call it signaling. It's two peers - they need to know the bare minimum details of each other. Most people will run a server up in the cloud, just like a little WebSocket server, and so you send up your details, I send up my details, and they get sent to each other. So there's like this little exchange. And you can do signaling via any protocol. Basically, I'm just sending a blob of key-value pairs that say "This is my IP, these are the codecs I support", and that's it. So it's just like "How do I get the minimum amount of information to each other?"
+
+I've seen people that do signaling via WebSockets, I've seen people do signaling via IPFS, I've seen signaling via HTTP. You can do really whatever you want.
+
+\[32:13\] So once you exchange that amount of details, you then go into this next step, which is like a full protocol called ICE, or internet connectivity exchange, where I have my list of details from that minimal blob that we've exchanged, and in time we just hit each other with pings and pongs to find the best route.
+
+**Jon Calhoun:** So when you're doing that pings and pongs and stuff - I assume it's trying to find the quickest connection, in the sense of like if you and I are on a local network, it'll try to use that instead of going out to the web, that sort of thing.
+
+**Sean DuBois:** Yeah, and that's the really cool magic about it, because there's also other attributes. What if we're on the local connection, but our local connection has a bunch of packet loss, but our web connection is actually better? So ICE can evaluate and say "Oh, even though this connected faster, maybe I wanna do this one", and then ICE also allows you to switch what they call "candidate pairs" at any time.
+
+So let's say me and you are talking via Wi-Fi, and then I walk outside - I need to switch the IP address I talk to you on. I now need to switch to cellular. And then when I walk back inside I switch to Wi-Fi. It has all these really cool things where it can measure the network cost, it can measure the roundtrip time, and all these interesting attributes to figure out what's the best candidate pair for this call.
+
+**Mat Ryer:** That's amazing. That's brilliant.
+
+**Sean DuBois:** Yeah.
+
+**Mat Ryer:** And does Pion do this?
+
+**Sean DuBois:** Yes. I just went and implemented a spec, which made it really easy. This is all magic that has existed in WebEx, and Skype, and everyone else. Everyone's doing this, so I'm lucky I get to just go to the IETF and read the plain text and I'm done.
+
+**Mat Ryer:** Yeah. You've kind of unlocked this though really for the Go community, because even if I'd just implemented the bits I needed, this isn't the kind of problem that you really have a chance in hell of being able to just build, and each person solve it. It's not like a router.
+
+**Sean DuBois:** Yeah. And it's fun, because different companies and different individuals have come and brought their little improvements. The first ICE implementation was super-rudimentary. It said "Whoever connected first, that's who we pick." But people now define how long they're willing to wait for an internet route, compared to a land route... All these different rules. It's like, "How can you build this best ICE agent implementation possible?"
+
+**Jon Calhoun:** When you were talking about connections that sort of -- you know, you jump from Wi-Fi to your cell network, and stuff... Does that mean that that's probably the type of technology that's being used behind Google Fi, that sort of thing? Or does that seem like a good use case for that, or would that be bad? I know that you can't say for sure what they're using, but...
+
+**Sean DuBois:** I have no idea, but I think Google Fi is probably a level lower. I know Duo and stuff like that is definitely using things. Because Duo is just WebRTC, and Hangouts is just WebRTC. But I think Fi is probably a layer lower. I know nothing about electronics or any of that stuff, so... I'm just a simple software engineer.
+
+**Jaana Dogan:** I think it's more of like an L2 type of switch...
+
+**Sean DuBois:** Yeah, that's my favorite part of WebRTC - the reactivity of it. Being able to choose "What's the best bit rate? What's the best connection route?" I like to picture it in my head like the little decentralization thing, where it shows a graph, and picking the best route to talk between two people.
+
+**Mat Ryer:** Yeah, it's nice that it does that for you... But what's that interface like for a Go programmer? How do you access that kind of capability? What does it literally look like in code?
+
+**Sean DuBois:** Yeah, so for WebRTC itself, all you do is you generate that bootstrap blob, and then you basically call "create offer", and I send my offer to the other side. Then the other side calls "create answer" and then sends the answer back, and you're done. You now have a full peer-to-peer communication that does all the things.
+
+\[35:57\] But we also expose all of the underlying technologies as their own individual packages. So this ICE things, it just implements -- I think it's an io ReadWriteCloser. So I push on a list of my remote's IP addresses, and I send my remote a list of my IP addresses, and then they just find each other, and that's it. You can slap that in front of anything.
+
+I forget -- I know some people have slapped VPNs on it. At WireGuard someone built an implementation, they call it WG VPN. If you get a chance, go look on the awesome-pion repo. But they took the Go implementation of WireGuard and married it with ICE, and now you have a NAT Traversing WireGuard implementation.
+
+**Jaana Dogan:** So in terms of packages, what is an application developer's responsibility, in terms of let's say the buffer size? Or what are some of the points left for optimization?
+
+**Sean DuBois:** Right now, what you do is you listen for data channels, you listen for the back pressure. Since it's like an async protocol, you can call `peerconnection.send` and send it a gigabyte, and then it will slowly make sure that data gets to the other side. But what you do is then WebRTC has this callback that says -- I forget the exact name, but basically you listen for the back pressure, and it says like "Okay, I'm done sending everything you've requested that I send. Go ahead and send more." And then WebRTC also has an API called getStats, where you can poll and see the quality of the connection. You can be like "Hey, this is like the average bandwidth between you and the other peer." So you can choose what you wanna do.
+
+**Jaana Dogan:** Yeah, so on the fly you optimize things... That sounds really cool.
+
+**Sean DuBois:** Yeah. The other interesting thing about WebRTC - or I think some people are frustrated by - is the... In Pion's implementation we took the WebRTC API that's dictated by the browser -- so the API isn't very Go-idiomatic. It's very callback-heavy, it feels very JavaScript. But it has two benefits. One is if you're already comfortable with writing WebRTC in the browser, you can go use Pion and you feel very comfortable.
+
+And two, we learned from all the research and all the hard lessons that people learned writing that API. I think it's so easy to jump in and build an API, but to have an API that has ten years of experience in the browser - we got a lot of stuff for free, because all of a sudden someone comes and they're like "I have this crazy edge case I need to solve." I'm like "Well, WebRTC luckily solves it in the corner, with this knob", and I would have never anticipated that. I would have painted myself into a corner.
+
+**Mat Ryer:** That's really interesting, yeah. Because I hear a lot the advice of "Don't copy other languages if you're doing a port" or something. But yeah, that's actually quite a compelling thing to think about, when people are forced to make that decision.
+
+**Jaana Dogan:** Yeah, I've worked on a lot of projects with multiple language support, and all this turns out to be the biggest advantage; you can easily see the future parity in all these edge cases, like utility libraries and so on, to build... And it's also somewhat easier to track progress. If there's a new API coming up, you can basically go and implement it in the other language. So there's definitely some advantages.
+
+**Jon Calhoun:** It's also -- like, if you happen to have some sort of bug or something, sometimes somebody who's worked on something similar in another language can more easily jump into your code and help out. Because there's been several projects I've seen -- Stripe is an example where their API libraries tend to be very similar across languages, even if it's not necessarily the most idiomatic for that language... And at times it's frustrating and at other times it's very helpful, because it's like "Okay, I feel at home, because I've used it in this other language before."
+
+So I can see the arguments from both sides, and I definitely think that there is an argument to be made for keeping the same design that's there if those benefits are gonna be useful.
+
+**Sean DuBois:** \[39:45\] And listen to this magic... Since Pion does the WASM, so you write -- if you call "create peer connection" and "create offer", "create answer", when you compile to WASM, it's basically just outputting the JavaScript that runs in the browser. It lets me write my Go code, but then evaluate against the browser's implementation. So I can check "Does Pion's pure Go implementation behave exactly the same as Pion, but then compiles and runs in the browser? ...so it allows me to ensure that I implement WebRTC the same way as the browser does, but in my pure Go implementation."
+
+**Mat Ryer:** That's great.
+
+**Jaana Dogan:** Talking about that compliance - what is the overall story like? Is it you run a bunch -- like, you convert them into unit tests, and compare the results? Or what else do you do?
+
+**Sean DuBois:** So the nice thing about WebRTC is it's 100% standardized. So you just go to the IETF and you look at the specs and you implement them. What I do right now is I run Pion against Chromium, I run Pion against Pion, and then we have all those individual parts. I think when you say WebRTC, there's very little new stuff that WebRTC adds. WebRTC is basically hooking all of these existing technologies. So if you can go prove that ICE and SCTP and RTP works right, there's really not much to do in WebRTC. I think the WebRTC package of Pion is probably one of the least complicated parts.
+
+**Jaana Dogan:** In some cases, some projects usually have some sort of like a testing binary, which makes a call against your peer, just runs some of the basic functionality and then checks the responses for compliance... I was wondering if WebRTC has something similar.
+
+**Sean DuBois:** Unfortunately not. We were going to do this at the IETF in Vancouver. Tim Patton had organized this thing where all the WebRTC implementations were gonna sit down at a table and we were gonna write a little test suite that would run implementations against each other, and then make it easier for people in the future to write new implementations. Because someone's trying to write an Erlang one, someone's trying to write a Ruby one... They should be able to write a new implementation. But unfortunately, with the whole Coronavirus thing the IETF got canceled and it didn't happen.
+
+It's something I definitely wanna do, because there's a lot of detective work that went into doing Pion... Because there's a lot of things that are specified, but there's a lot of things that are just edge cases in the Chromium codebase, and you look and it's just like "//do this because x", and it's not saved or written anywhere else.
+
+**Mat Ryer:** It's a shame that conference got canceled. It's a shame that there isn't a way to somehow communicate remotely.
+
+**Sean DuBois:** Not that I'm aware of. Yeah, someone needs to build something for that, I think.
+
+**Mat Ryer:** Yeah.
+
+**Jaana Dogan:** I think this is one of the cases where you actually wanna sit at the same table... You just really need that in-person connection.
+
+**Sean DuBois:** Yeah. I think the other thing is that the longer I've been doing this, the more I appreciate that it's all about personalities and relationships and emotions... Because if I go in and say "Look, Pion is the best. This is how it should work", it's gonna irritate a bunch of people and we're not gonna be able to work together. But at the same time, if I let someone else come and do that, then it'll hurt WebRTC... So it's all about coming together and finding out what compromises are we willing to make; what technical compromises are we willing to make for the sake of relationships, which ones are we not willing to make, and how do we get to middle ground? Because emotions are so tightly coupled with what we build.
+
+**Jon Calhoun:** Open source in general is like a challenge on that front. There are times where I'm amazed that certain projects survived, given that -- like, Linux is a pretty notorious one, in the sense that, you know, there were certain people who were very opinionated (that's how I'll put it), and I don't know if that would work nowadays. I feel like people just go find another project and be like "I'm not gonna mess with this."
+
+I'm not trying to say anything bad or anything like that, I'm just... It's interesting that now that we have a lot more open source stuff and it's a lot more common, that it almost feels like it's a completely different ball game as to how you manage it, managing that community. It's probably improved things in a lot of ways, and gotten a lot more perspective that wasn't there.
+
+**Jaana Dogan:** \[44:00\] And I think we are better in terms of identifying these issues in the beginning. Linux wouldn't be this successful probably if people were paying attention to these type of issues back then. So there's definitely some sort of change. People just don't necessarily support a project if they see that the leadership in the project is really toxic.
+
+**Sean DuBois:** Yeah. I don't think software matters that much. At the end of the day, every project is gonna come and go. Someday Linux won't matter, someday Pion won't matter, but people's emotions - that's their entire life. I want people to be happy, and I want people to come in and feel good about what they've done. So yeah, I'm totally willing to make technical compromises in this for people's emotions, because it matters more to me than a couple lines of code. We can write more unit tests, we can add more linters, but it's not worth being a jerk.
+
+**Mat Ryer:** That was well said. It's a shame that WebRTC isn't really a topic for new programmers to jump into... And maybe you'll tell me I'm wrong. Because I think that sort of attitude, where you're able to foster this community that is inclusive, like you've been talking about, that helps people -- that doesn't just treat people like some weird, remote resource, or something... I think that's great. But is it a good project for people that are new to programming to come and look at? Certainly to use, I suppose, but what about contributing to the project?
