Datasets:

NewEden

/

Numberphile-podcast-sharegpt

like 0

Follow

New Eden 26

[ { "from": "human", "prefix": "[ Author:", "value": " Various; Title: Numberphile; Tags: podcast, transcripts, chat ]", "loss": true }, { "from": "gpt", "prefix": "Brady Haran:", "value": " With much of the world currently locked down by the Coronavirus pandemic, mathematical modeling has become a household talking point. (music continues) To discuss this I'm speaking with Kit Yates, he's a mathematical biologist at the University of Bath. In this ever changing climate it's probably also worth noting we recorded this interview on the 30th of March, 2020.", "loss": true }, { "from": "human", "prefix": "Kit Yates:", "value": " By mathematical biologist I think people find that a bit of a strange one to deal with because I think people think math is quite pure and abstract and biology is pretty messy and real world and never the twain shall meet but really what I do day to day is take biological systems that I think are interesting so maybe anything from a swarm of locusts to the way that eggs get their patterning or the way that embryos develop and try to write down a system of equations or computer code which describe that to try to make predictions about those systems.", "loss": true }, { "from": "gpt", "prefix": "Brady Haran:", "value": " Kit you say that people don't often see the link between biological systems and mathematics I feel like that's changed pretty rapidly in the last few weeks. (laughter)", "loss": true }, { "from": "human", "prefix": "Kit Yates:", "value": " Right everyone is talking about exponential growth, about epidemiological models, about modeling and basic reproduction numbers, Boris Johnson was talking about the fast upward tick of the curve and yeah, even Conor McGregor, I saw the other days was talking about implementing half measures to halt exponential growth of the disease in Ireland so yeah everyone is talking about mathematical biology at the moment so it's a good time to be a mathematical biologist.", "loss": true }, { "from": "gpt", "prefix": "Brady Haran:", "value": " Is there a degree of that… I know I'm sure if you had a magic wand you'd make this all go away and what a terrible thing this is for the world but are mathematicians thinking, At Last! People get it.", "loss": true }, { "from": "human", "prefix": "Kit Yates:", "value": " Right, I think that's true I think people there's a good meme going round where it's just a guy standing at chalkboard drawing an exponential curve and students saying well, when are we ever gonna need this, right? and it's sort of the case it sort of feels like a little bit of vindication… although yeah of course ideally you would wish it away if you could do but I think mathematicians are trying to make the most of this opportunity and try and show people why maths really can be important. It can be a matter of life and death sometimes.", "loss": true }, { "from": "gpt", "prefix": "Brady Haran:", "value": " Do you feel like this has showed up sort of a problem with mathematical literacy or do you think people have kind of risen to the occasion and they're getting it, like how have you felt about how well the public are understanding some of these things before we do start talking about them ourselves.", "loss": true }, { "from": "human", "prefix": "Kit Yates:", "value": " Yeah I think people are doing an okay job actually I think the fact that people are actually talking about it and caring about you know how good these models are and asking these questions is a really good sign. I think its a little bit difficult to explain all the really complex models that are out there at the moment. But I think actually the basic mathematical models that underly epidemiological spread are actually not too difficult to understand in fairly straightforward terms and I think its our job as maths communicators to try and make those things as understandable as possible and I think there's a lot of good people out there working really hard to do that.", "loss": true }, { "from": "gpt", "prefix": "Brady Haran:", "value": " Where do we start? What is exponential growth?", "loss": true }, { "from": "human", "prefix": "Kit Yates:", "value": " Right so technically the mathematical definition is when something grows in proportion to its current size then that's exponential growth or equally it could be something decaying in proportion to its current size, that's also exponential growth so for exponential decay you can imagine taking a bag of M&Ms if you want, throw the M&Ms on your kitchen table on day one and eat all the ones that land on the M side upwards right so roughly half the M&Ms will disappear. Put them back in the bag save them for tomorrow, give them another shake up the next day, eat all the ones that have an M on them, and everyday you'll be eating roughly half of the M&Ms so the number of M&Ms is going to be decreasing in proportion the number of M&Ms you have left and therefore you'll be getting an exponential decay of M&Ms over time. So something is decreasing in proportion which means that the time it takes for halve or decrease by certain proportion will be the same no matter how much of that quantity you have. So that's exponential decay at least.", "loss": true }, { "from": "gpt", "prefix": "Brady Haran:", "value": " So Coronavirus, let's deal with that for a second. That's exponential growth.", "loss": true }, { "from": "human", "prefix": "Kit Yates:", "value": " At the moment, yeah, like so the way the simplest mathematical epidemiological models work is you break the population down into three different compartments. They'r called susceptible, so people who haven't had the disease. Infecteds are people who have the disease and can infect it to other people and then we have this removed category it's sort of a euphemistic term for people who have recovered but also people that have died.", "loss": true }, { "from": "gpt", "prefix": "Brady Haran:", "value": " Right this is the Sir model we hear talked about.", "loss": true }, { "from": "human", "prefix": "Kit Yates:", "value": " Right", "loss": true }, { "from": "gpt", "prefix": "Brady Haran:", "value": " S I R", "loss": true }, { "from": "human", "prefix": "Kit Yates:", "value": " We call it S I R but Sir, whatever you want to call it really.", "loss": true }, { "from": "gpt", "prefix": "Brady Haran:", "value": " Yeah", "loss": true }, { "from": "human", "prefix": "Kit Yates:", "value": " We at the very early stages you have a whole bunch of people who are susceptible to the disease and very few people who are infected and so each individual can go and infect a certain number of people who are susceptible, we call this the basic reproduction number of the disease. and if that number of people that they infect over the course of their infection period is greater than one then they will on average infect that many people who will then go infect that many people again so for COVID at the moment estimates of the basic reproduction number are between one and a half and four so it's quite a big window. The two and a half is generally sort of acceptable number, so the first people goes and infects two and half of the people, they're gonna infect two and half more and two and a half more and the exponential growth then occurs. You start to see the numbers growing exponentially.", "loss": true }, { "from": "gpt", "prefix": "Brady Haran:", "value": " The way people seem to be getting this across is… or the nature of it seems to be it starts off as a slow trickle and then suddenly becomes like this huge waterfall.", "loss": true }, { "from": "human", "prefix": "Kit Yates:", "value": " This is the thing about exponential growth, right, everyone… I think people confuse it with being big or fast all the time but actually if you think about the amount of money in your bank account or certainly my bank account, like it's growing exponentially you're getting interest proportional to how much money you've got. But the rate of interest is really low so you don't really see that exponential growth properly, it's just really really slow. If you left it long enough then great you'd get loads of money but who can leave their money in the bank account long enough. And that's sort of what you see the very start of the epidemic. You hardly notice these cases trickling in and then all of a sudden it seems to just absolutely take off when the number starts to really grow and that's the sort of danger of exponential growth is that if you don't recognize it early on then you can think, Oh this isn't a big problem, and actually you really need to be acting on it and taking steps which is what we're doing at the moment.", "loss": true }, { "from": "gpt", "prefix": "Brady Haran:", "value": " It's these grains of rice on a chess board is the famous story, isn't it?", "loss": true }, { "from": "human", "prefix": "Kit Yates:", "value": " Right, exactly, so uh… yeah the Emperor of China rewards this guy for saving his life and he comes along and says, you can have anything you want, and the guy this farmer says I'll just have you know one grain of rice for the first square on the chess board and two for the next and four for the next and if you could just complete that pattern that would be great and the King… the Emperor is like why yeah whatever that doesn't sound like very much you could have anything you wanted (laughter) and he chose this stupid amount of rice and then of course it's doubling every time you put rice on the next square it's more than rice than you've already put on the whole of the rest of the chess board so it turns out something like you'll have rice down to the knees or something across the whole surface of the Earth in the end. So yeah it's surprising, like counter intuitive if you like.", "loss": true }, { "from": "gpt", "prefix": "Brady Haran:", "value": " Do you think the public has gotten that now or do you think we got it too late or like just as an observer. I know you're not like a politician or like an expert but it seems like you know, are people understanding this?", "loss": true }, { "from": "human", "prefix": "Kit Yates:", "value": " Yeah I think the message is getting across to people I people are starting to realize as they are still seeing the numbers rise and rise, day on day we're still in this exponential growth phase so deaths are doubling every two to three days, cases are doubling every three to four days, so a little bit slower. But I think people are starting to get alarmed by the fact yeah we're having two hundred deaths a day and maybe more. And that's worrying them and they're seeing the numbers go up but not just go up in a linear fashion but actually sort of doubling every two or three days and I think people are starting to get the point that actually because exponential growth can be so fast anything that we can do early on can make a tremendous difference later on like we can really change the numbers later on if we can just make small changes even now. And there was this nice infographic that went around that showed that if one person infected two and a half people in five days then after thirty days they would have infected four hundred and six people, and actually if you can reduce that by fifty percent it changes from four hundred and six to fifteen people. And if you can reduce it even further by seventy-five percent which is there some of evidence that that's sort of what we're doing with our social distancing, you can reduce that number of total infected people to about two and a half including the first person who was infected. So you can really wipe out this epidemic if you can reduce this spread enough.", "loss": true }, { "from": "gpt", "prefix": "Brady Haran:", "value": " So, it's about getting that two point five number to something closer to your bank interest.", "loss": true }, { "from": "human", "prefix": "Kit Yates:", "value": " (laughter) right exactly if you can reduce I mean even your bank interest would be naught point you know half a percent per year right, even if we can reduce COVID spread to less than a hundred percent during the course of their infectious period then then you'll start to get this exponential decay, if you can get just go below replicating itself so just below zero percent then you'll start to see this exponential decline. But yeah, if we can bring it down to the rate of bank interest then we'll start to see very very slower exponential growth.", "loss": true }, { "from": "gpt", "prefix": "Brady Haran:", "value": " What kind of jobs are mathematicians doing at the moment? Like I can see the job of doctors and I can see the job of researchers trying to come up with vaccines. These graphs and models you're talking about have been known long before COVID came along and obviously they're being applied now. But are there day to day things mathematicians can be doing, like new work they can be doing to help in this battle?", "loss": true }, { "from": "human", "prefix": "Kit Yates:", "value": " Yeah absolutely. 'Cause every disease is different, right? And these simple models that I've sort of described, this S I R model, they can be made infinitely more complex. Right, so S I R is fine for some very simple diseases but actually there's extensions we can make to those S I R models to include things like a carrier class, so people who have the disease but aren't necessarily symptoms. So they're not necessarily you wouldn't class them as infected because you don't know their infected yet, but they've had this asymptomatic period, which is what is happening with COVID. And so they can be spreading it without you even knowing that they're infected and then can be a real problem and it has been a real problem with COVID, because we can't just isolate people as soon as they show symptoms we have to be isolating everyone in case they've got symptoms. So that's just one really simple extension you can make but adding in things like noise, so stochasticity or randomness in the way that people bump into each other. Taking account of the networks with which interact with each other can make a huge difference as well. You get these people are social hubs and therefore will spread the disease to lots of people compare to people who are stuck at home and don't go out so much. So you can make these arbitrarily complex and there's actually been a call by the Royal Society for Rapid Assistance in Modeling Pandemics where all mathematicians, especially applied mathematicians who have a familiarity with modeling can sign up and help out in this cause, so yeah there's to loads to be done at them moment and we're keeping really busy.", "loss": true }, { "from": "gpt", "prefix": "Brady Haran:", "value": " Kit, how does someone build in something like these you know hubs super spreader type as opposed to people who are locked away. Like I see that they exist but how would you put them into a model? These sort of unmeasurables or like I see could guess at what they are but then how do you?", "loss": true }, { "from": "human", "prefix": "Kit Yates:", "value": " So they're not necessarily unmeasurables, we actually know a lot about our social structure in part from using social networking data but also there's been a number of studies done and you could actually classify the way that people interact using a network. So people are nodes in that network and they're connections to other people are edges. So you can classify what's called the degree distribution of that network which tells you how many people are there who have a hundred contacts, how many people with ninety-nine, ninety-eight and so and down to how many people there with no contacts. And that distribution can tell you a great deal about the way that people interact so you can incorporate that network modeling into these S I R models you start the disease off in a particular node with a particularly connectivity. How many people do they actually interact with and you see how it spreads around the network. So, they may seem almost intangible but actually if we have good enough data and this is what it all comes down to really. If we have good enough data then we can actually capture these behaviors. But data at the moment on this emerging pandemic is really difficult to come by.", "loss": true }, { "from": "gpt", "prefix": "Brady Haran:", "value": " You talked about randomness and I know that's a real expertise of yours, can you give me an idea how randomness would feed into for example the coronavirus COVID situation. What are some of the random factors and how can mathematicians use them?", "loss": true }, { "from": "human", "prefix": "Kit Yates:", "value": " Well so it's sort of interesting that actually in the what we call the deterministic models. So the models that don't have noise in them. If your basic reproduction number is above one then your epidemic definitely takes off. Even if its only a tiny amount above one. And if its below one then it definitely dies off. But actually in models which build in noise or stochasticity you can tolerate being slightly above that threshold of R zero because in reality not everyone has the same number of contacts with everyone. There's noise and not every contact with someone else passes the