Lessons in innovative Olympic data journalism | #RISJSeminars with Alberto Cairo and Simon Rogers, The Data Journalism Podcast
2:30PM Jun 19, 2024
Speakers:
Caithlin Mercer
Keywords:
data
data visualization
olympics
work
visualization
journalism
graphics
ton
simon
question
add
sports
book
create
project
flourish
journalists
place
ai
talk
Welcome to the Global journalism seminars. This is the briefing. In just over a month's time, the Summer Olympics will return to Paris for the first time in 100 years. Thanks to the integration of advanced timing systems and AI driven analytics, journalists covering the Paris Olympics will have a wealth of data to analyze and present offering an opportunity for rich narratives that go beyond the immediate excitement of the competitions. We polled our journalists fellows to ask which area of data journalism do you think will have the most impact on storytelling during the Paris the Olympics, a third said athlete performance statistics and analytics, only 11% said engaging audiences through interactive features. Our guests today are Alberto Cairo and Simon Rogers boasts of the data drums and podcast. By day, Alberto was Knight chair in infographics and data visualization at the University of Miami School of Communication. And Simon is data editor at Google News labs. Their previous work includes a leading data visualization innovation at news outlets like comando and The Guardian, realized about opportunities to surprise and delight audiences with data journalism that goes far beyond the metal table. That's the briefing. Let's begin.
Welcome to the penultimate global journalism seminar of the 2023 24 academic year and I'm so very pleased to have joining us from Miami of Berger who is nature and infographics and data visualization at the University of Miami. Their school of communication. He previously worked as director of infographics and multimedia at editorial, global and Brazil and served as head of visualization at Spanish daily el mundo and He's authored several books on data visualization that are now used as textbooks in the field. And he's known for his advocacy for ethical inaccurate data representation. Welcome, Alberto.
Thank you. It's a pleasure to be here.
Thanks for giving me a chance to cough and joining us from San Francisco. Simon Rogers is data editor at Google News labs, where he works on innovative ways to use data in storytelling. Before that Simon was the founding editor of the Guardians data blog, and he once worked at Twitter as data editor. Simon is a champion for open and accessible data for all. And together Alberto environment present the data journalism podcast where they interview the world's top data journalists and explore how data is changing the world of journalism. And Simon we just have to ask for an introduction to our special star guest who is waiting fluffy
biscuits he's on the lookout for deliveries excellent and get your obviously dragons pretending to be postal workers. This is the problem so he's, he's he's there but he's now sleeping. So you know hopefully.
He's our most common guest.
Yes. Is this true?
Excellent. All right. Well, let's start for for those who aren't familiar with covering the Olympics, maybe there are people joining us today who are going to be covering their first ever what kind of data do the games generate? And where do newsrooms get it from Simon? Is this something I could lob in your direction?
Absolutely. I can. I can make a pass at it. Because it's complicated. And part of the reason it's complicated is because the IOC owns an enormous data set, which is obviously this real time feed of Olympic kind of achievement as it happens. And a lot of the newsrooms that with the bigger newsrooms obviously have that fee they subscribe and so on. And so for people like Reuters or Bloomberg, it's they've got a treasure trove of tastes. Like out there. I work at Google. And Google obviously shows Olympic results as they happen. So that kind of data is there is proprietary. If you don't have a big budget, maybe it's harder to get hold up. But there is a ton of other stuff out there. I think that makes the Olympics particularly interesting for the data storytelling, and hopefully we're getting some of that.
Yeah, how much roughly are we talking to you? No. Okay, no problem. But we can probably guess pricey ish. But you can still you could still start a Google sheet and collect your own data, as
well. That's, that's the thing. And I would add that sometimes the more read more the most interesting data about a particular story is not the data directly related to the story not not data related to the performance of athletes for example, but data that is related perhaps to the socio economic elements related to the to the Olympics and that type of data can be obtained. A I will not say easily but relatively easily, certainly much more recently than data related to the Olympics. And I think that there are a lot of stories that perhaps could be told a connecting the Olympics to the circumstances of the country where the Olympics take place, and also the circumstances of the places where the athletes come from.
What's the most creative non sports story that you've you've seen? told via data?
Yeah, absolutely. I think that, do you want to go first
Yeah, I will. I will go first I guess. Well, that's super difficult to answer because there's so much great work a been producing the civilization not only within the context of the news, but also in other in other in other areas. So not to do product placement here. But my most recent book, The Art of insight is essentially an exploration of the huge variety of expressions that visualization can adopt, right? The first, the first two sections of the book are titled, The pragmatists and the second one is about people who try to push the boundaries of a of data visualization in terms of artistic expression, and things like that and they are made you can find examples of examples of visualization that are absolutely amazing. For example, the one that opens up the book is an exploration of a how the COVID pandemic impacted the behavior of a common friend of Simon and I, a, who actually will who have collaborated with us in the past. We essentially create this visualization of the photographs that she took before, during and after the pandemic, and she represented the data related to those photographs using these flower like visualizations that are extremely expressive and evocative. There are also examples of people who use physically sation to represent data objects to represent data who model data on an on physical space, people who saw Nephi data transforming numbers into sound and music, which is an area that Simon and I have explored in the past helping develop a tool to certify data a call to turn. I mean, I could go on and on and on. I could share my screen and show you examples but that will take the entirety of the aim of the conversation.