+
+**Sean DuBois:** Yeah, so there is a huge backlog of problems that need to be improved. I think the other one that I encourage people to do more is like take Pion and go build something with it. What are you passions? Because I think open source projects are successful when they're built with passion, because that's what makes you wanna work on it on a Thursday night, compared to just going and playing video games, and watching a movie. Maybe you're not interested in contributing to Pion, but go build something with Pion. And if you tell me about it, I will spam Twitter, Hacker News, Reddit, Slack with it. I will promote your project for you, because I want Pion to be successful. I want to show people that Pion lets people build things.
+
+So I'm much happier seeing everyone with all their little projects than thousands of programmers descend on Pion. But I'd love to have you as well. Please, come contribute. No conversation occurs in private channels; everything's on GitHub. I try to make everything as easy as possible to get involved. That's my elevator pitch on new programmers... And that's the reason I'm here, why I was annoying you via email at one in the morning. I'm desperately trying to get into -- I've had a lot of luck getting people to talk to me in the WebRTC community, but in the Go community I haven't had as much luck, because WebRTC isn't hot right now. It's much easier -- if I was doing something with Kubernetes, or DevOps, or something like that, I'm sure I could get on to more conferences... But I have not had any luck with WebRTC. But this is the year that I annoy everyone and break in.
+
+**Jon Calhoun:** You'd think it would get more popular now that -- I feel like pair-programming applications and that sort of thing almost have to take off a little bit more, because there's not great options; at least it doesn't feel like there's great options for a lot of this stuff.
+
+I did have a question when you were talking about -- Matt was asking about beginners getting involved. I noticed that a lot of your repos have a difficulty tag that you put on all your issues... Has that been hard to manage? How has that helped is far as allowing people to get involved?
+
+**Sean DuBois:** I don't think it's really helped that much. It's really hard to just go to a random issue and pick it, because you're not passionate about the project. I think it's pretty idealistic to think "Today I'm gonna sit down and I'm gonna get involved, because I wanna get involved." I think you have to figure out what is the itch you're trying to scratch, because that's the only thing that's going to motivate you in the end. There's no reward at the end of this. You're not getting paid... It totally has to come from your own personal happiness. I tried that, and I do it a little bit, but I haven't seen it pay off that much.
+
+**Jon Calhoun:** \[48:03\] One of the things I think Mark did with Buffalo - I think he had people doing documentation type stuff. I think it was people who were learning Buffalo and then building stuff with it. Sort of like you said; you wanna build stuff with it first. And as things were less clear to them or they got confused, they'd sort of ask questions in the Slack, or whatever it was. And as that got hashed out and they actually got a good idea of what the description should look like, then it'd be like "Okay, well why don't you submit an issue that has this documentation improved a little bit? That way you can get involved." And it's not necessarily code, but it's still helping the project just as much, and it sort of gets you involved in that process, and everything. I think that can help, but I definitely don't know the right way to get developers involved in a project, because that's a challenging thing to do.
+
+**Sean DuBois:** I'm gonna have to pick Mark's brain on that, because I'm trying to do that right now. I give people access to the wiki and I say "Go edit it." But it's tough, because it's async; so if I'm not there to give immediate feedback and say "Great job. That's right", I lose them. I've had at least 3-4 people start this whole documentation push - they were learning WebRTC - and I'm gone for 24 hours because I've got work, or family stuff going on, and that green circle stays grey forever. It's tough to keep people. Yeah, I don't know.
+
+**Jaana Dogan:** One thing that works sometimes is weekly meetings, if you have time to go over "These are the certain things that we are thinking about", so people tinker and brainstorm about certain things. Everybody gets their point of action items, and just go and do... But it requires time, and it has to be consistent. You are creating that support system, and you can't really take it away... So it's really up to the community and how much time you have.
+
+**Sean DuBois:** That's a fantastic idea. I think I'm gonna do that. I'm gonna schedule like once every two weeks, "Here's an hour that I'm just gonna be in Hangouts", and people just join and talk about what's important to them. Because I think that's a big thing that's been lacking right now with the project. People are able to do their single-track things that they're excited about, but it's really hard to build multi-person momentum.
+
+**Break:** \[50:19\]
+
+**Mat Ryer:** Well, you said a lot of very popular things today, Sean, but it's time for our regular slot of the show... So sit down everyone, if you're stood up, and relax, or stay stood up -- just whatever you were gonna do. It's time for unpopular opinion!
+
+**Jingle:** \[51:54\]
+
+**Mat Ryer:** Alright, Sean, do you have anything unpopular to say?
+
+**Sean DuBois:** I don't know how much people are gonna care, but the one that always starts flame wars is I think that the GPL is a freedom-restricting license.
+
+**Mat Ryer:** Oh, really?
+
+**Sean DuBois:** Yes. Here's the one, if people care. I get the argument that I should use my power to ensure that someone else keeps their freedom, but at the same time the act of taking away someone's autonomy - that's freedom removing in the first place.
+
+This is a constant argument, and hopefully this is exciting enough for your audience, but that's the one I like to bicker with people about... Because I'm a big free software kind of guy, and it ties up with getting people involved, but I feel very strongly about what GNU is trying to do, and encourage individual ownership, and that we shouldn't have these giant companies own all software... But at the same time, the GPL is a very frustrating, tenuous topic for me. So if we wanted to argue about, that's the one that I'll get fired up and irritate people over.
+
+**Mat Ryer:** What license did you choose for Pion?
+
+**Sean DuBois:** MIT.
+
+**Mat Ryer:** So what led to that?
+
+**Sean DuBois:** For me, I think that people rise -- like, if someone's gonna do something bad, they're gonna do it no matter what. So even if I do the GPL, someone's gonna break the GPL, and they're gonna distribute stuff and they're gonna cheat system, and there's nothing I can do to prevent it. It's the same way I see fraud. You can add thousands of rules, but bad people are gonna be bad. But I want to encourage good people to be good.
+
+So I say "Hey, you have the full freedom to go do what you want. And if you don't wanna contribute back, that's fine. If you don't wanna share what you've done, that's fine, but I want to--" There's so many programmers that are using it, and they're able to get it approved because they can use it. I think that's what was important to me. You can add rail guards and try to make people better, but at the end of the day, people are gonna be who they are.
+
+**Mat Ryer:** So what if, in a hypothetical situation then, a company took Pion, and it was a startup, and they just built some phenomenal software that just went crazy, it became a unicorn, and people are making a lot of money on this technology - how would that feel for you?
+
+**Sean DuBois:** It's already happening. There's a couple big users of Pion that are using it, and some are coming out pretty soon, actually. These are like billion-dollar companies, with their big things. I wish I could get hired and get paid a million dollars to be a principal engineer there, but it's just not the way it works.
+
+For me, I'm gonna come and go; someday I will not be alive anymore, and was it really worth it? I would rather see the total good in the world increased, even if I miss out on something. It is what it is. If I hadn't done the GPL -- I'd much rather have the unicorn making people's lives better, than not Pion existing at all. Now, I think I would be hurt if I found out that the unicorn was putting people in cages. Then I'd probably be like "Man, I think I've made the world a worse place", but again, my arguments comes back to "What can I really do?" They're gonna do it no matter what. People are gonna do the wrong thing. I can help the good people, but I can't prevent the bad people.
+
+**Jon Calhoun:** I think that's something that's hard though, because we tend to focus on the bad... Versus just all sorts of cases where you just focus on the potentially bad things people can do and how to prevent it. And I see this all the time, with stuff where -- to give you an example, I have Go courses, and some of the things that are sometimes challenging is some people will try to buy, download everything, and then immediately ask for a refund.
+
+So you're like, okay, do I get rid of the refund policy because of that, or do I add some clause to it that's like, "If you do this, then you don't get a refund"? But then overall you're like, well, that means that anybody else who's a legitimate person doesn't -- you know, they don't wanna read through this clause of what applies, what doesn't... So I'm just like, "Alright, at the end of the day it's probably better off just to say the refund's there", and those bad people - I have to ignore them.
+
+\[56:08\] And it sucks, because when it does happen, you feel frustrated, and you're annoyed, and everything else. But at the end of the day, like you said, you can't focus on that, because they're probably gonna do it or find some way to do it. Because even if you don't give them a refund, they'll probably do a chargeback, or something. Like you said, they'll either commit fraud, or something. There's no way to stop that, so just focusing on it is probably not worth your effort.
+
+**Sean DuBois:** Yeah, I don't know what the right answer is. I guess at the end of the day you just have to find what's the total happiness you can do out of this day... And for me, I get my happiness out of helping/empowering the good people.
+
+**Mat Ryer:** Yeah, the really strange version of this is where you end up with software they've done extra work to make the software worse for some people... And it's kind of a crazy thing. It makes sense, because you think of course "Well, there's a light package that's free, and it's limited, you can't do as much with it, and then you pay to unlock more features, or whatever..." But it is a very strange thing that we're doing, which is putting effort in to making it worse.
+
+**Sean DuBois:** Yeah. I do hear arguments that for DRM it's worth having the basics... Like saying "Go get out your manual and type in the seventh letter of the ninth paragraph." But it's not worth doing the crazy DRM, because you just can't beat it. You can beat the 10-year-olds who are sharing the game between each other, but you can't beat the 25-year-old who's gonna sit and reverse-engineer and beat it no matter what. You can't beat them, but you can beat the easy ones. I don't know.
+
+**Jon Calhoun:** So we probably should have asked this earlier... Specifically what about the GPL is it that you don't like? What specifically does it force people to do, that they-- because I'm guessing not all the listeners have read the GPL or are that familiar with it.
+
+**Sean DuBois:** That if you make changes to the software, you can't distribute -- like, you have to distribute the source code. And I've seen that even go further with the Affero GPL, where if you are hosting something and someone uses your service, they have to have access to that service. But a part of me is like -- I don't think it's worth it, because I think people will then not build stuff with Pion. It will stop ideas from being built. So I would rather see people go do that, and go build something exciting, and then I think they'll do the right thing. I think the one that encouraged me was John Carmack's approach to open source, where he built Doom and Wolfenstein 3D, and then he just kind of threw the source code over the wall, to encourage people to learn and make things better. That's my outlook on it - just give people the freedom to do what they wanna do, and that's gonna bring out their best selves. So that's my issue with the GPL.
+
+The other one is I'm not a big fan of restraints. Just the idea that you have this big license is frustrating to me. I think that that encourages bureaucracy and that encourages more structure. It's not helping software developers, it's not helping people. It's encouraging these giant systems that I'm not really benefitting from.
+
+**Jon Calhoun:** Well, like you said before, I feel like it's also one of those -- if you're a big org, you can definitely tell which license it is, and what you have to do when you get it. But if you're an individual developer and you don't really know for sure, I feel like (like you said) it can scare them off, because they don't have the resources to actually figure out "Am I okay doing this or not?" They definitely don't have the money to go to court and defend themselves, or anything like that. So they're just like "I've gotta skip this, because I'm not really sure what this clause of the GPL means, or how it applies to my stuff."
+
+**Sean DuBois:** Yeah. And that's the same why I'm against patents as well, because patents I think only protect the big players. The big players amass these big war chests, and if you're a little guy, they come to you and say "Hey, sell to me, or I'm gonna crush you." So I think the GPL is the same way. It's the right intention, but what all of these systems do is they help the big players.
+
+\[59:54\] The big one that I heard is in the E.U, when we're adding all these protections, there's a reason that these big companies are encouraging it, because now it's building up their moat. So I can't go compete with a big company that's doing X, because they've built up this big moat, and there's all this regulation.
+
+So I sat down and I really wanna do the right thing. I'm adding regulation because I believe I'm making people's lives better. But at the end of the day, regulation is just abused by people with malevolent intent. And you can't beat them. Whatever system you make, they're going to abuse, so it's better not to have it at all.
+
+**Jaana Dogan:** Yeah, even if a large company's GPL is such a scary thing, because it's basically like software patents, you keep it as a leverage to sue another company in case you need to sue them, or for some other reason... Because you wanna destroy their business, or something.
+
+**Mat Ryer:** Yeah, I'm with you on patents, and stuff. I don't know if I'd advocate to abolish all laws, which seems to be what you're suggesting, Sean...
+
+**Sean DuBois:** \[laughs\] No, I'm not... I think where I come down -- I am all for laws that encourage community ownership and direct representation. But anything that empowers people war chests and large organizations, I am totally against. That's like a bigger ideological conversation, but a lot of my decisions are made off of that - how can I empower the little person? That's what drives everything - how do I encourage individuals to live a happy life. Because I think the U.S. was the best when there were more small businesses. You look back, and in the '40s and '70s you had more small businesses, and that directly ties to -- people had better lives because of that. That's my two cents.