disease on. So actually in reality you can actually have a non-zero probability of the disease dying out without having a big epidemic, even if you have this basic reproduction number being above zero in the deterministic model. Whereas yeah in the deterministic model it says it definitely takes off and then it dies out, but when you build in noise you can actually realize we can tolerate a little bit of spread above this basic reproduction number of one.", "loss": true }, { "from": "gpt", "prefix": "Brady Haran:", "value": " One of the things I've found interesting has been the use of graphs where the Y axis has been logarithmic I think, so they take away that dramatic curve and everything becomes straight lines. And that's been used a lot comparing the spread in different countries. But there's something about those graphs that I somehow find deceptive. Like I know you're a mathematician and you can straight away process but what do you think of the use of those graphs?", "loss": true }, { "from": "human", "prefix": "Kit Yates:", "value": " I don't think they're super helpful, actually. I was actually reading the comments below the line which you should obviously never do on the BBC article the other day which showed these log or what are called semi-log plots, so you have log axis so that you can really see how things are changing, the problem with having with a linear axis you don't really see what was happening early, it just looks like zero cases then all of a sudden it blows. Whereas the log you get this nice sort of straight line and it tells you what the exponent is. How quickly things are growing. Which is useful if you're studying the disease. But actually the comments below the line people like saying well there's clearly some mistake in this graph because the space between ten and a hundred is the same as the space between a hundred and a thousand, and that's the point of a log plot, but it's not actually that helpful unless you know something logarithms. So I, for me, I would steer away from those a little bit. Yeah.", "loss": true }, { "from": "gpt", "prefix": "Brady Haran:", "value": " I mean I'm no expert but I think I agree with you on that. There's something almost like it makes the growth so steady and calm when it's a straight line, you think, Oh well you know if we do something now it will be all right, but like it doesn't show how it's running away from you.", "loss": true }, { "from": "human", "prefix": "Kit Yates:", "value": " No and you compare the different number of countries cases and when you see that in a log plot you're like, oh, America's not very far ahead of us, I thought they were doing really badly, and actually like they are doing pretty, they've got a lot of more cases than us on an exponential, you know on sort of a linear scale it does look pretty bad. But on the log plot it doesn't really look like they're that far ahead of us but yeah it can be deceptive so I would steer away from using the log plots a bit I think.", "loss": true }, { "from": "gpt", "prefix": "Brady Haran:", "value": " As a mathematician but also as a human who you know who doesn't want Coronavirus to be terrible, what are some of the numbers and graphs and things that you're looking at most closely and that you're really interesting in and that you're really hoping change or get dealt with?", "loss": true }, { "from": "human", "prefix": "Kit Yates:", "value": " Right, so since we've had lockdown for about a week in the UK at the moment, and actually it's sort of too soon to tell what impact that's having, but even today we've been seeing the number of hospitalizations has gone down slightly, the daily hospitalizations that is. And similarly the number of cases that have been reported is going down. You can't read too much into that because actually not everyone is being tested. People are being urged to stay at home if they have mild symptoms, which is really sensible. And so… but we're testing everyone, so you can't read too much into it. But it's hinting that some of the early measure that we took like, telling people to work from home and banning gatherings are having an impact. It's a bit too early the impact of the lockdown yet and certainly like deaths is the figure that you really want to look at because deaths you know report all the deaths you don't miss any of those cases, right? But actually we won't really see the impact of the lockdown on deaths for probably a couple of weeks because there is this long period that you're ill for before you get seriously ill and then potentially go on and die. So but those are the really important figures is looking at case numbers and looking at numbers of deaths and once we start to control and bring those down then we know that the impact of the effects of the lockdown and the measures that we've been taking to control the disease.", "loss": true }, { "from": "gpt", "prefix": "Brady Haran:", "value": " I mean what numbers can we give mathematicians other than positive tests and deaths?", "loss": true }, { "from": "human", "prefix": "Kit Yates:", "value": " Yeah that's the tricky thing at the moment. There's a group led by a guy called John Edwards who are looking at other measures, so encouraging people to log how they are interacting with each other. So trying to figure out what the basic reproduction number would be if they were just talking to each other.", "loss": true }, { "from": "gpt", "prefix": "Brady Haran:", "value": " Of course another piece of data is people going into intensive care I guess and things of that and ventilation and stuff but…", "loss": true }, { "from": "human", "prefix": "Kit Yates:", "value": " yeah, so hospitalization data is something that's useful cause you don't go to hospital unless you're really serious and we can measure those numbers but it's not part of the data that's officially released by Public Health England everyday but it can give us an indication as it already is doing because the numbers are coming down a little bit per day that the effect the measures that we've been taking are generally having some effect at least.", "loss": true }, { "from": "gpt", "prefix": "Brady Haran:", "value": " So as mathematicians produce different models and things like that that they can then give to policy makers, what are the levers the policy makers have? Like is this just level of how draconian we're stopped from bumping into each other? Like what other levers can policy makers pull?", "loss": true }, { "from": "human", "prefix": "Kit Yates:", "value": " Yeah, I think unfortunately that's it basically. It can only be how much more lockdown we have at the moment. One thing that seems to be really important is getting this antibody test out to find out who's had the disease. So there was this talk early on from Boris Johnson about herd immunity. And so this is a mathematical idea really where if you have enough people in the population who are… who have had the disease and then are immune to it, then you can effectively wipe out the spread of that disease to the people who are more vulnerable and more at danger if they get the disease. And so actually there's a relationship directly between the basic reproduction number of the disease so how many infections one single infected person will spread to in the course of their infectious period and this ratio of how many people you have to have in order to get herd immunity and it's just one minus one divided by the basic reproduction number. So if the basic reproduction number is four then you have to have one minus a quarter or three quarters of the population immune to the disease before you effectively wipe it out. Ideally we would develop a vaccine so we could just get lots of people immune from the disease by vaccinating them but actually the alternative way to get herd immunity is by having disease. Obviously if every gets it then that's going to be awful but if we could get healthier demographics, younger demographics to have the disease, enough people to have it very mildly, that we could have that fraction of the population immune then potentially that's a strategy for taking the disease out. But in terms of what the government can actually do without putting huge numbers of life at risk, it's locking down further. Stopping people going out for exercise, putting road blocks in cities, stopping people from going on long journeys. So, yeah there's not much further we can go I guess but we could still go further. And unfortunately those are the only measures that are open to us without a vaccine at the moment.", "loss": true }, { "from": "gpt", "prefix": "Brady Haran:", "value": " This herd immunity idea was quite controversial at the time though isn't it? 'Cause there is there sort of feeling of lambs to the slaughter, let's deliberately let our people get it for the greater good. And I sometimes have felt this way myself that sometimes when we talk about it all so mathematically we kinda forget that these are all human beings who are like suffering as a result. As a mathematician how do you feel about that because I see the pleasure you get from the model and the mathematics of it but at the same time there's this very human element to it?", "loss": true }, { "from": "human", "prefix": "Kit Yates:", "value": " Yeah I think you're right, I think it was hugely controversial and it still hugely controversial and I'm not advocating this strategy of herd immunity that is one of a good way to go in this situation. Yeah I think that's partly why I decided to be a mathematical biologist in the first place because I thought I got to uni doing maths and I thought when's this going to be useful to people, right? And then I realized that actually you could use maths to describe biology and for me that was amazing because from a very early age I decided wanted to be a scientist. My mum died from cervical cancer and remember kids laughing at me at school when I was ten years old because I was like, I want to be a scientist, and they were like, I want to be footballer, I was like okay right. But yeah so like when I got to uni and did maths and found I could use maths to describe the real world and describe biology that was massive for me because I thought well I can actually have an impact on real people life and make a difference. And I think that's at the heart of all the modeling that we're trying to do is not to forget that actually these models describe real people. It's super important for us not to forget that.", "loss": true }, { "from": "gpt", "prefix": "Brady Haran:", "value": " I mean it's seems pretty obvious that this pandemic's gonna leave a scar on the world and have some kind long lasting effect. What about mathematically? Do you think it's going to have a mathematical legacy? Is it gonna like, cause I've never seen mathematics talked about so much with a global news event.", "loss": true }, { "from": "human", "prefix": "Kit Yates:", "value": " I think I hope it will draw people into mathematics and make people realize that you know if you become a mathematician you can have a real impact on the world, you're not just sat behind a desk writing a textbook for a subject which is dead and gone. You're actually living in the real world and you're doing things which can have a real important impact on people's lives so I hope that people, future generations of kids who are studying maths at school will say wow this is actually this living breathing exciting subject and I want to be part of that. And so I hope we'll get more people into mathematics and I also hope that people will take maths a little bit more seriously themselves. So maths generally only gets in the news when its the Fields Medalist times and you know people struggle to understand the very complicated maths that the Fields Medalists who have won the maths bowl but actually the moment we can tell people about relatively straight forward mathematical models which are having a huge impact on their everyday lives and suggesting how we have come to this unprecedented lockdown that we've never seen before. So yeah hopefully it will have a really big impact on people's lives.", "loss": true }, { "from": "gpt", "prefix": "Brady Haran:", "value": " Kit is there a cutting edge here though? Is there new mathematics to be discovered and found here? Or is this basically pretty relatively simple mathematics and statistics just being applied in clever and new ways with more data and more tweaks? Is there like a discovery or new types of mathematics that a genius could come along and change things?", "loss": true }, { "from": "human", "prefix": "Kit Yates:", "value": " I mean I have to say I'm a very sort of luddite mathematician. I like using tools that are already there. So I'm not working at the cutting edge of developing new mathematical tools. I like to know that someone knows that there's existence and uniqueness for certain problems but it doesn't have to me. So I'm probably the wrong person to ask that question to but…", "loss": true }, { "from": "gpt", "prefix": "Brady Haran:", "value": " Yeah.", "loss": true }, { "from": "human", "prefix": "Kit Yates:", "value": " But as far as I'm aware it's people applying tools in a new way. So it's tools that already exist, mathematical models that we've been developing for a number of years but actually applying them to new situations. Trying to get new insights from that.", "loss": true }, { "from": "gpt", "prefix": "Brady Haran:", "value": " Can you just lastly just tell me a little bit more about this Royal Society project? What is this? Is this like all hands to the pump type thing?", "loss": true }, { "from": "human", "prefix": "Kit Yates:", "value": " Right exactly, it's sort of like yeah… in Lord of the Rings when Aragorn runs in and says, you know, Gondor calls for aid, and everyone stands up says this is amazing. It's this call from the Royal Society, it came out over the weekend and they are asking people, mainly at Universities. Researching types of Universities to put proposals forward with their group and to volunteer amount of time can and what expertise they have including PhD students, postdocs and tenured faculty members to try and put forward their expertise and to offer it. And to hopefully try organize it. There's a lot of sort of individual modeling efforts people going off and doing their own thing. But actually we can work much better as a group as long as someone organizes us effectively and that's what the Royal Society are trying to do. So our research group, The Center for Mathematical Biology in Bath are putting together our proposal which we are going to send off tomorrow and we're going to contribute as much of our time as we can to help fight this.", "loss": true }, { "from": "gpt", "prefix": "Brady Haran:", "value": " And I know it's just a proposal but can you just so I understand how you can even help, give me some vague ideas of to what that proposal looks like? What you propose?", "loss": true }, { "from": "human", "prefix": "Kit Yates:", "value": " Yeah, so, yeah they're asking for ideas of what are your background expertise is and how you think that might be useful. So for me I have expertise in doing stochastic models, random models of multi-particle systems, multi-agent systems, so they're often called individual based models. And these can be really useful for understanding epidemic spread potentially at a very fine grained level. So looking literal interactions between people rather than thinking of everyone as this huge population in this S I R model. So that's the sort of thing that I'll be putting down. But we have other people in Bath who are traditional epidemiologists who have expertise in things like Household Models where you breakdown populations into households and you look at the effect of the different sizes of households and how people move between their local neighborhoods. There's all sorts of things that we're trying to do but that's the idea of the response in Bath at the moment.", "loss": true }, { "from": "gpt", "prefix": "Brady Haran:", "value": " If someone was listening to this who wasn't a mathematician and is sort of following what's going on with the degree of fear and bewilderment but also interest, what do you want them to understand? What would you say to them if you could say anything to them?", "loss": true }, { "from": "human", "prefix": "Kit Yates:", "value": " Right I think that if exponential growth teaches us anything about particularly about this disease spread, it's that the earlier you make your interventions and the tighter you do them, the more that you adhere to these rules that have been put in place, the bigger that you'll see this impact having within a few weeks, within even a few days it'll make a huge difference so now is the time to be so to speak doubling down on our efforts to lockdown so that we don't double up our number of cases if you'd like. So yeah really taking that exponential growth message to heart.", "loss": true }, { "from": "gpt", "prefix": "Brady Haran:", "value": " Thank you for your time and hopefully you get involved in more work to help us all out.", "loss": true }, { "from": "human", "prefix": "Kit Yates:", "value": " Great, thanks Brady it's been a pleasure.", "loss": true } ]

