| === woody86_ is now known as woody86 | |
| === santiago-ve is now known as Guest94206 | |
| === Guest94206 is now known as santiago-ve | |
| === woody86_ is now known as woody86 | |
| === kinnaz_ is now known as KiNnaZ | |
| === Rafik_ is now known as Rafik | |
| [11:32] * meborc is away: FPS - CS, ETQW and then a little teeworlds | |
| === johanvd_ is now known as johanvd | |
| === onsho_ is now known as onsho | |
| === hannes__ is now known as johannesbeere | |
| === oly- is now known as oly | |
| === oly is now known as oly- | |
| === WastePotato is now known as Guest2365 | |
| === oly- is now known as oly | |
| === ilia is now known as ilia_ | |
| === ilia_ is now known as ilia | |
| === popey changed the topic of #ubuntu-classroom to: Next Session: "Ask Mark" With Mark (sabdfl) Shuttleworth | Welcome to Openweek, questions in #ubuntu-classroom-chat please || Session details here: https://wiki.ubuntu.com/UbuntuOpenWeek | |
| === Silvy is now known as Fierelin | |
| === zever_ is now known as zever | |
| [14:46] <jcastro> win 38 | |
| [14:46] <soren> lose 97 | |
| [14:47] <jcastro> heh | |
| [14:47] * geser hands jcastro some / / / / / / / / / / | |
| [14:47] * sebner hands jcastro some gtk :P | |
| [14:47] <jcastro> A little over 10 minutes until Q+A with Mark Shuttleworth! | |
| [14:48] <sudobash> really wow | |
| [14:49] <sudobash> let me guess as soon as it hits 10 he comes in and there is a lock put on speaking in the channel which only allows voiced and ops to speek? | |
| [14:49] <lordnoid> there's a chat in #ubuntu-classroom-chat, so I don't really care. | |
| [14:49] <jcastro> Ask questions in #ubuntu-classroom-chat and I will paste the questions in here | |
| [14:50] <sudobash> lol i was right.... | |
| === rockstar` is now known as rockstar | |
| [14:54] <kosmos342> hi all | |
| [14:55] <thiebaude> hi | |
| [14:57] <kandj> hi sabdfl! | |
| [14:57] <sabdfl> howdy | |
| [14:57] <thiebaude> hi mark | |
| [14:58] <sudobash> whats up | |
| [14:58] <artir> all hail mark! | |
| [14:58] <snap-l> Hi sabdfl | |
| [14:58] <RainCT> Hi | |
| [14:58] <jcastro> good morning! | |
| [14:58] <amstella> good morning! | |
| [14:58] <thiebaude> hi jcastro | |
| === jcastro changed the topic of #ubuntu-classroom to: Current Session: Q+A with Mark Shuttleworth | Welcome to Openweek, questions in #ubuntu-classroom-chat please || Session details here: https://wiki.ubuntu.com/UbuntuOpenWeek | |
| [14:58] <HigH5> Good evening :) | |
| [14:58] <andylockran> howdy | |
| [14:59] <jcastro> Just a minute or so | |
| [14:59] <jcastro> And then we can begin | |
| [15:00] <jcastro> Ok thanks everyone for stopping by for another Ubuntu Open Week session | |
| [15:01] <jcastro> This 2 hour block will be a Question and Answer with Mark Shuttleworth, founder of Ubuntu | |
| [15:01] <sabdfl> alrighty | |
| [15:01] <jcastro> Question : First, thank you for all you do in Ubuntu. Intrepid is the best release yet and I will continue to show my support for it. Now for my question. The wireless applet is working beautifully now! How do you see the FCC's move to open up a wireless spectrum affecting Ubuntu? And do you think it will be a problem implementing it? Sorry for such a long question. Cheers | |
| [15:01] <sabdfl> thanks Jorge, it's a pleasure to be joining the Jorge-and-Jono show! | |
| [15:02] <sabdfl> much credit for the wireless applet belongs upstream, where NM is developed | |
| [15:02] <sabdfl> though the Ubuntu team has worked hard to smooth some rough edges, particularly on the new 3G pieces | |
| [15:02] <sabdfl> i do think regulatory authorities around the world are getting smarter about wireless spectrum management | |
| [15:03] <sabdfl> i think that's making it easier for wifi / bluetooth / 3g hardware manufacturers to support linux properly - i.e. with free software | |
| [15:03] <sabdfl> i don't know if there are plans to make the new "TV whitespace" bandwidth usable from PC's and Linux | |
| [15:03] <sabdfl> but since Linux now supports more devices than any other OS (stats are wonderful ;-)) the future looks bright | |
| [15:04] <sabdfl> did that answer your Q ok? | |
| [15:04] <sabdfl> Next! | |
| [15:05] <sabdfl> jcastro: next? | |
| [15:06] <sabdfl> gosh, that was easy :-) | |
| [15:06] <jcastro> QUESTION: Ubuntu tends to focus on the latest release, regardles of LTS. Most of the new fixes and features make it into the newest releases without getting many backports. Are there plans to address this perception? | |
| [15:06] <sabdfl> Looking at Hardy, there have been many, many updates | |
| [15:06] <sabdfl> most of those related to security or critical issues | |
| [15:07] <sabdfl> with a few SRU's (stable release updates - new versions) | |
| [15:07] <sabdfl> we need to be appropriately conservative with the primary repositories | |
| [15:07] <sabdfl> i believe there's also a good community of backporters | |
| [15:07] <sabdfl> perhaps jcastro can find me a URL for the backports archive? | |
| [15:08] <jcastro> It's in the software sources, you just check the box in the preferences | |
| [15:08] <sabdfl> and then there are hundreds of PPA's, with versions of packages there | |
| [15:08] <sabdfl> i think backports are very important and if there's something else we can do to assist the community processes around them, the CC and TB would be happy to listen | |
| [15:08] <sabdfl> jcastro: next? | |
| [15:09] <jcastro> System->Administration->Software Sources and check the appropriate box for those interested in checking out backports | |
| [15:09] <jcastro> QUESTION: Do you plan to introduce *official* ubuntu HCL (hardware compatibility list), somethinig like https://hardware.redhat.com/ ? If so, when we will see it and what hardware you will start certifying (servers/workstations/laptops/controllers) | |
| [15:09] <sabdfl> there is already such a list! | |
| [15:09] <sabdfl> let me look for the URL | |
| [15:10] <sabdfl> http://webapps.ubuntu.com/certification/list/?release=8.04%20LTS | |
| [15:10] <jcastro> http://hwdb.ubuntu.com/ | |
| [15:10] <sabdfl> the overall view is at http://webapps.ubuntu.com/certification/ | |
| [15:10] <sabdfl> we can add machines there if the manufacturer approaches us | |
| [15:11] <sabdfl> and there are many more on the way | |
| [15:11] <jcastro> QUESTION: Organisations like the BBC are releasing custom versions of their iPlayer application for platforms such as the iPhone, nokia phones .etc... Do Canonical aim to liase with the BBC (and other similar organisations) to bring apps to ubuntu, and to Linux as a whole? | |
| [15:11] <sabdfl> both servers and laptops / desktops | |
| [15:11] <sabdfl> the BBC is doing very interesting work with their content | |
| [15:12] <sabdfl> in 8.10 there is the beginnings of a framework to bring their content directly to FLOSS desktops | |
| [15:12] <sabdfl> i think that will expand, as they seem very committed to open access | |
| [15:12] <sabdfl> hopefully, it's also a catalyst for other content providers to get invovled | |
| [15:13] <sabdfl> there's a positive trend towards drm-free content, since the music industry realised that DRM was mostly a mistake | |
| [15:13] <jcastro> QUESTION: Canoncial is not that big currentlic compared to novell, redhat. Are there plans to get more employees: in general and/or developers(maybe MOTU) ? | |
| [15:13] <sabdfl> we're growing as fast as we can! | |
| [15:13] <sabdfl> we have a long hiring list | |
| [15:13] <sabdfl> ubuntu.com/employment is the tip of the iceberg | |
| [15:14] <sabdfl> but i'm conscious of the need to grow sustainably | |
| [15:14] <sabdfl> it feels like a delightful place to work and i want to protect and defend that | |
| [15:14] <sabdfl> since i'm here all the time :-) | |
| [15:14] <sabdfl> and i think the other team leads feel the same way | |
| [15:14] <sabdfl> we do like to hire from the community, because we know then that people are committed to the values of free software | |
| [15:15] <sabdfl> and can work on a global distributed basis | |
| [15:15] <sabdfl> i think we make a big difference despite being only 7% of the size of Red Hat | |
| [15:15] <sabdfl> i respect the other companies in the industry for their work, but i do think Canonical is special | |
| [15:15] <sabdfl> next? | |
| [15:15] <jcastro> QUESTION: What is for you the biggest Challenge and the biggest priority to be met by Ubuntu in each release? | |
| [15:16] <sabdfl> the biggest challenge is to make sure we deliver the very best free software release that we can | |
| [15:16] <sabdfl> we want to make sure that amazing work which has been done in the FLOSS ecosystem is available to every ubuntu user on demand, elegantly | |
| [15:16] <sabdfl> so we have to work hard to keep up with what's going on out there | |
| [15:16] <sabdfl> we also want to make sure we add value to the hard work of the debian community | |
| [15:17] <sabdfl> and make it as easy as possible for others to benefit from our efforts | |
| [15:17] <sabdfl> we also have to make tough decisions with every release, about what gets in and what does not | |
| [15:17] <sabdfl> those are the challenges | |
| [15:18] <sabdfl> the upside is we have a fantastic community that shares those goals | |
| [15:18] <sabdfl> and many hands make slightly lighter if more complex work :-) | |
| [15:18] <sabdfl> but definitely, more interesting work :-) | |
| [15:18] <sabdfl> next? | |
| [15:18] <jcastro> QUESTION: Ubuntu aims to be one of the user friendliest distributions out there. That, however, increasingly affects the speed of the system, since many services need be started and kept running in the background. Are there plans to increase the systems speed beyond the usual marketing "faster than ever" phrases? Thanks | |
| [15:18] <sabdfl> it's a good question | |
| [15:19] <sabdfl> we were all inspired by the work at Intel which demonstrated super-fast boot times on Linux | |
| [15:19] <sabdfl> now we want to turn that into something that can be maintained and managed | |
| [15:19] <sabdfl> there is an interesting tradeoff between ease-of-development and resource management | |
| [15:19] <sabdfl> for example, mono and python make for rapid development | |
| [15:19] <sabdfl> but do impose a performance overhead | |
| [15:20] <sabdfl> we also have to try and choose the sweet spot of hardware | |
| [15:20] <sabdfl> for example, 3D graphics are not around on older PC's | |
| [15:20] <sabdfl> do we aim to make Ubuntu work on those, or aim to lead the user experience front? | |
| [15:20] <sabdfl> fortunately, we have parts of the community that specialise in some of those areas, like Xubuntu | |
| [15:21] <sabdfl> and i feel that allows us to live on the other edge | |
| [15:21] <sabdfl> when it comes to actual raw performance, i've seen very good numbers from ubuntu | |
| [15:21] <sabdfl> for example, running virtual machines under KVM, Xen or VMWare, Ubuntu does extremely well | |
| [15:22] <jcastro> QUESTION: Are there any Internships possibilities for International Students/Graduates @ Canonical?? (And specially in the US??) | |
| [15:22] <sabdfl> if you have specific areas where you think we could re-engineer for lower overhead and better performance, raise them with the TB | |
| [15:22] <sabdfl> yes, there would be | |
| [15:22] <sabdfl> write to hr@canonical.com, especially if you can articulate an area that you think you would be able to make a specific contribution in the time available | |
| [15:22] <sabdfl> next? | |
| [15:22] <jcastro> QUESTION: in 2005 (at the early ubuntu times) i remember that the community was discussing to get out a repository tree which is on the constant bleeding edge. (no freeze times ect...) will we see such a constant bleeding edge Version of ubuntu any time ? | |
| [15:23] <sabdfl> i would like to! | |
| [15:23] <sabdfl> we used to describe that as "The Grumpy Groundhog", but we haven't implemented it yet | |
| [15:23] <sabdfl> PPA's have made some of that possible - there are lots of PPA's with daily builds of cool packages | |
| [15:24] <sabdfl> deb http://ppa.launchpad.net/bzr-nightly-ppa/ubuntu intrepid main | |
| [15:24] <sabdfl> try that one on for size :-) | |
| [15:24] <sabdfl> during Jaunty, i think we will make nightly builds of a couple of major floss packages available | |
| [15:24] <sabdfl> probably kernel, X, oo.o, firefox, and the like | |
| [15:25] <sabdfl> we haven't worked out all the details (come to UDS in Mountain View in December if you're interested) | |
| [15:25] <sabdfl> but the idea would be to make it easy for people to test today's code from upstream | |
| [15:25] <sabdfl> and to make it easy to verify if a bug has been fixed in the latest trunk | |
| [15:25] <sabdfl> if that experiment goes well, we can build on it | |
| [15:25] <sabdfl> next? | |
| [15:25] <jcastro> QUESTION: Have you had much success with distributing Ubuntu with Dell? | |
| [15:25] <sabdfl> yes, i think we can say that now :-) | |
| [15:26] <sabdfl> we have three different programs in place, for different markets | |
| [15:26] <sabdfl> there's the linux enthusiast program, where you can get one of their current top range of laptops or desktops with linux | |
| [15:26] <sabdfl> then there's a program in china, which makes Ubuntu an option across most of their range | |
| [15:27] <sabdfl> and then there are the Vostro and Inspiron Mini netbook ranges, which have special versions of Ubuntu with custom Dell software for specific areas of the user experience | |
| [15:27] <sabdfl> i would really credit Matt Domsch and John Hull at Dell with their insight into linux engineering | |
| [15:28] <sabdfl> and mario limonc.... (erk, spelling) is a champion of ubuntu there | |
| [15:28] <jcastro> mario limonciello | |
| [15:28] <sabdfl> they clearly see linux users as thought leaders | |
| [15:28] <sabdfl> and they know that it takes a lot of work to do something like linux well | |
| [15:28] <sabdfl> and they work very hard at it, so we appreciate the partnership | |
| [15:29] <sabdfl> (and the machines are selling well too) ;-) | |
| [15:29] <sabdfl> next? | |
| [15:29] <jcastro> Question: Many people here in south africa dont use ubuntu because they don't have internet access or broadband. IS there any development being made in thi regard? | |
| [15:29] <sabdfl> the broadband situation in south africa is poor | |
| [15:29] <sabdfl> the good news is that i hope it will be much improved by 2010 | |
| [15:30] <sabdfl> i do realise that we don't pay enough attention to the case where people don't have broadband | |
| [15:30] <sabdfl> we ship a lot of updates, and it's hard to know whether we are shipping too many | |
| [15:30] <sabdfl> for users with broadband, they are valuable | |
| [15:31] <sabdfl> for users without it, I think it would be useful if we could help them | |
| [15:31] <sabdfl> to decide which are more important | |
| [15:31] <sabdfl> the design of apt archives makes it possible to ship them round on dvd | |
| [15:31] <sabdfl> perhaps we could look into that? | |
| [15:31] <sabdfl> next? | |
| [15:31] <jcastro> QUESTION: How does ubuntu/canonical plan to work towards a fully free desktop, and are there plans for co-operation with gnewsense? | |
| [15:32] <sabdfl> there are so many definitions of "fully free" | |
| [15:32] <sabdfl> i think we have a basically good relationship with gNewSense | |
| [15:32] <sabdfl> and we are supportive of what they do, it's valuable and important work | |
| [15:32] <sabdfl> we failed to get real interest in a "radical freedom" version of Ubuntu, called Gobuntu | |
| [15:32] <sabdfl> i really wanted that to fly, but it didn't | |
| [15:33] <sabdfl> in a sense, it was making life harder for gNewSense, not easier, so we decided to stop work on it | |
| [15:33] <sabdfl> c'est la vie | |
| [15:33] <sabdfl> the area of Ubuntu which has a big freedom wart is that we allow binary drivers | |
| [15:33] <sabdfl> of course, i'm not aware of any major distro that *disallows* binary drivers, though it would be trivial to do it :-) | |
| [15:34] <sabdfl> but we are pragmatic, and that causes all of us some discomfort | |
| [15:34] <sabdfl> we engage with vendors and constantly make the point that we can serve them better if they adopt a more open approach | |
| [15:34] <sabdfl> and there are some vendors which have come around to that specifically because of ubuntu | |
| [15:35] <sabdfl> i hope we get some credit for that when it happens :-) | |
| [15:35] <sabdfl> at this stage there are no plans to try and compete with gNewSense in any way, we'll just support them as best we can | |
| [15:35] <sabdfl> next? | |
| [15:35] <jcastro> QUESTION: There are over 8 million Ubuntu users right now. This is 80 times what Mozilla had when they started. Are there plans for an Ubuntu foundation or similar nonprofit to take over some of Canonical's role? | |
| [15:35] <sabdfl> we already have the Ubuntu Foundation, which is primarily a fallback vehicle | |
| [15:36] <sabdfl> in case Canonical doesn't survive for any reason, there are funds in the Foundation to ensure we meet commitments around maintenance for LTS releases and normal releases | |
| [15:36] <sabdfl> we also keep the governance of the project (Community Council) separate from Canonical, though of course there are big overlaps | |
| [15:37] <sabdfl> we have a number of non-Canonical folks on the CC, and I would like more on the TB too, though it's tough to find folks with the intense experience and also the time needed to do justice to such an appointment | |
| [15:37] <sabdfl> i suspect that will get easier as other businesses come to build on Ubuntu as much as Canonical does | |
| [15:37] <sabdfl> i think it would be unhealthy to setup a Foundation to do the "nice" work while Canonical does the "business" work | |
| [15:38] <sabdfl> our goal at Canonical is to be "nice and businesslike" | |
| [15:38] <sabdfl> i don't see a fundamental conflict there - i think it's healthy for Canonical to have to live up to the values of the ubuntu project | |
| [15:38] <sabdfl> and wouldn't want to see a figleaf in the form of a non-profit in active tension with Canonical | |
| [15:39] <sabdfl> next? | |
| [15:39] <jcastro> QUESTION: Does Canonical ever make "Official Policy Announcements" on contentious issues? Two recent controversies were the hard drive wear issue and the issue with the Intel network cards being bricked. Are there official guidelines on what should be done when Ubuntu can possibly damage hardware? | |
| [15:40] <sabdfl> yes, we have a process for handling emergencies and screwups | |
| [15:40] <sabdfl> including making sure that we communicate clearly about what the situation is | |
| [15:40] <sabdfl> unfortunately we have that because there have been emergencies, and we have in the past occasionally screwed up | |
| [15:40] <sabdfl> but i think the policies are good | |
| [15:41] <sabdfl> i don't think such an issue is contentious - if we make a mistake, we need to sort it out, and keep people briefed | |