+
+**Mat Ryer:** Lovely.
+
+**Jaana Dogan:** I can't imagine publishing any of my open source projects if Google was telling me to use GPL. Because if I'm using MIT or Apache2... as soon as I leave, I still can take it and convert it into a business, or something. I don't own the IP, which is also a ridiculous thing in the U.S. If you're working for a company, the company owns the IP, not you as the main contributor. And on top of that, if it's been licensed as GPL, then Google just basically owns me until the end of my life. So that doesn't really give individuals any power at all.
+
+**Jon Calhoun:** That one's also so weird, in the sense that -- like, if you do it on company time, on company property, it's theirs. But if you can someone prove that it was on your own laptop, on your own time, not using -- like, there's all these hoops you have to jump through, and it's like, what person can really defend against Google?
+
+**Jaana Dogan:** Exactly. And it's more extreme than that. In order to have IP, you have to stop using old resources, including talking to people. So you can't really go and talk to your co-worker about your project if you wanna own the IP. It's basically not very practical.
+
+**Mat Ryer:** It's funny, yeah. I understand how that happens inside companies, but it is a shame. Another thing on patents that I've seen... I've written some patents, and one of the things -- at some point I found myself in a meeting with a lawyer who was basically saying "Other people can essentially just violate this patent, because it's about enforcement. So if no one's gonna go after them, then people are free to do it, basically." But you're an author of it, so you definitely have knowledge of this patent, so you absolutely cannot use any of that technology. So there is a kind of penalty to individual developers and individual IP creators if you do write a patent for the company.
+
+I always think companies ought to do something extra for people that are contributing, if they're doing patents, and stuff... But yeah, ideally we wouldn't have software patents. And I think anyone that loves open source can see that.
+
+**Jaana Dogan:** Large companies just kind of like give these grants, which is like $5,000 or $10,000, and there's all these lawyers... You just basically come up with an idea to the lawyer, and the lawyer just converts it into a 30-page patent, they submit it...
+
+**Mat Ryer:** That's mad.
+
+**Jaana Dogan:** It is mad, yeah. This entire system is just so broken...
+
+**Mat Ryer:** You just go into a meeting--
+
+**Jaana Dogan:** ...and it's all for leverage.
+
+**Sean DuBois:** \[01:04:09.02\] Yeah, you literally go into a meeting and somebody's like "What did you do in that project that can be patented?"
+
+**Jaana Dogan:** Exactly.
+
+**Sean DuBois:** It's not like you come up to somebody and say "We've found something novel." People literally hunt you down, like "Do you have anything that can be patented?" and you're like "I don't know."
+
+**Mat Ryer:** Yeah, "I don't know... I used paths. Has that been done, paths...?" \[laughter\] I'm saying that as a joke, but genuinely, I have a patent which essentially describes paths.
+
+**Jaana Dogan:** \[laughs\] Well, there's a patent for double-clicking... So I would completely believe that that would be legitimate.
+
+**Mat Ryer:** Yeah, a double-click... Imagine that, someone just got into a meeting and gone "What if we click twice?" I bet the bloke that invented the click was really annoyed, wasn't he?
+
+**Jon Calhoun:** I mean, Jaana is really hurting us. Now I can't write software that takes a double-click, because I know about the patent.
+
+**Mat Ryer:** Oh, Jaana... And everyone listening.
+
+**Jaana Dogan:** Yeah, I revealed a secret...
+
+**Sean DuBois:** I feel successful, because the GoTime channel on Slack has gone off about my GPL comments... So I think I did it.. That was my only goal.
+
+**Mat Ryer:** Yes, it turned out to be somewhat popular. And you also came out of this section looking like a saint, Sean... And that is not the purpose of Unpopular Opinion. It's meant to have the opposite effect. I want to damage careers, not help them...
+
+**Sean DuBois:** \[laughs\] No, I took months of preparation. I knew I was gonna be cornered.
+
+**Mat Ryer:** Yeah, exactly. You managed to do that trick you do in an interview, when you're asked for a weakness, and you go "You know what - I'm too good at work."
+
+**Sean DuBois:** "I care too much, man. That's the issue. I care too much."
+
+**Mat Ryer:** "Yeah, my code's just, if anything, too good. And it upsets people. It upsets all the people on the team." It's crazy.
+
+**Sean DuBois:** \[laughs\]
+
+**Mat Ryer:** Well, you can't say it's thieving. "Have you got a weakness?" "Yeah, thieving."
+
+**Sean DuBois:** "I don't really show up to work and I steal stuff from the supply closet."
+
+**Mat Ryer:** \[laughs\]
+
+**Jon Calhoun:** "And it's not really a weakness if you're not getting caught..." No, I'm just kidding. \[laughter\]
+
+**Mat Ryer:** Well, that's all the time we have today. Sean, thank you so much. Please do come back, and anyone interested in Pion, check it out. It's literally github.com/Pion, and you can start hacking on it.
+
+This has been great, thank you very much. I'm gonna wind up the show, unless there's other bits we wanna talk about.
+
+**Sean DuBois:** Thank you very much for having me. Hopefully, if you're interested in coming on, please jump in our Slack channel, open up our thing. I wanna have you involved and I wanna see if we can make either your life or other's a little better. And thank you very much for having me. I'd love to come back on, even if it's not about Pion.
+
+**Jon Calhoun:** What is the Slack channel.
+
+**Sean DuBois:** There's \#pion.
+
+**Jon Calhoun:** Just \#pion in the GopherSlack?
+
+**Sean DuBois:** Yup, exactly.
+
+**Mat Ryer:** Brilliant. Are you on Twitter, Sean?
+
+**Sean DuBois:** Yeah, @\_pion is the Twitter. I don't have one myself, but it's just me running it. So if you see any opinions on there, you can blame me.
+
+**Mat Ryer:** Okay, we will. \[laughter\] Thank you very much. Thanks for listening, we'll see you all next week!

What to expect when you’re NOT expecting_transcript.txt

@@ -0,0 +1,824 @@
+[0.00 --> 3.30]  I've done a few things with eBPF a little bit here and there.
+[3.60 --> 7.58]  Actually, Delve has a trace functionality, which works somewhat similar,
+[7.68 --> 12.14]  but it works at a higher level using ptrace and some of those other kind of syscalls.
+[12.46 --> 16.34]  And I've thought about experimenting a little bit on Linux systems that support it,
+[16.54 --> 19.54]  replacing that with like an eBPF-backed tracing system.
+[19.54 --> 24.40]  So Grant, if you ever want to send a pull request, we'd love to have it.
+[24.40 --> 34.52]  Yeah, I am happy to integrate it from the VS Code side with the visualization.
+[36.24 --> 37.44]  I would love that.
+[37.66 --> 39.88]  This is the most productive meeting I've ever been in.
+[40.50 --> 42.04]  It wasn't even meant to be a meeting.
+[44.24 --> 46.92]  BAM with 4Change Log is provided by Fastly.
+[47.22 --> 49.10]  Learn more at Fastly.com.
+[49.34 --> 51.64]  Our feature flags are powered by LaunchDarkly.
+[51.90 --> 53.70]  Check them out at LaunchDarkly.com.
+[53.70 --> 55.80]  And we're hosted on Linode cloud servers.
+[56.14 --> 59.72]  Get $100 in hosting credit at Linode.com slash changelog.
+[62.36 --> 66.22]  Whether you're working on a personal project or managing enterprise infrastructure,
+[66.22 --> 70.56]  you deserve simple, affordable, and accessible cloud computing solutions
+[70.56 --> 73.08]  so you can take your project to the next level.
+[73.54 --> 79.96]  Simplify your life with Linode's Linux VMs to develop, deploy, and scale your applications faster and easier.
+[79.96 --> 84.48]  Get started on Linode today with $100 in free credit for our listeners.
+[84.48 --> 88.28]  You can find all the details at Linode.com slash changelog.
+[88.40 --> 94.60]  Or if you're not at your desk, just text changelog to 474747 and get instant access to that $100.
+[95.14 --> 102.72]  Linode has 11 global data centers and provides 24-7, 365 human support with no tiers or handoffs,
+[102.92 --> 104.24]  regardless of your plan size.
+[104.24 --> 106.94]  In addition to shared and dedicated compute instances,
+[107.24 --> 112.42]  you can use that $100 credit on S3-compatible object storage, manage Kubernetes, and more.
+[113.20 --> 118.18]  Visit linode.com slash changelog and click on the Create Free Account button to get started.
+[118.36 --> 120.92]  Or just text changelog to 474747.
+[121.44 --> 123.08]  Get started today on Linode.
+[123.08 --> 134.66]  Let's do it.
+[135.28 --> 136.30]  It's go time.
+[137.06 --> 141.72]  Welcome to Go Time, your source for diverse discussions from around the Go community.
+[142.48 --> 147.04]  We're taking you back to GopherCon this week for our second live episode from the conference.
+[147.46 --> 150.56]  We hope you enjoy it, and if you do, please tell a friend about Go Time.
+[151.20 --> 152.86]  Okay, let's get straight into it, shall we?
+[153.08 --> 154.62]  Here we go.
+[168.32 --> 175.76]  Hello there, and welcome to a very special episode of Go Time, and it's a GopherCon mashup.
+[176.48 --> 181.20]  This is the lunchtime sessions for GopherCon, and also a Go Time episode.
+[182.10 --> 182.74]  So welcome.
+[182.74 --> 187.60]  Today, we're talking about what we do when things go wrong.
+[188.14 --> 193.56]  A manager once asked me to only write code that didn't have any bugs in it.
+[193.72 --> 195.88]  So that was an interesting thing.
+[196.08 --> 201.28]  We're going to find out today what are bugs, and what can we do to get rid of them,
+[201.38 --> 203.46]  or make sure they're never there in the first place.
+[203.46 --> 211.62]  And we'll look at tools and techniques around this, too, with our expert panel, who I'm going to introduce now.
+[212.08 --> 214.56]  We're joined by Hannah Kim from the Go team.
+[214.70 --> 215.06]  Hello, Hannah.
+[215.58 --> 215.96]  Hi.
+[216.34 --> 217.34]  Welcome to Go Time.
+[217.34 --> 222.40]  Yeah, I don't know if I'm an expert, but thank you very much for inviting me to Go Time.
+[222.48 --> 223.62]  It's so exciting.
+[223.96 --> 224.76]  Yeah, I'm so honored.
+[224.88 --> 226.48]  Yes, you're more than welcome to be here.
+[226.54 --> 229.98]  Sorry, I do sometimes accidentally say expert, and people don't like that.
+[230.08 --> 230.92]  It sets things up.
+[231.18 --> 233.14]  So don't worry.
+[233.22 --> 233.62]  No pressure.
+[234.42 --> 236.94]  We're also joined by Grant Seltzer-Richman.
+[237.06 --> 237.60]  Hello, Grant.
+[238.18 --> 238.48]  Hi.
+[238.66 --> 239.38]  Thanks for having me.
+[239.82 --> 240.68]  Oh, welcome, mate.
+[240.72 --> 241.40]  You're very welcome.
+[241.72 --> 242.80]  Are you having a good day so far?
+[242.80 --> 244.10]  So far, so good.
+[244.38 --> 246.64]  Go4Con is, so far, the talks have been great.
+[247.24 --> 247.76]  Great.
+[247.90 --> 249.34]  Well, hopefully we won't ruin it.
+[250.34 --> 253.02]  And we're also joined by Derek Parker.
+[253.40 --> 253.84]  Hello, Derek.
+[254.58 --> 255.38]  Hello, everybody.
+[256.10 --> 256.36]  Hello.
+[256.74 --> 260.18]  Welcome to Go Time slash Go4Con lunchtime session.
+[260.58 --> 261.06]  Thank you.
+[261.30 --> 266.28]  And good evening, good morning for me on the West Coast, having my second cup of coffee.
+[266.76 --> 267.12]  Great.
+[267.32 --> 267.88]  Enjoy it.
+[268.44 --> 272.62]  So, yes, I should say, Derek, you work at Red Hat.
+[272.62 --> 275.04]  And you actually created Delve, didn't you?
+[275.08 --> 275.94]  Which is a debugger.
+[276.68 --> 277.48]  Yeah, that's correct.
+[277.92 --> 278.12]  Okay.
+[278.28 --> 282.06]  So this will be good because we'll definitely dig into that a little bit more.