[ { "from": "human", "prefix": "Summary:", "value": " Math celebrity Hannah Fry tells us about her life as public intellectual and thought leader - plus her as-yet unreleased novel 'Free as a Bird', written as a schoolgirl.", "loss": true }, { "from": "gpt", "prefix": "Hannah Fry:", "value": " In America they call it a public intellectual, but I'm not using that because it's ridiculous.", "loss": true }, { "from": "human", "prefix": "Brady Haran:", "value": " (chuckles)", "loss": true }, { "from": "gpt", "prefix": "Hannah Fry:", "value": " The other one they do in America is they invite you to conferences and they're like, we just really wanted you as thought leader. (laughs)", "loss": true }, { "from": "human", "prefix": "Brady Haran:", "value": " A thought leader! We're reaching out to you as a thought leader. (laughs)", "loss": true }, { "from": "gpt", "prefix": "Hannah Fry:", "value": " What does that mean! (laughs)", "loss": true }, { "from": "human", "prefix": "Brady Haran:", "value": " Hi I'm Brady Haran, today's guest on the Numberphile podcast is the public intellectual and thought leader Hannah Fry. (music continues) Hannah's a mathematician at University College London, or UCL, and we'll be talking about her work shortly but in recent years Hannah's also become something of a celebrity. She's written books, fronted BBC TV shows, she makes radio programs and podcasts and occasionally when time permits even pops up in the odd Numberphile video. (music continues) So as you can guess there's a lot I wanted to ask her about.", "loss": true }, { "from": "gpt", "prefix": "Brady Haran:", "value": " Let's start close to the beginning. What were you like as a little girl? Like would I have known, ah she's gonna end up being, a mathematician?", "loss": true }, { "from": "human", "prefix": "Hannah Fry:", "value": " Hmm. Not early on, no, I don't think so. Not like as a kid. When I was about eleven I think that's when it sort of started to change really and I pinpoint it to one particular event which is (laughs) my Irish mother has a slightly skewed idea of what constitutes a fun summer holiday. So one summer she bought me a maths text book and insisted that I go through a single page in this text book every day before I was allowed to have any fun. (laughs) Which you can imagine my delight as an eleven year old.", "loss": true }, { "from": "gpt", "prefix": "Brady Haran:", "value": " Like to give you education, like to be wholesome or thinking that little Hannah would find this fun?", "loss": true }, { "from": "human", "prefix": "Hannah Fry:", "value": " Uh… I'm not sure really. She's just like really really into education. She's always has been. So, you know, her and her family really didn't get very much education. My dad left school when he was really young and I think that she was really just… had this totally inbuilt idea that education is the way that you kind of get yourself anywhere. So she was just very very strict about education the whole way through. So it's just kind of continuing that idea. But I think what happened though with that sort of summer is that then when I went back to school afterwards, actually I must've been ten 'cause I was still in primary school, so I must've been ten. Then when we were doing stuff in maths class I just had seen it all before really. I knew what was coming and I just totally got it. And at that point then it just becomes much more of a playground, right? When you're ahead of the class it's just way more fun and I think that, yeah, that's when I started playing with it a lot more.", "loss": true }, { "from": "gpt", "prefix": "Brady Haran:", "value": " So it's not like when you were doing those textbook exercises on holiday, you were enjoying that, but it gave you this skill later on that put you ahead of the curve which then allowed you to look at mathematics in a more fun way later, it's like set you up?", "loss": true }, { "from": "human", "prefix": "Hannah Fry:", "value": " Actually maybe I'm being a bit naive but I think everyone quite likes doing problems when they get them right, don't they? Everyone quite likes it when you get a puzzle correct. So it was only that first bit that's horrible and then when you're sort of in the exercises in the back of the textbook, the first couple are horrible 'cause you don't know what's happening but once you get it you're like, yeah, sailing. It's not like I was kind of getting up in the morning and self motivating myself to sit down and do some of this textbook. That I had someone standing over me… (laughs) so, yeah.", "loss": true }, { "from": "gpt", "prefix": "Brady Haran:", "value": " 'Cause a lot of the mathematicians I speak to when they talk about their childhood like confess to being a nerd, they're like, yeah I was bit of nerd.", "loss": true }, { "from": "human", "prefix": "Hannah Fry:", "value": " (laughs)", "loss": true }, { "from": "gpt", "prefix": "Brady Haran:", "value": " Were you a nerd?", "loss": true }, { "from": "human", "prefix": "Hannah Fry:", "value": " Yes! (laughs)", "loss": true }, { "from": "gpt", "prefix": "Brady Haran:", "value": " You were?", "loss": true }, { "from": "human", "prefix": "Hannah Fry:", "value": " Yeah. Totally wholeheartedly a complete nerd. I mean though what else can you say? (laughs) I was…", "loss": true }, { "from": "gpt", "prefix": "Brady Haran:", "value": " What…", "loss": true }, { "from": "human", "prefix": "Hannah Fry:", "value": " You picture a nerd.", "loss": true }, { "from": "gpt", "prefix": "Brady Haran:", "value": " Yeah?", "loss": true }, { "from": "human", "prefix": "Hannah Fry:", "value": " And that was me.", "loss": true }, { "from": "gpt", "prefix": "Brady Haran:", "value": " Like give me an example of something nerdy you would do besides being good at mathematics?", "loss": true }, { "from": "human", "prefix": "Hannah Fry:", "value": " (laughs) I took extra A level maths modules just because I wanted to.", "loss": true }, { "from": "gpt", "prefix": "Brady Haran:", "value": " Right?", "loss": true }, { "from": "human", "prefix": "Hannah Fry:", "value": " That's quite nerdy.", "loss": true }, { "from": "gpt", "prefix": "Brady Haran:", "value": " And was that because you were like wanting to be an alpha overachiever or because you just loved mathematics?", "loss": true }, { "from": "human", "prefix": "Hannah Fry:", "value": " Probably both. (laughs)", "loss": true }, { "from": "gpt", "prefix": "Brady Haran:", "value": " Right.", "loss": true }, { "from": "human", "prefix": "Hannah Fry:", "value": " Like I just once wrote a book in the summer.", "loss": true }, { "from": "gpt", "prefix": "Brady Haran:", "value": " What was it about?", "loss": true }, { "from": "human", "prefix": "Hannah Fry:", "value": " It (laughs) was so (laughs) so sh-t. (censor beep)", "loss": true }, { "from": "gpt", "prefix": "Brady Haran:", "value": " (laughs)", "loss": true }, { "from": "human", "prefix": "Hannah Fry:", "value": " Am I allowed to swear on this podcast?", "loss": true }, { "from": "gpt", "prefix": "Brady Haran:", "value": " I don't know.", "loss": true }, { "from": "human", "prefix": "Hannah Fry:", "value": " I just did.", "loss": true }, { "from": "gpt", "prefix": "Brady Haran:", "value": " You just did, alright.", "loss": true }, { "from": "human", "prefix": "Hannah Fry:", "value": " (laughs) It was an awful…", "loss": true }, { "from": "gpt", "prefix": "Brady Haran:", "value": " Yeah?", "loss": true }, { "from": "human", "prefix": "Hannah Fry:", "value": " (laughs) I can't believe I'm telling you this.", "loss": true }, { "from": "gpt", "prefix": "Brady Haran:", "value": " What was it about? Like what was…?", "loss": true }, { "from": "human", "prefix": "Hannah Fry:", "value": " (laughs) It's was called… it was about a guy who has a car crash…", "loss": true }, { "from": "gpt", "prefix": "Brady Haran:", "value": " Hmm.", "loss": true }, { "from": "human", "prefix": "Hannah Fry:", "value": " …and loses the ability to walk. And then learns to walk again after it makes friends with a bird in the garden.", "loss": true }, { "from": "gpt", "prefix": "Brady Haran:", "value": " How old were you at this point?", "loss": true }, { "from": "human", "prefix": "Hannah Fry:", "value": " Like maybe eleven or twelve?", "loss": true }, { "from": "gpt", "prefix": "Brady Haran:", "value": " Okay. Was it the bird that made him able to walk?", "loss": true }, { "from": "human", "prefix": "Hannah Fry:", "value": " Yeah!", "loss": true }, { "from": "gpt", "prefix": "Brady Haran:", "value": " Did it have magic powers or…?", "loss": true }, { "from": "human", "prefix": "Hannah Fry:", "value": " No! (laughs)", "loss": true }, { "from": "gpt", "prefix": "Brady Haran:", "value": " It wasn't super…", "loss": true }, { "from": "human", "prefix": "Hannah Fry:", "value": " I think it was like encouraging him to walk, so it would like sit on his shoulder and then like kind of (laughs)…", "loss": true }, { "from": "gpt", "prefix": "Brady Haran:", "value": " Alright. It could talk?", "loss": true }, { "from": "human", "prefix": "Hannah Fry:", "value": " No, no, but he was like it would be friendly to him. Try to get it to sort of sit on his hand and then the bird would just be slightly too far away so he's have to kind of reach for it.", "loss": true }, { "from": "gpt", "prefix": "Brady Haran:", "value": " What was hero's name? Do you remember?", "loss": true }, { "from": "human", "prefix": "Hannah Fry:", "value": " No. I can't remember. (laughs)", "loss": true }, { "from": "gpt", "prefix": "Brady Haran:", "value": " You really don't remember?", "loss": true }, { "from": "human", "prefix": "Hannah Fry:", "value": " No, I can't but my mum found this book when she was going through the attic a little while ago and she wrapped it up and gave it to me for me thirtieth birthday. So I've got it at home.", "loss": true }, { "from": "gpt", "prefix": "Brady Haran:", "value": " It still exists?!", "loss": true }, { "from": "human", "prefix": "Hannah Fry:", "value": " Yeah, yeah. And I even got a friend of mine (laughs) he was also like eleven, to illustrate it for me. So it's got like an illustrated cover and everything.", "loss": true }, { "from": "gpt", "prefix": "Brady Haran:", "value": " I so want have to see this book.", "loss": true }, { "from": "human", "prefix": "Hannah Fry:", "value": " (laughs)", "loss": true }, { "from": "gpt", "prefix": "Brady Haran:", "value": " What was it's name?", "loss": true }, { "from": "human", "prefix": "Hannah Fry:", "value": " It was called Free as a Bird.", "loss": true }, { "from": "gpt", "prefix": "Brady Haran:", "value": " Oh!", "loss": true }, { "from": "human", "prefix": "Hannah Fry:", "value": " It's so sh-t. (censor beep)", "loss": true }, { "from": "gpt", "prefix": "Brady Haran:", "value": " (laughs)", "loss": true }, { "from": "human", "prefix": "Hannah Fry:", "value": " (laughs) You know when you think you're like this real literary genius when you're like ten?", "loss": true }, { "from": "gpt", "prefix": "Brady Haran:", "value": " No, I'm impressed by that. I'm impressed by that. So you were hundred percent like science and math then? You were showing bit of a humanities bent at this point?", "loss": true }, { "from": "human", "prefix": "Hannah Fry:", "value": " No, I think I was just a nerd, mate. (laughs)", "loss": true }, { "from": "gpt", "prefix": "Brady Haran:", "value": " Alright.", "loss": true }, { "from": "human", "prefix": "Hannah Fry:", "value": " I think I was just a nerd.", "loss": true }, { "from": "gpt", "prefix": "Brady Haran:", "value": " At what point did it become evident that you were actually gonna be like a professional mathematician, like, this was actually gonna end up being my job?", "loss": true }, { "from": "human", "prefix": "Hannah Fry:", "value": " Oh I don't think that happened until it actually happened, really.", "loss": true }, { "from": "gpt", "prefix": "Brady Haran:", "value": " Right?", "loss": true }, { "from": "human", "prefix": "Hannah Fry:", "value": " I mean what I really wanted to do was work in Formula One, that was like my single minded ambition.", "loss": true }, { "from": "gpt", "prefix": "Brady Haran:", "value": " This is motor racing.", "loss": true }, { "from": "human", "prefix": "Hannah Fry:", "value": " Motor racing, exactly, that was I really wanted to work at. 'Cause my dad used to race motorbikes professionally.", "loss": true }, { "from": "gpt", "prefix": "Brady Haran:", "value": " Hmm.", "loss": true }, { "from": "human", "prefix": "Hannah Fry:", "value": " So like our family's just very into kind of racing and I just massively massively loved Formula One, I just wanted a way to work in it. And engineering just wasn't really on the table. I don't know… it just never occurred to me that that was a kind of career option. I went to all girls school and just no one ever spoke about it really. So, because I was good at maths I went and did maths at university. Although actually what I wanted to do when I was sixteen, although I was a massive nerd, what I really wanted to do was be a hairdresser.", "loss": true }, { "from": "gpt", "prefix": "Brady Haran:", "value": " Right?", "loss": true }, { "from": "human", "prefix": "Hannah Fry:", "value": " And my mum was like, okay (clears throat) what we'll do, you just do your A levels and then we'll talk about this again. So I did maths, physics and chemistry A levels. And then after I did my A levels I was like, you know I still would quite like to do a hairdressing course. And my mum was like, tell you what, just go off to university and do an undergraduate in maths or theoretical physics and then after that (laughs)…", "loss": true }, { "from": "gpt", "prefix": "Brady Haran:", "value": " Then we'll talk about hairdressing. (laughs)", "loss": true }, { "from": "human", "prefix": "Hannah Fry:", "value": " (laughs) We'll talk about it. And then, yeah, by the time I'd finished my PhD I was like, probably not gonna be a hairdresser.", "loss": true }, { "from": "gpt", "prefix": "Brady Haran:", "value": " Why do you think you wanted to be a hairdresser?", "loss": true }, { "from": "human", "prefix": "Hannah Fry:", "value": " It's just cool! Like, have you ever been and looked like watched what a hairdresser does? It's cool.", "loss": true }, { "from": "gpt", "prefix": "Brady Haran:", "value": " They're just like… cut and do things to people's hair.", "loss": true }, { "from": "human", "prefix": "Hannah Fry:", "value": " Yeah, I know, but they do like magic stuff, they'll like make it all… (laughs)", "loss": true }, { "from": "gpt", "prefix": "Brady Haran:", "value": " So it was the process, like you liked the idea of doing the processes that they do? The craft?", "loss": true }, { "from": "human", "prefix": "Hannah Fry:", "value": " Yeah, the craft of it and also but it was (laughs) I think, there's something quite nice about, you know, having a process that's quite short where you get to see the end result very quickly. It's kind of the reason why I liked doing Youtube videos because you have an idea, you execute the idea and then it's sort of out there, it's not like this really long thing that lasts for ages and ages before you know whether it's a good thing or a bad thing.", "loss": true }, { "from": "gpt", "prefix": "Brady Haran:", "value": " Short turnaround time?", "loss": true }, { "from": "human", "prefix": "Hannah Fry:", "value": " Yeah!", "loss": true }, { "from": "gpt", "prefix": "Brady Haran:", "value": " So when you started studying mathematics properly at university like did it take quickly? Was it like, yeah this is still holding my… well obviously it was 'cause it became your job but like you started to realize this was gonna be your job now? Like, you're all in?", "loss": true }, { "from": "human", "prefix": "Hannah Fry:", "value": " Yeah, so, I just… just always loved it. It's just so delicious and so… I mean I've always just played with it and massively massively enjoyed it. But I did actually make my dream, I did manage to get to Formula One. So after I finished my PhD, which was in aerodynamics, I was kind of incrementally getting closer and closer to working in Formula One.", "loss": true }, { "from": "gpt", "prefix": "Brady Haran:", "value": " So this was still… this was a plan? Like you were aiming for Formula One?", "loss": true }, { "from": "human", "prefix": "Hannah Fry:", "value": " Yeah.", "loss": true }, { "from": "gpt", "prefix": "Brady Haran:", "value": " You were like making choices that would get you closer?", "loss": true }, { "from": "human", "prefix": "Hannah Fry:", "value": " Yeah, I totally was. I was picking things I was enjoying doing at the same time.", "loss": true }, { "from": "gpt", "prefix": "Brady Haran:", "value": " Hmm.", "loss": true }, { "from": "human", "prefix": "Hannah Fry:", "value": " But was kind of edging towards it. And then I eventually got… I just hated it 'cause there wasn't enough maths in it. (laughs)", "loss": true }, { "from": "gpt", "prefix": "Brady Haran:", "value": " What did you do?", "loss": true }, { "from": "human", "prefix": "Hannah Fry:", "value": " So (laughs) I was doing… I worked for an aerodynamics consultancy firm.", "loss": true }, { "from": "gpt", "prefix": "Brady Haran:", "value": " Right?", "loss": true }, { "from": "human", "prefix": "Hannah Fry:", "value": " So we were designing the wings on cars like Ferrari or like Lotus or like…", "loss": true }, { "from": "gpt", "prefix": "Brady Haran:", "value": " Yeah.", "loss": true }, { "from": "human", "prefix": "Hannah Fry:", "value": " HRT as it was at the time.", "loss": true }, { "from": "gpt", "prefix": "Brady Haran:", "value": " Yeah.", "loss": true }, { "from": "human", "prefix": "Hannah Fry:", "value": " I had imagined that what this would be would be a group of people sitting around a chalkboard coming up with loads of really clever ideas, writing equations, you know just thinking about the physics of it and thinking about the mathematics of it.", "loss": true }, { "from": "gpt", "prefix": "Brady Haran:", "value": " Yeah?", "loss": true }, { "from": "human", "prefix": "Hannah Fry:", "value": " And it just wasn't like that at all, basically (chuckles).", "loss": true }, { "from": "gpt", "prefix": "Brady Haran:", "value": " What was it like?", "loss": true }, { "from": "human", "prefix": "Hannah Fry:", "value": " All of these… the really nice stuff has already been wrapped up in computer simulations. So you were just setting up runs on your computer overnight and then coming in the morning and like picking the picture that just had the least amount of blue in it. It's like chucking darts basically. And if you chuck enough darts something's gonna hit. It was just… boring really.", "loss": true }, { "from": "gpt", "prefix": "Brady Haran:", "value": " So you thought you'd becoming into work and say I've had a whole new idea for the way we should design the wing and we're gonna completely revolutionize it and we'd all go and… but it wasn't, it was just like, you were just like grunt work.", "loss": true }, { "from": "human", "prefix": "Hannah Fry:", "value": " Yeah exactly, exactly that, grunt work. So that I think was the point where I was like no, I'd rather be a professional mathematician thanks very much.", "loss": true }, { "from": "gpt", "prefix": "Brady Haran:", "value": " Do you still like Formula One?", "loss": true }, { "from": "human", "prefix": "Hannah Fry:", "value": " Yeah I do, although I haven't really watched it that much this season just because…", "loss": true }, { "from": "gpt", "prefix": "Brady Haran:", "value": " I've liked it this season.", "loss": true }, { "from": "human", "prefix": "Hannah Fry:", "value": " Have you?", "loss": true }, { "from": "gpt", "prefix": "Brady Haran:", "value": " Yeah.", "loss": true }, { "from": "human", "prefix": "Hannah Fry:", "value": " I know, I just haven't been around really.", "loss": true }, { "from": "gpt", "prefix": "Brady Haran:", "value": " Yeah?", "loss": true }, { "from": "human", "prefix": "Hannah Fry:", "value": " Just too much going on on Sundays.", "loss": true }, { "from": "gpt", "prefix": "Brady Haran:", "value": " Okay.", "loss": true }, { "from": "human", "prefix": "Hannah Fry:", "value": " Worked too much basically.", "loss": true }, { "from": "gpt", "prefix": "Brady Haran:", "value": " When you watch Formula One now do you look at it differently because of that time you spent working in the behind the scenes with the…?", "loss": true }, { "from": "human", "prefix": "Hannah Fry:", "value": " No, I think I still massively enjoy it.", "loss": true }, { "from": "gpt", "prefix": "Brady Haran:", "value": " Yeah?", "loss": true }, { "from": "human", "prefix": "Hannah Fry:", "value": " In the same way, I think you realize just how superficial the information that you get from the broadcast version is. Just how much more… unbelievably delicious technical detail there is going on behind the scenes that you just don't have access to.", "loss": true }, { "from": "gpt", "prefix": "Brady Haran:", "value": " Yeah.", "loss": true }, { "from": "human", "prefix": "Hannah Fry:", "value": " I mean ultimately the reason why I like Formula One is 'cause it's just gigantic maths competition, isn't it?", "loss": true }, { "from": "gpt", "prefix": "Brady Haran:", "value": " Yeah. Yeah, it is.", "loss": true }, { "from": "human", "prefix": "Hannah Fry:", "value": " (laughs)", "loss": true }, { "from": "gpt", "prefix": "Brady Haran:", "value": " But also a big like politics… competition.", "loss": true }, { "from": "human", "prefix": "Hannah Fry:", "value": " Yeah, completely.", "loss": true }, { "from": "gpt", "prefix": "Brady Haran:", "value": " It's like a big political thing as well and a big like business financial thing, it's like…", "loss": true }, { "from": "human", "prefix": "Hannah Fry:", "value": " Yeah, totally with a bit of danger thrown in. Bit of jeopardy.", "loss": true }, { "from": "gpt", "prefix": "Brady Haran:", "value": " Yeah! Yeah. Can I ask you a gender thing about university?", "loss": true }, { "from": "human", "prefix": "Hannah Fry:", "value": " Hmm.", "loss": true }, { "from": "gpt", "prefix": "Brady Haran:", "value": " Were there lots of women, girls on your course, or was it very male dominated?", "loss": true }, { "from": "human", "prefix": "Hannah Fry:", "value": " I would say when I went it was about… hmm… maybe a third were girls.", "loss": true }, { "from": "gpt", "prefix": "Brady Haran:", "value": " Oh okay.", "loss": true }, { "from": "human", "prefix": "Hannah Fry:", "value": " Maybe a bit less actually, between a quarter and a third for undergrad.", "loss": true }, { "from": "gpt", "prefix": "Brady Haran:", "value": " So it wasn't crazy? It wasn't like you were the lone woman in the course?", "loss": true }, { "from": "human", "prefix": "Hannah Fry:", "value": " No, no, not as an undergrad.", "loss": true }, { "from": "gpt", "prefix": "Brady Haran:", "value": " Yeah.", "loss": true }, { "from": "human", "prefix": "Hannah Fry:", "value": " There were other girls around.", "loss": true }, { "from": "gpt", "prefix": "Brady Haran:", "value": " Yeah.", "loss": true }, { "from": "human", "prefix": "Hannah Fry:", "value": " But then as you go up through the sort of stages, so Masters and then PhD and then postdoc, basically all the women sort of fall away.", "loss": true }, { "from": "gpt", "prefix": "Brady Haran:", "value": " Yeah?", "loss": true }, { "from": "human", "prefix": "Hannah Fry:", "value": " So then when I was doing my PhD there were… I would say maybe forty PhD students and there was me and two other girls.", "loss": true }, { "from": "gpt", "prefix": "Brady Haran:", "value": " Okay. When it becomes that much a minority does it affect… your enjoyment of it or your ability to do it or are you completely blind to it? Like what's it like when there's so few?", "loss": true }, { "from": "human", "prefix": "Hannah Fry:", "value": " So I personally never really minded that much. But I know that there was one girl who was doing a PhD… she was a couple years ahead of me and she dropped out and I had this conversation with her and she said that actually she just felt incredibly uncomfortable that she would walk into this PhD room, often when you're doing PhD you're not… you don't… you work at home a lot. You just sort of have days where you're working at home but really you're just sleeping.", "loss": true }, { "from": "gpt", "prefix": "Brady Haran:", "value": " (laughs) or watching Countdown. (laughs)", "loss": true }, { "from": "human", "prefix": "Hannah Fry:", "value": " (laughs) It happens quite a lot too.", "loss": true }, { "from": "gpt", "prefix": "Brady Haran:", "value": " Yeah.", "loss": true }, { "from": "human", "prefix": "Hannah Fry:", "value": " So you know, if there was no other girls in the office she'd be the only girl in the office and she said that she just felt massively intimidated in that environment. She just felt completely uncomfortable in that environment and it just really put her off. And I think there are some people who do get really put off by it.", "loss": true }, { "from": "gpt", "prefix": "Brady Haran:", "value": " Hmm.", "loss": true }, { "from": "human", "prefix": "Hannah Fry:", "value": " 'Cause ultimately, you know, people often ask the question of like why are there so few women in mathematics and like, you know, STEM subjects. And in a way, I mean I've got somethings to say about it, but in a way the people who you should really ask are the people who left.", "loss": true }, { "from": "gpt", "prefix": "Brady Haran:", "value": " Yeah.", "loss": true }, { "from": "human", "prefix": "Hannah Fry:", "value": " And ask them why they left. The one thing that I will say about it though is that… and actually I think sometimes you can use this to your advantage, but especially when I was younger and especially when I hadn't done much outreach work, people would often assume that you have no idea what you're talking about. So you'd kind of go into a room, especially if you didn't know anyone, and people'd be having a bit of a maths chat, you're at a conference or something, and people would just assume that you're a complete idiot.", "loss": true }, { "from": "gpt", "prefix": "Brady Haran:", "value": " Yeah.", "loss": true }, { "from": "human", "prefix": "Hannah Fry:", "value": " And that I just always really loved that when you could… (laughs) just say something and watch people's faces fall. (laughs)", "loss": true }, { "from": "gpt", "prefix": "Brady Haran:", "value": " Yeah. I always wished I could speak other languages…", "loss": true }, { "from": "human", "prefix": "Hannah Fry:", "value": " (laughs)", "loss": true }, { "from": "gpt", "prefix": "Brady Haran:", "value": " …so that moment when you're in the lift with two people speaking another language who think you don't know what's going and then you suddenly answer in their language. They're like, oh how embarrassing!", "loss": true }, { "from": "human", "prefix": "Hannah Fry:", "value": " (laughs)", "loss": true }, { "from": "gpt", "prefix": "Brady Haran:", "value": " That's like the mathematical equivalent of that.", "loss": true }, { "from": "human", "prefix": "Hannah Fry:", "value": " It totally is.", "loss": true }, { "from": "gpt", "prefix": "Brady Haran:", "value": " Yeah.", "loss": true }, { "from": "human", "prefix": "Hannah Fry:", "value": " I was teaching very young, at sort of twenty-two and I was teaching my first group of undergraduates and there was one year when… the first year where I wrote an exam for these undergraduate students and I went off to go to the exam to monitor the students during it and (laughs) they wouldn't let me in the hall because the invigilators were like, we're not allowing any students in. So I kind of argued my way in and they were like, do you have authority to read this exam?! And I was like, well I wrote it so… (laughs)", "loss": true }, { "from": "gpt", "prefix": "Brady Haran:", "value": " (laughs)", "loss": true }, { "from": "human", "prefix": "Hannah Fry:", "value": " I'm gonna go with probably yes.", "loss": true }, { "from": "gpt", "prefix": "Brady Haran:", "value": " (laughs) So, going back to leaving the wing factory, Formula One.", "loss": true }, { "from": "human", "prefix": "Hannah Fry:", "value": " Hmm?", "loss": true }, { "from": "gpt", "prefix": "Brady Haran:", "value": " How did you get from that back into mathematics? What was the path?", "loss": true }, { "from": "human", "prefix": "Hannah Fry:", "value": " It was really straightforward (laughs) because my PhD supervisor, Fellow of the Royal Society actually guy called Professor Frank Smith who is just… there are few people on earth who I admire more than him. I mean I was only there really for a couple of months, right, before I was like I'm done with this. So he had not only sort of set me up with the Formula One stuff 'cause he knew that I want to just do it, but when he knew that I wasn't enjoying it he also set me up with a postdoc. So he's a fluid dynamicist, but he'd been dipping his toe in the water of looking at human behavior and sort of that overlap between social sciences and mathematics, which was a very new area and he had this postdoc that came up and was like, wanted to know if I wanted to do it, and I did and… yeah that's it really.", "loss": true }, { "from": "gpt", "prefix": "Brady Haran:", "value": " So tell me about most of the work you've done. It does sound quite accessible for mathematics?", "loss": true }, { "from": "human", "prefix": "Hannah Fry:", "value": " Yeah! Completely.", "loss": true }, { "from": "gpt", "prefix": "Brady Haran:", "value": " Your work. How would you explain it?", "loss": true }, { "from": "human", "prefix": "Hannah Fry:", "value": " I mean it's super super super applied, right? It's like the most applied maths that you can get. So the idea is that you can look at humans as though they're a mathematical system and there are some things that sort of make sense, right? If you think about a crowd of people, there are analogies there with the way that fluids move, right? Or if you think about, you know, sort of systems of revolutions and you know politics, the way it ebbs and flows, there are sort chaotic moments in there, you know you can also think about the systems that work through history that kind of connect up with each other, the networks of people, the way that ideas spread, you know? Influence spreads through a population in a similar way to sort of epidemics do. There's lots of little things about people that if you take enough of a step back you can see them as mathematical objects really. So that's really the starting point. And it sounds like the different things that I've worked on, it sounds like they're really disconnected, so you know, I've worked with the police looking at burglaries, at riots, at terrorism, you know, I've worked with retailers looking at where people spend money in the sort of retail environment, looking at badger culls, right? (laughs) What other stuff? Lots of epidemic stuff, it sounds like those things are really different from one another but actually underneath them all they have this common theme which is that they're about patterns created by humans that move around in space and time and yeah, the sort of theoretical basis of that is actually just a really simple idea.", "loss": true }, { "from": "gpt", "prefix": "Brady Haran:", "value": " What mathematical tools or processes do you use to effect this? Like what kind of mathematics do you use?", "loss": true }, { "from": "human", "prefix": "Hannah Fry:", "value": " So there's tons of statistics, obviously, because if you're starting with a dataset and you're analyzing it, the first thing that you want to do is to really get a grip with what kind of patterns are in there. But then from there I mean you have all kinds of tools, you know, in your toolkit that you wanna use so lots of network stuff. That happens quite a lot. But you can build models that are essentially differential equations and then use them to sort of isolate certain situations. Lots of information theory as well, lots of entropy maximization where you're trying to fill in the gaps in your knowledge. All kinds of different things. All kinds of different things.", "loss": true }, { "from": "gpt", "prefix": "Brady Haran:", "value": " How do you find non-mathematical people respond to this idea of mathematically modeling human behavior? 'Cause even I feel this like inbuilt resistance to it where it's like, you can't do that we're all individuals and you know we can't predicted by mathematics. We're unpredictable and we're…", "loss": true }, { "from": "human", "prefix": "Hannah Fry:", "value": " I have free will!", "loss": true }, { "from": "gpt", "prefix": "Brady Haran:", "value": " Yeah, exactly, like and suddenly you're there saying, oh I can predict where the next burglary's gonna happen or where the next bomb's gonna go off or where people are gonna walk in a shopping center.", "loss": true }, { "from": "human", "prefix": "Hannah Fry:", "value": " Hmm.", "loss": true }, { "from": "gpt", "prefix": "Brady Haran:", "value": " Do you find a resistance to that? Are people very accepting to the fact that you're just treating them like data points or…?", "loss": true }, { "from": "human", "prefix": "Hannah Fry:", "value": " I think the difference is, it depends on what scale you're talking about. So, tomorrow morning if you're working in an office and you were doing your morning commute, it's certainly true that you can change your mind and you can decide to get a coffee or you can sleep in or you know there's all kinds of different things that you might do that are different, so predicting exactly what you're going to do, I'm not always gonna get it right, you know? There's limits to how accurate I can be in that regard but when you scale up to the size of a city, actually that randomness kind of ends up evening itself out and you can predict with quite a high degree of accuracy how many people are gonna pass through a particular station on a given morning. So it's really a think that shift is where people can still hold onto their free will but are still broadly predictable.", "loss": true }, { "from": "gpt", "prefix": "Brady Haran:", "value": " Have you read the Foundation series then?", "loss": true }, { "from": "human", "prefix": "Hannah Fry:", "value": " What? The Asimov stuff? (laughs)", "loss": true }, { "from": "gpt", "prefix": "Brady Haran:", "value": " Yeah, psychohistory.", "loss": true }, { "from": "human", "prefix": "Hannah Fry:", "value": " (laughs)", "loss": true }, { "from": "gpt", "prefix": "Brady Haran:", "value": " Hari Seldon is the king of your mathematical field.", "loss": true }, { "from": "human", "prefix": "Hannah Fry:", "value": " I know it's so true. It's so true.", "loss": true }, { "from": "gpt", "prefix": "Brady Haran:", "value": " Hmm. Okay.", "loss": true }, { "from": "human", "prefix": "Hannah Fry:", "value": " I like to think that the outcome of our stuff… is slightly well… I dunno… (laughs) less terrifying.", "loss": true }, { "from": "gpt", "prefix": "Brady Haran:", "value": " Is there a holy grail in your