And I think, like, in my addition to that would be particularly in relation to the Olympics. So the stuff that really resonates for me is stuff that kind of appeals to our humanity. So there are a revisional artists out there who might not be proud but who are really good at producing work that feels accessible, inhuman, that is not these numbers aren't abstract. And obviously the Olympics in a way is the most human data of all right, because it's about people's performance and some of the pieces the Olympic pieces I think, are really done this have kind of put you in the athletes place like could you do what Simone Biles does, right? Those kinds of things, or how would you do that? How, how does somebody you know, do the thing, the thing that she does so well, like, what are those moves, how to make it happen that the physics and geography of it and I think the times in particular perhaps because of the amazing people who are working there are particularly good at that.
You've got to I was I was kind of erasing my hand. I don't know.
Just talking but
anyway, so yeah, I mean, what time would have just described is a quite close to something that I have written down here in my notes, which is that the Olympics are intrinsically physical, but we often represent them through a graphics that are very abstract and sanitized like traditional data visualizations, bar graphs, line graphs, maps, and it is not that we need to stop doing that or there's certainly a place for traditional data visualizations in the representation of Olympics data, but I would like to see more a not necessarily graphics but representations that try to mimic that physical nature of the event. That's a reason why I refer before to data data physically, physical location, and then as Simon before mentioned, graphics that tried to put you in the place of the athletes. Not only that, but trying to use a game like approaches to the representation of the of the information. This is not something particularly new, by the way. When I was at Moodle, and I have a couple of examples here, very old graphics, very old, a meaning that they were made in 2004. So 20 years ago, but they still look relatively current. So there were graphics that made you play a with the with the application. So try to throw a try to shoot the arrow and hit the target or something like that, right. Besides explaining to you that particular sport. Then he led you put yourself in the place of the athlete, the athlete trying to hit the trying to hit the target and that type of engagement, putting the reader at the center or putting the reader in the control of the visualization that that is somehow I have a feeling that increases engagement on the part of the reader.
Let's take a cue from your book Alberto and deal with Chapter One pragmatism and then chapter two, let's let's go into the unexplored territory, but pragmatism, what are the what are the data visualizations that are expected and has to happen? Around the Olympics? And do you have any tips on how to make the expected have to happen? I'm thinking metal tables more engaging.
I think that's better, isn't it? Oh, yeah.
Either. Either of you. Please feel free to jump in and I
have some thoughts while but yeah, go ahead. Send me. I'll know what I'm thinking about this. Because I was uh, you inspired me to kind of look through a lot of the stuff we did back in the day when I was at The Guardian. And basically, think about the Olympics is really it's a set of moments. Right? At least for those moments. Produces expect stuff. So you've got the torch relay visualizations around the torch relay. There's a ton of data out where's it going? What's what's what's the you know, the demographics of the places the torch is going to the physical geography of, of getting it to the to the location, I think there's a ton of stuff there. And then you've got like, one thing we did was how Britain was different when it was hosting the Olympics back to the previous time in hosting, what's what's changed about the place. Now this is like an Olympic stage, right? This is stuff that you could just do stuff with. So that again, is not as another kind of moment and then when so by the time you can get into the sports you there's a lot of stuff you can do but the in terms of the expected over the metal tables. You know, the obvious thing when we've done the obvious is to look at what if, what if, how would you balance out those metal tables? What if What if they reflected population or their reflected GDP? Or yeah, like Fiji or something like that, which always gets a ton of males for a tiny place, right? You know, what have you can kind of add context to those basic things you've got to do. You've got to have mental total Sure. But then you can add context and information to like we even did something where we looked at the Team GB medalists and their kind of economic socio economic backgrounds. So you kind of building on the data that's already there. Maybe I'm jumping ahead for that question, though.
I have a little bit more to add to that. So it's in the answer to that. Question, which is that I mean, I come I come from a tradition of visualization that is very grounded on art and on illustration that began my career in 1997. So at that time, there was no talk about data visualization at all right? PII data visualization. Really, a became really popular in the news media around 2008 910, something like that, but I was already working in this industry for more than 10 years before that. And there is a huge tradition of illustration driven graphics that in part has been lost because of the a dominance of data visualization, again, the sanitized representation of information, the lack of physicality in the representation of information. I would like to see more graphics that use may actually send actual hand drawn illustrations, even to represent data so you do the metal metal table. You know, don't throw it in Adobe Illustrator and create these sort of like super shiny objects do it by hand I let the imperfection of the representation adds flavor. And that's personality to the display. A that you're creating.