+[282.28 --> 283.38]  It's such a great tool.
+[283.88 --> 286.02]  And, you know, it's a great way to get rid of bugs.
+[286.26 --> 287.32]  But maybe we could start.
+[287.76 --> 288.80]  What is a bug?
+[289.10 --> 294.68]  Why was it a little bit absurd that my manager asked me to only write code that didn't have bugs in?
+[295.02 --> 296.90]  I would say what is a bug?
+[296.90 --> 307.60]  Like a bug is, I guess, something unexpected in your code or something incorrect, right, is what it's typically is as how I think most people typically think of it.
+[307.72 --> 312.88]  It's something not only unexpected, but just incorrect, the wrong result or wrong something.
+[312.88 --> 325.96]  And the absurdity, I think, of that statement comes from, you know, how could you have the premonition not to, you know, sometimes it's maybe a little bit out of your control when a bug happens.
+[326.20 --> 329.44]  You know, like I would say like weird things can happen.
+[329.58 --> 338.88]  Like running on a different architecture could expose a bug that you wouldn't have seen on the CPU architecture that you're using or in a different environment or with different environment variables or whatever.
+[338.88 --> 350.62]  Whatever, you know, there could be so many things outside of your direct control that could expose a bug, which I think is part of the absurdity of that statement that please write code that does not have bugs.
+[351.58 --> 351.92]  Yes.
+[352.84 --> 353.32]  Yeah.
+[353.48 --> 354.90]  So, right.
+[355.00 --> 358.60]  It's just behavior that happens that we don't want to happen.
+[358.72 --> 363.20]  But of course, there's no way for the compiler to know that that shouldn't happen.
+[363.20 --> 372.04]  It's not like a type error where you can have a compiler look at the code and, you know, tell you or fail the compilation if things aren't right.
+[372.58 --> 381.96]  It's kind of they emerge sometimes from either different ways things are interacting or just sometimes it's, you know, made a mistake.
+[382.12 --> 382.82]  It happens too.
+[383.44 --> 383.56]  Yeah.
+[383.68 --> 385.16]  Sometimes bugs turn into features.
+[385.64 --> 387.76]  So, yeah, that's yeah, absolutely.
+[387.94 --> 392.28]  Especially when when I've done it, I will always pretend it was meant to be like that.
+[393.20 --> 394.06]  That's a great one.
+[394.70 --> 396.54]  So what do we mean by debugging then?
+[396.66 --> 401.52]  Is this just any method that anything you do to get rid of bugs?
+[401.52 --> 405.58]  Is that debugging or is debugging a more specific technical term?
+[406.20 --> 422.24]  So I suppose when you're, you know, using a debugger doing some type of debugging, you're trying to figure out what it is that's causing this unexpected behavior that, you know, everybody has an intention when they're writing their code.
+[422.24 --> 428.66]  So just like how you said that there's no way for the compiler to tell, you know, what you meant for your code to do.
+[428.74 --> 430.90]  It's doing exactly what you told it to do.
+[430.90 --> 443.24]  So, you know, let's say you're seeing some bugs, you know, output that's coming out of your program is not as expected or you're just seeing some errors that you didn't expect to happen.
+[443.76 --> 448.38]  The act of debugging is trying to figure out what's causing that bug.
+[448.56 --> 456.84]  You know, where in your code did you have some logic that isn't what you intended it to be and trying to figure that out so that you could then fix it.
+[456.84 --> 470.28]  Yeah. And when you come to debugging then with techniques and tools, do you have a particular favorite of what's your sort of go-to, what's the first thing you do when you've noticed something's wrong?
+[470.68 --> 471.80]  Does it depend or?
+[471.80 --> 489.88]  So I will say that I think there's a sort of a half joke within all of the tech industry that, you know, adding print statements to your code is wrong or like an amateur approach.
+[489.88 --> 501.62]  But to be honest, if your program compiles particularly quickly, there's no reason that adding a print statement should be looked at as, you know, like a dirty way of doing it.
+[501.86 --> 506.42]  So, you know, the feedback loop, you know, when you're debugging, you want to have a quick feedback loop.
+[506.56 --> 515.42]  So if it's a simple enough program where you could just add a print statement at, you know, a certain point that's printing out the contents of variables, I don't think there's anything wrong with that.
+[515.42 --> 521.34]  Because it's an intuitive interface, you know, that like, I want to know what happens at this point in the program.
+[521.74 --> 532.56]  If it's a particularly complex program or, you know, the compile time is long and you need a faster feedback loop or something like that, that's when I would typically use a debugger or some type of tracing tool.
+[533.42 --> 535.44]  But nothing wrong with print statements.
+[535.44 --> 547.84]  Yeah, I think that's a great, I'm really pleased you said that because I've met junior developers who feel like that's, they're just, they don't know how to use a debugger or they don't know what they're doing and they just put prints out.
+[548.00 --> 550.34]  But it is completely legitimate.
+[550.78 --> 553.86]  In fact, it tends to be my go-to thing is doing that.
+[554.20 --> 557.18]  And there's a particular verb that's very useful in Go.
+[557.18 --> 563.44]  If you use the FMPT package, you can do like percent plus V and then give it any type.
+[563.60 --> 570.72]  And it does a really good job of describing that type, even if it's a complicated kind of struct with nested data and all sorts.
+[571.22 --> 573.86]  And you see the field names too, which is quite useful.
+[574.34 --> 580.96]  Are there any other favorite techniques like that, the sort of simple, debuggable things?
+[581.66 --> 582.78]  Printing's a great one.
+[582.78 --> 592.62]  Certainly, you know, I'm sure both Derek and Hannah can talk about using Delve, but I certainly use Delve, you know, a full-fledged debugger.
+[592.94 --> 593.14]  Right.
+[593.28 --> 596.02]  So Delve's a kind of a different beast, really.
+[596.20 --> 600.08]  And the other one that we should talk about before we get on to Delve is test code.
+[600.16 --> 603.42]  Because actually test code is a great way to debug your code.
+[603.42 --> 615.58]  One trick that I find works really well is if somebody identifies there's a bug, if you ask them to write a failing test, if they can do that, you know, that is a great way.
+[615.68 --> 616.80]  You remove all ambiguity.
+[617.32 --> 619.64]  You're looking, you know, you speak in the same language.
+[619.78 --> 620.64]  You look at the code.
+[621.10 --> 625.46]  And if it's a failed test, you know, you've proven that there's a bug there.
+[625.46 --> 629.24]  And sometimes the test is wrong and some of the assumptions are wrong.
+[629.34 --> 630.56]  And that's one thing.
+[631.04 --> 633.64]  But usually it does kind of highlight the bug.
+[633.70 --> 640.46]  And then, of course, once it's fixed, you can put that test into your test suite and you kind of never get that same bug again.
+[641.22 --> 647.36]  Hannah, you work on the Go team and you're working on the VS Code plugin for Go, right?
+[647.36 --> 649.92]  So tell us about that then.
+[650.14 --> 662.22]  That plugs into the IDE, integrates, so turns visual code from being what might be just a sort of basic text editor and adds some kind of Go intelligence to it.
+[662.56 --> 663.16]  Is that fair?
+[663.72 --> 664.54]  Oh, yeah.
+[664.70 --> 667.04]  And also it has a debug integration too.
+[667.04 --> 678.78]  And also all kind of facilities that actually helps users to write a good test, like all kind of templating and autocomplete and that kind of features.
+[679.46 --> 681.08]  And also like a test command.
+[681.32 --> 685.36]  So with just one click, you can write a test on your package.
+[685.60 --> 685.72]  Yeah.
+[686.26 --> 687.74]  So, yeah.
+[687.74 --> 697.14]  But for debugging purpose, I think my go-to is still like a print app and log or that kind of thing.
+[697.64 --> 698.30]  So, yeah.
+[698.90 --> 699.12]  Yeah.
+[699.30 --> 701.64]  So maybe it's time then just to talk about Delve.
+[702.26 --> 707.18]  Maybe we could start by just for anybody that hasn't used a debugger before.
+[707.76 --> 709.38]  Maybe, Derek, this is one for you.
+[709.68 --> 711.84]  What is a debugger and what's it doing?
+[711.92 --> 712.56]  How does it work?
+[712.66 --> 713.34]  How do you use it?
+[713.34 --> 713.86]  Yeah.
+[713.86 --> 723.08]  So I actually heard like a really good explanation yesterday from Jason, who I was co-instructing our workshop with.
+[723.14 --> 724.50]  The format or is it a human?
+[725.38 --> 727.00]  You're not talking about the data format.
+[727.38 --> 727.52]  No.
+[728.96 --> 729.58]  It's a human.
+[730.14 --> 730.56]  Got it.
+[732.12 --> 732.34]  Okay.
+[732.80 --> 736.32]  That's a half joke, like what Grant did earlier.
+[736.92 --> 737.74]  It's just not funny.
+[738.70 --> 739.08]  Go on.
+[739.16 --> 739.32]  Sorry.
+[739.32 --> 745.44]  I like the way that Jason kind of explained like what a tool like a debugger is.
+[745.58 --> 751.72]  So it's actually kind of like, like I think that the name debugger is a little bit of a misnomer.
+[751.98 --> 755.38]  Like the tool itself doesn't actually fix the bug for you.
+[755.64 --> 760.30]  It's just a tool that you can use to understand your program, right?
+[760.30 --> 764.60]  It's just like a, it's a way to just understand what your program is doing.
+[764.74 --> 768.14]  And then once you figure out what's going wrong, you can fix the bug.
+[768.54 --> 770.80]  It doesn't fix the bug for you, but who's fault's that?
+[770.92 --> 773.94]  You're the creator and co-maintainer of Delve.
+[774.06 --> 776.08]  So really you've only got yourself to blame there.
+[776.32 --> 777.56]  In the next release, you know.
+[779.40 --> 780.94]  But it's actually an interesting point.
+[781.08 --> 782.42]  It can't be done automatically.
+[782.54 --> 785.52]  Otherwise, of course, the tooling would be doing it for you.
+[785.52 --> 791.74]  You know, you have to sort of tell it what correct is and you've already told it what incorrect is, right?
+[791.94 --> 797.38]  Yeah, I think it's approaching like any kind of debugging situation with the mindset of like,
+[797.88 --> 801.02]  how can I gain insight into what's actually happening?
+[801.02 --> 807.78]  And how can I do that in a way that will quickly allow me to figure out what's going wrong so that I can fix it?
+[807.78 --> 814.90]  And I think kind of like what Grant said earlier, whatever gives you the fastest feedback loop, that's kind of what you should pursue.
+[815.08 --> 821.68]  So whether that's, you know, I do print line debugging all the time as well, especially when like working on Delve,
+[821.70 --> 824.42]  it's hard to debug a debugger with a debugger sometimes.
+[824.42 --> 830.60]  So doing like print line debugging in those kind of situations and kind of whatever gives you the quickest feedback loop,
+[830.66 --> 834.34]  I think is the best tool to reach for in that situation.
+[834.34 --> 843.18]  Hmm. Okay. So if printing the results isn't working for you, then Delve allows you to kind of set a break point, doesn't it?
+[843.34 --> 847.50]  And what happens then in the program, the program stops at that point?
+[848.14 --> 853.00]  Yeah, exactly. So with a tool like Delve, like a traditional like source level debugger,
+[853.14 --> 856.60]  you're interacting with your program like in real time.
+[856.60 --> 863.62]  And that's kind of what's fun and interesting about using a debugger is you have the ability to stop what's happening,
+[863.62 --> 866.96]  inspect state, even change state if you want, continue.
+[867.64 --> 873.16]  So yeah, like for example, when you start up a new debug session and you set a break point, you continue to it.
+[873.20 --> 878.40]  You're telling the program, I want to stop at this specific location and just check out what's going on.
+[878.72 --> 880.60]  You know, see how I got here.
+[880.70 --> 882.16]  You can look at the stack trace.
+[882.26 --> 883.62]  You can see the value of variables.
+[883.62 --> 888.54]  And if you want to experiment a little bit, debuggers also can let you experiment.
+[888.80 --> 894.68]  So you can say, for example, like change the value of a variable and see if that gives you the result that you wanted.
+[894.84 --> 901.64]  So it gives you a little bit more of like real time interaction, like getting back to that quicker feedback loop,
+[901.70 --> 907.00]  you know, without like changing a variable in the code, recompiling it, rerunning it, trying to hit that bug again,
+[907.00 --> 912.06]  or seeing if you don't, you can kind of do some interactive stuff within like a debug session.