| [15:41] <sabdfl> next? | |
| [15:41] <jcastro> QUESTION: I want to see Ubuntu Desktop become more profitable so it stays in focus, any chance of integrating search to the default menus and making money the same way Firefox does? | |
| [15:42] <sabdfl> yes, that's a possibility. there are ways to introduce services into the desktop and server | |
| [15:42] <sabdfl> it's important that we be tasteful when we do so | |
| [15:42] <sabdfl> i hope you find the landscape sysinfo work tasteful, for example | |
| [15:43] <sabdfl> we can consider the search idea, thanks for the suggestion! | |
| [15:43] <sabdfl> next? | |
| [15:43] <jcastro> QUESTION: What is a main reason for the major Linux distributions (especially, Ubuntu) to ship Gnome by default? Is there anything other than preferring LGPL (of GTK) over GPL (of QT)? Do you think it's feasible to ask NOKIA to release QT under LGPL to make KDE more attractive to enterprises? What are the other steps that can be done to push KDE ahead to be default desktop? | |
| [15:43] <sabdfl> good question | |
| [15:43] <sabdfl> some days, i think it's good we have multiple desktop environments, and at least two major ones | |
| [15:44] <sabdfl> because it drives innovation | |
| [15:44] <sabdfl> other days, it drives me nuts, because we have to write software twice | |
| [15:44] <sabdfl> and get criticised no matter what we do! | |
| [15:44] <sabdfl> i do think that a big driver of GNOME and Gtk in the *early* days was licensing | |
| [15:44] <sabdfl> and we don't know what Nokia will do with Qt licensing | |
| [15:45] <sabdfl> perhaps that will make it easier for converge | |
| [15:45] <sabdfl> personally, i would like to see a real effort to focus the innovation of the free software community around a common set of libraries | |
| [15:45] <sabdfl> and i'm delighted that GNOME and KDE are co-hosting their conferences this year | |
| [15:45] <sabdfl> that can *only* improve dialogue and communication | |
| [15:46] <sabdfl> we chose GNOME in 2004 because they had a real commitment to release cycles, and a real commitment to usability | |
| [15:46] <sabdfl> now KDE has adopted similar ideas | |
| [15:46] <sabdfl> and soon it might be just as effective a counterpart | |
| [15:46] <sabdfl> hopefully, they find Kubuntu a good way to get their code out there | |
| [15:46] <sabdfl> next? | |
| [15:46] <jcastro> Question: RE: Bug #1. MS Office is the anchor application in the Windows experience for a lot of people. OpenOffice and Evolution provide a good analog to Office but lack feature parity and innovation. What is Canonical doing about inspiring vision in these important packages? | |
| [15:47] <sabdfl> neither is a big focus for us, i'm afraid | |
| [15:47] <sabdfl> we do participate in OpenOffice | |
| [15:47] <sabdfl> i was on the advisory board of OO.o till they dropped it recently | |
| [15:48] <sabdfl> and we try to champion better processes for collaboration with the community there | |
| [15:48] <sabdfl> they have definitely improved | |
| [15:48] <sabdfl> soi'm excited about OO.o 3.0 | |
| [15:49] <sabdfl> as for inspiring vision, i don't know that it would be well received in those upstream communities unless we were also willing to write a lot of code there | |
| [15:49] <sabdfl> we focus our code on integration, making sure that the pieces work well together, rather than the pieces themselves | |
| [15:49] <sabdfl> though that is changing with our new desktop experience engineering team | |
| [15:49] <sabdfl> even there - the emphasis is on how the whole desktop fits together, the whole experience rather than any particular part | |
| [15:50] <sabdfl> next? | |
| [15:50] <jcastro> Question:1. Many NEW users of Ubuntu aren't used to IRC is there any thoughts of using Jabber or other protocols ? 2. Will we ever see a over the internet sent video or audio conference of UDS or open week with possibility to participate? | |
| [15:50] <sabdfl> jabber is a wonderful protocol for point-to-point communications | |
| [15:50] <sabdfl> person to person | |
| [15:50] <sabdfl> or system to system | |
| [15:50] <sabdfl> or person to system | |
| [15:50] <sabdfl> but it's not great, iirc, for group communications | |
| [15:51] <sabdfl> most of our teamwork is group oriented, hence the preference for IRC | |
| [15:51] <sabdfl> if you think we could use Jabber effectively, that would be interesting for the TB to hear | |
| [15:51] <sabdfl> especially if it comes with code and infrastructure! | |
| [15:51] <sabdfl> w.r.t. video and audio | |
| [15:51] <sabdfl> we have streamed audio from most UDS's in the past | |
| [15:52] <sabdfl> we can look into video too | |
| [15:52] <sabdfl> there is some discussion of video recording this year's UDS, but not streaming real-time video AFAIK | |
| [15:52] <sabdfl> also, you can participate in UDS via IRC while listening to the streaming audio | |
| [15:52] <sabdfl> jcastro: is that a fair summary? | |
| [15:53] <jcastro> yep | |
| [15:53] <sabdfl> ok | |
| [15:53] <sabdfl> next! | |
| [15:53] <jcastro> We also have developer interviews which we put up on youtube and in ogg during UDS | |
| [15:53] <jcastro> QUESTION: Following codeweavers lame duck challenge givaway of crossover office etc, could you ever see a day were that software was incorporated into Ubuntu through a partnership agreement, potentially that could widen the actractability of Ubuntu if it gave users (particularly business users) the opportunity to use the software they're used to even though OpenOffice offers similar features, users are loathe to change | |
| [15:54] <sabdfl> we would not include non-free apps in Ubuntu, so MS would have to do more than make it easy to use with Linux :-) | |
| [15:54] <sabdfl> in principle though, i would like people to be able to build solutions on top of ubuntu | |
| [15:54] <sabdfl> if that means ubuntu+codeweavers+msoffice, that's fine by me | |
| [15:55] <sabdfl> it's one step closer for that organisation to ubuntu+oo.o | |
| [15:55] <sabdfl> same goes for Firefox-on-Windows | |
| [15:55] <sabdfl> slowly slowly catchee monkey :-) | |
| [15:55] <sabdfl> i know that scott richie goes great work with WINE in Ubuntu | |
| [15:55] <sabdfl> and hope that CodeWeavers feel's they can easily offer solutions on top of Ubuntu | |
| [15:56] <sabdfl> the more people feel they can be part of a broad ecosystem the better | |
| [15:56] <sabdfl> next? | |
| [15:56] <jcastro> QUESTION: do you you have any plans to advertise ubuntu on tv, i still find far to many people dont have a clue what ubuntu or linux is in a business environment or at home it seems to make ubuntu profitable this needs to be addressed. | |
| [15:56] <sabdfl> the cost of tv advertising might slow down our profitability too, though :-) | |
| [15:57] <sabdfl> again, i think it's best to grow sustainably | |
| [15:57] <sabdfl> we focus on building something that works really well for well-informed people, and letting them share their knowledge along with the software | |
| [15:57] <sabdfl> i would love to see ubuntu, or any free software brand, become widely recognised in the regular world | |
| [15:57] <sabdfl> and not just at Hogwarts :-) | |
| [15:58] <sabdfl> i'm grateful to Mozilla for doing that with Firefox! | |
| [15:58] <sabdfl> so, perhaps in time we'll see TV advertising, but not now | |
| [15:58] <sabdfl> next? | |
| [15:58] <jcastro> Question : Re - Open Air : Thank you for your response. Google is extremely excited about this because it could open up "Free" wifi and with mobile going big, I think this could have a huge impact. Do you think we should begin researching this to stay ahead? Maybe this could be a good niche to get in to those areas that don't have internet available. | |
| [15:59] <sabdfl> yes, i would be interested to see the results of that research | |
| [15:59] <sabdfl> who is making the devices, what is the bandwidth going to be used for, how will it fit into existing network protocols etc | |
| [15:59] <sabdfl> it's amazing what has been achieved with the small bit of "unlicensed" spectrum | |
| [16:00] <sabdfl> when i meet regulators i love to tease them with that - and encourage them to find other opportunities to stimulate that innovation | |
| [16:00] <sabdfl> the challenge is that spectrum is fragmented differently in every region | |
| [16:00] <sabdfl> the wifi bandwidth was "unlicensed" because it is the same as microwaves | |
| [16:01] <sabdfl> so people thought there would be too much interference there to do anything useful | |
| [16:01] <sabdfl> heh. openness surprises people sometimes :-) | |
| [16:01] <sabdfl> next? | |
| [16:01] <jcastro> Question: Has your realtionship with Linspire or Freespire inspired or affected you or your work in any way on the last two releases? | |
| [16:01] <sabdfl> not afaict. | |
| [16:01] <sabdfl> next? | |
| [16:01] <jcastro> QUESTION: What in your opinion is the biggest challenge to acceptance of the Ubuntu desktop? | |
| [16:02] <sabdfl> compatibility with ipods | |
| [16:02] <sabdfl> people can do most everything else they need to do, realistically | |
| [16:02] <sabdfl> the one thing I hear a lot about is the ability to manage an ipod | |
| [16:02] <sabdfl> so, patches welcome! | |
| [16:02] <sabdfl> next? | |
| [16:02] <jcastro> Question: Why is i18n a secondary (or worse) priority project? Will this situation radically change in a short time, for example before Jaunty? We translators got only promises before intrepid, before hardy, but the LP translations database is still slow, undermaintaned and i18n problems in packages are largely ignored - this sucks a lot more than it should. | |
| [16:03] <sabdfl> yes, it does | |
| [16:03] <sabdfl> i've personally spent a few weeks / months of my life working on LP translations | |
| [16:03] <sabdfl> it's important to me that we do well, and there are still lots and lots of areas we can improve | |
| [16:03] <sabdfl> i know that team works hard, and i think they are making steady progress now | |
| [16:03] <sabdfl> it's very challenging - HUGE data sets | |
| [16:04] <sabdfl> i didn't help them by designing it in a complex way to try and make LP work for both distros and upstreams | |
| [16:04] <sabdfl> but i think they are getting over the hump | |
| [16:04] <sabdfl> so hope it will improve | |
| [16:04] <sabdfl> it's crazy to me that we still don't have an efficient bi-directional flow of translations with upstream | |
| [16:04] <sabdfl> until i sit and look at the challenges | |
| [16:05] <sabdfl> at which point i realise - it's very difficult :-) | |
| [16:05] <sabdfl> next? | |
| [16:05] <jcastro> QUESTION: Becouse of the economic downturn more companies are looking to cut costs and linux is looking great for that. What benefits and disadvantages would it be to hire Canonical for a project compaired to offshoring to India or China? | |
| [16:05] <jcastro> Lots of questions about the economy | |
| [16:05] <jcastro> So I guee one round-up answer would be best | |
| [16:05] <sabdfl> this will be a very challenging year, economically, imo | |
| [16:05] <sabdfl> i don't think there will be many upside surpises | |
| [16:06] <sabdfl> we just had a lovely one in the US election, though ;-) | |
| [16:06] <sabdfl> but economically, yes, this will be a tough year | |
| [16:06] <sabdfl> i think a lot of companies will look to restructure their systems for cost-effectiveness as a result | |
| [16:06] <sabdfl> ubuntu makes a superb platform for common server tasks, and common desktop tasks | |
| [16:07] <sabdfl> so, if you have thousands of desktops doing little but the web, and thousands of servers doing basic web app serving, file and print, mail, etc, then you can benefit substantially with a move to ubuntu | |
| [16:07] <sabdfl> from windows, unix or a more expensive linux | |
| [16:07] <sabdfl> we offer full commercial support 24x7 from a Canadian call center (Go Montreal) | |
| [16:08] <sabdfl> and we negotiate so folks only pay for support on the servers / desktops that are in production | |
| [16:08] <sabdfl> which can result in big saving | |
| [16:08] <sabdfl> s | |
| [16:08] <sabdfl> so, i think this will be a good year for ubuntu and canonical | |
| [16:08] <sabdfl> next? | |
| [16:08] <jcastro> QUESTION: Has there been any result from your call to syncronise releases of major distros? | |
| [16:08] <sabdfl> hmm.... officially, no | |
| [16:09] <sabdfl> in practice, i'm glad the meme of regular release cadence is gaining ground | |
| [16:09] <sabdfl> not because it's my idea (it wasn't), but because it's a *good* idea | |
| [16:09] <sabdfl> and because the more projects do it, the healthier the whole FLOSS ecosystem becomes | |
| [16:09] <sabdfl> as it happens, i think all the rapid-release distros (Ubuntu, Fedora, OpenSuse) used 2.6.27 in this round | |
| [16:10] <sabdfl> and as it happens, i think all the LTS releases (Ubuntu LTS, SLES, RHEL) will be in 2010 | |
| [16:10] <sabdfl> so *maybe* we'll get good syncronisation by good fortune | |
| [16:10] <sabdfl> i would really like to see this embraced by the community as a best practice | |
| [16:10] <sabdfl> because it works so well | |
| [16:10] <sabdfl> next? | |
| [16:10] <jcastro> QUESTION (REPEATED): Is the quest for 'free software laptop' still going on? Are there any manufacturers ready to listen? | |
| [16:11] <sabdfl> yes and no | |
| [16:11] <sabdfl> there are LOTS of manufacturers working with Linux now | |
| [16:11] <sabdfl> but afaik none that have made "uber-freedom" the total goal | |
| [16:11] <sabdfl> i.e. nobody is working with CoreBoot (the old LinuxBIOS) and nobody is pushing for open source firmware either | |
| [16:11] <sabdfl> which is a pity! | |
| [16:12] <sabdfl> but i do put the idea on the table often, so perhaps it will stick some day | |
| [16:12] <sabdfl> the key thing is that we have to be able to deliver a much better end-user experience, otherwise it's not interesting to the manufacturers | |
| [16:12] <sabdfl> and so far, we haven't been able to make that case | |
| [16:12] <sabdfl> next? | |
| [16:12] <jcastro> Question: are there any plans to offer a cloud based Ubuntu OS? | |
| [16:12] <jcastro> There was no escaping this question. :) | |
| [16:13] <sabdfl> We have Ubuntu JEOS ("Just Enough OS") for people who are working in virtualised environments | |
| [16:13] <sabdfl> that's now evolved to be useful on EC2, VMWare, Xen and KVM | |
| [16:13] <sabdfl> i would like to see projects like Eucalyptus gain traction, to standardise the cloud around known API's with free software management | |
| [16:14] <sabdfl> and we're doing our part to make that happen | |
| [16:14] <sabdfl> next? | |
| [16:14] <jcastro> QUESTION: Re the Ubuntu mobile and UMPC projects, when can we expect to see devices like Ubuntuphones? | |
| [16:14] <sabdfl> i don't have a good answer for you i'm afraid | |
| [16:14] <sabdfl> linux is a perfect platform for consumer electronics, and the rate of adoption there is fantastic | |
| [16:15] <sabdfl> but it's highly fragmented | |
| [16:15] <sabdfl> there are many different projects / environments / frameworks for "linux on smartphones" | |
| [16:15] <sabdfl> limo, openmoko, maemo, moblin, android etc | |
| [16:16] <sabdfl> we have partnered with moblin from Intel because we think they are committed to (a) a great user experience and (b) open processes and governance | |
| [16:16] <sabdfl> intel has really set the pace with hardware enablement in free software | |
| [16:16] <sabdfl> they invest a lot in it, and they invest it well - they work like an open source project to a much greater extent than any other manufacturer | |
| [16:17] <sabdfl> (hmm, maybe Sun would say they pioneered that, i don't know) | |
| [16:17] <sabdfl> w.r.t. Intel support for graphics, wireless, chipsets etc, they are a leader in the industry | |
| [16:17] <sabdfl> so i'm hopeful they will bring that same focus and approach to moblin, and so far they seem willing to do that | |
| [16:17] <sabdfl> BUT | |
| [16:17] <sabdfl> it's early days yet | |
| [16:18] <sabdfl> so, i'm very much looking forward to moblin 2 | |
| [16:18] <sabdfl> we will package that up in ubuntu and try to do it justice | |
| [16:18] <sabdfl> hopefully, it becomes a magnet for more developers and turns into a great community like GNOME or KDE | |
| [16:19] <sabdfl> a single-vendor solution isn't ideal, but it's not a bad place to start if the goal is a successful open community | |
| [16:19] <sabdfl> sometimes it takes one focused push from one organisation, even if you want to have a broader ecosystem | |
| [16:19] <sabdfl> i think the winning ecosystem in Linux on mobiles will need to have participation from multiple hardware vendors | |
| [16:19] <sabdfl> and multiple operators, and multiple OSV's | |
| [16:19] <sabdfl> so, early days :-) | |
| [16:19] <sabdfl> phew | |
| [16:19] <sabdfl> i'm rambling | |
| [16:19] <jcastro> Question: Professional as well as independent producers (think youtube demographic) both need for Ubuntu to come with higher-end Multimedia Production tools (then what is available). What can be done by Canonical to encourage/subsidize developers who wish to meet this need by creating a suite of multimedia tools, such as a high end Non-Linear Video Editor (Final Cut or Premiere level of quality). There are currently | |
| [16:20] <sabdfl> summary - i think moblin is headed in a good direction, and am supporting that | |
| [16:20] <jcastro> some projects that attempt to meet this need, but does not meet user expectations in usability or features. | |
| [16:20] <sabdfl> is joe jackson around? | |
| [16:21] <jcastro> doesn't appear to be | |
| [16:21] <sabdfl> Ubuntu Studio is a lovely project, and the team behind it has a rare combination of style and skills | |
| [16:21] <sabdfl> i would like to hear their response to this question! | |
| [16:21] <sabdfl> i do agree that high-end content editing would be a natural place for linux | |
| [16:22] <sabdfl> i don't know what the state of the art in those fields is, though, or what we could do to improve it | |
| [16:22] <jcastro> collabora has recently announced a fulltime developer for pitivi, which is a video editor, if that helps answer the question | |
| [16:22] <sabdfl> i'm open to suggestions! | |
| [16:22] <sabdfl> next? | |
| [16:22] <jcastro> QUESTION: With much of the world on dialup, is there any focus going towards a debian based package managment system similiar to delta rpm's? | |
| [16:22] <jcastro> some questions about using p2p for updates like bittorrent, etc. | |
| [16:22] <jcastro> rolling all those into one. :) | |
| [16:22] <sabdfl> afaik, there has been some talk about a next-gen .deb format | |
| [16:22] <sabdfl> but it's not been concentrated | |
| [16:22] <sabdfl> i may be out of that loop, though | |
| [16:23] <sabdfl> i would certainly love to see it happen, because the current approach is wasteful on bandwidth | |