We were discussing that before we went live how you've become more attractive in an age of AI to what is handcrafted.
Yeah, I've always been interested in handcrafted things. I still draw maps by hand for example. And then But then I call them in Photoshop. So I use, you know, I use a sort of like half and half in terms of techniques. As a reader, right, this is not as a designer or as a data journalist, but as a reader. I like to see things that have personality and uniqueness. So every bar graph is identical to any other type of bar graph unless that you add something that is truly yours, either in terms of the data that you use, or the angle that you use to analyze the data, or in terms of the standard you use. To represent the data is style is really important. In data visualization, there has been for quite a while, what I call in my most recent book, The Art of inside the Tahitian tradition, after Edward Tufte to you is a very a influential data visualization historian. And the problem with that is that Tufte he has a very personal view of what data is salutation should be highly sanitized, hyper precise, you know, always represented the data first and there is value in that tradition. But at the same time that traditional also overlooks the importance of, again, unique uniqueness and style, in graphics and personality and imperfection. Sometimes in order to represent let's say, the uncertainties of the data, the perfection of the representation, it mimics that sort of underlying uncertainty as well and hand drawn illustrations and drawn graphics, or graphics or tried to mimic that style. That imperfect style, created by the human hand, have a lot of value in my opinion, and B they become really warm. So to speak. Data visualization is very cold, very cool, but things that are made by hand stand crafted, they become more valuable.
So that's such a kind of I'd be sending plus ones emojis to that person I agree about a lot of stuff this is a problem. But um but you know that feeling of of data being personally human is so important. Because basically, you know, most people you're who you're drawing for Who are you designing who you keep telling datasource for? Is it for your colleagues, because you want to show off, essentially, or is it for for your readers and users alike and how accessible Do you want to be? And so like this, this cover is a really beautiful handrolled graphics, things like the stuff the Manage calibrators like which are incredibly powerful visuals, which are really basically her you know, but they're very accurate. You know, you can you can measure those that had her charts against like bar charts and line charts and so on and they are exactly accurate but they're also human and accessible and that's with this like the Olympics is the most human accessible thing there is it's in our living rooms we're seeing this the whole time we kind of suddenly discovered lose beat we didn't know in each field. So yeah, what is that about them that makes them who they are and how they get there.
I'm hearing that there's room for emotion in Yeah. And visualization. The tough tn numbers clinical model is not fit for a digital age as much as it was.
I think I think there's like there's a tendency to you know, you as your work being designed by computer or has it been signed by you as a journalist or you tell you tell the story of his computer time that story, in which case you know what you're doing? Yeah, I think I think it's bringing yourself to the data isn't abstract. It's not like something that has nothing to do with people that you humanity generally. So bringing that in specially with a story like the Olympics because it covers so many different right. It's about it's about socio economics, it's about sports. It's about human achievement. It's about politics. It's about cash, you know, it's about everything.
Yeah,
so getting that across and you can remember there was a thing The Guardian did a few years ago, where the gamified and this little Olympics game, like could you be an Olympic runner? You were like, there was no eight bit a better Olympics. Okay, great. It's fun. And it's there's also real data behind it. So yeah, yeah. Okay,
um, one more question under the pragmatic column and then let's go and look at the weird and wonderful, the weirder, the more wonderful that will be. But if you were sitting together in a newsroom, now, you've got 37 days before the Paris Olympics kickoff. What what is the team that you bring together and what are the questions? That you're asking as a team to plan the data visualizations that will surprise and delight your audiences?
I love this question, because I was. There's also the inference that there is a perfect team. I think if we do that, then we would, we would know everything. I think I think it's very personal to whoever's there. What do you think about I think it would be like, you definitely have to have people like you'd have your your Jamba Murdock's, who are kind of basically data, visual data journalists, I think you'd have, you'd have to have a coder and a design and add somebody who's good. Good. At Python and everything. What do you think about what am I missing? Yeah,
I mean, a couple of developers obviously to the with the with the coding, I mean, it happens. It all depends on what you really want to focus on. If you want to do a pure data driven operation, then obviously a front end or to front end designers, someone who want people to a person or two who can analyze data and who are a have domain specific knowledge because that's super important to analyze data. You cannot analyze data in the abstract, you need to know where the data comes from, what is generated, what is being represented, who is being counted. So you know, you need analysts who have that type of knowledge. A back end, again, developers you need developers as well. But again, it greatly depends again on what the type of graphics that I want that you want to create. For example, if the team were mine, I would add one or two visual artists, who, again, who can add this flavor, this style, this personality to the representation of the data and by visual visual artists. I mean, people who draw and who paint are mortal and who can bring new ideas to the table, something that has not been done before, or people who do sculptures. I mean, I don't know it might be possible to the data, sculptures, all these things I try to bring people again creatives who can help us again, push the boundaries a little bit. And then obviously, you need reporters. So you need people with a journalistic eye. Now that this doesn't mean that you need I'm talking more about tasks, rather than people it might be possible that a person is really good as a developer and also as a front end designer, and then that person can fulfill two different roles at once.