+[912.60 --> 915.26]  And you can step then, can't you, through the instructions.
+[915.26 --> 921.54]  So you can advance the program step by step slowly and keep an eye on things just so you understand what's happening.
+[921.64 --> 926.74]  It sort of like puts it into slow motion and lets you do that carefully, right?
+[927.36 --> 927.60]  Yeah.
+[927.78 --> 928.44]  Yeah, absolutely.
+[928.60 --> 928.86]  Exactly.
+[928.86 --> 930.70]  And how does that actually work?
+[930.78 --> 931.36]  What's going on?
+[931.44 --> 936.68]  Do you have to build the binary specifically with that debug information added to it?
+[936.76 --> 938.08]  Or can you just debug anything?
+[938.98 --> 939.16]  Yeah.
+[939.30 --> 943.32]  So all binaries include information called, it's called dwarf information.
+[943.76 --> 950.00]  And that's like a standard format of debug information.
+[950.10 --> 957.78]  Basically, it tells tools like Delve how to find variables, how to unwind the stack, how to do all kinds of things.
+[957.78 --> 961.28]  So Go, by default, will build that into all binaries.
+[961.58 --> 963.74]  You have to opt out of it specifically.
+[964.38 --> 969.20]  And the only reason why you would opt out of it would be maybe you really, really care about binary size.
+[969.30 --> 975.24]  And you want to get out every last bit that you can to shrink your binary as much as possible.
+[975.44 --> 978.00]  But by default, that information is going to be in there.
+[978.00 --> 992.82]  The only other thing that like Delve does by default, and I would recommend folks do if they're going to try to debug their Go applications, processes, whatever, is Go also by default will put in optimizations.
+[992.82 --> 999.42]  So like if you're familiar with like GCC or some other compiler, you have to explicitly tell it what level of optimization you want.
+[999.54 --> 1003.32]  And you kind of have to opt into some of the more extreme optimizations.
+[1003.90 --> 1005.10]  But Go does that by default.
+[1005.58 --> 1008.86]  And that's great for when you're building a production binary, you want to ship it off.
+[1008.92 --> 1010.72]  It's going to be fast and performant and all that.
+[1010.72 --> 1017.58]  But it could hamper debugging a little bit because of like inlining functions can get weird sometimes.
+[1017.74 --> 1018.94]  Delve handles it really well now.
+[1019.08 --> 1026.02]  And the Go compiler has gotten a lot better at providing information for telling debuggers how to handle that.
+[1026.14 --> 1032.22]  But there's still certain weirdness there that you can run into when trying to debug and optimize binary.
+[1032.50 --> 1035.66]  So that's the only caveat that I would like explicitly mention.
+[1036.94 --> 1037.74]  That's interesting.
+[1037.74 --> 1045.76]  And so, Hannah, when you talk about the VS Code plugin and it has debugger support, does it support Delve?
+[1046.90 --> 1049.44]  Well, actually, Delve is behind the scene.
+[1049.94 --> 1054.68]  Actually, like other editors like Golan, they also use Delve.
+[1055.10 --> 1062.20]  So basically this idea, like when user requests to debug their code, it actually formulates all this Delve command
+[1062.20 --> 1066.38]  and then invoke Delve and ask Delve to answer the question.
+[1066.38 --> 1074.12]  And then most of modern IDs, they have a kind of like all this local variable, global variable or step trace.
+[1074.44 --> 1076.24]  So it just asks Delve.
+[1076.48 --> 1081.84]  And then all the information are visible through all this UI.
+[1082.14 --> 1089.04]  And then users can actually step through all this, yeah, step through the program using the UI.
+[1089.04 --> 1092.84]  And again, we ask Delve to do all this job.
+[1093.26 --> 1100.92]  So IDEA's work is kind of like provide the best experience, user experience, and visualize the data coming in and coming out.
+[1101.32 --> 1107.22]  Like, I mean, all the information between this Delve and the front end.
+[1107.22 --> 1110.04]  So that's really nice then.
+[1110.22 --> 1119.58]  So you don't have to learn these complicated commands and you don't have to know about the Dwarf data or anything like that because it's integrated.
+[1119.78 --> 1120.80]  Yeah, ideally.
+[1121.54 --> 1121.68]  Yeah.
+[1122.06 --> 1127.82]  Because it's integrated, you get to just do it right in your code, the same place where you're writing the code.
+[1127.88 --> 1128.72]  So that's really cool.
+[1128.82 --> 1132.02]  How do you actually do that then in VS Code?
+[1132.46 --> 1133.54]  How do you set a breakpoint?
+[1133.54 --> 1138.16]  So actually what the nice thing about Delve is Delve has the API.
+[1138.66 --> 1145.16]  So like all this method and instruction, they are actually, they can be invoked through the RPC.
+[1145.94 --> 1150.32]  And so we just launched the headless Delve server.
+[1150.86 --> 1158.22]  And then from the VS Code, we also use, so from the VS Code, we just invoke this RPC.
+[1158.22 --> 1166.82]  And there is some recent movement about the debug adapter protocol that is a kind of like a standardized all this.
+[1167.14 --> 1173.28]  So not Delve is Go debugger, but there is GDB, there is LLDB, and there is a JavaScript debugger.
+[1173.44 --> 1175.70]  And like there are all kinds of debuggers, right?
+[1175.76 --> 1177.32]  And then we have VS Code.
+[1177.32 --> 1185.90]  So VS Code team, yeah, they try to standardize the interaction between debugger, just general debugger and editor.
+[1186.40 --> 1188.48]  So it's called the debug adapter protocol.
+[1189.52 --> 1194.94]  And like VS Code Go extension speaks Delve adapter protocol.
+[1195.22 --> 1201.70]  And there is a small tiny Delve adapter, a debug adapter that actually talks Delve RPC.
+[1201.70 --> 1208.52]  So it's a little bit complicated, but we try to minimize, like simplify this, like communication path.
+[1208.68 --> 1215.54]  So that next version, I hope like the communication is more efficient and like, yeah.
+[1216.06 --> 1219.20]  So yeah, that is the general direction we are heading at.
+[1219.58 --> 1220.68]  That's really cool then.
+[1220.76 --> 1221.20]  That's nice.
+[1221.30 --> 1225.84]  It's nice because we don't, as users of this, we don't sort of have to worry about that, right?
+[1225.84 --> 1229.46]  That's like something that happens behind the scenes.
+[1229.66 --> 1232.54]  We get to just use the VS Code interface.
+[1233.42 --> 1234.76]  So that, yeah, that's really great.
+[1245.12 --> 1248.84]  How much time does your team spend building and maintaining internal tooling?
+[1249.12 --> 1253.14]  I'm talking about those behind the scenes apps, the ones no one else sees.
+[1253.14 --> 1259.62]  The S3 uploader you built last year for the marketing team, that quick Firebase admin panel that lets you monitor key KPIs.
+[1259.92 --> 1264.90]  Maybe even the tool your data science team hacked together so they can provide custom ad spend analytics.
+[1265.48 --> 1267.42]  Now, these are tools you need so you build them.
+[1267.62 --> 1268.56]  And that makes sense.
+[1269.10 --> 1275.62]  But the question is, could you have built them in less time, with less effort, and less overhead and maintenance required?
+[1275.94 --> 1278.12]  And the answer to that question is, yes.
+[1278.56 --> 1279.86]  That's where Retool comes in.
+[1279.86 --> 1283.72]  Rohan Chopra, engineering director at DoorDash, has this to say about Retool.
+[1284.10 --> 1284.34]  Quote,
+[1284.34 --> 1292.58]  The tools we've been able to quickly build with Retool have allowed us to empower and scale our local operators, all while reducing the dependency on engineering.
+[1293.02 --> 1293.36]  End quote.
+[1293.84 --> 1300.10]  Now, the internal tooling process at DoorDash was bogged down with manual data entry, missed handoffs, and long turnaround times.
+[1300.10 --> 1309.64]  And after integrating Retool, DoorDash was able to cut the engineering time required to build tools by a factor of 10x and eliminate the error-prone manual processes that plague their workflows.
+[1310.04 --> 1314.14]  They were able to empower backend engineers who wouldn't otherwise be able to build frontends from scratch.
+[1314.54 --> 1319.52]  And these engineers were able to build fully functional apps in Retool in hours, not days or weeks.
+[1319.52 --> 1323.68]  Your next step is to try it free at retool.com slash changelog.
+[1323.84 --> 1326.28]  Again, retool.com slash changelog.
+[1326.28 --> 1350.96]  I'm interested.
+[1351.60 --> 1354.72]  Grant, your talk is, is your talk on Friday?
+[1354.72 --> 1356.28]  Uh, yes.
+[1356.48 --> 1358.60]  Friday, I think something around four o'clock.
+[1359.06 --> 1359.30]  Mm-hmm.
+[1359.72 --> 1362.00]  That's meaningless to me because I'm in the different time zone.
+[1362.30 --> 1363.26]  Oh, Eastern time.
+[1363.56 --> 1364.06]  Yeah, okay.
+[1364.46 --> 1366.24]  I think we should get rid of time zones, by the way.
+[1366.28 --> 1367.00]  I think that was a bug.
+[1367.34 --> 1369.34]  I think there's a bug in that, to be honest.
+[1369.76 --> 1371.80]  We should all just follow New York time.
+[1372.22 --> 1372.76]  If you like.
+[1372.98 --> 1376.16]  I mean, don't, you know, jump to an assumption there.
+[1376.48 --> 1380.82]  I mean, Greenwich, I live near Greenwich, which is actually where they invented time, I think.
+[1381.20 --> 1382.28]  So, do you know what I mean?
+[1382.84 --> 1384.24]  I live near Greenwich Village.
+[1384.72 --> 1384.92]  Yeah.
+[1385.88 --> 1386.24]  Okay.
+[1388.48 --> 1390.44]  I wasn't going to say there's a line and you've crossed it.
+[1390.64 --> 1394.72]  I was going to say in Greenwich, there's a line that's like the Meridian zero line.
+[1394.72 --> 1394.92]  Yeah.
+[1394.98 --> 1396.48]  You can sort of go and cross it.
+[1396.48 --> 1396.90]  I gotcha.
+[1397.40 --> 1398.14]  Yeah, it's all right.
+[1398.14 --> 1400.86]  So, yeah.
+[1400.90 --> 1403.56]  I was going to ask, what is your talk about on Friday?
+[1403.56 --> 1407.50]  So, my talk is about tracing Go programs with eBPF.
+[1408.30 --> 1416.72]  So, eBPF has been talked about a lot at various different conferences for the past two or three years.
+[1416.84 --> 1417.96]  It's been gaining a lot of momentum.
+[1417.96 --> 1422.68]  And it's a feature of the Linux kernel.
+[1422.68 --> 1424.24]  So, it's certainly Linux specific.
+[1424.24 --> 1432.02]  But what it allows you to do is add ad hoc, add logic to the Linux kernel.
+[1432.36 --> 1433.56]  And I know that's very abstract.
+[1433.56 --> 1447.44]  But the way that it's often put, Brendan Gregg, a leader in the eBPF space, likes to put it as eBPF does to the Linux kernel what JavaScript does to HTML.
+[1447.44 --> 1451.40]  So, you can attach eBPF programs.
+[1451.54 --> 1460.32]  You could think of them as scripts and attach them to various hooks, such as to network sockets every time a packet comes in and have some logic.
+[1460.64 --> 1466.72]  Or to kernel probes, which, you know, every time source code is executed within the Linux kernel itself.
+[1467.40 --> 1473.60]  And in particular, what my talk is about is attaching eBPF programs to something called uProbes.
+[1473.60 --> 1480.88]  And uProbes attach to what essentially is like source code symbols.
+[1481.22 --> 1492.32]  So, if you have a Go program that has a function called test function, you can attach a uProbe to that and attach an eBPF program to that uProbe.
+[1492.32 --> 1508.76]  So, that every time a process executes that function, so if you run that program and it's, you know, a running service or whatever else, you could have essentially a script respond to that function every time it's called.
+[1509.04 --> 1510.98]  So, you could print out what the arguments are.
+[1511.12 --> 1513.92]  You could have some logic for inspecting another area of memory.
+[1513.92 --> 1522.40]  You know, and it's useful for debugging, for monitoring, potentially for fuzzing or fault injection as well.