field? Like what's your Riemann Hypothesis or the like the ultimate thing that could be done in your field that would make that person the King or Queen of this thing? Like what's the ultimate?", "loss": true }, { "from": "human", "prefix": "Hannah Fry:", "value": " Well so, you know, a lot of the statistical methods are stepping aside for a lot of machine learning techniques now. Which are just incredibly power at following these really really subtle patterns. So, you know, for example, I mean this is not exactly my area, but just for example, there are these really beautiful mathematical models that can predict earthquakes and aftershocks, right? In earthquakes. And they've been around for a long time and we, you know, really know the equations very well. Those equations instantly you find a similar pattern in burglaries which is why I know about them. But now, in the last year or so it's been shown that machine learning techniques are actually better at making these predictions than the beautifully elegant mathematical models are. And it's a similar thing with predicting burglaries or predicting whether someone's gonna go onto commit a crime in future, you can come up with these really well crafted theoretically sound mathematical models. But sometimes all you want is that prediction and so there are occasions when the machine learning techniques are the better route to go down. But some people sort of say that, you know, as artificial intelligence gets better if you have artificial general intelligence then you won't need the mathematical models at all. You know if you have something that can look at a situation, work out exactly what's going on, predict the future perfectly, and not lose any context or nuance in the process then… who needs maths? (chuckles)", "loss": true }, { "from": "gpt", "prefix": "Brady Haran:", "value": " How do you respond to that?", "loss": true }, { "from": "human", "prefix": "Hannah Fry:", "value": " Umm.", "loss": true }, { "from": "gpt", "prefix": "Brady Haran:", "value": " As a mathematician? (chuckles)", "loss": true }, { "from": "human", "prefix": "Hannah Fry:", "value": " I think A, it's a long way away…", "loss": true }, { "from": "gpt", "prefix": "Brady Haran:", "value": " Yeah?", "loss": true }, { "from": "human", "prefix": "Hannah Fry:", "value": " So I think I still got a job.", "loss": true }, { "from": "gpt", "prefix": "Brady Haran:", "value": " Yeah.", "loss": true }, { "from": "human", "prefix": "Hannah Fry:", "value": " And B, I think that actually you lose something when you throw away the mathematical model, which is that you lose an understanding of the mechanisms behind those patterns. So it's all very well saying the statistic shows that this area is the one where burglaries are gonna happen this evening, but if you don't know why and if you don't know what it is in your model that's driving that… that result, then it's very difficult to go in and unpick it and make a positive change for the causes behind it.", "loss": true }, { "from": "gpt", "prefix": "Brady Haran:", "value": " Like I don't really understand how a car engine works at a proper level but I just know that if it breaks down, I go and get it fixed, like so do you always need to know how something works as long as it works?", "loss": true }, { "from": "human", "prefix": "Hannah Fry:", "value": " I think yeah, I think you do actually. I think if you don't understand how it works then you can't be sure that it's completely robust. You can't be sure that what you've created isn't just making it's decision based on something that isn't part of the situation at all. So to give you an example, right? There was this driverless car that was built in like the Nineties, I think it was called ALVINN. And it was amazing, it was an amazing driverless car, it drove thousands of miles, had a camera on top of it, a neural network running behind the scenes that was working out where to drive, it was brilliant, right? Until they asked ALVINN to drive over a bridge. At which point it went completely made and almost killed the driver… or it would be the person inside the car.", "loss": true }, { "from": "gpt", "prefix": "Brady Haran:", "value": " Right.", "loss": true }, { "from": "human", "prefix": "Hannah Fry:", "value": " And they worked out what had actually been happening was all of those thousands of miles that it had been driving, it had learned to just drive with the grass on the lefthand side… righthand side I guess 'cause it was probably in America. But there had always been a verge, it was always nice and simple, and then as soon as there was a bridge and there was no grass it was kind of essentially hunting for grass, right?", "loss": true }, { "from": "gpt", "prefix": "Brady Haran:", "value": " Right.", "loss": true }, { "from": "human", "prefix": "Hannah Fry:", "value": " And you can end up with a similar situation if you're hunting for criminals or you're working out where burglaries are happening in the city. You want to be sure that it's really picking up on the signals that you want it to be picking up.", "loss": true }, { "from": "gpt", "prefix": "Brady Haran:", "value": " I wanna ask you something about doing mathematics or being a mathematician. So take away all the other things you do on the side.", "loss": true }, { "from": "human", "prefix": "Hannah Fry:", "value": " (chuckles)", "loss": true }, { "from": "gpt", "prefix": "Brady Haran:", "value": " You know, extracurricular activities and the administration of being an academic and even take away teaching. Just like when you're doing your research.", "loss": true }, { "from": "human", "prefix": "Hannah Fry:", "value": " Mhm?", "loss": true }, { "from": "gpt", "prefix": "Brady Haran:", "value": " What does that look like? Like do you come into work with a notepad and a pencil and think, okay, today I'm gonna have an idea or do you go for a walk around a park? For you what does the process of doing new mathematics look like?", "loss": true }, { "from": "human", "prefix": "Hannah Fry:", "value": " So to be honest, it's a lot a lot of stuff on the computer where you're crunching through data looking for stuff, doing statistics on a machine essentially. Or when you're building models inside a machine but if you are sort of back to like the pen and paper mathematics where you're designing something. Designing a model from scratch or you're doing some analysis on your model to try and work out where it's limits are or where the weaknesses are or, you know, how it works in particular situations. Then you are literally just in a room with the doors closed and a pen and paper and scribbling and then, you know, when you get stuck you go for a walk to think it through.", "loss": true }, { "from": "gpt", "prefix": "Brady Haran:", "value": " Do you have like a happy place or an inspiration place or a way that you give your brain a little kick start to start thinking about…", "loss": true }, { "from": "human", "prefix": "Hannah Fry:", "value": " Oh yeah, always on a walk.", "loss": true }, { "from": "gpt", "prefix": "Brady Haran:", "value": " A walk?", "loss": true }, { "from": "human", "prefix": "Hannah Fry:", "value": " Always on a walk. Yeah, isn't it the same for you? I always have my best ideas when I'm walking.", "loss": true }, { "from": "gpt", "prefix": "Brady Haran:", "value": " I have my best ideas in the shower.", "loss": true }, { "from": "human", "prefix": "Hannah Fry:", "value": " Do you?", "loss": true }, { "from": "gpt", "prefix": "Brady Haran:", "value": " Or water or a swimming.", "loss": true }, { "from": "human", "prefix": "Hannah Fry:", "value": " When you're working on a particularly creative project you just find yourself fully immersed in water most of the time?", "loss": true }, { "from": "gpt", "prefix": "Brady Haran:", "value": " (laughs) Yeah it's very hard to edit underwater.", "loss": true }, { "from": "human", "prefix": "Hannah Fry:", "value": " (laughs)", "loss": true }, { "from": "gpt", "prefix": "Brady Haran:", "value": " But like I'll go for a swim and have all these amazing ideas and by the time I've like dried off and gotten home and it's time to work again like all the inspiration and motivation's gone. But I find water very inspiring. But yours is a walk? But you walk around like an urban environment or does it have to be like a natural environment?", "loss": true }, { "from": "human", "prefix": "Hannah Fry:", "value": " No it can be an urban environment too, but if I'm working on something and like really kind of embedding myself in it then I try to take my doggie for a nice long walk everyday. But the reason for it though is exactly what you said about swimming. That you can write stuff down when you're swimming. Where is if you go for a walk, you can like have a phone with you and kind of take notes and stuff.", "loss": true }, { "from": "gpt", "prefix": "Brady Haran:", "value": " Professionally, academically, what's been your proudest moment? Like what's been the paper so far…", "loss": true }, { "from": "human", "prefix": "Hannah Fry:", "value": " Hmm.", "loss": true }, { "from": "gpt", "prefix": "Brady Haran:", "value": " …that's the one that you're like that was really good. That was the one that you would like, you know, show another mathematician…", "loss": true }, { "from": "human", "prefix": "Hannah Fry:", "value": " (laughs)", "loss": true }, { "from": "gpt", "prefix": "Brady Haran:", "value": " …look, you know…", "loss": true }, { "from": "human", "prefix": "Hannah Fry:", "value": " Hey.", "loss": true }, { "from": "gpt", "prefix": "Brady Haran:", "value": " I was pretty good that.", "loss": true }, { "from": "human", "prefix": "Hannah Fry:", "value": " (laughs) Okay so, (laughs) the truth is that the one that I like the most in that regard is also the one that no one will ever read. (laughs)", "loss": true }, { "from": "gpt", "prefix": "Brady Haran:", "value": " Right?", "loss": true }, { "from": "human", "prefix": "Hannah Fry:", "value": " I mean it's almost certainly got zero citations, but it took me the longest to write and its a paper that I wrote with my PhD supervisor, Frank Smith, and it's where we have taken a very old model of retail behavior and we have bent it out of all of recognition into the most delicious mathematical problem imaginable and just played with it to our hearts content. That's my favorite.", "loss": true }, { "from": "gpt", "prefix": "Brady Haran:", "value": " So this isn't one that like Westfield are gonna use 'cause…", "loss": true }, { "from": "human", "prefix": "Hannah Fry:", "value": " No. (laughs)", "loss": true }, { "from": "gpt", "prefix": "Brady Haran:", "value": " …it's applicable to their shopping centers?", "loss": true }, { "from": "human", "prefix": "Hannah Fry:", "value": " God no.", "loss": true }, { "from": "gpt", "prefix": "Brady Haran:", "value": " This is one where you guys just have gone like freestyle and played with math?", "loss": true }, { "from": "human", "prefix": "Hannah Fry:", "value": " I think we turned shopping centers into a one dimensional continuous line.(laughs)", "loss": true }, { "from": "gpt", "prefix": "Brady Haran:", "value": " Right? (laughs) Do you remember what it's called or…?", "loss": true }, { "from": "human", "prefix": "Hannah Fry:", "value": " I think it's called Rate Effects on the Growth of Centers.", "loss": true }, { "from": "gpt", "prefix": "Brady Haran:", "value": " Okay.", "loss": true }, { "from": "human", "prefix": "Hannah Fry:", "value": " As I said no one has ever read it. (laughs)", "loss": true }, { "from": "gpt", "prefix": "Brady Haran:", "value": " We'll link to it.", "loss": true }, { "from": "human", "prefix": "Hannah Fry:", "value": " But I like it.", "loss": true }, { "from": "gpt", "prefix": "Brady Haran:", "value": " We're gonna link to it in the notes from this show.", "loss": true }, { "from": "human", "prefix": "Hannah Fry:", "value": " (laughs)", "loss": true }, { "from": "gpt", "prefix": "Brady Haran:", "value": " So people can go and have a look.", "loss": true }, { "from": "human", "prefix": "Brady Haran:", "value": " Obviously in the last few years a lot of people will know that you're like on TV a lot now and on the internet and on Youtube and radio and podcasts and books and all sorts of things and we'll talk about a few of them probably shortly, how did that all start? How did you become like… outreach person as well?", "loss": true }, { "from": "gpt", "prefix": "Hannah Fry:", "value": " Wasn't just an accident, really? It was never an intention. It was never sort of my ambition but I did this (chuckles) very stupid talk in a pub which was about sexual contact networks, right? It's called the Friendship Paradox, do you know about this? It's the idea that when you have a network that structured like a sexual contact network or like a friendship network, so actually the way that Twitter's structured is very similar, there's a strange paradox which is this, your friends are more popular than you, essentially, so on Twitter both the people you follow and the people that follow you will on average have more followers than you do.", "loss": true }, { "from": "human", "prefix": "Brady Haran:", "value": " (laughs) So you're always bottom of the heap?", "loss": true }, { "from": "gpt", "prefix": "Hannah Fry:", "value": " Yeah, I think it's for ninety-five percent of users this is true. Right, which is a really strange idea. It's a really kind of weird idea. Actually…", "loss": true }, { "from": "human", "prefix": "Brady Haran:", "value": " That doesn't seem possible!", "loss": true }, { "from": "gpt", "prefix": "Hannah Fry:", "value": " It doesn't seem possible does it? The thing is that actually the reason why it happens is because a slightly odd quirk of when you use the mean, it illustrates the downside of using the mean, because you have people on Twitter like Katy Perry right, who've got ten million scrillion billion followers. So if you follow her that number so massively dwarfs everyone in your network who's like you, that the people who you follow are then way more popular than you are. But it also happens in the reverse, all it takes is a couple of people who are really popular who follow you and then it skews your numbers too. But it actually has a really interested implication when it comes to targeting intervention programs for halting the spread of sexually transmission diseases, this is a really nice idea.", "loss": true }, { "from": "human", "prefix": "Brady Haran:", "value": " Okay. This is Mick Hucknall the singer.", "loss": true }, { "from": "gpt", "prefix": "Hannah Fry:", "value": " Yeah, he's also I would say not the most attractive man that's ever walked the earth.", "loss": true }, { "from": "human", "prefix": "Brady Haran:", "value": " No. (sighs) Well there are three thousand people that disagree, but alright.", "loss": true }, { "from": "gpt", "prefix": "Hannah Fry:", "value": " (laughs)", "loss": true }, { "from": "human", "prefix": "Brady Haran:", "value": " (laughs)", "loss": true }, { "from": "gpt", "prefix": "Hannah Fry:", "value": " So I did this stupid talk about that.", "loss": true }, { "from": "human", "prefix": "Brady Haran:", "value": " Yeah.", "loss": true }, { "from": "gpt", "prefix": "Hannah Fry:", "value": " In a pub.", "loss": true }, { "from": "human", "prefix": "Brady Haran:", "value": " Yeah.", "loss": true }, { "from": "gpt", "prefix": "Hannah Fry:", "value": " Kind of like a comedy night thing.", "loss": true }, { "from": "human", "prefix": "Brady Haran:", "value": " Yeah, yeah.", "loss": true }, { "from": "gpt", "prefix": "Hannah Fry:", "value": " And someone was there who then invited me to go and give another talk that was filmed.", "loss": true }, { "from": "human", "prefix": "Brady Haran:", "value": " Right.", "loss": true }, { "from": "gpt", "prefix": "Hannah Fry:", "value": " Which did really well on the internet.", "loss": true }, { "from": "human", "prefix": "Brady Haran:", "value": " Okay.", "loss": true }, { "from": "gpt", "prefix": "Hannah Fry:", "value": " And then someone saw that and then asked me to do another talk, which was a TEDx talk which I did the Maths of Love for.", "loss": true }, { "from": "human", "prefix": "Brady Haran:", "value": " Okay.", "loss": true }, { "from": "gpt", "prefix": "Hannah Fry:", "value": " Never expecting anything to happen from it but it ended up being promoted as a proper TED talk.", "loss": true }, { "from": "human", "prefix": "Brady Haran:", "value": " Okay.", "loss": true }, { "from": "gpt", "prefix": "Hannah Fry:", "value": " And then really it… that sort of got millions of views and everything really came from there.", "loss": true }, { "from": "human", "prefix": "Brady Haran:", "value": " Okay. And then more and more people started asking you to do more and more stuff.", "loss": true }, { "from": "gpt", "prefix": "Hannah Fry:", "value": " Yeah, exactly.", "loss": true }, { "from": "human", "prefix": "Brady Haran:", "value": " You said yes to it. So clearly there's something you like about it. So what's been you feelings about now going into this outreach side of things in such a big way?", "loss": true }, { "from": "gpt", "prefix": "Hannah Fry:", "value": " Well I think that initially… I was a real scaredy cat when I was younger, right? I never like pushed myself or anything, just like your classic nerd. You know, never went anywhere, never travelled anywhere, never took any risks at all, and so when this stuff started happening and it was people coming to me asking me to do something, right? I just sort of thought, well, I'm just gonna say