| [16:23] <sabdfl> next? | |
| [16:23] <jcastro> question: What is your favorite Music Artist / Band? Have you had any thoughts of incorporating a Pandora Music Player, from the Music Genome Project, into the Desktop? | |
| [16:23] <sabdfl> cool idea | |
| [16:23] <sabdfl> i like that we have magnatunes :-) | |
| [16:24] <sabdfl> i like Elbow | |
| [16:24] <sabdfl> "The seldom seen kid" | |
| [16:24] <sabdfl> and lots of R&B | |
| [16:24] <sabdfl> i think we could do more with music in the desktop | |
| [16:25] <sabdfl> i was a bit bummed when Best Buy bought Napster, but that's a different story ;-) | |
| [16:25] <sabdfl> next? | |
| [16:25] <jcastro> QUESTION: The hardware database mentioned earlier seems limited to certifying whole machines. It seems like it would be more useful for most of us if we had a listing of individual hardware products that were known to work (or not work), particularly video and wireless cards... | |
| [16:25] <jcastro> QUESTION: .. Yet few wireless vendors would see the point in submitting their hardware for certification unless there were already a database to be added to. Is there a plan to get that sort of certification? | |
| [16:25] <sabdfl> i think jcastro pointed to the hardware database earlier | |
| [16:26] <sabdfl> we try to aggregate the information folks send us about their hardware | |
| [16:26] <sabdfl> it's difficult to do component-level certification | |
| [16:26] <sabdfl> because often, something breaks at the system level | |
| [16:26] <sabdfl> we do work with component manufacturers, though, if there is a machine that needs to be enabled | |
| [16:26] <sabdfl> next? | |
| [16:26] <jcastro> QUESTION: Do you agree on the cricitism raised e.g. by Greg Kroah-Hartman in his Linux Plumbers keynote at http://video.google.com/videoplay?docid=3385088017824733336 that Canonical does not contribute enough back to upstream? Do you plan to increase the number of the developers to answer this criticism? | |
| [16:27] <sabdfl> i think Greg is well-meaning | |
| [16:27] <sabdfl> and his numbers are accurate enough | |
| [16:28] <sabdfl> i think the approach he took was appalling | |
| [16:28] <sabdfl> as it reminded me how easily we take friends and turn them into strangers | |
| [16:29] <sabdfl> we each care about specific things, and we spend our time and energy on those things | |
| [16:29] <sabdfl> Greg was saying "the things I care about are the only things that matter" | |
| [16:29] <sabdfl> well | |
| [16:29] <sabdfl> they may be the only things that matter to Greg, but they aren't necessarily the things that matter most to me, or to other members of the Canonical team, or to other members of the Ubuntu community | |
| [16:30] <sabdfl> in ubuntu, we invest a vast amount of energy (time, money, love, attention, reputation) into free software | |
| [16:31] <sabdfl> we care deeply about making sure that anybody anywhere can get it, use it legally, and use it safely and securely | |
| [16:31] <sabdfl> our users appreciate it - THEY know that this is a big investment and gift | |
| [16:31] <sabdfl> i think someone in Greg's presentation asked "who here uses Ubuntu" and half the room put their hands up | |
| [16:32] <sabdfl> and Tridge spoke wisely when he said "look, I use Ubuntu heavily, I get all the updates free, it makes me productive, and I've never contributed a patch to Ubuntu" | |
| [16:32] <sabdfl> in other words, the ecosystem is broader, deeper and richer than Greg was making out | |
| [16:32] <sabdfl> and those relationships are more complex than Greg was making out | |
| [16:33] <sabdfl> i believe Ubuntu and Canonical are making a very big difference in free software, and that has little to do with how many patches in the kernel have an @canonical.com email address associated with them | |
| [16:33] <sabdfl> i'm not going to hire lots of people to keep greg happy :-) | |
| [16:33] <sabdfl> i do continue to hire people to make free software into an amazing experience for me, my friends, my colleagues and peers, and everyone else i meet in a plane, train or on the street :-) | |
| [16:34] <sabdfl> when we open source launchpad, we will have more than 100 people doing nothing but hack on free software | |
| [16:34] <sabdfl> i think that's a great contribution for a company of 200 people | |
| [16:34] <sabdfl> and perhaps some day greg too will feel that way | |
| [16:34] <sabdfl> next! | |
| [16:34] <jcastro> QUESTION: why not release updated disc images with new backported drivers and software updates not only for LTS releases? usually, ubuntu have some problems, that you can solve installing an update, but it will be much easier for newbie, inestead of downloading image and updates to download updated images... and,will not it be good for ubuntu not to be so "bleeding-edge"? | |
| [16:35] <sabdfl> choices, choices, choices :-) | |
| [16:35] <sabdfl> we have to balance resources, and we've come up with the process of LTS+point-releases, and standard regular releases | |
| [16:35] <sabdfl> that allows us to keep getting new wonderful code into releases for testing and for use | |
| [16:36] <sabdfl> and also allows us to meet the needs of people who want something that will stay relatively stable for longer periods | |
| [16:36] <sabdfl> we could do more if we had more hands ;-) | |
| [16:36] <sabdfl> next! | |
| [16:36] <jcastro> QUESTION: There have ben a few articles about Ubuntu getting slower with each new release. With Windows 7 being the first Windows release to be faster than it's predecessor, are there some similar plans for Ubuntu on enhancing it's performance? | |
| [16:36] <sabdfl> it's easy to improve off a low base ;-) | |
| [16:37] <sabdfl> i hope we can rise to the challenge of really fast boot times in Jaunty | |
| [16:37] <sabdfl> i know the Ubuntu team is very interested in the challenge | |
| [16:37] <sabdfl> as for broader performance, i don't know that we are doing anything that makes the system slow | |
| [16:38] <sabdfl> though there are more pieces running by default these days than there were a while ago | |
| [16:38] <sabdfl> let's see when Win7 releases, shall we? | |
| [16:38] <sabdfl> next! | |
| [16:38] <jcastro> QUESTION: What's Your opinion about the overcrowded bug tracker? Seems that users are generating reports faster than developers could resolve them, see: http://keithcu.com/wordpress/?p=24 and currently, there are almost five times this much open bugs. This causes them to ignore most of the bugs (or just mines? :)) What can be the solution? | |
| [16:39] <sabdfl> interesting | |
| [16:40] <sabdfl> we definitely need to figure out how to scale our bug processes | |
| [16:40] <sabdfl> it's a huge privilege to have so many users | |
| [16:40] <sabdfl> and a great privilege that many of them will take the time and effort to report bugs | |
| [16:40] <sabdfl> we can't hire twice as many developers when we get twice as many users, though | |
| [16:40] <sabdfl> we need to figure out how to involve the community to help triage and isolate the bug cases | |
| [16:41] <sabdfl> and how to share the load across the whole ecosystem | |
| [16:41] <sabdfl> some people think "ubuntu shoudl fix every bug that's reported in ubuntu" | |
| [16:41] <sabdfl> well, we certainly feel some responsibility for every bug | |
| [16:41] <sabdfl> but we also know that many of the bugs reported in ubuntu also exist in every other distro and upstream | |
| [16:42] <sabdfl> so we need to get efficient about making sure that the bugs are reported accurately upstream | |
| [16:42] <sabdfl> we build some plugins for bugzilla and trac under the GPL that make it easier to coordinate the response to those bugs betwen the distro and upstream | |
| [16:42] <sabdfl> so if you work upstream on a project that uses bugzilla and/or trac, please consider installing those! | |
| [16:43] <sabdfl> https://help.launchpad.net/Bugs/BugzillaPlugin | |
| [16:43] <sabdfl> not sure where the Trac one went :-) | |
| [16:43] <sabdfl> next? | |
| [16:43] <jcastro> Question: There are a lot of bugs concerning the audio subsystem in Hardy and Intrepid. The combination of OSS ALSA and PA sound systems seems to have made it painful and confusing to new users. Many audio applications do not work as expected. Are there any plans for Canonical to get involved in these projects and hire audio engineers/developers to manage the Ubuntu Audio Team and attempt and fix these issues? | |
| [16:44] <sabdfl> good question | |
| [16:44] <sabdfl> i'm hoping that PulseAudio is The Answer to linux audio issues | |
| [16:44] <sabdfl> this has been a long, long journey | |
| [16:44] <sabdfl> i don't have a specific plan or agenda in the audio space though | |
| [16:44] <sabdfl> if you think we need one, please put something together for the TB | |
| [16:45] <sabdfl> next! | |
| [16:45] <jcastro> QUESTION: Will it be much longer until launchpad is released open source ? | |
| [16:45] <sabdfl> it will be done by November 2009 | |
| [16:45] <sabdfl> the team is working through a list of issues already | |
| [16:46] <sabdfl> license checks, separating out pieces that are not related to the web service, determining the right approach for various issues | |
| [16:46] <sabdfl> it may happen sooner, but you have a concrete commitment that it will be done by then | |
| [16:46] <sabdfl> there are still a lot of unanswered questions | |
| [16:47] <sabdfl> we will see how that release goes, and probably tweak processes after that | |
| [16:47] <sabdfl> next? | |
| [16:47] <jcastro> QUESTION: Any plans to making Launchpad more social? Perhaps people will accidentally start working on bugs and answering questions, if we integrate a Laconica microblogger and some other features that are considered cool these days? | |
| [16:47] <sabdfl> :-) | |
| [16:47] <sabdfl> i would like that | |
| [16:47] <sabdfl> i did some of the work on the maps feature | |
| [16:47] <sabdfl> and i know some members of the LP team want more social features | |
| [16:47] <jcastro> (ed. note - the laconica guy will be at FOSSCamp and there is a strong presence of ubuntu folks on identi.ca itself) | |
| [16:47] <sabdfl> after all, open source is a social exercise | |
| [16:48] <sabdfl> if you have specific social ideas for LP, please hop onto #launchpad and discuss, or send them to kiko | |
| [16:48] <sabdfl> or beuno, or me | |
| [16:48] <sabdfl> next! | |
| [16:48] <jcastro> QUESTION: NVidia's drivers are notoriously misfitting on the Linux desktop, and this is getting very visible with the awesome kernel mode-setting and Fedora's Plymouth graphical boot. What does the Ubuntu team at Canonical do to grow hardware support in cases like NVIdia's? | |
| [16:48] <sabdfl> remind nvidia about the benefits of open source drivers :-) | |
| [16:49] <sabdfl> the kernel modesetting work from Intel is fantastic | |
| [16:49] <sabdfl> i do think nvidia and ati will get that part quickly | |
| [16:49] <sabdfl> the harder stuff is the deep integration of GL and X | |
| [16:49] <sabdfl> stuff like GEM will be harder to drive adoption of | |
| [16:50] <sabdfl> the main focus i think needs to be on driving adoption of the linux desktop in large-scale markets | |
| [16:50] <sabdfl> because that drives the agenda at hardware manufacturers | |
| [16:50] <sabdfl> next! | |
| [16:50] <jcastro> QUESTION: Do you have plans on driving X.org out or reducing it to minimum at some point in future? | |
| [16:50] <sabdfl> driving it out? | |
| [16:50] <jcastro> This is probably about that prototype X server called wayland | |
| [16:50] <sabdfl> i think the X guys have made big strides since Keith &co engaged | |
| [16:51] <sabdfl> and Arjan's work on the fast boot showed that they can still improve a lot | |
| [16:51] <sabdfl> i don't know enough about X.org alternatives to have a cogent view on them i'm afraid | |
| [16:51] <jcastro> QUESTION: How successful has getting boxed version of Ubuntu into the USA retailer Best Buy been for Canonical? | |
| [16:51] <sabdfl> i would hope though that the X.org folks are continuing to make X leaner and meaner | |
| [16:52] <sabdfl> i don't have any stats on the Best Buy initiative, i'm afraid | |
| [16:52] <sabdfl> but i'm glad whenever floss shows up in a new channel | |
| [16:52] <jcastro> QUESTION: Will we have a new theme for jaunty? | |
| [16:52] <sabdfl> there are lots of roads, hopefully they all lead to linux eventually! | |
| [16:52] <sabdfl> hmm.... maybe! | |
| [16:52] <sabdfl> we have been hiring, for months, designers and user experience guys | |
| [16:53] <sabdfl> it has taken much longer than i hoped | |
| [16:53] <sabdfl> that team will be in place in jan / feb, i imagine | |
| [16:53] <sabdfl> whether their initial work will make a dramatic visual impact on Jaunty, i don't know | |
| [16:53] <sabdfl> i know that other work, on the user experience front, will land | |
| [16:53] <sabdfl> but i'll keep some surprises in store till later ;-) | |
| [16:53] <sabdfl> next! | |
| [16:53] <jcastro> QUESTION: Qt 4.5 will ship QGtkStyle by default, which will make it use GTK to provide a native look for Qt application in GNOME. With this, you write applications once, and they look native on both GNOME and KDE. Are there plans for some ubuntu-apps to be written in Qt 4.5? | |
| [16:54] <jcastro> (one more after this) | |
| [16:54] <sabdfl> that's very interesting | |
| [16:54] <sabdfl> it would certainly be a relief to be able to write once and deliver into both kubuntu and ubuntu simultaneously | |
| [16:54] <sabdfl> i'll have to ask the team themselves what they think of this option, though | |
| [16:55] <sabdfl> personally, i would like to see a real effort to bring the toolkits together | |
| [16:55] <sabdfl> as painful as that may sound | |
| [16:55] <sabdfl> it would make it much more efficient to work in desktop linux | |
| [16:55] <sabdfl> next! | |
| [16:55] <jcastro> QUESTION: how robust is the laptop certification process between Canonical and its partners? should customers expect 0% system breakage (in terms of hardware or software)? | |
| [16:56] <sabdfl> they should expect it, and we strive to deliver it | |
| [16:56] <sabdfl> see above for how we handle emergencies and screwups :-) | |
| [16:56] <sabdfl> next! | |
| [16:56] <jcastro> QUESTION: There have been rumors that you personally aren't quite comfortable with Wine. Could you make a statement as a lead in to my talk up next? ;) | |
| [16:56] <sabdfl> i'm perfectly comfortable with WINE | |
| [16:56] <sabdfl> next! | |
| [16:56] <jcastro> that's it | |
| [16:56] <jcastro> we are out of time | |
| [16:56] <sabdfl> phew | |
| [16:57] <jcastro> thanks Mark for answering questions, and thanks everyone for getting involved! | |
| [16:57] <sabdfl> my fingers ache but i'm happy to meet you all! | |
| [16:57] <sabdfl> thanks for the great questions | |
| [16:57] <jcastro> over 300 people participated! | |
| [16:57] <sabdfl> thanks jcastro for the quick fingerwork :-) | |
| [16:57] <sabdfl> cheers all | |
| [16:57] <jcastro> ok, next up we have Scott Ritchie with WINE | |
| [16:57] <jcastro> in about ~3 minutes | |
| [16:57] * jcastro runs to the restroom while he can | |
| [17:00] <YokoZar> Hey everyone, thanks for sticking around. I promise to be even more interesting than Mark Shuttleworth. | |
| [17:01] <YokoZar> So, I'm the Wine guy. I maintain both the "official" Ubuntu Wine packages and the "unofficial" ones over at WineHQ (or the Wine Team PPA -- same bits). | |
| [17:01] <YokoZar> I'm also a Wine developer. About a month ago I was at the Wine developer conference, where I got way more involved than I expected to be. | |
| [17:02] <YokoZar> Feel free to ask questions in #ubuntu-classroom-chat at any time, by the way. | |
| [17:02] <YokoZar> I found myself leading an hour-long discussion about all the ultimate usability goals we (downstream) needed to integrate Wine well into the desktop. | |
| [17:02] <YokoZar> In practical terms these translate into feature requests across multiple projects -- to make it easy to configure a Windows Application by right clicking on it and selecting Properties, for instance, we need new code from both Wine and Gnome (or KDE). | |
| [17:03] <YokoZar> Such a feature would be useful to a human being who wanted to have a "full screen" Windows application run in only a Window. While the other Wine developers were definitely interested in seeing such an eventual thing happen, none of them wanted to write code outside Wine to handle it. | |
| [17:03] <YokoZar> They were glad, however, to meet halfway and write the code for other software to interface with Wine via command line or D-Bus message. So, I volunteered to make the appropriate Gnome interface and actually see that it gets out there to end users. | |
| [17:04] <YokoZar> So, now I'm going to become a Gnome developer. It's exciting, although I'm currently unemployed and am desperately trying to find more time to work on Ubuntu-related projects -- the desktop integration projects alone could take me months of full time work. | |
| [17:04] <YokoZar> johnflux: QUESTION: What is the current programming focus of wine? | |
| [17:05] <YokoZar> Wine isn't specifically "focused" on any particular thing. There are many developers working on many different applications. | |
| [17:06] <YokoZar> But you don't write a patch to "make program x work", you write a patch to "make windows function foo needed by program x work the way it's supposed to" | |
| [17:06] <YokoZar> Eventually when you do enough of that applications just start working, especially if you implemented the windows function correctly. | |
| [17:07] <YokoZar> Wine developers do focus on particular applications, of course, but there's also a lot of "collateral damage" -- apps that just start working on accident, since development was done the right way. | |
| [17:07] <YokoZar> Related: johnflux: QUESTION: what are the main current hurdles left? | |
| [17:08] <YokoZar> There are many. Many applications don't work, even more don't work perfectly. Some whole areas are unimplemented, like Direct3D 10 | |
| [17:08] <YokoZar> Wine is nowhere near "almost done", yet we're quite near 95% of applications being useful enough to run | |
| [17:09] <YokoZar> Wine development was frustrating for many years, because any one thing left broken or unimplemented could keep an application from working entirely. Wine's finally getting "enough" such that many applications are starting to actually work | |
| [17:10] <YokoZar> sudobash: question: Is it true that Wine is built around Windows 3.11 for its emultaor like abilites? | |
| [17:10] <YokoZar> No | |
| [17:11] <YokoZar> Wine is 100% microsoft-free. There is no "built around" a version of Windows | |
| === ChanServ changed the topic of #ubuntu-classroom to: /cs topic Current Session: Wine | Welcome to Openweek, questions in #ubuntu-classroom-chat please || Session details here: https://wiki.ubuntu.com/UbuntuOpenWeek | |