+[1523.20 --> 1528.92]  So, does the original function still run and you're just sort of intercepting it or do you replace it?
+[1529.62 --> 1530.74]  That's a really good question.
+[1530.86 --> 1532.24]  So, it does still run.
+[1532.48 --> 1535.82]  It doesn't stop the program from running at all.
+[1535.96 --> 1536.98]  It doesn't affect the process.
+[1537.16 --> 1542.70]  It runs in its own virtual machine inside the Linux kernel, actually.
+[1542.70 --> 1559.04]  So, the difference between, or, you know, I guess there's a lot of difference in terms of the underlying technology, but the advantage to eBPF, which I guess could also be seen as a disadvantage compared to buggers, is that it's not stopping the program.
+[1559.36 --> 1562.10]  It's not attaching to the process.
+[1562.64 --> 1570.84]  You can have a running program that is completely unaware of the fact that it's, you know, being inspected via eBPF.
+[1570.84 --> 1575.10]  Because you're doing it down at the low levels of the operating system, I guess.
+[1575.64 --> 1576.02]  Exactly.
+[1576.80 --> 1581.84]  So, what use cases are there for that then, from a kind of debugging practical standpoint?
+[1582.12 --> 1583.28]  What sorts of things can you do?
+[1583.76 --> 1586.10]  So, I have a project I'll shamelessly plug in.
+[1586.16 --> 1587.24]  I'm almost at 100 stars.
+[1587.44 --> 1588.42]  Go star it.
+[1588.60 --> 1589.54]  It's called Weaver.
+[1589.92 --> 1592.98]  It's on my GitHub, GrantSeltzer slash Weaver.
+[1592.98 --> 1598.48]  And what I'm trying to do is have strace-like functionality.
+[1599.06 --> 1604.38]  Strace is another tracing tool that you run a program, it'll print out every time a system call is executed.
+[1604.80 --> 1608.68]  Where I'm trying to have a functionality like that for Go programs.
+[1608.68 --> 1614.56]  Well, where you run a Go program, and every time any function inside of that is called.
+[1614.88 --> 1622.52]  So, you know, all of your functions in all of your packages, every time they're called, it will print out a line with,
+[1622.64 --> 1627.12]  it was called at this timestamp, and the arguments passed had these values.
+[1627.52 --> 1629.98]  And its return was X, Y, or Z.
+[1629.98 --> 1638.88]  So, the application there is, you know, for debugging purposes, let's say you want to know why you're getting some garbled output.
+[1639.30 --> 1647.42]  And you want to know at what point down the function call stack, a function was getting some weird output.
+[1647.64 --> 1655.68]  And you see somewhere along the line, like, this function for, like, printf, or, like, you know, a wrapper around printf is getting really weird output.
+[1655.68 --> 1659.28]  So, then you might want to start inspecting at the function that called that one.
+[1659.98 --> 1666.98]  It's also useful for, you know, not that I'm saying that this is the greatest idea.
+[1667.48 --> 1670.34]  It's still a developing ecosystem.
+[1670.34 --> 1680.92]  But you could attach these to services running in production because it has such a minimal effect on the performance of the service.
+[1680.96 --> 1683.18]  And you could attach it to running programs as well.
+[1683.98 --> 1684.50]  Hmm.
+[1684.94 --> 1686.34]  That is really interesting.
+[1686.34 --> 1689.04]  I mean, can you interact?
+[1689.04 --> 1695.22]  Can you, like, I guess you can't change things in these little eBPF programs, can you?
+[1695.78 --> 1700.02]  So, I have a little example of how you can, actually, in my talk.
+[1700.40 --> 1704.52]  There is a really good talk that was given at some security conference.
+[1704.80 --> 1705.94]  I can link it later.
+[1706.04 --> 1709.90]  But of how you can write essentially malicious code with eBPF.
+[1709.90 --> 1715.78]  But even for non-malicious purposes, you could actually write to memory from eBPF.
+[1715.84 --> 1719.72]  So, you can change the value of parameters, which I do in my talk.
+[1720.26 --> 1720.48]  Wow.
+[1720.88 --> 1723.56]  And would you recommend that or not sure yet?
+[1723.96 --> 1734.90]  I think it has its use cases if you are trying to do, let's say, something like fault injection, where, you know, you have processes that are communicating with one another.
+[1734.90 --> 1738.12]  And you have a function that is pulling in from another endpoint.
+[1738.76 --> 1750.72]  And if you don't want to have that external dependency, you could have an eBPF program that writes some garbled data to, you know, to a particular function and, you know, see how your program reacts to it.
+[1750.72 --> 1768.88]  And you could also, if you have a compiled and running service and you want to see if, you know, a particular fix to your source code will fix the issue, you can write, you know, insert a small eBPF program that writes correct data, you know, if it was getting incorrect data.
+[1769.02 --> 1774.08]  And if that fixes your whole issue, you know, that's, you know, the symptom of it.
+[1774.14 --> 1776.28]  But I guess it depends on a case-by-case basis.
+[1776.46 --> 1778.04]  Certainly not in production, I'll say that.
+[1778.16 --> 1780.60]  And so these scripts, what language are they?
+[1780.72 --> 1782.26]  Does it have its own little language?
+[1782.40 --> 1784.58]  Is it something that would be familiar to us?
+[1784.90 --> 1786.28]  So it would be familiar to you.
+[1786.38 --> 1787.38]  It's essentially C.
+[1787.60 --> 1789.16]  It's a subset of C.
+[1789.32 --> 1792.42]  There are some restrictions to it, but it essentially looks like C.
+[1792.48 --> 1794.86]  I guess the language could be called BPF.
+[1794.86 --> 1800.14]  And it's, you know, there's a verifier within the Linux kernel when you try and load the bytecode.
+[1800.30 --> 1802.76]  It's an LLVM-backed compiler.
+[1804.10 --> 1805.60]  It's a really interesting thing.
+[1805.60 --> 1807.42]  Do you think there's work?
+[1807.54 --> 1811.88]  What's sort of next before we can start using that kind of technique?
+[1812.06 --> 1815.22]  Is it because it feels like it's quite a new thing on the scene?
+[1815.30 --> 1815.92]  Has it been around?
+[1816.26 --> 1817.00]  Where did it come from?
+[1817.00 --> 1824.62]  So the original technology of it, I think, was, I don't even want to guess, like early 2000s.
+[1824.68 --> 1828.18]  It used to be strictly for network packet processing.
+[1828.86 --> 1833.82]  But I would say it's been within the past two years or so that the ecosystem has really developed.
+[1833.98 --> 1835.66]  There's a group of startups.
+[1835.82 --> 1840.98]  I know Facebook does a lot of EBPF stuff, and they've contributed to the community quite a bit.
+[1840.98 --> 1847.64]  I would say there's no better time to start doing it than right now because the ecosystem definitely is developing.
+[1847.82 --> 1862.42]  But there's a really strong community of people who, you know, really help one another, you know, try and figure this all out and define what good EBPF code looks like and what the ecosystem related to how it's related to Go looks like.
+[1863.30 --> 1867.66]  So I think it's best to get in at the ground floor, so to speak.
+[1868.66 --> 1869.66]  Very interesting.
+[1869.66 --> 1871.26]  Yeah, it's definitely something to play with.
+[1871.34 --> 1884.56]  It sounds like one of those things that can be extremely powerful, but also a bit like in C and C++, you can do like operator overloading and things, which is, if used correctly, it can be great.
+[1885.08 --> 1893.76]  But as soon as it's abused, you end up not knowing what an add means in the code, you know, what a plus symbol is doing to things.
+[1894.44 --> 1894.78]  Fair enough.
+[1894.92 --> 1898.38]  There's probably one of those things you would end up using it very cautiously, I suppose.
+[1898.38 --> 1899.90]  Yeah, that's fair.
+[1900.04 --> 1910.86]  I will say that the goal of my talk is to show how accessible the technology is and you don't have to have any expertise in the low levels of Linux or even of Go for that matter.
+[1910.98 --> 1919.46]  Like you could really start playing around with it and it can make a whole new class of problems much more accessible to so many more people.
+[1919.46 --> 1921.84]  Have you ever used it, Derek?
+[1922.12 --> 1922.86]  Are you aware of it?
+[1924.12 --> 1929.34]  Yeah, I've done a few things with EBPF a little bit here and there.
+[1929.34 --> 1938.74]  Actually, Delve has a trace functionality, which works somewhat similar, but it works at a higher level using ptrace and some of those other kind of syscalls.
+[1939.56 --> 1947.92]  And I've thought about experimenting a little bit, replacing on Linux systems that support it, replacing that with like an EBPF backed tracing system.
+[1947.92 --> 1953.00]  So Grant, if you ever want to send a pull request, we'd love to have it.
+[1954.26 --> 1955.28]  Yeah, I would.
+[1955.52 --> 1963.10]  Yeah, I am happy to integrate it from the VS Code side with the visualization.
+[1964.74 --> 1966.16]  I would love that.
+[1966.40 --> 1968.62]  This is the most productive meeting I've ever been in.
+[1969.10 --> 1970.76]  It wasn't even meant to be a meeting.
+[1970.76 --> 1985.20]  Yeah, and the benefit like of what Grant was talking about doing it, like the EBPF route versus, so Go does it kind of at a higher level using like ptrace syscalls and various other like syscalls on different platforms like Windows and stuff like that.
+[1985.58 --> 1993.78]  But the fundamental problem of why it's like slower than the approach that Grant described with EBPF is like EBPF stays all within the kernel.
+[1993.78 --> 1998.90]  So there's no context switching from kernel space to user space back to kernel space back to user space.
+[1998.90 --> 2001.44]  That context switch can get expensive.
+[2002.08 --> 2012.38]  So when Delve traces in kind of a more portable way, it traces in such a way where there's, you know, you do switch from the kernel to user space back to the kernel back to the user space.
+[2012.68 --> 2019.22]  And typically, you don't really see like that much of a slowdown if you're just tracing a program locally or something like that.
+[2019.22 --> 2026.92]  But certainly there is a performance hit there that could be like alleviated by switching to EBPF where appropriate, where possible.
+[2026.92 --> 2031.18]  But usually people are debugging not in production.
+[2031.64 --> 2032.80]  But I mean, does it change?
+[2032.88 --> 2033.68]  Does that change at all?
+[2033.76 --> 2037.46]  Or is this, we're not going to, we're still going to keep doing how we're doing it?
+[2037.82 --> 2038.06]  Do you know what I mean?
+[2038.66 --> 2045.98]  I think with EBPF, you could make the case that it's easier and a little bit safer and more rational to do in a production environment.
+[2045.98 --> 2052.86]  I would say I wouldn't recommend doing like a, like a delve trace on a production system unless you really, really had to.
+[2053.28 --> 2058.88]  For example, yeah, there's, you're just going to run into some performance penalties there is really kind of the biggest issue.
+[2058.88 --> 2072.20]  What up, Gophers?
+[2072.48 --> 2074.48]  Jared Santo here, your humble producer.
+[2075.16 --> 2078.28]  I want to take a quick moment to tell you about Changelog++.
+[2078.68 --> 2084.88]  It's our membership program where you can directly support GoTime and all the podcasts we create here at Changelog.
+[2084.88 --> 2090.02]  Ditch the ads, get closer to the metal and enjoy supporting GoTime into the future.
+[2090.60 --> 2093.46]  Learn more at changelog.com slash plus plus.
+[2093.78 --> 2096.56]  Once again, that's changelog.com slash plus plus.
+[2096.74 --> 2097.40]  Check it out.
+[2097.80 --> 2098.90]  We'd love to have you with us.
+[2098.90 --> 2119.56]  Hannah, you mentioned earlier that Delve has an API, an RPC API.
+[2119.96 --> 2120.60]  What is that?
+[2120.66 --> 2121.54]  What does that look like?
+[2121.70 --> 2122.90]  How do you consume that?
+[2122.96 --> 2125.88]  How does VS Code, is it an HTTP API?
+[2126.16 --> 2127.58]  Are there, is it protobuf?
+[2127.58 --> 2128.86]  How does it actually work?
+[2129.78 --> 2131.46]  Yeah, so Derek is here.
+[2131.96 --> 2135.02]  So it's a little bit weird to answer the question.
+[2135.28 --> 2139.76]  So I think, yeah, that is a JSON RPC, like one, right?
+[2139.88 --> 2144.48]  So it's just like the JSON streaming between client and server.
+[2144.74 --> 2145.72]  Like it's a simple one.