yes, because I'd rather live in a world where I said yes and it went badly than I said no and never and didn't know how it would go. So I just had a year where my motto was to say yes, basically. And then actually all of the stuff went quite well and now I'm in this situation where you know, I get to like… have someone pay for me to go to Dubai and fly in a helicopter over the desert at like sunrise and watch people who've built jetpacks strap them to their back and like chuck themselves out of a helicopter and fly around the desert. I mean like who the hell wouldn't want that?", "loss": true }, { "from": "human", "prefix": "Brady Haran:", "value": " So is it just the experience of doing it that's motivating you, do you like have this underlying I must do outreach for the people. I must bring math to the masses or is that not really like motivating you?", "loss": true }, { "from": "gpt", "prefix": "Hannah Fry:", "value": " I think it's a combination of the two really, I mean on the one hand it's like, you know, you only get one life and like, why on earth would I ever turn this stuff down? It's amazing to have these opportunities, but I think on the other hand I mean I really do like whole-heartedly believe that maths has got this reputation that it doesn't deserve and I think it's a massive shame that there is this just glorious living breathing playground that is just so incredibly wonderful and people just think it deserves to sit in a dusty textbook and be ignored and you know, I don't think that you can force people to like the subject. I don't think you can make people into mathematicians who were never going to be mathematicians but I do also think that actually the whole of society has something to gain by just looking at it with a little bit more understanding about what they're actually talking about.", "loss": true }, { "from": "human", "prefix": "Brady Haran:", "value": " 'Cause this side of things really become a big part of your life now. I imagine it must take a lot of your time because of the some of things you've been involved with, the books and the excellent TV programs. How has that affected your ability to still be a working mathematician?", "loss": true }, { "from": "gpt", "prefix": "Hannah Fry:", "value": " So actually the biggest thing that affected me being a working mathematician was having a baby. (laughs)", "loss": true }, { "from": "human", "prefix": "Brady Haran:", "value": " Right. (chuckles) Okay. Yes.", "loss": true }, { "from": "gpt", "prefix": "Hannah Fry:", "value": " So, I'm half and half, right, so I'm half proper academic and I'm half a outreach person, I guess.", "loss": true }, { "from": "human", "prefix": "Brady Haran:", "value": " (chuckles) Outreach person, is that what you would call it?", "loss": true }, { "from": "gpt", "prefix": "Hannah Fry:", "value": " Outreach person. Well I dunno, what'd'you call it?", "loss": true }, { "from": "human", "prefix": "Brady Haran:", "value": " I dunno.", "loss": true }, { "from": "gpt", "prefix": "Hannah Fry:", "value": " In America they call it a public intellectual (with an American accent).", "loss": true }, { "from": "human", "prefix": "Brady Haran:", "value": " A public intellectual?", "loss": true }, { "from": "gpt", "prefix": "Hannah Fry:", "value": " But I'm not using that 'cause it's ridiculous.", "loss": true }, { "from": "human", "prefix": "Brady Haran:", "value": " You're too late, you said it.", "loss": true }, { "from": "gpt", "prefix": "Hannah Fry:", "value": " No! (laughs)", "loss": true }, { "from": "human", "prefix": "Brady Haran:", "value": " And you said it in an American accent. (laughs)", "loss": true }, { "from": "gpt", "prefix": "Hannah Fry:", "value": " (laughs)", "loss": true }, { "from": "human", "prefix": "Brady Haran:", "value": " Okay so half half. So how do you feel about that? How do you feel about being a half academic? 'Cause it seems like something that, you know, when I hear you talk about your academic life, I see the pleasure that it gave you.", "loss": true }, { "from": "gpt", "prefix": "Hannah Fry:", "value": " Mhm.", "loss": true }, { "from": "human", "prefix": "Brady Haran:", "value": " And the happiness and the pride, and having to start squashing that more and more 'cause of the other stuff, you know, you think you got the balance right? How do you feel about it?", "loss": true }, { "from": "gpt", "prefix": "Hannah Fry:", "value": " Yeah, I mean I think so, I'm about to go off on maternity leave again and I think that, it's inevitable actually that, this is sort of the main thing, right? I think like lots of females find this the… when they have children there's a period of time where it's very very hard, right? You can't just pick up and put down research. It's really really hard, you kinda it takes momentum and it takes, you know, being embedded in the community and being at the conferences and hearing all of the current ideas and you know, being totally immersed in it. So a lot of female academics have this where they have their first child and they're away for a year and then they come back for a year and then they're away for a year and it's three years really out, so I think really that's the thing that's stilted stuff, right? That aside, the being half and half, in some ways it quite helps with the research because you meet people you would never meet otherwise. You have the opportunity for collaboration you would never have otherwise. And also you have the opportunity for engaging people in your projects, right? In a way you wouldn't have otherwise, so just to give you an example, earlier this year, this project that we did with the BBC, that was also with the University of Cambridge and the London School of Hygiene and Tropical Medicine. It was a program called the Pandemic, right? Now in epidemiology there's one big problem which is that you can come up with these glorious mathematical models but if you want them to really work, if you want them to be able to predict how an epidemic might spread, you have to have a real idea of where people are traveling, how far they're moving and how many people they're coming into contact with. And the only information that we had up until this point is a paper survey that was conducted ten years ago where it was like how many people do you reckon you met this week? Right, which is… there's limitations to how well, you know, that can actually be employed. So, because I had this sort of platform, this opportunity came up where we got the BBC to pay for an app to be built, people could download to their smartphones, that would collect their data for them and we've created this unbelievably glorious dataset that now is, you know, gonna be released to the academic community and make a massive difference, you know, for years to come really and that's something that you couldn't do unless you had this double edge to your bow, what's the phrase?", "loss": true }, { "from": "human", "prefix": "Brady Haran:", "value": " Oh.", "loss": true }, { "from": "gpt", "prefix": "Hannah Fry:", "value": " Double sword to your…", "loss": true }, { "from": "human", "prefix": "Brady Haran:", "value": " Second string to your bow, is it?", "loss": true }, { "from": "gpt", "prefix": "Hannah Fry:", "value": " There we go, we'll go with it.", "loss": true }, { "from": "human", "prefix": "Brady Haran:", "value": " One of those.", "loss": true }, { "from": "gpt", "prefix": "Hannah Fry:", "value": " (laughs)", "loss": true }, { "from": "human", "prefix": "Brady Haran:", "value": " How has being like a more public figure changed your life away from the academic side of things? Has it been… like, 'cause you know I imagine you would get recognized more than you used to 'cause your in television shows and things like that like… has it been like a real positive experience for you being a public intellectual? Or…", "loss": true }, { "from": "gpt", "prefix": "Hannah Fry:", "value": " (laughs)", "loss": true }, { "from": "human", "prefix": "Brady Haran:", "value": " Has it been… has it been… (laughs)", "loss": true }, { "from": "gpt", "prefix": "Hannah Fry:", "value": " (laughs) D'you know…", "loss": true }, { "from": "human", "prefix": "Brady Haran:", "value": " A Public Intellectual is probably the title of the podcast now.", "loss": true }, { "from": "gpt", "prefix": "Hannah Fry:", "value": " (laughs) Shut up, no you can't do that. (laughs)", "loss": true }, { "from": "human", "prefix": "Brady Haran:", "value": " You're totally screwed. (laughs)", "loss": true }, { "from": "gpt", "prefix": "Hannah Fry:", "value": " You can't do it. The other one they do in America is they invite you to conferences and they're like, we just really wanted you as a thought leader. (laughs)", "loss": true }, { "from": "human", "prefix": "Brady Haran:", "value": " A thought leader? We're reaching out to you as a though leader. (chuckles)", "loss": true }, { "from": "gpt", "prefix": "Hannah Fry:", "value": " (laughs) What does that mean!", "loss": true }, { "from": "human", "prefix": "Brady Haran:", "value": " I dunno.", "loss": true }, { "from": "gpt", "prefix": "Hannah Fry:", "value": " What does that mean.", "loss": true }, { "from": "human", "prefix": "Brady Haran:", "value": " I don't know.", "loss": true }, { "from": "gpt", "prefix": "Hannah Fry:", "value": " Nothing, I think. Don't call me that. (laughs)", "loss": true }, { "from": "human", "prefix": "Brady Haran:", "value": " Alright. (laughs)", "loss": true }, { "from": "gpt", "prefix": "Hannah Fry:", "value": " (laughs)", "loss": true }, { "from": "human", "prefix": "Brady Haran:", "value": " So has it been like quite positive? Do you like the sort of glitzy celebrity side of it?", "loss": true }, { "from": "gpt", "prefix": "Hannah Fry:", "value": " I think the glitzy celebrity side is where you going down red carpets, right? You're getting papped, you're appearing in Hello magazine, I mean those are celebrities, right?", "loss": true }, { "from": "human", "prefix": "Brady Haran:", "value": " Okay.", "loss": true }, { "from": "gpt", "prefix": "Hannah Fry:", "value": " Whereas if you're just on BBC 4… (laughs)", "loss": true }, { "from": "human", "prefix": "Brady Haran:", "value": " Yeah.", "loss": true }, { "from": "gpt", "prefix": "Hannah Fry:", "value": " Or like… (laughs)", "loss": true }, { "from": "human", "prefix": "Brady Haran:", "value": " Okay.", "loss": true }, { "from": "gpt", "prefix": "Hannah Fry:", "value": " Radio 4 in the morning.", "loss": true }, { "from": "human", "prefix": "Brady Haran:", "value": " or Numberphile. (laughs)", "loss": true }, { "from": "gpt", "prefix": "Hannah Fry:", "value": " Oh, yeah. (laughs)", "loss": true }, { "from": "human", "prefix": "Brady Haran:", "value": " (laughs)", "loss": true }, { "from": "gpt", "prefix": "Hannah Fry:", "value": " Exactly.", "loss": true }, { "from": "human", "prefix": "Brady Haran:", "value": " Yeah. (laughs)", "loss": true }, { "from": "gpt", "prefix": "Hannah Fry:", "value": " And getting the Tube still. I mean…", "loss": true }, { "from": "human", "prefix": "Brady Haran:", "value": " Okay.", "loss": true }, { "from": "gpt", "prefix": "Hannah Fry:", "value": " It's not like you know…", "loss": true }, { "from": "human", "prefix": "Brady Haran:", "value": " You don't still get the Tube do you?", "loss": true }, { "from": "gpt", "prefix": "Hannah Fry:", "value": " Sometimes.", "loss": true }, { "from": "human", "prefix": "Brady Haran:", "value": " (laughs)", "loss": true }, { "from": "gpt", "prefix": "Hannah Fry:", "value": " (laughs) Actually I quite hate the Tube. Uber's ruined my life. (laughs)", "loss": true }, { "from": "human", "prefix": "Brady Haran:", "value": " Alright.", "loss": true }, { "from": "gpt", "prefix": "Hannah Fry:", "value": " But… no, I really get the Tube, I do get the Tube.", "loss": true }, { "from": "human", "prefix": "Brady Haran:", "value": " Yeah?", "loss": true }, { "from": "gpt", "prefix": "Hannah Fry:", "value": " But it's been almost universally positive. It's been almost universally positive.", "loss": true }, { "from": "human", "prefix": "Brady Haran:", "value": " Yeah.", "loss": true }, { "from": "gpt", "prefix": "Hannah Fry:", "value": " Like you get letters of support from people who just make a massive difference. You know, things that you… aren't even really aware that your having an impact on people and then they contact you, it just makes an enormous difference.", "loss": true }, { "from": "human", "prefix": "Brady Haran:", "value": " Yeah. That's nice.", "loss": true }, { "from": "gpt", "prefix": "Hannah Fry:", "value": " There's downsides too. I mean obviously.", "loss": true }, { "from": "human", "prefix": "Brady Haran:", "value": " I'll ask you a question, but it'll be hard because it will make you sound unhumble so I'll ask you to be humble and self-deprecating. Answer both sides of it. When it comes to outreach what do you think you're good at, and what part of it do you think you're still not good out? Like you when you look back at stuff… yeah?", "loss": true }, { "from": "gpt", "prefix": "Hannah Fry:", "value": " Ooh, good question.", "loss": true }, { "from": "human", "prefix": "Brady Haran:", "value": " Like what's your strength and your weakness?", "loss": true }, { "from": "gpt", "prefix": "Hannah Fry:", "value": " Okay, so… I think that it's quite new to have someone who's not a middle aged man. (laughs)", "loss": true }, { "from": "human", "prefix": "Brady Haran:", "value": " Right.", "loss": true }, { "from": "gpt", "prefix": "Hannah Fry:", "value": " (laughs) I think that's quite a big part of it.", "loss": true }, { "from": "human", "prefix": "Brady Haran:", "value": " Yeah.", "loss": true }, { "from": "gpt", "prefix": "Hannah Fry:", "value": " I think also if I'm honest…", "loss": true }, { "from": "human", "prefix": "Brady Haran:", "value": " But you can't say I'm good at being… a young woman.", "loss": true }, { "from": "gpt", "prefix": "Hannah Fry:", "value": " I'm good at not being (laugh) I'm good at not being a man.", "loss": true }, { "from": "human", "prefix": "Brady Haran:", "value": " That's just what you are.", "loss": true }, { "from": "gpt", "prefix": "Hannah Fry:", "value": " Okay.", "loss": true }, { "from": "human", "prefix": "Brady Haran:", "value": " (laughs)", "loss": true }, { "from": "gpt", "prefix": "Hannah Fry:", "value": " Alright I think also the fact that I'm not a genius helps, a lot, actually. I think the fact that…", "loss": true }, { "from": "human", "prefix": "Brady Haran:", "value": " You're like a professional mathematician though it's not like you're like…", "loss": true }, { "from": "gpt", "prefix": "Hannah Fry:", "value": " Yeah but I'm not like a good one. (laughs)", "loss": true }, { "from": "human", "prefix": "Brady Haran:", "value": " (laughs)", "loss": true }, { "from": "gpt", "prefix": "Hannah Fry:", "value": " You know I'm not like Marcus du Sautoy who's like professor at Oxford.", "loss": true }, { "from": "human", "prefix": "Brady Haran:", "value": " Okay.", "loss": true }, { "from": "gpt", "prefix": "Hannah Fry:", "value": " And professor twice 'cause he's a proper professor and he's also a professor… an outreach professor.", "loss": true }, { "from": "human", "prefix": "Brady Haran:", "value": " Okay.", "loss": true }, { "from": "gpt", "prefix": "Hannah Fry:", "value": " I'm not like David Spiegelhalter who's like… have you read someone of the papers that man was writing when he was twenty-three?", "loss": true }, { "from": "human", "prefix": "Brady Haran:", "value": " What so you think you're seen as having like a common touch?", "loss": true }, { "from": "gpt", "prefix": "Hannah Fry:", "value": " Yeah I think so. I think so. I mean basically I'm just a normal person who just really likes maths, basically.", "loss": true }, { "from": "human", "prefix": "Brady Haran:", "value": " Alright.", "loss": true }, { "from": "gpt", "prefix": "Hannah Fry:", "value": " So I think that helps.", "loss": true }, { "from": "human", "prefix": "Brady Haran:", "value": " Yeah.", "loss": true }, { "from": "gpt", "prefix": "Hannah Fry:", "value": " And then I also think that I don't take myself too seriously, I'm quite happy with people take out the piss out of me.", "loss": true }, { "from": "human", "prefix": "Brady Haran:", "value": " Hmm.", "loss": true }, { "from": "gpt", "prefix": "Hannah Fry:", "value": " Which I think helps to be a bit more approachable.", "loss": true }, { "from": "human", "prefix": "Brady Haran:", "value": " Especially as a public intellectual.", "loss": true }, { "from": "gpt", "prefix": "Hannah Fry:", "value": " (laughs)", "loss": true }, { "from": "human", "prefix": "Brady Haran:", "value": " (laughs) Yeah?", "loss": true }, { "from": "gpt", "prefix": "Hannah Fry:", "value": " As speaking as a thought leader and an intellectual, yeah. (laughs)", "loss": true }, { "from": "human", "prefix": "Brady Haran:", "value": " (laughs) Okay, yeah. Right.", "loss": true }, { "from": "gpt", "prefix": "Hannah Fry:", "value": " I think that stuff helps.", "loss": true }, { "from": "human", "prefix": "Brady Haran:", "value": " Yeah.", "loss": true }, { "from": "gpt", "prefix": "Hannah Fry:", "value": " But in terms of what I'm still not good at, oh my god. Like everything.", "loss": true }, { "from": "human", "prefix": "Brady Haran:", "value": " Really?", "loss": true }, { "from": "gpt", "prefix": "Hannah Fry:", "value": " Yeah. On everything.", "loss": true }, { "from": "human", "prefix": "Brady Haran:", "value": " Oh wow.", "loss": true }, { "from": "gpt", "prefix": "Hannah Fry:", "value": " Actually it's