| [17:11] <YokoZar> However we do try and make certain windows functions behave like Windows (which version depends on how you've configured Wine), so that applications work | |
| === ChanServ changed the topic of #ubuntu-classroom to: Current Session: Wine | Welcome to Openweek, questions in #ubuntu-classroom-chat please || Session details here: https://wiki.ubuntu.com/UbuntuOpenWeek | |
| [17:11] <YokoZar> Wine got started over 15 years ago in the Windows 3.1 days, and that code's still there | |
| [17:12] <YokoZar> In fact there are quite a few applications that work well in Wine but refuse to run in Vista, especially old 16-bit apps. | |
| [17:12] <YokoZar> Milyardo: QUESTION: What is the WINE Team's relation with Microsoft? Does the WINE team get help from Microsoft? Has the WINE team even given Microsoft any insight into their own product? | |
| [17:13] <YokoZar> The closest the Wine team's interaction with Microsoft ever gets is reading MSDN documentation to figure out how to implement a function | |
| [17:14] <YokoZar> Unlike SAMBA, Microsoft doesn't help Wine in any way. Often times that MSDN documentation is flat-out wrong as well. | |
| [17:15] <YokoZar> We suspect that many Wine developers know the internals of how Windows functions behave better than most Microsoft developers, though Wine developers are prohibited from seeing any Microsoft code (and if you ever have, eg downloading the stolen source code on the internet a while back, you can't ever contribute to Wine) | |
| [17:15] <YokoZar> Microsoft is also rumored to prohibit their employees from even looking at GPL/LGPL'ed code, so it's likely they don't look at Wine either. | |
| [17:16] <YokoZar> So Wine is way different from Mono and Samba - Microsoft basically ignores us, and in so doing they hope their customers ignore us too. | |
| [17:17] <YokoZar> johnflux: QUESTION: have you tried to get a job to work on wine with Canonical? :) worth a shot | |
| [17:17] <YokoZar> Yes. I'd really like to. I'm gonna try hobknobbing some more at the next Ubuntu Developer Summit. Hopefully I'll have some UI work done to show off, as Canonical has recently started employing Gnome people. | |
| [17:18] <YokoZar> I'm also a part time teacher, and doing Ubuntu training work has also fascinated me. I live in the US, however, and the closest Ubuntu-certified training partner is all the way in Canada ;) | |
| [17:19] <YokoZar> I'm also considering doing a community fundraiser as an experiment, in part to replenish the Wine developer fund that we used up sponsoring people like me to the last Wine conference | |
| [17:19] <YokoZar> joemoe: QUESTION: Will Wine ever be able to match like, Windows XP software compatibility? | |
| [17:20] <YokoZar> Yeah. It's a very feasible goal. Wine development has been quickening, for much the same reason Wikipedia began growing faster; the more useful Wine and Ubuntu get, the more people use us, and then the more developers we get | |
| [17:21] <YokoZar> That said, focus will still be on applications rather than targeting the API as a whole. There's little need to implement foox.dll if only Windows-internal applications use it | |
| [17:22] <YokoZar> QUESTION: aren't there any 'big' general areas currently being worked on? e.g. directx 10 or something | |
| [17:22] <YokoZar> Yes, that was one of the "major features" outlined by Alexandre for the next release | |
| [17:23] <YokoZar> Basically there are 5 major features that will take forever to do | |
| [17:23] <YokoZar> Any one of them would trigger a new release freeze | |
| [17:23] <YokoZar> They were: Direct 3D 10, Win-64 bit support, a Quartz display driver for Mac, USB support (eg for newer iPods), and one other I can't remember off the top of my head | |
| [17:24] <YokoZar> Of those the most likely to happen within the next 6 months is the USB support | |
| [17:24] <YokoZar> GFORGX: QUESTION: will there be something like Qt's QGtkStyle for Wine? It'd be great if Windows apps'll look native in GTK environment. | |
| [17:25] <YokoZar> Unfortunately, it's impossible to render Wine in GTK. Windows widgets don't map neatly to GTK, and Wine needs to control it with it's own drawing code | |
| [17:25] <YokoZar> However, it is possible to theme Wine itself | |
| [17:25] <YokoZar> Getting Wine to read the system theme and generate it's own theme on the fly is theoretically possible, but really hard (and if you do it wrong it often looks very ugly) | |
| [17:26] <YokoZar> Google tried that with Picassa for a bit before ultimately giving up and just going with the default Wine blue-boring-windows theme | |
| [17:26] <YokoZar> The proposal I gave at Wineconf, however, seems like the way forward: we make a Windows-style .theme file for each of the default themes, and then have Wine read this whenever we set the system theme. | |
| [17:27] <YokoZar> This way you set Ubuntu to use Human and then Wine reads the (otherwise ignored) Human-for-Windows theme included with it, and then renders that way. | |
| [17:27] <YokoZar> I'm happy to say work on this has now started | |
| [17:28] <YokoZar> johnflux: QUESTION: Is wine being developed faster or slower than new windows code is being added? | |
| [17:29] <YokoZar> I'd say faster, much faster. The worry is that Microsoft will keep changing the API so much that Wine can't keep up. But Microsoft are very limited in how they can do that, as they have to maintain backwards compatibility so your old applications work. | |
| [17:29] <YokoZar> This is why, for instance, many programs still worked on Windows 95 even as late as 2001. Being "Windows-95 compatible" would have made most applications work on Wine that late as well | |
| [17:31] <YokoZar> Microsoft has to focus on developers, and developers don't like changing the APIs they use for no reason, especially if it means their application can only be used on the latest version of Windows. I suspect it'll be a few years before we even start seeing a substantial amount of "Vista or later" applications | |
| [17:31] <YokoZar> KDesk: QUESTION: What is the difference between Wine, Cedega and CrossOver? | |
| [17:32] <YokoZar> Cedega is unrelated at this point. It is a proprietary fork of a very old version of Wine. From what I read on the forums, free Wine works better than Cedega in almost all cases, even on applications that Cedega specifically "supports" | |
| [17:33] <YokoZar> Every now and again Transgaming (makers of Cedega) likes to claim that they are "sending code to Wine" but their contributions have been largely worthless | |
| [17:34] <YokoZar> Crossover is based on free, LGPL Wine. It's basically a tested version of Wine with a few small proprietary UI enhancements in front of it | |
| [17:34] <YokoZar> Codeweavers makes Crossover, and they've been good patrons of the project. They pay most of the development staff, and everything gets pushed into Wine | |
| [17:35] <YokoZar> Sometimes Crossover will work better than Wine for it's specific supported application (say, Office), but this is mainly because they put some ugly specific-workaround that might isn't suitable for Wine as a whole as it could break other applications | |
| [17:35] <YokoZar> mbt: QUESTION: 16-bit apps don't run on 64-bit systems. Does Wine intend on providing some sort of real emulation layer to support old Windows/DOS applications that depend on real-mode? | |
| [17:35] <YokoZar> Well, there's libx86 for one | |
| [17:36] <YokoZar> Which Wine does link against and use | |
| [17:36] <YokoZar> Although that did break with Wine 1.1.7, but that's being fixed | |
| [17:37] <YokoZar> But yes, that is something done, although it's obviously not a huge priority | |
| [17:37] <YokoZar> joemoe: QUESTION: Do you know if Linux will ever get native support for running PE libraries and executables? | |
| [17:37] <YokoZar> That's exactly what Wine is | |
| [17:38] <YokoZar> Wine is a native program for loading PE libraries and executables. | |
| [17:38] <YokoZar> joemoe: QUESTION: Is there any way someone with moderate development skills can make Wine better? | |
| [17:39] <YokoZar> Yes. Quite a few ways | |
| [17:39] <YokoZar> You can help me with the UI stuff if you like | |
| [17:39] <YokoZar> But if you're a C hacker, you can do a lot for Wine | |
| [17:39] <YokoZar> There's tons of areas of the code that are basically "todo, someone please just implement me" | |
| [17:40] <YokoZar> The best thing you can do is just send an email to wine-devel and say "hey I'd like to help point me to something" | |
| [17:40] <YokoZar> Laney: QUESTION: Do you have stairs in your house? | |
| [17:40] <YokoZar> Yes, I am protected. | |
| [17:40] <YokoZar> johnflux: QUESTION: Can you explain what you mean by 'USB support' a bit more? what does this mean for wine to support USB? | |
| [17:41] <YokoZar> So, when you plug your iPod touch in it's not just a generic mass storage device | |
| [17:41] <YokoZar> it loads it's own special driver | |
| [17:41] <YokoZar> Wine can't understand that driver yet | |
| [17:41] <YokoZar> That's basically it - we need code to understand the special instructions devices like that are saying. | |
| [17:42] <YokoZar> Interestingly, with support like that it's possible that arbitrary USB devices can be made to work on Linux, like key fobs | |
| [17:42] <YokoZar> Or, more relevantly, printers | |
| [17:42] <YokoZar> There might come a time in the future where you'd be able to make a printer work on Linux by running the driver through Wine. | |
| [17:43] <YokoZar> This is something you can't do with most drivers since Wine has to be a user-level process, but printers is an interesting exception | |
| [17:43] <YokoZar> sudobash: Question: Do you have a list of software that you have tested and feel confident that it can be used by intermediate to advanced users with no problem? | |
| [17:43] <YokoZar> Check out appdb.winehq.org | |
| [17:43] <YokoZar> That's basically our best guess at what works | |
| [17:44] <YokoZar> Even working applications can run into issues if you have bad (especially ATI) drivers though | |
| [17:44] <YokoZar> I will say that I recently put Ubuntu on my girlfriend's EeePC, and installing iTunes 7 was a matter of double clicking | |
| [17:45] <YokoZar> Works fine for her, even with her purchased music | |
| [17:45] <YokoZar> (iTunes 8 can't install though) | |
| [17:45] <YokoZar> And she's a complete newbie. But now she's becoming more in to the Ubuntu thing | |
| [17:45] <YokoZar> Wine was her gateway drug | |
| [17:45] <YokoZar> robin0800: Question: Where can I find a list of applications that definatly work 100% in wine? | |
| [17:46] <YokoZar> Appdb is your best bet still | |
| [17:46] <YokoZar> We always need help with AppDB | |
| [17:46] <YokoZar> It's very easy to become a "maintainer" and just test your application with the latest Wine version | |
| [17:46] <YokoZar> So please do that if you care about an app and it doesn't have a maintainer | |
| [17:47] <YokoZar> alucardni: QUESTION: Do you think there will be a day when we won't need WINE anymore?? | |
| [17:47] <YokoZar> No | |
| [17:47] <YokoZar> There will always be legacy Windows applications that people won't rewrite | |
| [17:47] <YokoZar> In many cases they might not even have the code | |
| [17:47] <YokoZar> Especially old games | |
| [17:48] <YokoZar> Warcraft 3 will never get a Linux port, for instance. But it runs great in Wine, and it will continue to do so 10 years from now | |
| [17:48] <YokoZar> rzr: QUESTION: i know this is outtopic but are there similar project that aims to support macosx/cocoa binaries over gnu/linux ? | |
| [17:48] <YokoZar> I'm pretty sure there isn't | |
| [17:48] <YokoZar> at least not one that's gotten anywhere near off the ground | |
| [17:48] <YokoZar> lordnoid: QUESTION: Are virtualisation apps like VirtualBox and VMWare competition for Wine? | |
| [17:49] <YokoZar> Yes, since they share a similar goal - running Windows applications | |
| [17:49] <YokoZar> But virtualization has inherent limitations, especially the need to purchase and run a copy of Windows | |
| [17:50] <YokoZar> Wine might actually help virtualization technology. If you have a Windows application that works in Wine, for instance, you can easily redistribute it for free as a Virtual Appliance | |
| [17:50] <YokoZar> If Wine didn't work, such an application would require a license for Vista Ultimate with every distribution | |
| [17:52] <YokoZar> Also keep in mind that some virtualization solutions actually use Wine | |
| [17:52] <YokoZar> For instance Parallels has "Direct3D 8 support" because it's actually using Wine's Direct3D 8 code to translate Direct3D into OpenGL | |
| [17:53] <YokoZar> mirza: QUESTION: How goes cooperation with ReactOS project if there is any ? | |
| [17:53] <YokoZar> ReactOS uses Wine code | |
| [17:53] <YokoZar> The Wine project itself is careful to not use any ReactOS code that might have been tainted, however | |
| [17:54] <YokoZar> Some parts of Wine are directly from reactOS, for instance the Wine task manager | |
| [17:55] <YokoZar> If I skipped any questions on accident, please feel free to reask | |
| [17:55] <YokoZar> Or follow up :) | |
| [17:55] <YokoZar> We've got a couple of minutes left | |
| [17:56] <YokoZar> I'd like to make an announcement too. I'm making a video detailing the current state of Wine, and where it can be improved in terms of user experience | |
| [17:57] <YokoZar> If you double click an exe file in Intrepid without having Wine installed, for instance, it acts like it's an archive and spits out some completely useless error message | |
| [17:57] <YokoZar> samgee: QUESTION: I thought ReactOS had cleared up all their 'tainted' issues? | |
| [17:58] <YokoZar> I believe they have, but most of the ReactOS developers work these days seems to be on the ReactOS specific parts (rather than the Wine parts), eg their kernel | |
| [17:58] <YokoZar> MadsRH: QUESTION: Are there any plans for 9.04 to present the user with a "You need to install Wine to run this application" dialog when running a exe file or similar? | |
| [17:58] <YokoZar> Yes, that's right on my todo list. You can check my wiki page for other projects https://wiki.ubuntu.com/ScottRitchie | |
| [17:59] <YokoZar> Hopefully I'll find time (or employment) to complete them all before 9.04 | |
| [17:59] <YokoZar> Some of those projects are over a year old when I went to the Ubuntu Developer Summit for Hardy | |
| [18:00] <YokoZar> I do still need to clean that wiki page up a bit. It kind of looks like the pen board I have on my wall - large todo list | |
| [18:00] <jcastro> ok that's about it for time | |
| [18:00] <jcastro> thanks scott! | |
| [18:00] <YokoZar> You are welcom sir :) | |
| === jcastro changed the topic of #ubuntu-classroom to: Current Session: Verifying Stable Update (SRU) bugfixes | Welcome to Openweek, questions in #ubuntu-classroom-chat please || Session details here: https://wiki.ubuntu.com/UbuntuOpenWeek | |
| [18:01] <jcastro> take it away sbeattie! | |
| [18:01] <sbeattie> jcastro: should we take a couple minute break? | |
| [18:01] <jcastro> sure, if you'd like | |
| [18:03] <sbeattie> Alright, let's get started. Questions in #ubuntu-classroom-chat please, feel free to ask as we go along. | |
| [18:04] <sbeattie> I'm from the ubuntu QA team, here to talk a little bit about the process we go through to verify bugfixes for updates. | |
| [18:05] <sbeattie> First, a little overview of the stable release updates (SRU) process | |
| [18:06] <sbeattie> SRUs are updates to packages in released versions of the distro (e.g. Intrepid, Hardy) to fix particularly annoying bugs | |
| [18:06] <sbeattie> The process is that we get a proposed fix for a bug from a developer or upstream. | |
| [18:07] <sbeattie> Once we have that, either the SRU team (for packages in main) or MOTU SRU (for universe) will approve the fix, then the archive admins will upload the package to the aprropriate -proposed repository | |
| [18:08] <sbeattie> Once that happens, then verification of the fix occurs, and the package can go into the updates repository | |
| [18:09] <sbeattie> The annoying cdrom eject behavior in intrepid is being fixed via this process, for example. | |
| [18:09] <sbeattie> The process is described in detail at https://wiki.ubuntu.com/StableReleaseUpdates | |
| [18:10] <sbeattie> Security updates go through a slightly different but similar process, due to often being embargoed such that they can't be publicly discussed until a certain date | |
| [18:12] <sbeattie> There are two things we're looking for when a package moves into -proposed before it can go into -updates: verification of the fix itself and regressions | |
| [18:12] <sbeattie> Both are important to find; we want to find regressions so that we don't break things for existing users | |
| [18:13] <sbeattie> But we also want to verify that the fix does actually improve the situation for our users; if it doesn't help we're wasting their time and bandwidth as well as ours. | |
| [18:14] <sbeattie> The current set of outstanding SRUs are tracked (unfortunately) in 3 different locations: | |
| [18:14] <sbeattie> The master page at http://people.ubuntu.com/~ubuntu-archive/pending-sru.html | |
| [18:14] <sbeattie> A page specifically dedicated to packages in main at http://people.ubuntu.com/~sbeattie/sru_todo.html | |
| [18:15] <sbeattie> And a corresponding one for universe at http://qa.ubuntuwire.com/sru/todo.html | |
| [18:15] <sbeattie> (each contains slightly different information, which is why they all exist, but my intention is to merge them all eventually) | |
| [18:16] <sbeattie> The process we follow when verifying is roughly as follows: | |
| [18:16] <sbeattie> - set up a known, relatively pristine, environment | |
| [18:17] <sbeattie> - try to reproduce the issue in the most recent published version of the software | |
| [18:17] <sbeattie> - enable the -proposed repository and install the proposed update for that software | |
| [18:17] <sbeattie> - try to reproduce the issue with the updated software | |
| [18:18] <sbeattie> - report results on the SRU bug report | |
| [18:18] <sbeattie> I'll cover these steps in a little more detail with a walkthrough of a particular bug | |
| [18:19] <sbeattie> The first step is to make sure we have a relatively pristine version of the distro the fix is targeted for | |
| [18:19] <sbeattie> we want to ensure we eliminate outside effects, and are just seeing the changes due to the package update | |
| [18:20] <sbeattie> virtualization is useful here, particularly with snapshotting capabilities | |
| [18:20] <sbeattie> so kvm, virtualbox, xen, vmware et al are nice tools to have for an isolated environment | |
| [18:21] <sbeattie> Generally, we also want to make sure the image is up-to-date with regard to updates | |