+[2146.24 --> 2151.18]  And yeah, that is a kind of like another JSON RPC 2 based protocol.
+[2151.18 --> 2154.12]  So just do JSON message exchange.
+[2154.56 --> 2158.36]  So you start the program, start the debugger, Delve.
+[2158.82 --> 2162.04]  And does it then return back some endpoint for you to hit?
+[2162.38 --> 2163.70]  Or how does it work?
+[2163.70 --> 2163.84]  Yeah.
+[2163.96 --> 2170.28]  So like, yeah, just to connect it to the socket, the network socket, like the port and then create a socket.
+[2170.68 --> 2173.62]  And yeah, communication over the socket.
+[2173.62 --> 2174.02]  Yeah.
+[2174.82 --> 2175.50]  Very cool.
+[2176.26 --> 2186.50]  Well, you see, I asked that because that's quite interesting because I think there's a whole space of tooling, particularly like static analysis or even other sort of runtime tools like debuggers.
+[2186.50 --> 2194.16]  And there's sort of a lot of choice for how to build those things so that they can be easily consumed by plugins and things.
+[2194.32 --> 2198.26]  So that's quite interesting to always is quite interesting to hear about that.
+[2198.90 --> 2203.74]  When did the VS Code plugin officially get taken up by the Go team?
+[2203.84 --> 2206.06]  Because it used to just be something else before, didn't it?
+[2206.86 --> 2207.08]  Yeah.
+[2207.20 --> 2211.00]  So it was originally owned by Microsoft, Microsoft team.
+[2211.00 --> 2218.72]  And I think VS Code Go was one of the earliest language supporting plugins VS Code team offered.
+[2219.26 --> 2223.34]  And then for a while, like it was in the maintenance mode.
+[2223.56 --> 2229.24]  And this year, actually, we got the responsibility to maintain.
+[2229.56 --> 2235.64]  So I think there was a blog post from blog.golang.org about this transition.
+[2235.64 --> 2242.18]  So now like the tool team inside of like Go team in Google.
+[2242.44 --> 2242.60]  Yeah.
+[2242.70 --> 2244.34]  We are maintaining this plugin.
+[2245.00 --> 2247.24]  How many is on the Go tool team?
+[2247.66 --> 2248.26]  Tool team.
+[2248.68 --> 2248.92]  Hmm.
+[2249.36 --> 2251.30]  I remember when there was just the Go team.
+[2251.74 --> 2253.90]  And now there's like there's a security team.
+[2254.00 --> 2255.30]  There's a team for tools.
+[2256.06 --> 2256.28]  Right.
+[2256.36 --> 2257.14]  It's really growing.
+[2257.68 --> 2258.02]  Yeah.
+[2258.24 --> 2258.94]  So, yeah.
+[2259.28 --> 2261.60]  So, yeah, there is a high demand.
+[2261.60 --> 2264.88]  So, yeah, we need a lot of work to do.
+[2265.60 --> 2269.86]  And so there is like currently the Go Place team.
+[2270.10 --> 2273.44]  That is like Go Place is one of the biggest project.
+[2273.82 --> 2275.24]  I don't know if you heard about it.
+[2275.48 --> 2275.82]  Yes.
+[2275.84 --> 2281.42]  That is the language like service implementation for the Go language.
+[2282.02 --> 2287.00]  And, yeah, the Go tool team is basically provide the best developer experience,
+[2287.00 --> 2290.92]  including the debug support or language intelligence support.
+[2291.38 --> 2296.20]  And, yeah, so VS Code Go is kind of like one of the projects.
+[2296.60 --> 2301.34]  And currently, like we are based on the New York and like a handful number of like a few
+[2301.34 --> 2307.10]  of us are working on various aspects of this developer experience improvement.
+[2307.10 --> 2307.50]  Yeah.
+[2308.60 --> 2313.08]  Well, we all appreciate all the work, of course, because it's very nice for us to just we just
+[2313.08 --> 2315.80]  get to use it and it hopefully makes our lives easier.
+[2316.00 --> 2318.60]  So I do like to thank people that have contributed.
+[2318.72 --> 2319.86]  This goes for everyone on this call.
+[2320.38 --> 2325.38]  Can you give us any spoilers about things that you're working on now that we might see soon?
+[2326.06 --> 2326.90]  Won't tell anyone.
+[2328.66 --> 2329.06]  I will.
+[2330.34 --> 2332.48]  That is technically, that is legally watertight.
+[2334.70 --> 2336.24]  So what's coming next?
+[2337.10 --> 2337.52]  Yeah.
+[2337.78 --> 2349.56]  So, yeah, we are currently like working really hard to use GoPlace as a default Go intelligent,
+[2349.92 --> 2352.54]  yeah, Go language service.
+[2353.06 --> 2358.98]  And also we are now currently working on, like, I think I talked about it, right?
+[2358.98 --> 2364.64]  The debug adapter plugin, adapter protocol, so that we can simplify and then provide like
+[2364.64 --> 2370.42]  a more performant, like a debugging experience from the VS Code users.
+[2371.12 --> 2371.80]  So, yeah.
+[2372.06 --> 2376.58]  So that is, yeah, they are the two big main projects I'm currently working on.
+[2377.26 --> 2378.04]  Oh, great.
+[2378.14 --> 2378.82]  Yeah, sounds good.
+[2379.02 --> 2380.74]  And what about for Delve?
+[2380.74 --> 2386.18]  I mean, is that pretty much kind of done or is there a roadmap there?
+[2386.82 --> 2388.94]  I'm interested in what's coming next for that too.
+[2390.50 --> 2394.14]  Yeah, it's one of those things where it's still constantly evolving.
+[2394.14 --> 2396.72]  We have a few kind of big things planned.
+[2396.96 --> 2399.96]  We always work to keep up to date with the latest Go release.
+[2400.16 --> 2401.94]  So Go 1.16 is coming out soon.
+[2402.06 --> 2407.20]  With each release like that, there's subtle things that may change in the runtime or how
+[2407.20 --> 2409.96]  the binaries are put together that Delve kind of has to adapt to.
+[2409.96 --> 2415.68]  So we continuously work on supporting the latest release, making sure that, you know, by the
+[2415.68 --> 2418.94]  time that release comes out, there's a Delve version that can support and debug it.
+[2419.30 --> 2421.04]  So that's always kind of a big thing for us.
+[2421.16 --> 2425.16]  We also have a few kind of interesting like features coming up down the line.
+[2425.42 --> 2430.78]  So my co-maintainer is working on a feature where you can, during an interactive debug session,
+[2430.78 --> 2435.78]  you can create and produce a core dump from the process that you're debugging.
+[2435.92 --> 2439.38]  So it's similar to like G-Core, if folks have ever used something like that.
+[2439.38 --> 2441.00]  But works a little bit differently.
+[2441.26 --> 2443.50]  So that's kind of a cool feature that's coming up.
+[2443.94 --> 2449.28]  Another big push that we're kind of trying to do is improve the overall architecture support.
+[2449.56 --> 2455.26]  So right now, Delve actually only supports a subset of all of the architectures that Go
+[2455.26 --> 2456.18]  can actually run on.
+[2456.50 --> 2462.46]  And it supports the main ones that folks actually use, you know, AMD 64, ARM 64, things like that.
+[2462.58 --> 2467.88]  But there are some kind of outlier architectures that Go supports that Delve doesn't yet that
+[2467.88 --> 2469.56]  we're also working on as well.
+[2469.74 --> 2472.84]  So there's a pull request up right now for supporting 32-bit ARM.
+[2473.06 --> 2478.62]  We're looking at supporting like PowerPC 64 and S390X, which are kind of weird architectures.
+[2479.12 --> 2482.58]  But those are kind of some of the bigger things that we have on the roadmap so far.
+[2483.04 --> 2484.18]  What about Apple Silicon?
+[2484.18 --> 2486.26]  Yeah, so that's an interesting one.
+[2486.36 --> 2490.92]  Because with Delve, we actually have like a so we have a few different backends that Delve
+[2490.92 --> 2491.48]  can actually use.
+[2491.56 --> 2496.90]  So there's a native backend, which we actually we wrote and maintain and we can actually interact
+[2496.90 --> 2497.70]  with other backends.
+[2497.82 --> 2502.84]  So like the talk that I'm giving tomorrow is on using Mozilla RR as a backend to do like
+[2502.84 --> 2503.84]  record replay debugging.
+[2503.84 --> 2511.26]  So with that, Delve on macOS actually uses LODB server as the backend.
+[2511.44 --> 2518.26]  We have a native Mac backend, but it turns out that the documentation for the mock kernel
+[2518.26 --> 2519.70]  is horrendous.
+[2519.84 --> 2524.52]  And trying to figure out like how to actually work and interact with that kernel means like
+[2524.52 --> 2529.98]  when I wrote the original backend for Mac, it was digging through like the open source
+[2529.98 --> 2534.90]  kernel to figure out some of these like ptrace commands and some of these weird stuff that
+[2534.90 --> 2538.74]  I that I had to do because the documentation is subpar for that kind of thing.
+[2538.86 --> 2541.84]  So all that to say, we use LODB server on the backend.
+[2542.00 --> 2545.54]  So there's some kind of changes that we have to do internally with Delve.
+[2545.68 --> 2550.96]  But some of the heavy lifting we kind of get for free by using LODB server, which, you
+[2550.96 --> 2555.16]  know, Apple is certainly going to make work on their silicon.
+[2555.58 --> 2557.46]  So, okay, great.
+[2557.46 --> 2557.78]  Wow.
+[2557.94 --> 2562.66]  So Delve really is kind of a big thing because I always think of it as this little tool.
+[2563.52 --> 2564.64]  I mean, how big is it?
+[2565.08 --> 2566.80]  Big in terms of what metric?
+[2567.26 --> 2567.70]  Size.
+[2568.08 --> 2569.94]  It's a fairly, I mean, no, I don't know.
+[2570.14 --> 2575.58]  The scope of the actual source code and all that stuff, it's definitely grown and it's
+[2575.58 --> 2580.54]  grown a little bit in complexity over the years as we've introduced like different backends
+[2580.54 --> 2581.56]  and things like that.
+[2581.62 --> 2586.90]  The goal has kind of always been to keep it as simple and straightforward from a code perspective
+[2586.90 --> 2587.88]  on this possible.
+[2588.06 --> 2592.48]  But, you know, over time, obviously things get more complicated and you have to deal
+[2592.48 --> 2593.62]  with weird situations.
+[2594.32 --> 2598.86]  But yeah, I mean, from just, you know, like the perspective of the code and stuff like
+[2598.86 --> 2601.68]  that, the project itself, it's a fairly big project at this point.
+[2602.32 --> 2603.24]  Yeah, it sounds like it.
+[2603.24 --> 2610.20]  When new features come to Go, like say, generics lands in Go, what will that mean for Delve?
+[2610.54 --> 2614.16]  You know, is there things you're just going to get for free or will there be times when
+[2614.16 --> 2618.30]  certain language features are added that that creates a lot of work for you?
+[2618.64 --> 2619.20]  It depends.
+[2619.36 --> 2625.08]  So a lot of that we would get for free a little bit by the kind of debug information that's
+[2625.08 --> 2627.92]  provided from like Go binaries and things like that.
+[2627.92 --> 2633.64]  So it would kind of be up to the Go compiler and linker to produce the correct information
+[2633.64 --> 2636.64]  that Delve needs to be able to debug that stuff properly.
+[2637.22 --> 2641.24]  And with big new features like that, sometimes the support is there, sometimes it's not.
+[2641.40 --> 2646.72]  You know, sometimes we have to work with folks upstream to get that in or submit some patches
+[2646.72 --> 2648.22]  ourselves and things like that.
+[2648.52 --> 2652.10]  But a lot of it comes with like just coordination with the Go team.
+[2652.10 --> 2657.70]  There's certain things that are Go specific that, you know, we've had to work really closely
+[2657.70 --> 2659.76]  with the Go team to be able to achieve.
+[2660.06 --> 2661.66]  Like, for example, function calls.
+[2661.76 --> 2665.30]  This is something that actually requires support from the Go runtime.
+[2665.76 --> 2668.50]  And we had to work with the Go team to kind of make that happen.
+[2668.60 --> 2669.54]  It was a coordinated effort.
+[2669.74 --> 2671.34]  So sometimes there's more coordination.
+[2671.56 --> 2672.70]  Sometimes we get stuff for free.