not really a star rating, it's a score out of ten.", "loss": true }, { "from": "human", "prefix": "Brady Haran:", "value": " Right, on anything you'll do like a…?", "loss": true }, { "from": "gpt", "prefix": "Hannah Fry:", "value": " Every talk. Every radio program.", "loss": true }, { "from": "human", "prefix": "Brady Haran:", "value": " Youtube videos?", "loss": true }, { "from": "gpt", "prefix": "Hannah Fry:", "value": " Uh… yeah.", "loss": true }, { "from": "human", "prefix": "Brady Haran:", "value": " Does that mean that you've got star ratings for your Numberphile appearances?", "loss": true }, { "from": "gpt", "prefix": "Hannah Fry:", "value": " Oh yeah.", "loss": true }, { "from": "human", "prefix": "Brady Haran:", "value": " (gasps)", "loss": true }, { "from": "gpt", "prefix": "Hannah Fry:", "value": " Yeah, although, it's slightly different when you're edited. Your performance isn't totally… do you know what I mean?", "loss": true }, { "from": "human", "prefix": "Brady Haran:", "value": " Yeah.", "loss": true }, { "from": "gpt", "prefix": "Hannah Fry:", "value": " It's not like totally… but yeah totally with the idea… yeah completely.", "loss": true }, { "from": "human", "prefix": "Brady Haran:", "value": " How have you been going on Numberphile so far?", "loss": true }, { "from": "gpt", "prefix": "Hannah Fry:", "value": " (laughs) So some of them I'm really happy with.", "loss": true }, { "from": "human", "prefix": "Brady Haran:", "value": " What's your highest star rated one so far?", "loss": true }, { "from": "gpt", "prefix": "Hannah Fry:", "value": " Umm… I think it was the Rock Paper Scissors actually.", "loss": true }, { "from": "human", "prefix": "Brady Haran:", "value": " Okay, yeah?", "loss": true }, { "from": "gpt", "prefix": "Hannah Fry:", "value": " Which was my first one.", "loss": true }, { "from": "human", "prefix": "Brady Haran:", "value": " Yeah, that was good. That was the first time I ever met you.", "loss": true }, { "from": "gpt", "prefix": "Hannah Fry:", "value": " Yeah, I think it was.", "loss": true }, { "from": "human", "prefix": "Brady Haran:", "value": " Yeah. We've just gone downhill since, have we?", "loss": true }, { "from": "gpt", "prefix": "Hannah Fry:", "value": " I know. (laughs) No! (laughs)", "loss": true }, { "from": "human", "prefix": "Brady Haran:", "value": " (laughs)", "loss": true }, { "from": "gpt", "prefix": "Hannah Fry:", "value": " You know, I 'm not giving myself tens or whatever.", "loss": true }, { "from": "human", "prefix": "Brady Haran:", "value": " No, no, no.", "loss": true }, { "from": "gpt", "prefix": "Hannah Fry:", "value": " But the thing is I think it's incredibly important to look back at what you've done and… acknowledge where you can improve.", "loss": true }, { "from": "human", "prefix": "Brady Haran:", "value": " Yeah.", "loss": true }, { "from": "gpt", "prefix": "Hannah Fry:", "value": " So I still do… I mean I've given hundreds of public lectures now, right, hundreds of public talks, and I still do this where it's like, I have a note… you know, little notes… in every folder where I save the talk and I just say what worked and what didn't work.", "loss": true }, { "from": "human", "prefix": "Brady Haran:", "value": " Alright.", "loss": true }, { "from": "gpt", "prefix": "Hannah Fry:", "value": " So that next time… yeah.", "loss": true }, { "from": "human", "prefix": "Brady Haran:", "value": " What's been your… your biggest crash so far, and what's your like highest rating? What's the high and the low so far?", "loss": true }, { "from": "gpt", "prefix": "Hannah Fry:", "value": " Ooh that's a good question.", "loss": true }, { "from": "human", "prefix": "Brady Haran:", "value": " Have you ever got a ten or is ten not allowed?", "loss": true }, { "from": "gpt", "prefix": "Hannah Fry:", "value": " I've got like close to a ten a couple of times.", "loss": true }, { "from": "human", "prefix": "Brady Haran:", "value": " Hmm.", "loss": true }, { "from": "gpt", "prefix": "Hannah Fry:", "value": " So one actually, I did an interview on Shaun Keaveny's show.", "loss": true }, { "from": "human", "prefix": "Brady Haran:", "value": " He's great.", "loss": true }, { "from": "gpt", "prefix": "Hannah Fry:", "value": " He's brilliant, I like him a lot.", "loss": true }, { "from": "human", "prefix": "Brady Haran:", "value": " Yeah, yeah.", "loss": true }, { "from": "gpt", "prefix": "Hannah Fry:", "value": " And that got a really high mark, but it got a really high mark because I managed to like keep my nerves under control.", "loss": true }, { "from": "human", "prefix": "Brady Haran:", "value": " Yeah?", "loss": true }, { "from": "gpt", "prefix": "Hannah Fry:", "value": " And I was really nervous about it, it's like Shaun Keaveny, do you know?", "loss": true }, { "from": "human", "prefix": "Brady Haran:", "value": " Yeah.", "loss": true }, { "from": "gpt", "prefix": "Hannah Fry:", "value": " And I hadn't done this program before and I hadn't done loads of radio… chatty radio shows before. But I… yeah, managed to not sound like an idiot.", "loss": true }, { "from": "human", "prefix": "Shaun Keaveny:", "value": " Our next guest is an expert in fluid dynamics, understands the trigonometry of love and can tell us a thing or two about ear-worms. Here to talk about… I've gone a little bit William Hague.", "loss": true }, { "from": "gpt", "prefix": "Hannah Fry (in clip):", "value": " (laughs)", "loss": true }, { "from": "human", "prefix": "Shaun Keaveny (in clip):", "value": " Here to talk about… Theon! Here to talk about the next series of the Curious Case of Rutherford and Fry, it's Hannah Fry, good morning!", "loss": true }, { "from": "gpt", "prefix": "Hannah Fry (in clip):", "value": " Hello, good morning.", "loss": true }, { "from": "human", "prefix": "Shaun Keaveny (in clip):", "value": " Thanks for coming in.", "loss": true }, { "from": "gpt", "prefix": "Hannah Fry (in clip):", "value": " What an intro.", "loss": true }, { "from": "human", "prefix": "Shaun Keaveny (in clip):", "value": " It was quite good.", "loss": true }, { "from": "gpt", "prefix": "Hannah Fry (in clip):", "value": " I enjoyed that.", "loss": true }, { "from": "human", "prefix": "Shaun Keaveny (in clip):", "value": " I know it's hugely… we start well and then it goes downhill. (laughs)", "loss": true }, { "from": "gpt", "prefix": "Hannah Fry (in clip):", "value": " Well I think if anything it peaked up as the voice… you know…", "loss": true }, { "from": "human", "prefix": "Shaun Keaveny (in clip):", "value": " As it went a bit William Hague, yeah…", "loss": true }, { "from": "gpt", "prefix": "Hannah Fry:", "value": " 'Cause the thing is on those kind of shows, right, you have to be really careful that you are jumping in exactly the right amount. You can't just sit there and wait for them to talk to you. (pause) Because it has to be a conversation, so you kind of have to jump in a little bit.", "loss": true }, { "from": "human", "prefix": "Brady Haran:", "value": " Were you doing it via phone or were you in the studio with him?", "loss": true }, { "from": "gpt", "prefix": "Hannah Fry:", "value": " I was in the studio, yeah.", "loss": true }, { "from": "human", "prefix": "Brady Haran:", "value": " Okay.", "loss": true }, { "from": "gpt", "prefix": "Hannah Fry:", "value": " Which makes it easier.", "loss": true }, { "from": "human", "prefix": "Brady Haran:", "value": " Yeah.", "loss": true }, { "from": "gpt", "prefix": "Hannah Fry:", "value": " But you also you can't jump in when they're talking when it's not appropriate.", "loss": true }, { "from": "human", "prefix": "Brady Haran:", "value": " Yeah.", "loss": true }, { "from": "gpt", "prefix": "Hannah Fry:", "value": " So it's really hard to like get that balance right.", "loss": true }, { "from": "human", "prefix": "Brady Haran:", "value": " Okay.", "loss": true }, { "from": "gpt", "prefix": "Hannah Fry:", "value": " But then also I just explained my thing that was going on… quite well… whatever.", "loss": true }, { "from": "human", "prefix": "Brady Haran:", "value": " Yeah.", "loss": true }, { "from": "gpt", "prefix": "Hannah Fry:", "value": " But I mean I've definitely bombed a lot of talks. (laughs)", "loss": true }, { "from": "human", "prefix": "Brady Haran:", "value": " Like do you ever get a zero or a one, what's like, what's the lowest mark you can get?", "loss": true }, { "from": "gpt", "prefix": "Hannah Fry:", "value": " Oh… no because I try… I always really prepare.", "loss": true }, { "from": "human", "prefix": "Brady Haran:", "value": " Okay.", "loss": true }, { "from": "gpt", "prefix": "Hannah Fry:", "value": " So I never get a zeroes and ones. That would be like…", "loss": true }, { "from": "human", "prefix": "Brady Haran:", "value": " A zero would just be falling over on the stage and then not giving a talk at all?", "loss": true }, { "from": "gpt", "prefix": "Hannah Fry:", "value": " (laughs) Or I… just going in and shouting at everyone.", "loss": true }, { "from": "human", "prefix": "Brady Haran:", "value": " Yeah.", "loss": true }, { "from": "gpt", "prefix": "Hannah Fry:", "value": " Going like, hey y'all! (growls) (laughs)", "loss": true }, { "from": "human", "prefix": "Brady Haran:", "value": " (laughs)", "loss": true }, { "from": "gpt", "prefix": "Hannah Fry:", "value": " And stalling up. (laughs)", "loss": true }, { "from": "human", "prefix": "Brady Haran:", "value": " Okay. (laughs)", "loss": true }, { "from": "gpt", "prefix": "Hannah Fry:", "value": " But I think that you know, I've definitely got low marks where I just haven't thought enough about what the audience were wanting.", "loss": true }, { "from": "human", "prefix": "Brady Haran:", "value": " So a low mark is at times where you feel like you've just missed the audience? You've missed your…", "loss": true }, { "from": "gpt", "prefix": "Hannah Fry:", "value": " Yeah, or maybe I've been a bit sloppy and I haven't prepared as well as I could have done. So one thing that I'm still not very good at, right? That Matt Parker is amazing at actually, instantly. Is that, if the audience is in a really good mood and they're going with me I can accelerate that when I'm stage this is. I know how to accelerate that and, you know, get it to feedback on itself. But sometimes the audience just isn't in the right mood and what Matt can do is he can turn the audience to being into him.", "loss": true }, { "from": "human", "prefix": "Brady Haran:", "value": " Win them. Win them back.", "loss": true }, { "from": "gpt", "prefix": "Hannah Fry:", "value": " He can win them, right? And I just haven't quite worked out how to do that yet. I haven't worked out that when you've got a bad audience, not to just be like okay I'm just going to say what I was gonna say and that's it and then I haven't got anything else… but rather be like, what is it about you that you need, you know, do you need me to be more goofy, do you need me to be more energetic, what is it that you need? And give them that and then turn them to your side, I haven't worked that out.", "loss": true }, { "from": "human", "prefix": "Brady Haran:", "value": " Let's talk about your books for a second. You've done a few now, you had a few like light hearted fun ones. One about love and stuff and one about Christmas and now you've written a more hardcore one.", "loss": true }, { "from": "gpt", "prefix": "Hannah Fry:", "value": " Yeah.", "loss": true }, { "from": "human", "prefix": "Brady Haran:", "value": " Tell me about the… how they were different. How writing them was different?", "loss": true }, { "from": "gpt", "prefix": "Hannah Fry:", "value": " So the love one (laughs) well, in a way, I sort of don't want anyone to ever read that book (laughs) because… I'll tell you why (laughs)", "loss": true }, { "from": "human", "prefix": "Brady Haran:", "value": " (laughs) Okay. Link in the show notes, people. (laughs)", "loss": true }, { "from": "gpt", "prefix": "Hannah Fry:", "value": " (laughs) I know you're not supposed to say this about your own work.", "loss": true }, { "from": "human", "prefix": "Brady Haran:", "value": " Yeah?", "loss": true }, { "from": "gpt", "prefix": "Hannah Fry:", "value": " But what happened with that book, right, was that I'd just done this talk, this TEDx talk, got an email from TED saying, we're putting it on the main channel, we're making it a proper TED talk. It's completely unexpected, right? And then a couple of weeks later, they said, actually what we're gonna do, we're not gonna put it online yet, because we want you to write a book to go with it.", "loss": true }, { "from": "human", "prefix": "Brady Haran:", "value": " Hmm.", "loss": true }, { "from": "gpt", "prefix": "Hannah Fry:", "value": " And then we'll launch them both together.", "loss": true }, { "from": "human", "prefix": "Brady Haran:", "value": " Hmm.", "loss": true }, { "from": "gpt", "prefix": "Hannah Fry:", "value": " And I was like… my god this is amazing! I mean I never would have written a book if someone hadn't come to me and said you need to write a book, right?", "loss": true }, { "from": "human", "prefix": "Brady Haran:", "value": " Yeah.", "loss": true }, { "from": "gpt", "prefix": "Hannah Fry:", "value": " Never in a million years. And then I kinda thought about it and was like okay I'm gonna do it. And they were like, great! We need the whole book delivered in six weeks time.", "loss": true }, { "from": "human", "prefix": "Brady Haran:", "value": " (gasps) (laughs)", "loss": true }, { "from": "gpt", "prefix": "Hannah Fry:", "value": " (laughs) And I'm like full-time academic. Got a full-time job, you know, it's like right, that's not really gonna be possible. So I tried to negotiate with them and they were like (tsks) yeah… thing is… we're TED… so…", "loss": true }, { "from": "human", "prefix": "Brady Haran:", "value": " Yeah. Yeah.", "loss": true }, { "from": "gpt", "prefix": "Hannah Fry:", "value": " Do you wanna do it or not?", "loss": true }, { "from": "human", "prefix": "Brady Haran:", "value": " Yeah.", "loss": true }, { "from": "gpt", "prefix": "Hannah Fry:", "value": " So I did it. I mean it was the most horrific experience of my life.", "loss": true }, { "from": "human", "prefix": "Brady Haran:", "value": " Bit of it… you felt like it was a bit rushed?", "loss": true }, { "from": "gpt", "prefix": "Hannah Fry:", "value": " Yeah I do think it was a bit rushed, yeah.", "loss": true }, { "from": "human", "prefix": "Brady Haran:", "value": " How many stars does it get in your notepad?", "loss": true }, { "from": "gpt", "prefix": "Hannah Fry:", "value": " Oh god… I mean… a lot less than it gets on Amazon. Put it that way. (laughs)", "loss": true }, { "from": "human", "prefix": "Brady Haran:", "value": " Right. (laughs) Is that a humble brag? I can't quite decide. (laughs)", "loss": true }, { "from": "gpt", "prefix": "Hannah Fry:", "value": " (laughs)", "loss": true }, { "from": "human", "prefix": "Brady Haran:", "value": " (laughs)", "loss": true }, { "from": "gpt", "prefix": "Hannah Fry:", "value": " Blimey (laughs). It might be.", "loss": true }, { "from": "human", "prefix": "Brady Haran:", "value": " It gets very high ratings on Amazon! (laughs)", "loss": true }, { "from": "gpt", "prefix": "Hannah Fry:", "value": " What I'm saying… no… (laughs)", "loss": true }, { "from": "human", "prefix": "Brady Haran:", "value": " But I think it's rubbish!", "loss": true }, { "from": "gpt", "prefix": "Hannah Fry:", "value": " No it doesn't, no it doesn't, it doesn't, it doesn't but what I'm saying is that people are still being generous.", "loss": true }, { "from": "human", "prefix": "Brady Haran:", "value": " Okay.", "loss": true }, { "from": "gpt", "prefix": "Hannah Fry:", "value": " That's what I mean.", "loss": true }, { "from": "human", "prefix": "Brady Haran:", "value": " Okay.", "loss": true }, { "from": "gpt", "prefix": "Hannah Fry:", "value": " I mean like, don't get me wrong, I mean I think the idea is there, like taking maths and something love and smashing them together. I still would like stand by that idea and I think there was like the kind of beginning of something much better there. But it was just such a hard process. I'm not a natural writer and… getting the ideas… it wasn't my research area either, right? You know, getting that up to the point where it's like, yeah I'm really proud of this and I'm ready to release it to the world, it just wasn't… I just needed more time and just didn't have it.", "loss": true }, { "from": "human", "prefix": "Brady Haran:", "value": " Hmm. Hmm.", "loss": true }, { "from": "gpt", "prefix": "Hannah Fry:", "value": " The Christmas one on the other hand, I think I did have more time with that and I did that with… as a collaboration with my colleague Thomas Oléron Evans, who is basically a wonderful human being. And we just had loads of funs writing it and just being really silly and… playful and whatever and coming up with stupid ideas and seeing them through. So that one's kind of more like what I wanted Love to be, if you know what I mean.", "loss": true }, { "from": "human", "prefix": "Brady Haran:", "value": " Yeah.", "loss": true }, { "from": "gpt", "prefix": "Hannah Fry:", "value": " But I really like that book, it's really… it's just a stupid little fun book.", "loss": true }, { "from": "human", "prefix": "Brady Haran:", "value": " Yeah.", "loss": true }, { "from": "gpt", "prefix": "Hannah Fry:", "value": " But then the latest one, the Hello World one…", "loss": true }, { "from": "human", "prefix": "Brady Haran:", "value": " Yeah.", "loss": true }, { "from": "gpt", "prefix": "Hannah Fry:", "value": " I