| [18:22] <sbeattie> Then we try to reproduce the bug in question. | |
| [18:22] <sbeattie> We want to do this so that we know what we're testing for when we install the proposed package. | |
| [18:23] <sbeattie> Often just doing this will offer us more insights into the actual problem being solved than just a cursory read of the bug report. | |
| [18:23] <sbeattie> If we can't reproduce the initial problem, we have much less confidence that the update actually is going to fix the issue | |
| [18:24] <sbeattie> (sometimes, the bug has already been fixed and launchpad's status is out of date for whatever reason) | |
| [18:25] <sbeattie> hardware specific bugs offer their own special problems; if we don't have the hardware, it's much harder to reproduce | |
| [18:25] <sbeattie> There's not much we can do except prompt on the bug report for people affected by the bug to do the testing and install the package ourselves to look for regressions. | |
| [18:26] <sbeattie> These are often kernel or X problems, though webcams are particular issue for intrepid | |
| [18:27] <sbeattie> The example bug we'll look at is an SRU for update-manager in intrepid, bug 40058 | |
| [18:27] <sbeattie> https://bugs.launchpad.net/ubuntu/+source/update-manager/+bug/40058 | |
| [18:27] <sbeattie> update-manager is a key tool, the default gui for performing updates; it's important we don't break this for our users! | |
| [18:28] <sbeattie> The bug is that changelogs aren't being shown to users | |
| [18:28] <sbeattie> particularly if source repositories aren't enabled in apt | |
| [18:29] <sbeattie> In this case, the dev has been nice enough to put in an explicit TESTCASE section | |
| [18:30] <sbeattie> That's not always the case, so sometimes you get to figure out how to test for the condition | |
| [18:31] <sbeattie> Normally, we would wait to enable the -proposed repository until after we've reproduced the bug, but in this case we'll do it now as part of the reproduction procedure | |
| [18:32] <sbeattie> You can do this through the menus, via System -> Admin -> Software Sources menu | |
| [18:32] <sbeattie> or invoke the tool directly 'sudo software-properties-gtk' | |
| [18:33] <sbeattie> From there, under the 'updates' tab menu, tick the "proposed updates" box | |
| [18:33] <sbeattie> alternately, you can edit or script changes to your /etc/apt/sources.list file directly | |
| [18:34] <sbeattie> you'll want to ensure that apt's information is update, via 'sudo apt-get update' | |
| [18:35] <sbeattie> (If you're specifically editing /etc/apt/sources.list, you'd want something similar to "deb http://us.archive.ubuntu.com/ubuntu/ intrepid-proposed universe main multiverse restricted" as a line in there) | |
| [18:36] <sbeattie> Again, we'd normally do this after reproducing the bug on the original software. | |
| [18:36] <sbeattie> As part of reproducing, we want to double-check exactly what version of the software we're testing. | |
| [18:37] <sbeattie> In this case, we'll look to see what version of update-manager and update-manager-core we have installed | |
| [18:37] <sbeattie> from the command line, we can do: apt-cache policy update-manager update-manager-core | |
| [18:37] <sbeattie> The relevant portions of that are: | |
| [18:38] <sbeattie> Version table: | |
| [18:38] <sbeattie> 1:0.93.34 0 | |
| [18:38] <sbeattie> 500 http://us.archive.ubuntu.com intrepid-proposed/main Packages | |
| [18:38] <sbeattie> *** 1:0.93.32 0 | |
| [18:38] <sbeattie> 500 http://us.archive.ubuntu.com intrepid/main Packages | |
| [18:38] <sbeattie> 100 /var/lib/dpkg/status | |
| [18:38] <sbeattie> (note that we're also seeing the proposed version as available but not installed_ | |
| [18:38] <sbeattie> (the *** indicates which version is installed) | |
| [18:39] <sbeattie> Okay, so we've got the original version installed, let's run it and see if we can reproduce | |
| [18:39] <sbeattie> we'll open update-manager, either via the menu or via 'sudo update-manager' directly | |
| [18:40] <sbeattie> scroll (if necessary) to the proposed updates section | |
| [18:40] <sbeattie> if you're not seeing some proposed updates available, then either something's gone wrong in the process of enabling the -proposed repository, or you already had -proposed enabled | |
| [18:41] <sbeattie> Tick the 'description of update' toggle for one of the proposed updates and you should see something like | |
| [18:41] <sbeattie> http://people.ubuntu.com/~sbeattie/OW2008/Update_Manager_unfixed.png | |
| [18:42] <sbeattie> If so, then yay, we've reproduced the bug. | |
| [18:42] <sbeattie> so then we move on to seeing if the bug still exists in the proposed version. | |
| [18:43] <sbeattie> we'll go ahead and install the update-manager and update-manager-core packages from proposed | |
| [18:44] <sbeattie> again, we don't want to install all the packages in proposed here, because we want isolate interference to ensure we know what we're looking at | |
| [18:45] <sbeattie> since we're testing update-manager we could untick all the other updates, but exiting and running 'sudo apt-get install update-manager update-manager-core' is a bit quicker | |
| [18:45] <sbeattie> then we try to reproduce the problem again | |
| [18:46] <sbeattie> as before, we want to run 'apt-cache policy update-manager update-manager-core' to ensure we have the right packages installed | |
| [18:47] <sbeattie> Now version 1:0.93.34 of each should be installed | |
| [18:47] <sbeattie> we'll run update-manager again | |
| [18:47] <sbeattie> and look at the details for some of the proposed updates | |
| [18:48] <sbeattie> should see something like: http://people.ubuntu.com/~sbeattie/OW2008/Update_Manager_proposed.png | |
| [18:48] <sbeattie> Yay, looks like the bug is fixed in the proposed package. | |
| [18:49] <sbeattie> We'll poke around checking out the details for multiple packages to make sure they're all showing changelogs | |
| [18:49] <sbeattie> Once we've done that, we can look for regressions as well. | |
| [18:50] <sbeattie> Some things to look for in terms of regresions: | |
| [18:50] <sbeattie> - install requirements; does it depend on other packages in proposed, particularly ones not from the same source package? | |
| [18:50] <sbeattie> if it does, and both packages aren't moved into updates at the same time, the package will be uninstallable. | |
| [18:51] <sbeattie> - possible differences in behavior for fresh install versus update from buggy version; sometimes the fixes only show up for new installations, particularly for packaging problems | |
| [18:51] <sbeattie> Things to do: | |
| [18:52] <sbeattie> just generally play with the app; helps here if you're familiar with it. | |
| [18:52] <sbeattie> For update-manager we'll go ahead and install some updates, and see if things still work properly | |
| [18:53] <sbeattie> we can also look on https://testcases.qa.ubuntu.com/ for things to test | |
| [18:53] <sbeattie> The security team has some regression scripts in a bzr tree: | |
| [18:53] <sbeattie> https://code.launchpad.net/~ubuntu-bugcontrol/qa-regression-testing/master | |
| [18:53] <sbeattie> Finally, you can also enable the proposed repository for your day-to-day desktop: https://wiki.ubuntu.com/Testing/EnableProposed | |
| [18:54] <sbeattie> That comes at some risk, if a bad update comes through. But we want to know about that as early as possible! | |
| [18:54] <sbeattie> Finally we report our results on the SRU bug on launchpad | |
| [18:55] <sbeattie> Both failures and successes are important to report! | |
| [18:55] <sbeattie> note both the before and after versions of the software you tested | |
| [18:55] <sbeattie> also report anything out of the ordinary or unexpected you had to do in testing | |
| [18:56] <sbeattie> it's also helpful to update the description of the bug if elements are missing (eg. you created a TESTCASE) | |
| [18:57] <sbeattie> When doing all this, if you run into problems, you can always solicit for help on #ubuntu-bugs and #ubuntu-testing, as well as asking directly on the bug report | |
| [18:57] <sbeattie> We could use help in doing verifications and looking for regressions | |
| [18:58] <sbeattie> In particular, hardy SRUs have lagged recently due to focus on the intrepid release | |
| [18:58] <sbeattie> There's an SRU verification team on launchpad at https://launchpad.net/~sru-verification | |
| [18:58] <sbeattie> And this coming monday, November 10th, we'll be focusing a testing day on SRU verifications as well. | |
| [18:59] <sbeattie> That's all I have, if there's quick questions I can take them here or over in #ubuntu-testing | |
| [19:02] <jcastro> ok thanks sbeattie! | |
| === jcastro changed the topic of #ubuntu-classroom to: Current Session: usb-creator | Welcome to Openweek, questions in #ubuntu-classroom-chat please || Session details here: https://wiki.ubuntu.com/UbuntuOpenWeek | |
| [19:02] <jcastro> And now we have Evan Dandrea, who will be discussing usb-creator | |
| [19:02] <jcastro> a new feature in 8.10! | |
| [19:03] <jcastro> take it away evand! | |
| [19:03] <evand> My name is Evan Dandrea, and I work on the Installer Team for Canonical. I am the primary author of usb-creator. | |
| [19:03] <evand> Is there a volunteer to field questions from -chat? | |
| [19:04] <jcastro> evand: looks like you're on your own. :D | |
| [19:04] <evand> hooray :) | |
| [19:04] <evand> ok | |
| [19:04] <evand> I'm going to (hopefully) briefly speak about what usb-creator is, where it stands, and where it's going, as well as how you can help make it awesome, and then field questions. Though if you have a question about a specific point I make, please feel free to interrupt. | |
| [19:05] <evand> usb-creator was started out of the need to provide an installation medium for systems without a CD-ROM drive and as a way of providing a much faster installation. | |
| [19:05] <evand> It was decided to create a desktop program that writes a CD or ISO file to a USB disk rather than autogenerate images on http://cdimage.ubuntu.com because space and bandwidth on cdimage are already maxed out, and because a tool would still be needed to write the generated image to a disk. | |
| [19:05] <evand> Incidentally, this is part of the reason why hybrid CDs do not solve the fundamental problem that usb-creator was created for. A tool is still needed to write the image to disk and to create a persistence file, if requested. | |
| [19:06] <evand> For more details on this, the original specification for usb-creator can be found here: | |
| [19:06] <evand> https://wiki.ubuntu.com/USBInstallationImages | |
| [19:06] <evand> There was much more work than just writing usb-creator, as proper support for USB disk installs has been a long standing series of bugs. | |
| [19:07] <evand> Briefly, this included making sure that the installer did not let you install to a USB disk if it was also the installation source media, adding root-by-uuid support to GRUB and grub-installer so that disk ordering issues, such as when you remove a USB disk after install, should no longer break the bootloader configuration, and making sure the installer selected a reasonable default when installing from a USB disk (e.g. not hd0). | |
| [19:08] <evand> jcastro: can you +v homy ? | |
| [19:08] <evand> Right, so that's briefly why we have the tool. | |
| [19:09] <evand> thanks | |
| [19:09] <evand> If you have 8.10 installed, you should have a "Create a USB disk" entry in System -> Administration | |
| [19:10] <evand> If you run this and insert a CD and USB disk, or select an ISO file from your desktop, it will write the contents of that CD image to the USB disk, making some modifications to the bootloader configuration | |
| [19:11] <evand> these are just to add things like persistence support, removing the "remove the CD and press enter to continue" message, and a hint to the alternate installer that you're using a USB disk as an install source | |
| [19:11] <evand> It also, if you elect to specify a persistence size, writes a persistence file to the USB disk | |
| [19:11] <homy> QUESTION: <weboide> does it support any cdimage of ubuntu and any distro? | |
| [19:11] <evand> this will allow you to make changes to the booted usb disk and have them save even after you've rebooted | |
| [19:12] <evand> It supports any Ubuntu derived distribution | |
| [19:12] <evand> It *might* work with Debian CDs, but I honestly haven't tested that | |
| [19:13] <evand> It will work with both desktop and alternate Ubuntu, Kubuntu, Mythbuntu, and Xubuntu CDs for both i386 and amd64 archs. | |
| [19:13] <homy> QUESTION: <janitux> the persistence is only for files inside the user's home, or it can save changes like apps or changes in the theme? | |
| [19:14] <evand> It's the entire filesystem, so you can install programs, change the wallpaper, write documents, etc | |
| [19:14] <evand> casper (the live CD initramfs) supports making only home persistent, but usb-creator only supports making all of / persistent at the moment | |
| [19:15] <evand> Continuing, once usb-creator finishes, you can reboot your computer with the USB disk inserted and boot from it | |
| [19:16] <evand> you may have to enter your boot menu to access it | |
| [19:16] <evand> it operates much like a regular Ubuntu CD does | |
| [19:16] <evand> except, as mentioned, your changes are saved, assuming you enabled persistence | |
| [19:17] <evand> Onto future plans... | |
| [19:17] <evand> It has always been the plan to support usb-creator on GNOME, KDE, and Windows, but it was only feasible to develop the initial GNOME frontend in the 8.10 cycle. | |
| [19:17] <evand> That said, KDE and Windows frontends are planned for 9.04, with a branch already created by a contributor to start work on the former. | |
| [19:18] <evand> I have not planned a Mac frontend and backend at this time as it is my understanding that Apple's firmware is broken and does not properly support booting "legacy" operating systems from USB devices. | |
| [19:19] <evand> However I cannot test this beyond my somewhat aging 2006 iMac and have been unable to dig up much information on the problem. If anyone has experience in this area or knows more about the issue, please let me know. | |
| [19:19] <evand> In addition to new frontends and backends, after discussions with probono and the ubuntu-devel mailing list, it was decided that we should embed gnome-app-install into usb-creator to allow the user to select additional software to include on the USB disk: | |
| [19:20] <evand> The full thread on that can be found here: https://lists.ubuntu.com/archives/ubuntu-devel/2008-October/026698.html | |
| [19:22] <evand> Obviously, bug fixing is going to be a major focus in further development. | |
| [19:22] <evand> A number of early users ran into problems when using usb-creator because it only wrote a bootloader to the partition it was being installed on, rather than the MBR of the device. | |
| [19:23] <evand> This was decided as part of the goal as being as unintrusive as possible (it also only formats the device if it really has to and will happily coexist with other files on a vfat partition). | |
| [19:23] <evand> Unfortunately, many scenarios require that a bootloader be written to the MBR, and during the final release candidate freeze, I changed usb-creator to always write a bootloader to the MBR, but it was unfortuantely too late to get the change in the 8.10 final release. | |
| [19:24] <evand> However, you can expect it and additional bug fixes in a Stable Release Update of usb-creator soon, as well as a backport of that update to 8.04.1. | |
| [19:24] <evand> Some thoughts on getting involved... | |
| [19:25] <evand> There has already been a few new branches of usb-creator created to fix bugs, add new functionality, and even provide new frontends. I encourage anyone interested in helping to develop usb-creator to branch off of trunk and start poking around the source code: | |
| [19:25] <evand> bzr branch lp:usb-creator usb-creator.trunk | |
| [19:25] <evand> I lurk in #ubuntu-installer and can answer any usb-creator development questions you have there, or via private message. | |
| [19:26] <evand> If you encounter any bugs in usb-creator, please file them here: | |
| [19:26] <evand> http://launchpad.net/ubuntu/+source/usb-creator/+bugs | |
| [19:27] <evand> Please look to see if your bug has already been filed though. | |
| [19:27] <evand> At a minimum I will need a detailed description of the problem in order to fix it. | |
| [19:28] <evand> The output of hal-device as an attachment and sudo parted /dev/sdb print (where sdb is your usb disk) often helps | |
| [19:29] <evand> One final note... | |
| [19:29] <evand> I suspect many of you are not aware that Canonical is selling USB disks with Ubuntu (and 3GB of persistent storage) on them via its international shop (http://shop.canonical.com) and very soon on its US shop (http://usshop.ubuntu.com). | |
| [19:30] <evand> Having been tasked with generating the images for these disks using usb-creator, I've spent a fair amount of time using them and can attest to their build quality. I've never owned a usb disk (and I've had a fair number of them) that felt as solid as these do. | |
| [19:30] <evand> It makes no difference to me if you buy one, but I figured I'd mention them as I was thoroughly impressed with the design and would pick up one myself if I didn't already have a test copy :) | |
| [19:31] <evand> Apologies for kind of speeding through this, but I wanted to leave plenty of time for questions as this is a very new project and I suspect there may be quite a few. | |
| [19:31] <evand> So, any questions? | |
| [19:33] <evand> QUESTION: < Inteysedut> "starting up" followed by freeze what can I do? | |
| [19:33] <evand> (turning that into a question myself ;) | |
| [19:34] <evand> Please file a bug. Attach the output of `ps auxf` and strace -p $PID -o strace.log where pid is the process ID of install.py | |
| [19:35] <evand> the strace command will run as long as usb-creator is running, so just let it run for about a minute or so and then Ctrl-C it | |
| [19:36] <evand> jcastro: can you +v homy again? | |
| [19:36] <evand> QUESTION: Is it working in kubuntu now? | |
| [19:37] <evand> There is no Kubuntu frontend yet. | |
| [19:37] <evand> It's in development for 9.04 | |
| [19:37] <evand> There was an issue running usb-creator from Kubuntu (it was missing dependencies) | |
| [19:37] <evand> That should have been fixed in usb-creator 0.1.10 | |
| [19:39] <evand> If it still doesn't work in Kubuntu, please attach ~/.usb-creator and /var/crash/*usb-creator* to a new bug report. | |
| [19:40] <homy> QUESTION: <johnsgruber> How new does a computer need to be to be able to boot from USB? | |
| [19:40] <evand> It just needs to be able to boot from USB disks | |
| [19:40] <evand> you can usually find this out by entering the boot menu on your computer | |