+[2673.76 --> 2675.04]  Oh, cool.
+[2675.82 --> 2679.54]  Okay, well, it's time for our regular slot.
+[2679.54 --> 2682.48]  It's time for Unpopular Opinions.
+[2699.00 --> 2702.08]  So who wants to kick us off with an unpopular opinion?
+[2702.92 --> 2707.68]  I will say that when you mentioned this, I was going to say that print statements are okay
+[2707.68 --> 2710.94]  for debugging, but I don't think that will be that unpopular.
+[2710.94 --> 2719.64]  So I will say that baseball is the by far most exciting sport in the world.
+[2720.00 --> 2720.44]  Baseball?
+[2720.54 --> 2721.20]  Which one's that?
+[2721.40 --> 2724.58]  It's the one with all the bases and the ball.
+[2724.98 --> 2725.56]  Clues in the name.
+[2726.48 --> 2726.88]  Absolutely.
+[2727.30 --> 2728.42]  Clever name now, actually.
+[2728.52 --> 2731.00]  I genuinely didn't actually make that link.
+[2731.00 --> 2736.92]  Well, baseball, it gives us lots of metaphors, doesn't it?
+[2737.14 --> 2739.62]  It contributes the most metaphors, but I don't know.
+[2740.12 --> 2742.64]  Hannah, is baseball a good sport?
+[2742.64 --> 2743.56]  Hannah, do you agree?
+[2743.68 --> 2744.78]  Is baseball a good sport?
+[2745.42 --> 2750.32]  So other than US and some Asian countries, who play baseball?
+[2750.90 --> 2751.70]  Yeah, I don't know.
+[2752.26 --> 2752.86]  Latin America?
+[2753.56 --> 2753.72]  Yeah.
+[2753.84 --> 2754.68]  Oh, yeah.
+[2755.22 --> 2756.12]  And in Europe?
+[2756.38 --> 2757.12]  No, not really.
+[2757.12 --> 2760.94]  No, we have kind of versions, different versions of it.
+[2761.18 --> 2761.60]  I don't know.
+[2761.74 --> 2764.54]  But yeah, that is potentially unpopular.
+[2764.78 --> 2765.40]  We'll just...
+[2765.40 --> 2767.44]  But they are missing the best sport, right?
+[2767.84 --> 2768.52]  Apparently so.
+[2768.80 --> 2769.04]  Yeah.
+[2769.28 --> 2770.66]  That's what we've heard.
+[2770.92 --> 2771.78]  According to Grant.
+[2772.02 --> 2772.18]  Yeah.
+[2772.80 --> 2774.86]  Derek, is baseball the best sport?
+[2775.38 --> 2777.24]  Best or most exciting?
+[2777.40 --> 2780.16]  I would refute most exciting.
+[2780.34 --> 2782.78]  I think football is pretty exciting.
+[2783.10 --> 2784.52]  I get excited watching...
+[2784.52 --> 2786.04]  I don't know if you consider this a sport,
+[2786.04 --> 2789.44]  but I like watching poker championships and stuff,
+[2789.54 --> 2790.40]  and that's pretty exciting.
+[2790.66 --> 2792.70]  So it depends on your metric.
+[2793.22 --> 2795.80]  I watch the StarCraft online, the StarCraft championships.
+[2796.50 --> 2796.70]  Yeah.
+[2796.84 --> 2797.20]  There you go.
+[2797.26 --> 2797.60]  That's exciting.
+[2798.34 --> 2798.54]  Yeah.
+[2798.98 --> 2799.16]  Yeah.
+[2799.26 --> 2802.48]  But I just don't go outside, so I've certainly never played baseball.
+[2803.44 --> 2805.16]  I think baseball is exciting.
+[2805.66 --> 2805.84]  Yeah.
+[2805.92 --> 2809.14]  Especially because when you watch baseball,
+[2809.38 --> 2811.96]  you eat hot dog, you drink beer.
+[2812.28 --> 2813.62]  How cool is it, right?
+[2813.70 --> 2815.06]  That's the most exciting one.
+[2815.06 --> 2817.16]  Like soccer, you have to watch.
+[2818.28 --> 2819.52]  Basketball, you have to watch.
+[2819.98 --> 2821.92]  Baseball, it's so slow and relaxing.
+[2822.10 --> 2822.80]  Best sport.
+[2823.02 --> 2823.12]  I agree.
+[2823.12 --> 2824.70]  You don't have to pay attention to it.
+[2824.76 --> 2825.60]  That's how good it is.
+[2825.78 --> 2826.52]  That's how exciting.
+[2827.04 --> 2827.14]  Yeah.
+[2827.64 --> 2829.24]  You can just focus on your hot dog.
+[2830.40 --> 2832.38]  Do we have any other unpopular opinions?
+[2832.58 --> 2834.40]  And by the way, we test these on our Twitter,
+[2834.82 --> 2836.06]  at GoTimeFM.
+[2836.06 --> 2840.06]  So we'll find out if that is indeed popular or not.
+[2840.32 --> 2840.74]  Any others?
+[2840.94 --> 2841.82]  We've got a couple of minutes.
+[2842.92 --> 2845.06]  I have an opinion.
+[2845.48 --> 2848.56]  The world will be better if everybody uses Linux.
+[2849.44 --> 2849.70]  Oh.
+[2850.84 --> 2851.52]  Now that's okay.
+[2851.78 --> 2851.94]  Oh, yeah.
+[2852.28 --> 2853.16]  It's controversial.
+[2853.54 --> 2853.88]  Oh, yeah.
+[2853.88 --> 2857.54]  All the EVPF, like all this Ptrace,
+[2857.70 --> 2860.52]  they are not available in other platforms.
+[2860.52 --> 2864.60]  But what about every other app in the world?
+[2865.94 --> 2867.76]  But I suppose if everyone was using it,
+[2868.30 --> 2869.40]  if everyone's using it,
+[2869.50 --> 2871.42]  they would work too, wouldn't they?
+[2872.26 --> 2872.90]  That's a fair one.
+[2872.98 --> 2873.06]  Yeah.
+[2873.68 --> 2875.22]  Derek, do you have an unpopular opinion?
+[2875.62 --> 2876.38]  I don't think so.
+[2876.62 --> 2877.98]  Do you agree with the Linux one?
+[2878.32 --> 2879.78]  I think the world could be a better place
+[2879.78 --> 2880.74]  if everybody used Linux.
+[2882.56 --> 2884.30]  I'm not as creative as everybody else.
+[2884.36 --> 2885.82]  I don't have anything off the top of my head.
+[2886.58 --> 2887.36]  That's unpopular.
+[2887.58 --> 2888.64]  You created Delve.
+[2889.44 --> 2890.02]  You've done it.
+[2890.42 --> 2892.40]  You've accidentally fulfilled
+[2892.40 --> 2893.64]  your contractual obligations
+[2893.64 --> 2896.86]  to provide an unpopular opinion for us.
+[2897.06 --> 2899.96]  We are running out of time.
+[2900.30 --> 2902.06]  So I really only have time to say
+[2902.06 --> 2904.14]  thank you so much for doing this.
+[2904.22 --> 2905.12]  It was a great conversation.
+[2905.54 --> 2906.98]  I wish we could spend more time.
+[2907.06 --> 2908.74]  And in fact, we'll invite you back
+[2908.74 --> 2909.92]  at some point to come and do
+[2909.92 --> 2911.12]  another GoTime episode.
+[2911.64 --> 2914.04]  Thank you very much to my guests,
+[2914.18 --> 2915.22]  Hannah, Grant and Derek.
+[2915.78 --> 2916.72]  It's been a pleasure.
+[2917.66 --> 2918.02]  Goodbye.
+[2918.94 --> 2919.64]  Thank you so much.
+[2919.64 --> 2920.36]  Thank you.
+[2920.62 --> 2920.98]  Thank you.
+[2920.98 --> 2922.22]  Yeah, thank you for inviting me on.
+[2922.22 --> 2927.88]  If you enjoyed this episode,
+[2928.20 --> 2930.28]  subscribe now at GoTime.fm.
+[2930.68 --> 2932.42]  Hey, we are getting close to the end of the year
+[2932.42 --> 2934.10]  and you may be dusting off the old blog
+[2934.10 --> 2936.88]  to write that epic best of or worst of post.
+[2937.38 --> 2939.76]  If so, we'd love if you'd include GoTime
+[2939.76 --> 2940.54]  in your list of favorites.
+[2941.06 --> 2942.72]  Let us know on Twitter when you publish.
+[2942.72 --> 2944.74]  I can pretty much guarantee you a retweet
+[2944.74 --> 2945.96]  from at GoTime.fm.
+[2946.42 --> 2948.74]  Music for GoTime is produced by the Mysterious One,
+[2948.96 --> 2949.78]  Breakmaster Cylinder,
+[2949.96 --> 2951.62]  and we're brought to you by awesome sponsors.
+[2952.12 --> 2954.78]  Thanks again to Fastly, Linode, and LaunchDarkly.
+[2955.22 --> 2956.26]  That's our show.
+[2956.52 --> 2957.40]  On the next episode,
+[2957.52 --> 2960.06]  Ellen Corbis joins Matt, Chris, and Natalie
+[2960.06 --> 2962.46]  to discuss Go in other spoken languages.
+[2962.56 --> 2963.46]  It's a good one,
+[2963.62 --> 2964.62]  so stay tuned for that
+[2964.62 --> 2965.96]  next week.
+[2992.46 --> 3022.44]  Thank you.
+[3022.46 --> 3052.44]  Thank you.
+[3052.46 --> 3055.00]  You'd have told me you were going to do that and then I'd have done an index earlier.
+[3056.16 --> 3056.94]  As it is.
+[3057.10 --> 3057.24]  Yeah.
+[3057.52 --> 3057.82]  Nothing.
+[3058.52 --> 3064.52]  All my jokes are from my kids because they ask Alexa for jokes all the time and they're
+[3064.52 --> 3066.96]  just the dumbest jokes ever.
+[3067.48 --> 3070.20]  And those are literally the only ones I know because they'll ask me like six times.
+[3070.26 --> 3073.22]  I have a bunch of kids so each kid will ask me the same joke because they just learned
+[3073.22 --> 3074.16]  it from their brother or sister.
+[3074.72 --> 3076.02]  They think I've never heard it.
+[3076.02 --> 3076.90]  I've heard them all six times.
+[3078.76 --> 3080.54]  What has four wheels and flies?
+[3081.50 --> 3082.18]  A garbage truck.
+[3082.58 --> 3082.76]  See?
+[3083.12 --> 3083.62]  They're not funny.
+[3084.24 --> 3085.38]  But I knew it immediately.
+[3086.06 --> 3086.76]  It's not funny.
+[3086.88 --> 3087.22]  What does it fly?
+[3087.80 --> 3088.46]  It has flies.
+[3088.76 --> 3089.60]  Oh, it has flies.
+[3089.96 --> 3090.18]  Yeah.
+[3090.36 --> 3092.14]  What has four wheels and flies?
+[3092.56 --> 3093.10]  Oh, right.
+[3093.66 --> 3094.10]  There you go.
+[3094.16 --> 3094.48]  Ah, see?
+[3094.52 --> 3095.10]  It is funny.
+[3096.04 --> 3097.50]  It's better than you thought it was.
+[3098.00 --> 3098.16]  Yeah.
+[3098.66 --> 3099.76]  I need to explain it.
+[3099.82 --> 3102.62]  But yeah, once it's explained, I'm all over it.
+[3102.94 --> 3105.28]  Did you know that ducks can float?
+[3105.62 --> 3105.96]  Can they?
+[3106.44 --> 3106.72]  Yeah.
+[3106.86 --> 3107.50]  I didn't know that.
+[3107.64 --> 3108.00]  What do you mean?
+[3108.28 --> 3108.84]  Of course they do.
+[3108.90 --> 3110.16]  They're always on the water, aren't they?
+[3110.66 --> 3111.52]  Is this another joke?
+[3112.08 --> 3113.00]  No, it's not a joke.
+[3113.18 --> 3113.52]  Flunt back?
+[3113.52 --> 3114.02]  Ducks can float.
+[3114.52 --> 3114.70]  Yeah.
+[3114.70 --> 3116.46]  I think it's pretty cool.
+[3117.08 --> 3118.16]  Are they floating or swimming?
+[3118.60 --> 3119.86]  I guess a little bit of both.
+[3120.38 --> 3120.60]  Hmm.