sort of see that to be honest as my first book.", "loss": true }, { "from": "human", "prefix": "Brady Haran:", "value": " Right.", "loss": true }, { "from": "gpt", "prefix": "Hannah Fry:", "value": " Because six weeks to write a book is just like a pamphlet, not a book. And then the other one was just me mucking around with someone.", "loss": true }, { "from": "human", "prefix": "Brady Haran:", "value": " They're sort of more like stocking filler type fun books?", "loss": true }, { "from": "gpt", "prefix": "Hannah Fry:", "value": " Yeah, totally, I mean they're very thing, right? They're like twenty-five thousand words or whatever.", "loss": true }, { "from": "human", "prefix": "Brady Haran:", "value": " Yeah.", "loss": true }, { "from": "gpt", "prefix": "Hannah Fry:", "value": " Whereas Hello World is like… I mean it's still quite, well I hope anyway, that it still got that chatty style and there's still like some silly bits in there, it's not completely serious, not like it's really serious heavy read. But it's not just much more the things that I'm actually interested in, in terms of research that I do. And also I just think it's got some really good stories in it.", "loss": true }, { "from": "human", "prefix": "Brady Haran:", "value": " Where did you write it? Where do you sit and write?", "loss": true }, { "from": "gpt", "prefix": "Hannah Fry:", "value": " On a walk, mate.", "loss": true }, { "from": "human", "prefix": "Brady Haran:", "value": " You write it on a walk?", "loss": true }, { "from": "gpt", "prefix": "Hannah Fry:", "value": " Yeah. (laughs)", "loss": true }, { "from": "human", "prefix": "Brady Haran:", "value": " But you can't type on a walk.", "loss": true }, { "from": "gpt", "prefix": "Hannah Fry:", "value": " You can dictaphone though can't you?", "loss": true }, { "from": "human", "prefix": "Brady Haran:", "value": " Oh, you dictaphone it!", "loss": true }, { "from": "gpt", "prefix": "Hannah Fry:", "value": " Yeah, so I'd go for a walk with my doggie. For a really long walk and I'd dictaphone loads.", "loss": true }, { "from": "human", "prefix": "Brady Haran:", "value": " Oh.", "loss": true }, { "from": "gpt", "prefix": "Hannah Fry:", "value": " And then I come back and I type it up and make it into sentences and then pad it out and do the same next day.", "loss": true }, { "from": "human", "prefix": "Brady Haran:", "value": " Have you still got the dictaphone recordings?", "loss": true }, { "from": "gpt", "prefix": "Hannah Fry:", "value": " Probably, yeah?", "loss": true }, { "from": "human", "prefix": "Brady Haran:", "value": " Will you give us like… fifteen seconds so we can hear what that sounds like?", "loss": true }, { "from": "gpt", "prefix": "Hannah Fry:", "value": " (laughs) Yeah but my sentences are really slow.", "loss": true }, { "from": "human", "prefix": "Brady Haran:", "value": " That's okay!", "loss": true }, { "from": "gpt", "prefix": "Hannah Fry:", "value": " So I'll be like…", "loss": true }, { "from": "human", "prefix": "Brady Haran:", "value": " I'd love to hear what it sounds like, you know, for a peak behind the curtain.", "loss": true }, { "from": "gpt", "prefix": "Hannah Fry:", "value": " (laughs) There'd be heavy breathing as I'm walking up hill. (laughs)", "loss": true }, { "from": "human", "prefix": "Brady Haran:", "value": " Yeah!", "loss": true }, { "from": "gpt", "prefix": "Hannah Fry:", "value": " (laughs)", "loss": true }, { "from": "human", "prefix": "Brady Haran:", "value": " I'm gonna lobby you for that. I'm gonna lobby you for a little…", "loss": true }, { "from": "gpt", "prefix": "Hannah Fry:", "value": " Okay, I'll see if I can find one.", "loss": true }, { "from": "human", "prefix": "Brady Haran:", "value": " If you can find some, just give us a little snippet and we'll just play it now so people can get a feel for that.", "loss": true }, { "from": "gpt", "prefix": "(low quality dictaphone recording of Hannah Fry plays):", "value": " On the one hand it means that we don't… we kind of feel surprised when things get stuff wrong… and on the hand it means that we are dismissive… when things go wrong. That's what we're just really bad at that stuff. Because we're not really good at calculating, but I sort of think actually the more interesting and more accurate way to frame it (bird chirps) is to have that as a secondary idea. (audio distortion) And instead have how much power should we give the… the machine. (breathing and audio distortion) (bird chirping) Part one is the algorithm gets final say, part two is that the human gets final say. (bird chirping)", "loss": true }, { "from": "human", "prefix": "Child on Recording:", "value": " Bye bye doggie! (laughs) (audio distortion) (recording ends)", "loss": true }, { "from": "gpt", "prefix": "Brady Haran:", "value": " The parts of the book that I've read, what I like about it, is it does sound like you're talking to me, like it's like, oh yeah this is… obviously…", "loss": true }, { "from": "human", "prefix": "Hannah Fry:", "value": " Because I was. (laughs)", "loss": true }, { "from": "gpt", "prefix": "Brady Haran:", "value": " Yeah! That… it all makes more sense now. That's why I was gonna say you're so good at writing in your voice, your chatty voice, it's because you are, it is your chatty voice. How has it been going, like, how was it received? I know it's been nominated for some awards and stuff…", "loss": true }, { "from": "human", "prefix": "Hannah Fry:", "value": " Yeah it has.", "loss": true }, { "from": "gpt", "prefix": "Brady Haran:", "value": " …so it's obviously, you know…", "loss": true }, { "from": "human", "prefix": "Hannah Fry:", "value": " I has. No, it's been really good actually. It's been really amazing. 'Cause I think… (laughs) one thing that I do find about it is that, I think because I have done quite silly stuff in the past, like the love stuff and, you know, telly stuff, whatever, I think one thing that I've noticed is that people think I don't know what I'm talking about.", "loss": true }, { "from": "gpt", "prefix": "Brady Haran:", "value": " Right?", "loss": true }, { "from": "human", "prefix": "Hannah Fry:", "value": " It's sort of back to what we were saying earlier.", "loss": true }, { "from": "gpt", "prefix": "Brady Haran:", "value": " Yeah.", "loss": true }, { "from": "human", "prefix": "Hannah Fry:", "value": " About walking into a room and people thinking you're an idiot. I don't think people think I'm an idiot but I think they don't…", "loss": true }, { "from": "gpt", "prefix": "Brady Haran:", "value": " Yeah.", "loss": true }, { "from": "human", "prefix": "Hannah Fry:", "value": " I think people are a bit like why is this TV presenter writing a book about AI?", "loss": true }, { "from": "gpt", "prefix": "Brady Haran:", "value": " Yeah.", "loss": true }, { "from": "human", "prefix": "Hannah Fry:", "value": " You know, she's jumping on the bandwagon or whatever.", "loss": true }, { "from": "gpt", "prefix": "Brady Haran:", "value": " Yeah, yeah.", "loss": true }, { "from": "human", "prefix": "Hannah Fry:", "value": " But it's not really, I mean it's stuff that I've been working on for a really long time. It's been nominated for two awards, right, shortlisted for two awards. And one of them was the Royal Society, which is… like a really big deal because when it comes to science writing, you just don't really get a bigger institution than the Royal Society.", "loss": true }, { "from": "gpt", "prefix": "Brady Haran:", "value": " Hmm.", "loss": true }, { "from": "human", "prefix": "Hannah Fry:", "value": " And you know the books that they pick are like, these weighty, you know…", "loss": true }, { "from": "gpt", "prefix": "Brady Haran:", "value": " Tomes.", "loss": true }, { "from": "human", "prefix": "Hannah Fry:", "value": " Exactly.", "loss": true }, { "from": "gpt", "prefix": "Brady Haran:", "value": " (chuckles)", "loss": true }, { "from": "human", "prefix": "Hannah Fry:", "value": " I mean they really are it's like, you know, a history of engineering, right?", "loss": true }, { "from": "gpt", "prefix": "Brady Haran:", "value": " Yeah. (laughs)", "loss": true }, { "from": "human", "prefix": "Hannah Fry:", "value": " Like that was one of the other books. Or like, what I find quite intimidating books.", "loss": true }, { "from": "gpt", "prefix": "Brady Haran:", "value": " Yeah.", "loss": true }, { "from": "human", "prefix": "Hannah Fry:", "value": " And why it was a real surprise it was because exactly as you said the book that I've written is really chatty. It's not written in the style of a scientist, really.", "loss": true }, { "from": "gpt", "prefix": "Brady Haran:", "value": " Yeah.", "loss": true }, { "from": "human", "prefix": "Hannah Fry:", "value": " So it's really that was really amazing to get that recognition. And then the other one, the other big prize that it's been nominated for is this non-fiction, like, the biggest non-fiction award which is anything in the English language so I think there's only a couple of English authors on there, which is frankly absurd. I mean science books almost never even get on this list so it's…", "loss": true }, { "from": "gpt", "prefix": "Brady Haran:", "value": " Brilliant.", "loss": true }, { "from": "human", "prefix": "Hannah Fry:", "value": " It's completely absurd.", "loss": true }, { "from": "gpt", "prefix": "Brady Haran:", "value": " Brilliant.", "loss": true }, { "from": "human", "prefix": "Hannah Fry:", "value": " But you know, it's actually I did try really hard. (laughs)", "loss": true }, { "from": "gpt", "prefix": "Brady Haran:", "value": " Well, there you go, you did something right. So you get to dip your oar into a lot of the different types of outreach now. You've got like, you know, radio and podcasts which are really successful, you've done these television shows, a book, occasionally you even get to be in Numberphile videos…", "loss": true }, { "from": "human", "prefix": "Hannah Fry:", "value": " (laughs)", "loss": true }, { "from": "gpt", "prefix": "Brady Haran:", "value": " …which is amazing.", "loss": true }, { "from": "human", "prefix": "Hannah Fry:", "value": " You know, actually though…", "loss": true }, { "from": "gpt", "prefix": "Brady Haran:", "value": " (laughs)", "loss": true }, { "from": "human", "prefix": "Hannah Fry:", "value": " …in terms of people like coming and talking to me in the street, it's almost always Numberphile people.", "loss": true }, { "from": "gpt", "prefix": "Brady Haran:", "value": " Ah, there you go.", "loss": true }, { "from": "human", "prefix": "Hannah Fry:", "value": " They're the best people.", "loss": true }, { "from": "gpt", "prefix": "Brady Haran:", "value": " Well, keep doing them then. So of all of those, and now there is no offense to be taken here, what gives you the most pleasure? Like, what are the ones like if I said, alright you can't do any of it anymore except one, what's the one that like, you know, feels most in your wheelhouse? That you enjoy doing?", "loss": true }, { "from": "human", "prefix": "Hannah Fry:", "value": " So I… and you know, well I actually have this conversation quite often with my husband who's like, you need to give up something. (chuckles)", "loss": true }, { "from": "gpt", "prefix": "Brady Haran:", "value": " Yeah?", "loss": true }, { "from": "human", "prefix": "Hannah Fry:", "value": " (laughs)", "loss": true }, { "from": "gpt", "prefix": "Brady Haran:", "value": " He's like, stop those Numberphile videos, mate!", "loss": true }, { "from": "human", "prefix": "Hannah Fry:", "value": " (laughs) But the problem is that actually, each of them have their own merits.", "loss": true }, { "from": "gpt", "prefix": "Brady Haran:", "value": " Hmm.", "loss": true }, { "from": "human", "prefix": "Hannah Fry:", "value": " And I think that they all sort of add to this bigger picture, so you know, the radio stuff, I just absolutely love the team that I have. So, you know, it's a podcast called the Curious Cases of Rutherford and Fry, with Adam Rutherford the geneticist. And it's just me, him and this producer, and we just have the best time making these programs. It's just I wouldn't give it up for anything and just absolutely love it. The telly stuff is like, those are times where you get these incredible experiences. I mean it's hard work don't get me wrong, right?", "loss": true }, { "from": "gpt", "prefix": "Brady Haran:", "value": " Yeah.", "loss": true }, { "from": "human", "prefix": "Hannah Fry:", "value": " And you do not make your fortune in television. (chuckles)", "loss": true }, { "from": "gpt", "prefix": "Brady Haran:", "value": " No.", "loss": true }, { "from": "human", "prefix": "Hannah Fry:", "value": " Believe me.", "loss": true }, { "from": "gpt", "prefix": "Brady Haran:", "value": " That's about the access to…", "loss": true }, { "from": "human", "prefix": "Hannah Fry:", "value": " Yeah but it's about the experiences and the access and then…", "loss": true }, { "from": "gpt", "prefix": "Brady Haran:", "value": " Hmm.", "loss": true }, { "from": "human", "prefix": "Hannah Fry:", "value": " You know, being part of something that eventually gets seen by so many people and that just looks really beautiful and glossy on the surface and…", "loss": true }, { "from": "gpt", "prefix": "Brady Haran:", "value": " Hmm.", "loss": true }, { "from": "human", "prefix": "Hannah Fry:", "value": " Yeah, having that role, I think, is really… I wouldn't want to give that up.", "loss": true }, { "from": "gpt", "prefix": "Brady Haran:", "value": " Yeah.", "loss": true }, { "from": "human", "prefix": "Hannah Fry:", "value": " And then the Youtube stuff is where you can actually be yourself, right?", "loss": true }, { "from": "gpt", "prefix": "Brady Haran:", "value": " Yeah.", "loss": true }, { "from": "human", "prefix": "Hannah Fry:", "value": " And then the books that's where you get to… really author something.", "loss": true }, { "from": "gpt", "prefix": "Brady Haran:", "value": " Then your husband says, well you gotta give one of 'em up! (laughs)", "loss": true }, { "from": "human", "prefix": "Hannah Fry:", "value": " (laughs) Yeah.", "loss": true }, { "from": "gpt", "prefix": "Brady Haran:", "value": " (laughs)", "loss": true }, { "from": "human", "prefix": "Hannah Fry:", "value": " You can sleep when you're dead.", "loss": true }, { "from": "gpt", "prefix": "Brady Haran:", "value": " (laughs)", "loss": true }, { "from": "human", "prefix": "Hannah Fry:", "value": " (laughs)", "loss": true }, { "from": "gpt", "prefix": "Brady Haran:", "value": " So, like to finish up then, like, in… five years time what's the next thing?", "loss": true }, { "from": "human", "prefix": "Hannah Fry:", "value": " Oh I don't know. I don't know. I think everything's worked out quite well so far by being micro-ambitious.", "loss": true }, { "from": "gpt", "prefix": "Brady Haran:", "value": " Yeah?", "loss": true }, { "from": "human", "prefix": "Hannah Fry:", "value": " So just looking at what's immediately in front of you, and doing the very best that you can at that and trying your hardest.", "loss": true }, { "from": "gpt", "prefix": "Brady Haran:", "value": " Is mathematics still a big part of it? Like do you think in ten, fifteen years you're still gonna be someone who's like churning out papers and researching, or like…?", "loss": true }, { "from": "human", "prefix": "Hannah Fry:", "value": " I hope so, I hope so. So I mean as I said, you know, I'm just about to go on maternity leave again, so like in the last couple of years I haven't done as much maths as I'd want to. And I won't again for another year or so.", "loss": true }, { "from": "gpt", "prefix": "Brady Haran:", "value": " Hmm.", "loss": true }, { "from": "human", "prefix": "Hannah Fry:", "value": " Because, you know, the new baby. But when I come back from that maternity leave, yeah I really want to get back into it, because there's just nothing in the world gives me more pleasure that getting totally stuck into a totally delicious maths problems.", "loss": true }, { "from": "gpt", "prefix": "Brady Haran:", "value": " Well… Hannah Fry, public intellectual. (gentle music fades in)", "loss": true }, { "from": "human", "prefix": "Hannah Fry:", "value": " No! (laughs)", "loss": true }, { "from": "gpt", "prefix": "Brady Haran:", "value": " Thought leader.", "loss": true }, { "from": "human", "prefix": "Hannah Fry:", "value": " (laughs) Brady!", "loss": true }, { "from": "gpt", "prefix": "Brady Haran:", "value": " Thank you for your time today. (laughs)", "loss": true }, { "from": "human", "prefix": "Hannah Fry:", "value": " (laughs)", "loss": true }, { "from": "gpt", "prefix": "Brady Haran:", "value": " Thank you very much.", "loss": true } ]

[{"from":"human","prefix":"Brady Haran:","value":" So Matt I know you're like stand-up comedy. You'r(...TRUNCATED)

[{"from":"human","prefix":"Brady Haran:","value":" Do you recognize this tune?","loss":true},{"from"(...TRUNCATED)

[{"from":"human","prefix":"Brady Haran:","value":" Today we're talking with set theorist Asaf Karagi(...TRUNCATED)

[{"from":"human","prefix":"Summary:","value":" Edward Frenkel's fascinating mathematical journey - f(...TRUNCATED)

[{"from":"human","prefix":"Summary:","value":" Mathematical greatness can strike at any time - even (...TRUNCATED)

[{"from":"human","prefix":"Summary:","value":" David Eisenbud's entertaining stories about mathemati(...TRUNCATED)

[{"from":"human","prefix":"Summary:","value":" From Illinois to Cambridgeshire, Holly Krieger’s pa(...TRUNCATED)

[{"from":"human","prefix":"Summary:","value":" Experimental physicist Phil Moriarty works with tempe(...TRUNCATED)