| [19:41] <evand> and if it has an option like "USB device" then you should be able to use usb-creator just fine | |
| [19:42] <evand> I suspect any computer made within the past 3-4 years will be ok, but I could be horribly, horrbily wrong :) | |
| [19:42] <evand> horribly* | |
| [19:42] <homy> QUESTION: <knome> Even if I couldn't boot from USB stick, I can create a CD with a boot menu, right? | |
| [19:43] <evand> Not using usb-creator, but yes, there are instructions on the wiki for customizing Ubuntu CDs... | |
| [19:43] <evand> https://help.ubuntu.com/community/LiveCDCustomization | |
| [19:44] <evand> oh | |
| [19:44] <evand> I think I misread the question | |
| [19:44] <evand> It depends on what your BIOS supports | |
| [19:44] <evand> But yes, you might be able to get away with chainloading to the usb disk | |
| [19:44] <homy> QUESTION: <ilia> Can I somehow use this project results to install from locally downloaded iso image, booting from locally installed grub, i.e. I can't boot from CD/USB but already have some ext3 partition to put iso on | |
| [19:45] <evand> usb-creator only looks for USB disks (and SD cards soon) | |
| [19:46] <evand> but, you could dump the contents of an Ubuntu CD to an ext3 partition | |
| [19:46] <homy> QUESTION: <janitux> Can i create an iso based on the usb stick? | |
| [19:46] <evand> and install a bootloader to that, assuming you point the bootloader to the kernel and initrd | |
| [19:46] <evand> you'd be better off with a vfat partition as then you could use syslinux | |
| [19:48] <evand> janitux: I'm not sure I understand your question, but the wiki page I posted above will let you create a custom CD and you should be able to use a USB disk that usb-creator made as a source for that. | |
| [19:48] <homy> Question: <sudobash> For a USB Live Virus Scanning Tool what distro would you recommend and if there is a specific tool like the ClamAV Live CD/USB I would be interested... | |
| [19:49] <evand> Of course I would recommend Ubuntu ;), but I am biased. In future versions of usb-creator you'll be able to add the clamav packages from the archive, but if you're looking for a solution now, or something more tightly integrated to the USB disk (such as having a clamav scan in the boot menu) I'd suggest looking at one of the existing options | |
| [19:50] <evand> I'm not very familiar with whats out there, otherwise I'd suggest one | |
| [19:50] <jcastro> (10 minute warning!) | |
| [19:50] <evand> thanks jcastro | |
| [19:50] <evand> Any last questions? | |
| [19:51] <homy> SEMI-QUESTION: <knome> I was mostly referring to http://www.gnu.org/software/grub/manual/html_node/Making-a-GRUB-bootable-CD_002dROM.html , because I've been recently working with a laptop which can't boot from USB, but the user wants a USB-based Ubuntu installation | |
| [19:52] <homy> QUESTION: <komputes> Creating a persistent install on USB key - Can I make an installation on Ubuntu on a HDD, boot from live cd afterwards and create a disk image of the installation using dd, take the outputted ISO and pass it onto a key using usb-creator. | |
| [19:53] <evand> knome: Could work, but I have not tried it myself. The image usb-creator makes is like any other live CD produced by Ubuntu, so any bootloader should work, provided you point it at your kernel and initrd off the USB disk | |
| [19:54] <homy> QUESTION: <Inteysedut> Can i create a usb stick with two or more distros to boot from, given the space on my drive is big enough? ..lets say id like xubuntu for old PCs at work, and my plain ubuntu for faster machines on the other end of hte office? | |
| [19:54] <evand> komputes: sort of. You will need to prepare an initramfs, but you could take an existing system as the source for the squashfs | |
| [19:55] <evand> See the wiki link I posted above on live CD customization | |
| [19:55] <evand> Inteysedut: This would be something you would have to do yourself, usb-creator does not support this, but I believe there's at least one project to put multiple distros on a single medium | |
| [19:56] <evand> I'd look into them as a starting point | |
| [19:57] <evand> Ok, I'm going to wrap it up here as we're almost out of time. If you have any additional questions, feel free to email the Installer Team mailing list ubuntu-installer@lists.ubuntu.com, or ask in #ubuntu-installer. | |
| [19:57] <jcastro> ok thanks evan! | |
| [19:57] <evand> Thanks for your time and good luck with all of your USB disk projects! | |
| === jcastro changed the topic of #ubuntu-classroom to: Current Session: Cruft. What is it and why it sucks and cruft workshop | Welcome to Openweek, questions in #ubuntu-classroom-chat please || Session details here: https://wiki.ubuntu.com/UbuntuOpenWeek | |
| [19:58] <jcastro> we'll take a few minute break before NCommander takes the floor | |
| [19:58] * NCommander nods to jcastro | |
| [20:01] <NCommander> Hello, and welcome to the Cruft 101 lesson here in #ubuntu-classroom | |
| [20:01] <NCommander> I'm your host, NCommander, also known as Michael Casadevall and sonicmctails. I'm an MOTU, and the founder of the ubuntu-cruft-buster team on Launchpad | |
| [20:02] <NCommander> As you are all still here, I expect your dying to learn the ins and outs of cruft in the Ubuntu archive | |
| [20:02] * NCommander waits for jcastro to -m the room | |
| === nizarus_ is now known as nizarus | |
| [20:02] <NCommander> Ok, you can all talk freely :-) | |
| [20:02] <NCommander> If you have any questions at a given moment, feel free to ask them | |
| [20:03] <NCommander> Anyway, archive cruft is the Ubuntu equivalent to things like junk and other old pieces one might find in the closest | |
| [20:03] <NCommander> *closet | |
| [20:03] <NCommander> They don't directly hurt anyone, but it is undesirable, and removing cruft is a priority of all Ubuntu Developers, some more so than others | |
| [20:04] <NCommander> Cruft, roughly defined, is a package or packages that are no longer used or needed in the Ubuntu archive | |
| [20:04] <NCommander> In most cases, these are software libraries, but normal programs also are subject to cruft removal. | |
| [20:05] <NCommander> Cruft accumulation is a normal part of software development, as software grows and matures, cruft will be generated | |
| [20:05] <NCommander> But how it is generated is something we'll come back to in a minute | |
| [20:06] <NCommander> To understand cruft, you need to understand what reverse-dependencies, or rdepends are | |
| [20:06] <NCommander> For anyone who has ever installed a piece of software on Ubuntu, you know that one program is often dependent on other bits of software | |
| [20:06] <NCommander> These dependent bits are a packages dependencies | |
| [20:07] <NCommander> rdepends are packages that depend on one specific package | |
| [20:07] <NCommander> For instance, X11 is an rdepends of GNOME, or XFCE, or KDE (this is on a very simple level, but the concept applies with concept management) | |
| [20:08] <NCommander> To remove cruft from an archive, that piece of cruft must have no reverse-dependencies that aren't also being removed | |
| [20:09] <NCommander> rdepends can be checked with the apt-cache command | |
| [20:09] <NCommander> When you run apt-cache rdepends on a package, it lists all the reverse dependencies of that package, or none if there are none | |
| === Guest2365 is now known as WastePotato | |
| [20:10] <NCommander> A good example is libwxgtk2.4, an old wxGTK library which is very close to be ready for removal from the archive is libwxgtk2.4-1 | |
| [20:10] <NCommander> On my system, I get the following output: | |
| [20:10] <NCommander> mcasadevall@blacksteel:~/src/sponsoring$ apt-cache rdepends libwxgtk2.4-1 | |
| [20:10] <NCommander> libwxgtk2.4-1 | |
| [20:10] <NCommander> Reverse Depends: | |
| [20:10] <NCommander> python-wxgtk2.4 | |
| [20:10] <NCommander> libwxgtk2.4-dev | |
| [20:10] <NCommander> libwxgtk2.4-1-contrib | |
| [20:10] <NCommander> ctsim | |
| [20:11] <NCommander> As the wxgtk2.4 packages are slated for removal, that leaves ctsim the only package that would have to be migrated to a new wxGTK to remove 2.4 as cruft | |
| [20:12] <NCommander> Now, by this point, you must be wondering how cruft ends up in the archive | |
| [20:12] <NCommander> There are usually two main ways, when a package is updated, and a library's soname is bumped, OR when a libraries API changes to the point that we need a seperate package for that library | |
| [20:13] <NCommander> sonames define the ABI, or Application Binary Interface of a library | |
| [20:13] <NCommander> In laymans terms, the ABI is how other applications access functions and code of a library | |
| [20:14] <NCommander> As long as an ABI is unchanged, any programs that link against that library will continue to work even when that library is updated | |
| [20:14] <NCommander> I realize I'm going kinda fast, does anyone have any questions? | |
| [20:14] <NCommander> (before I continue) | |
| [20:15] <NCommander> Ok, I'm going to assume no questions and continue | |
| [20:15] <NCommander> ABIs change whenever a upstream maintainer does something to change the prototypes of a library | |
| [20:16] <NCommander> Prototypes are entry points into a library, they say how a program can access it, and how many bits of data it will accept via pointers. | |
| [20:17] <NCommander> When these prototypes change, programs calling the old prototype fails | |
| [20:17] <NCommander> (this sometimes shows up as Can not resolve symbol error) | |
| [20:18] <NCommander> Prototypes change because a maintainer adds or removes a feature, or changes the size of the data a function can take (i.e., replacing a int with a long) | |
| [20:18] <NCommander> When this happens, the library maintainer is supposed to bump the soname to show that an ABI break has happened, and any packages that rdepend on it must be transitioned | |
| [20:19] <NCommander> Now if your exceptionally lucky, the upstream maintainer has not changed the Application Programming Interface, or API | |
| [20:19] <NCommander> APIs are like blueprints, they tell a program, at compile time how to call a libraries methods and functions | |
| [20:20] <NCommander> A more realistic scenario is that whatever application rdepended on your library, it will likely need tweaking (or in some cases, full out porting) | |
| [20:21] <NCommander> Actually doing porting is outside the scope of this discussion, and in most cases unneccessarily fortunately | |
| [20:21] <NCommander> (exceptions do exist, for instance gtk1.2 -> 2) | |
| [20:22] <NCommander> Whenever an ABI break happens, we need to do a transition to move all packages from the old version of the library to the new | |
| [20:22] <NCommander> These happen in two forms, NBS (Not Built From Source), or multiple libraries packaged seperately | |
| [20:22] <NCommander> NBS is the most common, so we will look at that first | |
| [20:23] <NCommander> Now for those of you who attended Packaging 101, you may be wondering how a package in Ubuntu can't be built from source | |
| [20:23] <NCommander> The answer is your right, no package can exist without its assiocated source package. NBS more properly defined are Packages that are no longer built from current source packages | |
| [20:24] <NCommander> For instance, lets say I have liba0 in the archive | |
| [20:24] <NCommander> (all library packages have a libname0, and a libname-dev package, or something similar) | |
| [20:24] <NCommander> If the upstream maintainer releases a new version with a soname break | |
| [20:24] <NCommander> I'll now have a liba1 and a liba-dev, instead of a liba0, and a liba-dev | |
| [20:25] <NCommander> Now since liba-dev depends on the new liba1, installing it pulls in the new package | |
| [20:25] <NCommander> The old library, since it is no longer built from the liba source package becomes a NBS library | |
| [20:26] <NCommander> In some cases, doing the transition is as trival as a no-changes upload to cause Launchpad to rebuild your package against the new API | |
| [20:26] <NCommander> *ABI | |
| [20:27] <NCommander> In other cases, you'll likely have to tweak the package to get it build, or (more likely) package a newer upstream or sync a newer upstream from Debian | |
| [20:27] <NCommander> Once all rdepends on a NBS library are removed | |
| [20:28] <NCommander> The library magicially go poof via an NBS cleaner script run by your handy dandy archive administrators | |
| [20:28] <NCommander> (or in other words, they remove themselves) | |
| [20:28] <NCommander> In some cases however, NBS is not possible or apporiate | |
| [20:29] <NCommander> This is normally the case for packages with a tremendous number of rdepends, or ones who have API breaks to the point that each rdepend requires manual porting to get working | |
| [20:29] <NCommander> (gtk1.2, and wxgtk2.4/2.6 are good examples of these in the archive) | |
| [20:30] <NCommander> In that case, what will happen, you will end up with multiple binary and -dev packages | |
| [20:30] <NCommander> To do the transition, each rdepend must be downloading, its build-dependencies changed, build testing, installed, etc. | |
| [20:30] <NCommander> In some cases, this will be as change it and build it, with no other necessary changes | |
| [20:31] <NCommander> In others, it may require working with upstream authors to port their code (or if the package is dead upstream, doing the work yourself for a cruft free future) | |
| [20:32] <NCommander> Any questions so far? | |
| [20:33] <NCommander> I'll assume no, and continue onward | |
| [20:33] <NCommander> Once you have finished removing all the rdepends of a package, you need to actually remove it from the archive | |
| [20:34] <NCommander> Unlike NBS packages, oldlibs do not automagicially remove themselves when their dependencies are removed | |
| [20:34] <NCommander> Manual intervention from the archive administrators are required | |
| [20:34] <NCommander> This is known as a Removal Request | |
| [20:35] <NCommander> Removal Requests state if a source or binaries are being remove, what distributions to remove them from (usually only the current development), and why, as well as confirmed there are no rdepends left of that package | |
| [20:36] <NCommander> As with all packages related tasks | |
| [20:36] <NCommander> Removal requests must be done or at least ACKed by an MOTU for universe/multiverse, or a core developer for main/restricted | |
| [20:36] <NCommander> (it should be noted that a direct removal from main is extremely rare, usually the package will be demoted to universe, then removed) | |
| [20:37] <NCommander> Removals are a two step package | |
| [20:37] <NCommander> er | |
| [20:37] <NCommander> process :-) | |
| [20:38] <NCommander> Beside the actual removal of the package from Ubuntu, the blacklist file may also be updated | |
| [20:38] <NCommander> As anyone running Jaunty knows, packages are pulled from Debian for roughly the first three-four months of a development build cycle (this is known as autosync) | |
| [20:39] <NCommander> The autosync blacklist does extactly what it says on the tin | |
| [20:39] <NCommander> If a package is blacklisted, unless an MOTU or coredev do an explicate sync request that package will not be pulled over from Debian | |
| [20:40] <NCommander> Speaking of Debian, I should briefly bring up how Debian does transitions | |
| [20:40] <NCommander> Just like Ubuntu, Debian goes through transitions (all distributions do) | |
| [20:40] <NCommander> Unlike Ubuntu, rebuilding binary packages on Debian do not require a no-changes source upload | |
| [20:41] <NCommander> Debian does whats called a binNMU, or binary Non-Maintainer Upload | |
| [20:41] <NCommander> These packages are versioned with a +bX in the deb | |
| [20:41] <NCommander> For most intents and purposes, you can pretend these do not exist, but if your looking for hints on how to transition a package, seeing if Debian has done any recent binNMUs on said package or its rdepends is a good place to start | |
| [20:41] <NCommander> (the changelog is also another good place to start) | |
| [20:42] <NCommander> I've covered a lot of information thus far in the last 43 minutes | |
| [20:43] <NCommander> I'd like to open the floor to any questions (I've covered all the basics, but I bet a bunch of you are scratching your heads) | |
| [20:44] <sudobash> 3+. | |
| [20:44] <NCommander> sudobash, shoot | |
| [20:44] <sudobash> sorry accident... | |
| [20:45] * NCommander facepalms | |
| [20:45] <NCommander> Seriously, no questions? I finished my basic overview of cruft, and I was willing to work on doing some hands on cruft removal if anyone was interested | |
| [20:46] <NCommander> Anyway, if you have any questions about cruft, feel free to ping me on IRC< or any of the cruft buster team members | |
| [20:48] <artir> QUESTION: When removing cruft packages from user's pcs, what is used: apt-get remove or apt-get remove --purge ? | |
| [20:48] <NCommander> artir, excellent question | |
| [20:49] <NCommander> If your removing a cruft package directly, apt-get remove *cruftlib* works like you expect it to | |
| [20:49] <NCommander> If a package has no dependencies is installed, and it was installed as part of another package, the next time apt-get remove --purge is run, it is removed | |
| [20:50] <NCommander> artir, does that answer your question? | |
| [20:50] <artir> yep, thanks :) | |
| [20:51] <NCommander> MadsRH, you had a question? | |
| [20:52] <MadsRH> QUESTION: Not a very technical question, but I was just looking at "Remove Cruft" in the menu and why did you decide to use that icon? I really don't see the link to removing cruft. Is it just a temporary icon or is it there to stay? | |
| [20:52] <NCommander> MadsRH, where specifically is the Remove Cruft? | |
| [20:52] <NCommander> In Synaptic? | |
| [20:52] <jcastro> I think he's mixing up what your topic is with the cruft remover app that shipped with intrepid | |
| [20:52] <NCommander> Oh! | |
| [20:53] <MadsRH> system -> admin. -> remove cruft | |
| [20:53] <NCommander> gtk-cruft-removal (now gtk-system-cleaner) is something completely different ;-) | |
| [20:54] <MadsRH> Sorry :-[ Just jumped in! | |
| [20:54] <NCommander> MadsRH, thats fine | |
| [20:57] <NCommander> I guess that does it on cruft | |
| [20:58] * knome applauds for NCommander | |
| [20:58] <NCommander> I did have some work showing actual examples of cruft hands of, if people are interested in that | |
| [20:58] <RainCT> nice explanation, NCommander :) | |
| [20:58] <charlie-tca> Thank you | |
| [20:59] <artir> so the biggest problem to deal with cruft is to know what is, actually, cruft | |
| [20:59] <NCommander> artir, with some experience, it becomes more clear | |
| [21:00] <NCommander> also, there is an oldlibs category for packages that are cruft | |
| [21:00] <NCommander> but still have rdepends thus unremovable | |
| === Silvy is now known as Fierelin | |
| [21:06] <charlie-tca> <gscholz> QUESTION: Is the cruft workshop already running? (five past nine) | |
| [21:07] <NCommander> charlie-tca, No it isn't, it didn't seem like anyone was active (this is more interactive than the previous) | |
| [21:08] <charlie-tca> I just forwarded from gscholz | |
| [21:08] <NCommander> I'd be glad to do the workshop, but its not something I can just stand here and do, its something thats sorta interactive and such :-/ | |
| [21:09] <NCommander> well | |
| [21:09] <NCommander> The people demand it | |
| [21:09] <jcastro> heh | |
| [21:09] <NCommander> Thus I shall talk | |
| [21:09] <fabrice_sp> I wiould be interested, but if I am the only one :-/ | |
| [21:09] * NCommander clears throat | |
| [21:09] <fabrice_sp> really :-) | |
| [21:09] <NCommander> Lets take a look at the oldlibs cruft sections of both Debian and Ubuntu | |
| [21:10] <charlie-tca> I'll forward the questions if you want | |
| [21:10] <NCommander> charlie-tca, yes, please | |
| [21:10] <NCommander> http://packages.ubuntu.com/jaunty/oldlibs/ - Ubuntu's oldlibs | |
| [21:10] <NCommander> and packages.debian.org seems to be down :-/ | |
| [21:10] <NCommander> (or lagging | |
| [21:11] <NCommander> http://packages.debian.org/unstable/oldlibs/ | |
| [21:11] <NCommander> and there we go | |
| [21:11] <fabrice_sp> question: how is this list populated? Automatically? | |
| [21:11] <NCommander> fabrice_sp, its when the section in the control file is set to oldlibs | |
| [21:11] <NCommander> Usually the Debian QA team will set it, and then that change trickles down to Ubuntu | |
| === andre___ is now known as andre__ | |
| [21:12] <NCommander> oldlibs is sorta the long term storage of cruft | |
| [21:12] <NCommander> Its things we want to toss but can't get rid of due to rdepends | |
| [21:13] <NCommander> There is actually a good example of cruft here, liblzo1 | |
| [21:13] <NCommander> http://packages.ubuntu.com/jaunty/oldlibs/liblzo1 | |
| [21:14] <NCommander> http://packages.debian.org/unstable/oldlibs/liblzo1 | |
| [21:14] <NCommander> The changelog the best reason to see why a library has ended up here | |
| [21:15] <NCommander> CHecking the rdepends on liblzo1, on jaunty, we get: | |
| [21:15] <NCommander> Reverse Depends: | |
| [21:15] <NCommander> transcode | |
| [21:15] <NCommander> liblzo-dev | |
| [21:15] <NCommander> liblzo-dev | |
| [21:15] <NCommander> dact | |
| [21:15] <NCommander> we can safely ignore liblzo-dev since its part of the packaging itself | |
| [21:15] <NCommander> Which leaves dact, and transcode as the two real rdepends | |
| [21:16] <charlie-tca> <canta> why is cruft a problem? | |
| [21:16] <NCommander> canta, another good question. Cruft is a problem because it increases the size of the archive, the amount of maintaince required to maintain it, and the work the security team needs to do to keep things going | |
| [21:17] <NCommander> I'm currently working on migrating the dact package | |
| [21:17] <persia> Also, there's usually bugs fixed in newer libraries and still extant in the older libraries, but perhaps hard to fix, if say, the fix required an API change. | |
| [21:18] <charlie-tca> <gscholz> QUESTION: Sorry I missed the last session. How did you get the backward dependencies for liblzo? | |
| [21:19] <NCommander> gscholz, apt-cache rdepends liblzo1 | |
| [21:19] <NCommander> canta, why don't you ask your question here | |
| [21:19] * NCommander thinks its saner if we base this all in a single change | |
| [21:19] <fabrice_sp> question: how do you know what is the new version? From packages.ubuntu.com? | |
| [21:20] <NCommander> fabrice_sp, yeah, I searched for liblzo, then found it | |
| [21:20] <NCommander> Its usually pretty obvious what the new package is once you see it | |
| [21:21] <NCommander> now dact is actually a great example | |
| [21:21] <NCommander> I recommend anyone who actually wants to experience doing a transition now grab the source package (please note, I'm doing this on intrepid/jaunty. If your using hardy, your millage may vary :-)) | |
| [21:23] <NCommander> You'll also want the devscripts, and build-essential packages (and apt-get build-dep dact && apt-get install liblzo2-dev) | |
| [21:24] <NCommander> Once you grab the source package, and its build-deps, I recommend creating a new changelog entry with dch -i | |
| [21:24] <NCommander> For my changelog, I used this: * Changed rdepend liblzo -> liblzo2 to remove dependency on oldlibs | |
| [21:26] <NCommander> then open debian/control, find the liblzo-dev dependency, and change it to liblzo2-dev | |
| [21:26] <NCommander> Save, and build | |
| [21:27] <NCommander> If done correctly, you should have a dact_0.8.41-4ubuntu1 package | |
| [21:28] <canta> and how does it help if i do something alike with such packages? i cant put them into the ubuntu-archive | |
| [21:28] <NCommander> canta, huh? | |
| [21:28] <NCommander> I don't understand | |
| [21:28] <NCommander> If you referring to getting the package into the Ubuntu archive itself, there is a process known as sponsoring to do extactly that | |
| [21:29] <canta> So i can easily "pay" the community back if i would do transitions? | |
| [21:30] <NCommander> canta, its a good way to get involved, As we're very close to archive open, we don't have any ones going on, but usually a large transition or two happens from time to time where people need to others to test packages, and help build them | |
| [21:31] <fabrice_sp> Shouldn't we also update the Maintainer field and the Standard-version one? Or send the debdiff to Debian? | |
| [21:31] <NCommander> fabrice_sp, normally for no-code changes transitions, its a bug on the BTS, to make the change. Generally speaking, you don't post a debdiff to the BTS unless your doing a non-maintainer upload in debian | |
| [21:32] <NCommander> fabrice_sp, and yes, you would do all the normal work you do when you normally create a patch | |
| [21:32] <fabrice_sp> ok | |
| [21:33] <gscholz> QUESTION: So this was quite easy, but only because the library API did not change from liblzo1 to liblzo2? | |
| [21:33] <NCommander> gscholz, the changelog of liblzo says the API did change, but dact doesn't use any changed APIs | |
| [21:33] <NCommander> (aka, a recompile will fix it) | |
| [21:34] <gscholz> OK | |
| [21:34] <NCommander> while we've been talking, I actually been testing dact against a bunch of files and checking to see if it works right | |
| [21:34] <NCommander> (it does :-)) | |
| [21:34] <NCommander> SO this change will land in jaunty hopefully relatively soon | |
| [21:35] <NCommander> Now, assuming transcode doesn't need any magic to get going against liblzo2, the next step would be to file a removal request | |
| [21:35] <NCommander> Now as we can't actually zap a package from the archive, I'll simply find an old request to show you the elements of a request | |
| [21:36] <NCommander> if LP will work today | |
| [21:36] * NCommander grumbles | |
| [21:38] <NCommander> https://bugs.edge.launchpad.net/ubuntu/+source/ncurses4.2/+bug/269722 | |
| [21:38] <ubot5`> Launchpad bug 269722 in ncurses4.2 "Request for Removal: unneeded oldlibs" [Undecided,Fix released] | |
| [21:38] <NCommander> That's what a removal request by a non-MOTU looks like (as I was at the time) | |
| === Lilith is now known as Fierelin | |
| [21:40] <fabrice_sp> (Motu since 5 hours! Congrats :-) ) | |
| [21:41] <NCommander> Thats transitions in a nutshell | |
| [21:41] <gscholz> QUESTION: We had the same topic yesterday. When you work on Ubuntu like this, how does the result go back to debian (which is "package-upstream")? | |
| [21:43] <NCommander> ah | |
| [21:43] <NCommander> ANother good question | |
| [21:43] <NCommander> Usually with oldlibs transitions, its file a bug and a patch (if needed) and leave it as so | |
| [21:43] <NCommander> SOmetimes Debian will match us on a transition ... and sometimes they won't | |
| [21:45] <gscholz> QUESTION: I guess Debian should not spend the same afford? So you will make a patch and send it to Debian? | |
| [21:46] <gscholz> QUESTION: I am not quite clear about, how communication works here. | |
| [21:47] <NCommander> gscholz, Debian uses BTS (Bug Tracking System) to handle bugs | |
| [21:47] <NCommander> You can use the 'reportbug --BTS=debian' command to file bugs on their tracker | |
| [21:47] <gscholz> OK | |
| [21:48] <fabrice_sp> Question: I was trying to the same for transcode, but the resulting package FTBFS, so in this case, it's patching time? Or I should report upstream? | |
| [21:49] <fabrice_sp> s/to/to do/ | |
| [21:49] <NCommander> fabrice_sp, patching time :-) | |
| [21:49] <fabrice_sp> :-) | |
| [21:51] <fabrice_sp> Another question: How do we know if someone is already working on trying to remove oldlibs? Looking for bugs in launchpad? | |
| [21:52] <persia> Just as a side note, if you patch code to migrate to a new version of a library, upstream is usually glad to receive the patch, and will probably include it in the next release. | |
| [21:53] <NCommander> fabrice_sp, LP bugs is the offical mechanism to know who's doing what | |
| [21:53] <NCommander> fabrice_sp, just check the package, and see if there is a bug about removal or something similar | |
| [21:54] <jcastro> (5 minute warning) | |
| [21:54] <NCommander> jcastro, we're done | |
| [21:55] <jcastro> \o/ | |
| [21:55] <jcastro> thanks NCommander! | |
| [21:55] <jcastro> and congratulations on becoming a MOTU! | |
| [21:55] <fabrice_sp> Thanks NCommander! | |
| [21:55] <charlie-tca> good job, NCommander | |
| [21:55] <jcastro> we'll take 5 before the next session | |
| [21:56] * mneptok turns on the elevator music | |
| [21:56] <NCommander> jcastro, what's up next? | |
| [21:56] <gscholz> Thanks NCommander | |
| === jcastro changed the topic of #ubuntu-classroom to: Current Session: Ubuntu and your Money | Welcome to Openweek, questions in #ubuntu-classroom-chat please || Session details here: https://wiki.ubuntu.com/UbuntuOpenWeek | |
| [21:57] <jcastro> This one is called Ubuntu and Your Money | |
| [21:57] <jcastro> starring Kurt von Finck | |
| [21:57] <jcastro> aka mneptok | |
| [21:57] <mneptok> \o/ | |
| [21:57] <jcastro> "How to make smart buying decisions as a free software user. Covers hardware, software, support, and services. " | |
| [21:57] <jcastro> I have been looking forward to this one! | |
| [21:58] <NCommander> oooh | |
| [21:58] <mneptok> jcastro: i was asked not to dance half naked on the tables. still looking forward to it? | |
| [21:58] <NCommander> sounds interesting | |
| [21:58] <mneptok> everyone ready? | |
| [21:58] <jcastro> ok, I will paste questions from #ubuntu-classroom-chat | |
| [21:58] <jcastro> so please post your questions there | |
| [21:59] <jcastro> mneptok: take it away! | |
| [21:59] <mneptok> LET'S ROCK! | |
| [21:59] <mneptok> Hello everyone, my name is Kurt von Finck. I'm an Ubuntu Member, as well as a Senior Ubuntu System Support Analyst for Canonical. I'll be discussing how, as a user of Free Software, you can make informed and smart buying decisions. | |
| [21:59] <mneptok> I want to point out that although I am a Canonical employee, I am speaking today solely as a community member. Canonical does not endorse hardware products that have not been through the certification process, and certainly does not make specific vendor recommendations. I'll be using some examples during the session, and these are just that, examples. | |
| [21:59] <mneptok> Please pay attention to the methodology and principles involved, and do not assume that because I use Dell, Valusoft, Newegg, Amazon, or some other vendor as an example that this is an official endorsement by Canonical. | |
| [22:00] <mneptok> You will hear me discuss some strategies you might use to advance the cause of Free Software. Clearly, Canonical would benefit from this, as well. But rest assured, my interest and purpose here is to inform and educate, and thus benefit all users of Free Software, not just Ubuntu users or Canonical's business interests. | |
| [22:00] <mneptok> Also, I will be talking about Canonical's support offerings. For this section of the session, you may expect that I am acting as an official Canonical spokesperson. But please, remember that for other sections, I'm a community member. | |
| [22:00] <mneptok> Finally, about questions. I'd like to save these until the end (please jcastro), and I will NOT be entertaining "what printer do I buy, mneptok?!?!" type questions. :) Sorry, but I think that's a rat's nest I'll sidestep. But feel free to ask about strategies for printer shopping (or similar). | |
| [22:01] <mneptok> Phew! Caveats are out of the way! Let's go! | |
| [22:01] <mneptok> First, I want to be clear that this session is not intended to provide an in-depth look at hardware and software support. It's a basic "best practices" guide to making informed decisions. Also, a lot of this content may be moot in a few years. This is more a "Current State Of Affairs" talk than it is a timeless reference guide. | |
| [22:02] <mneptok> Let's start with hardware, which is traditionally the biggest hurdle new users of Free Software face. Being fairly active on IRC, I see a lot of new Ubuntu users coming to grips with the fact that the world that they have entered is rather different from the world of Windows which most of them have left. Let's do a quick comparison. | |
| [22:02] <mneptok> In the world of Windows, you are strictly a consumer. Products are created for you, and you pay to be able to have the right to use them. This is a very traditional business model, and the idea of how profit can be created from such a model is well understood. Thus, most hardware manufacturers understand how they can profit from the Windows ecosystem, and they actively participate in it. | |
| [22:03] <mneptok> Also, Windows has a vast majority of the market share for operating systems. This also encourages vendors to focus on providing solutions for Windows, as their potential profit center is much larger. | |
| [22:03] <mneptok> However, as we all know, hardware does not "Just Work" on Windows. Sure, almost every device you can find has a Windows driver, but a substantial portion of these have a *not very good* Windows driver. The web is rife with examples of users encountering the frustration this causes. That new inkjet printer just doesn't provide output that looks quite right. The USB PCI card driver works, but somehow disables the internal USB. The ... | |
| [22:03] <mneptok> ... webcam operates, but the picture quality is far below what the hardware is capable of. | |
| [22:03] <mneptok> If you have used Windows for any length of time, I'm sure you know the, "I know this is a good piece of hardware, but the driver renders it almost useless," sentiment. | |
| [22:04] <mneptok> (That's almost worth a ;) but really it deserves a :/ ) | |
| [22:04] <mneptok> In the Free Software world things are quite different. Vendors that understand the open source model and embrace it are at a distinct advantage. First, if they open their drivers or hardware specifications, they allow a horde of skilled, yet unpaid, volunteers to work to refine and debug drivers. Second, their hardware drivers, if released under a GPLv2 compatible license, are allowed directly into the Linux kernel. In the case ... | |
| [22:04] <mneptok> ... of the BSDs (Free, Open, Net, etc), the driver needs to be BSD compatible. | |
| [22:05] <mneptok> Sadly, few vendors have taken this approach. But some have. Thus the situation roughly looks like this. If you use Windows, almost every device will work. However, few will work to their potential. If you use Linux, fewer devices work. But those that do work tend to do so fairly well. Make sense? | |
| [22:05] <mneptok> Having your driver in the mainline kernel has distinct advantages. Let's use Intel as a victim ... err ... example here. :) | |
| [22:05] <mneptok> A few years ago, Intel decided that they were first and foremost a hardware company. Thus, keeping drivers closed and secret was not in the best interest of their hardware business, which seeks as broad a base of adoption as possible. Intel released full design specs for their chipsets: CPUs, bridgesets, graphic adapters, and 802.11x, for example. | |
| [22:06] <mneptok> The license Intel chose allows these open drivers to compiled into and distributed with the Linux kernel. This has a few effects. | |
| [22:06] <mneptok> First, it means Linux developers started poking through the driver code and making improvements and tweaks, and/or filing bug reports against code they could see. This led to more robust drivers. Second, it means that if you have an Intel graphics adapter or wireless chipset, your devices "Just Work" in Linux. I have a Lenovo laptop with the Intel GMA965 graphics adapter and 4965BGN wireless chips. When I install Ubuntu, my ... | |
| [22:06] <mneptok> ... wireless card works immediately, and I have 3D and compositing immediately. | |
| [22:06] <mneptok> Contrast this to Broadcom, who regard themselves as purveyors of both hardware and software. They write their own drivers, and will not share the specs necessary to allow others to do driver development. Broadcom chooses not to write Linux or BSD drivers. Thus, getting Broadcom wireless working in Free Software has always been painful, and usually involves somehow "wrapping" the Windows driver so it works. Ugly. | |
| [22:07] <mneptok> So, if you choose an Intel wireless chipset, you install Ubuntu, another Linux, or a type of BSD and everything works right away. If you choose a Broadcom chipset, you have no such guarantee. In fact, it is far more likely you will have problems than that you will eventually make things work. | |
| [22:08] <mneptok> I am NOT disparaging Broadcom here. They have made a business decision. From OUR seats this decision seems flawed. But none of us are truly in a position to say for sure what the best business plan is for any company. Perhaps time will vindicate their position. Perhaps it will affirm ours. But until then, let's not cast aspersions needlessly. Let's keep a civil relationship with these businesses, so they continue to find our ... | |
| [22:08] <mneptok> ... input credible and valuable. 'Nuff said. | |
| [22:08] <mneptok> So, now that you have an idea of how this works, let's look at what manufacturers have made business decisions that make their products attractive to Free Software users. | |
| [22:09] <mneptok> If you are buying a new machine on which to run Ubuntu it's tough to do much better than buying from an OEM that has partnered with Canonical. These vendors send their products through certification, and thus you may depend that they run Ubuntu without issue. A Dell, System76, Sylvania, or other product with Ubuntu pre-installed will give you a great "out-of-the-box" experience, for obvious reasons. | |
| [22:09] <mneptok> You can look in on the Canonical certification process at: http://webapps.ubuntu.com/certification | |
| [22:10] <mneptok> This holds true for other Free Software offerings. eRacks, a server vendor, offers a variety of Free Software operating system options on their products; including Ubuntu, CentOS, OpenBSD, and others. It's a fairly safe bet that eRacks has tested their equipment and has determined that the products they sell you work with the OSes they offer as options. | |
| [22:10] <mneptok> So, if you're buying a new machine, and want the easiest path to a system supported by Free Software, choose a vendor who has chosen Free Software. But what about if you're choosing a vendor without Free Software offerings, building your own, or adding components to an existing machine? Whose products do you choose? Again, choosing a vendor who has chosen to support Free Software is the best course of action. | |
| [22:11] <mneptok> Intel is the poster child here. If you are buying a new machine you can't do much better than choosing Intel graphics and wireless. Intel's realization that they are first and foremost a hardware company, and that they trust the worldwide community of developers to help create robust driver sets for their products, makes them the clear choice of Free Software users. | |
| [22:12] <mneptok> However, Intel does not provide much in the way of aftermarket products. Meaning, it's relatively easy to choose a motherboard with integrated Intel devices, but less easy to locate add-in Intel devices. The exception, mini-PCI Express wireless devices, proves the rule. | |
| [22:12] <mneptok> http://www.amazon.com/Intel-wireless-3945ABG-Network-Connection/dp/B000EDQOK8/ref=sr_1_1?ie=UTF8&s=electronics&qid=1225993082&sr=8-1 | |
| [22:12] <mneptok> http://www.amazon.com/Wireless-Wifi-Link-Mini-Card/dp/B000QAY00K/ref=sr_1_2?ie=UTF8&s=automotive&qid=1225994628&sr=8-2 | |
| [22:13] <mneptok> You can add-in Intel wireless, but Intel provides no add-in graphics. So, if you're looking at adding graphics and wireless options that are supported by Free Software, things become a little more complicated. | |
| [22:13] <jcastro> andresmujica: please hold questions until the end | |
| [22:13] <mneptok> For graphics, the two big players are nVidia and AMD/ATI. It used to be (just a few months ago) that nVidia was a somewhat better choice. nVidia does not offer Free drivers. They offer the same kind of driver support for Linux that they offer for Windows. Namely, they provide a closed-source driver that enables 3D and other advanced features. | |
| [22:14] <mneptok> This approach is commonly called a "binary blob." The vendor supplies a driver, but does not provide access to the source code. Many Free Software advocates find this approach reprehensible, for obvious reasons. However, being pragmatic, users have needs, and if nVidia's binary approach suits your needs, then by all mean avail yourself of it. | |
| [22:14] <mneptok> ATI used to use the same methodology. However, their binary drivers were not as robust as nVidia's in many cases, and that made nVidia somewhat of a better choice. However, the game has changed, as AMD has announced that they will be opening the source to their drivers over the coming months and years. Thus, while at this time last year nVidia seemed the better choice for add-in graphics, this seems to have changed. | |
| [22:15] <mneptok> Understand, though, that binary blobs means work for the end user after installation. Ubuntu will not distribute code without source as part of the base distribution, and thus you'll need to add restricted drivers after you install to enable things like 3D (necessary for things like Compiz). | |
| [22:16] <mneptok> Also, be sure to look at the binary blobs before choosing a card. The latest and greatest card may not have support in the drivers provided by the most recent Ubuntu release. Choose a card you can confirm has support in the current Ubuntu restricted packages. | |
| [22:16] <mneptok> As concerns wireless, the best bet for aftermarket PCI cards are those based on the Atheros chipsets. Atheros is now opening their drivers, and prior to this decision the MadWiFi project reverse engineered Atheros drivers while Atheros turned a blind eye to the potential legal issues (thanks Atheros!). If you're shopping for wireless, and Intel has no options for you, look at the MadWiFi site and compare some of your choices to ... | |
| [22:16] <mneptok> ... the Compatibility section of their site. | |
| [22:16] <mneptok> http://madwifi-project.org/wiki/Compatibility | |
| [22:17] <mneptok> As an example, if you are looking to purchase a Cardbus wireless adapter for use with Ubuntu, the SMC SMCWCB seems to be a good choice according to MadWiFi user reports. Newegg has this card for ~US$26. | |
| [22:18] <mneptok> http://www.newegg.com/Product/Product.aspx?Item=N82E16833129133 | |
| [22:18] <mneptok> Carefully read MadWiFi's compatibility listings. Then look at your preferred vendor's site or retail location and ensure you choose something reported to have worked. Do not trust brand names. Trust specific models and chipsets. This holds true for most devices, not just wireless. | |
| [22:18] <mneptok> On to printing! | |
| [22:18] <mneptok> This one is actually pretty easy. OpenPrinting. | |
| [22:19] <mneptok> http://www.linuxfoundation.org/en/OpenPrinting | |
| [22:19] <mneptok> OpenPrinting is a project to document the current status quo for printing from Free operating systems. The project was recently brought under the aegis of the Linux Foundation. Their printer database is the best resource for Free Software users looking to purchase a printer they can depend upon. | |
| [22:19] <mneptok> http://openprinting.org/printer_list.cgi | |
| [22:19] <mneptok> Determine your needs (e.g. laser vs inkjet, scanning, faxing). Create a short list from your favorite vendor (e.g. Amazon, Newegg, Best Buy). Then look up the models on your short list in the OpenPrinting database. Your choices will narrow, and eventually become clear to you. | |
| [22:20] <mneptok> Other sites exist that provide a database of devices that work (or do not) in Ubuntu. One example is the Linux USB Device database, which catalogs USB devices and their current support status in Linux: | |
| [22:21] <mneptok> http://www.qbik.ch/usb/devices/ | |
| [22:22] <mneptok> What about software? Of course, most Ubuntu users do not purchase much software, and instead use the Free Software offerings provided by the package system. If you find yourself considering a purchase of commercial software, ensure you know the vendor's position with regards to Free Software. | |
| [22:22] <mneptok> In some cases, using an RPM (Red Hat Package Manager) based distribution may be preferable, as this is what the software vendor uses. In some cases (e.g. IBM's db2), Ubuntu may be a better choice as the software vendor has taken steps to ensure compatibility with Ubuntu. | |
| [22:23] <mneptok> This is a good rule of thumb no matter what OS you use, be it Free Software or not. Know the product. Know the vendor's preferences for OS versions. Know your support options. | |
| [22:24] <mneptok> What about buying Ubuntu? As many of you may know, Ubuntu is now being offered for purchase at Best Buy. Why would you pay for what you can download for free? Well, precisely because of those support options. Ubuntu is being offered at Best Buy by a company named Valusoft. Valusoft offers 60 days of *starter* support if you buy their CD at Best Buy. | |
| [22:24] <mneptok> Let's be fair. Valusoft is not going to help you configure a multi-domain mail and web server. :) But they'll help you get things working as best as Ubuntu can on your hardware. They can answer basic questions about installation. Spending that US$20 may be a good idea for someone entirely new to Linux. | |
| [22:25] <mneptok> Which brings me to other types of support. When you choose Free Software, if you need support, make sure you know what your options are. For instance, if you choose to pay for an entitlement to Red Hat Enterprise Linux, you may rest assured that Red Hat will provide you with support. If you choose Fedora or CentOS, you will need to rely upon either the user community or third-party expertise for support. | |
| [22:26] <mneptok> Ubuntu falls somewhere in the middle. Like Fedora or CentOS, use of the software does not cost you. Neither does availaing yourself of the free community resources (e.g. IRC or the Ubuntu Forums). You may also purchase commercial support and consulting services from qualified third parties. Some of them are listed in the Ubuntu Marketplace: | |
| [22:26] <mneptok> http://webapps.ubuntu.com/marketplace/ | |
| [22:26] <mneptok> However, you are also able to purchase support from the corporate entity that helps drive the project, namely Canonical. We have desktop and server contracts, and 9*5 and 24*7 contracts. Our contracts cover 10 cases or 1 year, whichever comes first. And we only guarantee support for packages in the Main repository. Other work is on a "best effort" basis. | |
| [22:27] <mneptok> As an example, I am currently helping a customer sort out issues with the RealPlayer plugin for Firefox. It may be that "it's broken, and I can't help you further," is the final answer. But this would not be the case with the Totem plugin, which is in Main. However, I'm not going to hang up on a paying customer. I won't let him stand between me and customers with problems with supported packages, but I'll make the effort to help ... | |
| [22:27] <mneptok> ... him. | |
| [22:28] <mneptok> Also, Canonical support is not a consulting service. We offer break/fix support. So, "What is Apache and how do I configure it?" is not a support question. "I am running Apache and mod-mneptok reliably crashes the daemon. This is an Apache module you provide by default," will really set off the alarm bells in our office. :) | |
| [22:29] <mneptok> Canonical support contracts are available in the Canoncial web shop at: | |
| [22:29] <mneptok> http://shop.canonical.com | |
| [22:29] <mneptok> WHEW! Didjya get all that? :D | |
| [22:29] <jcastro> time for questions? | |
| [22:29] <mneptok> In conclusion: | |
| [22:29] <mneptok> - Do research. Know what you're buying. Rely upon the experiences of other users to guide your buying decisions. | |
| [22:30] <mneptok> - Do not expect to grab any old printer (or mp3 player or other device) and be guaranteed of a satisfactory experience in Linux. | |
| [22:30] <mneptok> - Know your support options. Know what is supported. | |
| [22:30] <mneptok> And finally, SPEAK UP! :) Tell vendors you use Ubuntu (or Red Hat or Fedora or NetBSD or SuSE). Tell them you make choices based on an OEMs commitment to providing support for Free Software users. Tell them Canonical (or Red Hat or Novell) has ways to help them get certified and working. Tell them it's time to be open! | |
| [22:30] <mneptok> Thanks a lot for your attention! I hope this was somewhat helpful to everyone. Let's open the floor to questions! Jorge! Who's my first victim?! >:) | |
| [22:31] <jcastro> QUESTION: What happened with the recently Broadcom linux drivers? http://www.broadcom.com/support/802.11/linux_sta.php | |
| [22:31] <mneptok> these drivers, like the nVidia drivers, are binary blobs. | |
| [22:32] <jcastro> (just tell me "next" when you want another question) | |
| [22:32] <mneptok> a user of these chipsets will face the same hurdles. you'll need to install Ubuntu, then use the wired connection to run updates and install restricted drivers. | |
| [22:33] <mneptok> if you're OK with that, Broadcom's slightly more open attitude may be palatable to you. | |
| [22:33] <mneptok> next q? | |
| [22:33] <jcastro> QUESTION: Apart from the FSF database are there any other resources to find out which hardware works with Free Software? So not just "works with Ubuntu", which doesn't say anything about blobs or proprietary software. | |
| [22:34] <mneptok> there's the OpenPrinting db for printers, and the USB Device DB for USB. i use these extensively. | |
| [22:34] <mneptok> be aware, the FSF db will not list devices supported by blobs. so that may be a bit misleading as to "what works" and "what works with Free Software." | |
| [22:35] <mneptok> that answer your question? | |
| [22:35] <mneptok> (if so, next!) :) | |
| [22:35] <jcastro> QUESTION: notwithstanding the fact that i love wikis, don't you think perhaps hardware compatibility might deserve some more organized infrastructure? we have a "hardware compatibility report" utility currently in Ubuntu, could that be expanded? If users were able (and encouraged by default) to give their hardware's | |
| [22:35] <jcastro> compatibility a "scoe" and comments, which would then end up on some public DB, hwhich other users could then browse, could that | |
| [22:35] <jcastro> not provide some good community buying advice? | |
| [22:35] <jcastro> long question! | |
| [22:35] <mneptok> that smells like LjL. hence, i will have to put on my Batgirl costume to answer ... | |
| [22:36] <jcastro> heh, it is | |
| [22:36] <LjL> :> | |
| [22:36] <mneptok> i agree, a centralized database would be an excellent thing. | |
| [22:37] <mneptok> i'd like to see such a thing for all Free Software variants. imagine checkboxes [x] Ubuntu || [x] Fedora || [ ] NetBSD | |
| [22:37] <mneptok> the FSF runs something like this, but as i said, "what works" is sorta different than "what works with pure Free Software" | |
| [22:38] <mneptok> i imagine most users want "what works," which is a choice i believe they are rightly entitled to make. | |
| [22:38] <mneptok> so yes, such an idea has a lot of merit, IMO. perhaps we could start such a thing focused on Ubuntu, but expand it to embrace other communities. | |
| [22:39] <mneptok> any more questions? | |
| [22:39] <jcastro> <QUESTION> what about the diference between madwifi and ndiswraper ... | |
| [22:39] <mneptok> MadWifi is a project to reverse engineer drivers for Atheros chipsets. | |
| [22:40] <mneptok> the drivers supplied by the MadWifi project are designed and built for Linux | |
| [22:41] <mneptok> ndiswrapper is not a driver. it's a wrapper, or a layer, that acts between the Windows driver and the Linux kernel. | |
| [22:41] <mneptok> when you use MadWifi drivers you are only using software designed for Linux. when you use ndiswrapper, you use a driver designed for Windows, with what amounts to a traffic cop standing between that driver and the Linux kernel. | |
| [22:42] <mneptok> next q? | |
| [22:42] <jcastro> QUESTION: which languages can I get Canonical support in? Can I call you up in Dutch? | |
| [22:43] <mneptok> currently we officially offer support in English, French, Spanish, and German | |
| [22:43] <mneptok> so yes, you can speak Deutsch. ;P | |
| [22:44] <mneptok> (not Dutch) ;) | |
| [22:44] <jcastro> QUESTION: Is Canonical approaching hardware manufacturers and encourage them to open their? | |
| [22:44] <jcastro> I assume he means "drivers" at the end of that | |
| [22:45] <mneptok> i can't say for sure what we're doing in that regard, because that touches on a part of Canonical's business that i do not work in directly. | |
| [22:45] <mneptok> and if i did, i imagine i couldn;t say anything, anyway. ;) | |
| [22:45] <jcastro> For lists of supported hardware on Ubuntu see https://wiki.ubuntu.com/HardwareSupport - To help debugging and improving hardware detection, see https://wiki.ubuntu.com/DebuggingHardwareDetection | |
| [22:45] <jcastro> Someone asked about a wiki page for hardware support | |
| [22:45] <mneptok> but here's my best assumption, based on what i know of the industry ... | |
| [22:45] <jcastro> so I am putting that in there for reference | |
| [22:46] <mneptok> say Mnep-O-Net makes wireless controllers. | |
| [22:46] <mneptok> and they release only binary drivers for Windows. and those chipsets are in *everything* | |
| [22:46] <mneptok> now, Canonical approaches this company and says, "You need to make Linux drivers available, or open your drivers." | |
| [22:47] <mneptok> as CEO, i'd ask "why? what we're doing now seems to be working just fine." | |
| [22:47] <mneptok> and what could Canonical say to that? | |
| [22:47] <mneptok> (answer: "pretty please? sabdfl will take you on a jet ride. honest!") | |
| [22:48] <mneptok> not compelling (although sabdfl looks great at 20K feet, man) | |
| [22:48] <mneptok> BUT ... | |
| [22:48] <mneptok> say Canonical approaches someone like ... hmmm .... Dell. | |
| [22:49] <mneptok> and then it's Dell that asks, because they want to keep using the same supply chain for Linux models as they do for Windows models. | |
| [22:49] <mneptok> now the Mnep-O-Net CEO asks "why?" and Dell says "because we'd like to keep buying 50 million units per year, if we can. if not, Intel over there has ....." | |
| [22:50] <mneptok> my guess is that this is the approach our team is taking. i have no evidence to base this on, but it seems logical. | |
| [22:50] <mneptok> make sense? | |
| [22:50] <jcastro> QUESTION: Are there any other support options beside telephone (chat, remote desktop, ...)? | |
| [22:51] <mneptok> we handle cases via phone and the web. we have a web support portal that allows customers to create and update cases, and analysts to repsond to them. | |
| [22:51] <mneptok> we do not currently use IRC, IM, or other real time Internet methods. | |
| [22:52] <mneptok> any more questions? | |
| [22:52] <jcastro> time for one more! | |
| [22:52] <mneptok> woohaa! | |
| [22:53] <jcastro> wow, I guess you were so thorough that there are no more questions | |
| [22:54] <mneptok> or people fled when they realized who i was :) | |
| [22:54] <mneptok> well, thanks everyone! | |
| [22:54] <charlie-tca> Good information. Thank you. | |
| [22:54] <samgee> thanks mneptok | |
| [22:54] <mneptok> i'm an IRC junkie. feel free to find me in Ubuntu channels. | |
| [22:55] <mneptok> i may not reply right away (paying customers and all) but i will reply. | |
| [22:55] <jcastro> ok thanks Kurt! | |
| [22:55] <samgee> whoo, free support :) | |
| [22:55] <jcastro> thanks everyone for joining in today | |
| [22:55] <mneptok> thanks for a great OpenWeek Jorge! | |
| === jcastro changed the topic of #ubuntu-classroom to: Current Session: See you at 1500UTC for the next session! | Welcome to Openweek, questions in #ubuntu-classroom-chat please || Session details here: https://wiki.ubuntu.com/UbuntuOpenWeek | |
| [22:55] <jcastro> we have no sessions until 1500 tomorrow | |
| [22:55] <jcastro> but people are more than welcome to hang out if they want | |
| [22:56] <LjL> mneptok: that costume seems to have some sort of hole by the way | |
| [22:56] <LjL> though i didn't want to interrupt you before | |
| [22:56] <jcastro> ls -al | |
| [22:56] <jcastro> oops | |
| [22:56] <mneptok> LjL: "speed holes" | |