Okay, so in this complete Narnia, worlds that were
Can I just kind of just add to that, though, yeah, go on. So sorry. Sorry. So I think this is all great. I think the fun component is almost like the most important thing, but also what's important for you as an organization. So if you're a small newsroom, you don't have resources. What is important for you, is it to relay metal information? Is it to have like, a commentary take on it? Is it to have like a visual fun, something that will go viral or something that people will love on Tik Tok? Or is it what is it that you're trying to do? And I think sometimes people just dive into into ENT and you know, a million flowers bloom without really thinking about what's strategic for them. So stepping back, which I'm guilty, I'm totally guilty of that. So I'd like to stepping back and thinking what am I actually trying to do? Do I want to be like, the social the social impacts organization? Are we the social Olympics campaign everybody looks at and just really focused on that do that properly and well, I'm not trying to do everything, because you can't do everything successfully because you're setting yourself up for the disaster, I would say. So.
What are we trying to achieve? What are other people doing that that that and that we don't want to replicate? Yeah,
I mean, say you say there's like an Olympic medal feed that you can get, and it's fine. It's not the most beautiful thing. Well, it's fine. Do you want to spend the time and effort to make that really beautiful? Or do you want to do something a particular angle or focus on like, you know, medalists heights or something I don't know some of that's fun. Who said that? I think having those conversations before you start will really shame kind of team you have like like a better said having like having a visual artists or somebody could do it. You can just draw stuff that's, that's That's lovely. Maybe that'll have a bigger impact for you. If you get like a ton of social engagement, that maybe that's more valuable for you than getting like a 2% uplift on Olympian pages. Yeah.
So maybe it's defining the questions before you define that. Yeah. Good. Good point,
about the type of organization that you are. If I had to put together a team in design, thinking about what are you a general newspaper or words publication or are you Simon said before a commentary application, I would think differently about this project. If I were working, let's say for a Harper's Magazine, or that they're gonna do a lot or the New Yorker data visualization, but if they did, I will produce different kinds of visualizations for The New Yorker than I would do for let's say, a Mundo which is a general or, or employees in Spain, or more general readership newspaper, okay,
but if you're if your focus is like Gen Z, Gen Z, I can't say see Gen. Zed, young people, like maybe you're wasting your time doing things that weren't great on desktop. You know, maybe that's a complete waste of time. It's nice because we all want that feeling that we get from the paper. Full page. But ironically, like something suddenly that's in print on a paper for paper, it's beautiful, might have more impact for that audience than some of the works on the desktop in ways that people just don't use.
Okay, so we're a magical mobile first young people targeting app news product funded heavily by Jeff Bezos. So we've asked these questions and we bought together a team with a couple of developers couple of front end designers, couple analysts, clinical journalists, couple visual artists. What are the questions we need to ask now? As we have our first team meeting, about how we approach the Olympics at I'm going to call it Dream Dream news outlet.
I think there is something to thinking about what can I do right now there's going to have an impact now why did you know obviously the stuff we're gonna be doing down the road, but But you kind of want to establish yourself as the place people go to a whatever that kind of baby is that global youth focused? Olympic journalism, whatever it is, you want to if you want to stop yourself that what can I do on both runs now? What can I do that simple, but effective? Like right now? Is it like, where the money's coming from? Go do something because something today about how the Olympics are being are being funded and where's that money's come from where it goes. Power sponsorships, stuff like that. There's stuff that's happening right now. Yeah, all what training has happened? Yeah. Who is how much money is each government putting into training and stuff and there's just like, there's a ton of pre stuff you could do. You could do the historic, the historic stuff. Right? You've got like, you know, all those metal tables going back decades, right. All of that you pretty much grab that from Wikipedia right now. Yes, do with it. That would be my first question. Then start planning ahead. Like what is the stuff you're gonna do during the games themselves that is unique to you? As that's different types versus doing that makes sense, rather than just taking a wire feed.
How do we find the pack? Yeah,
like, people are looking for you. Nice to tell that story.
I was going to add that a good approach that meetings sort of like in two different directions. One of them is to first think about what needs to be done and what can be automated on the one hand, so it's like what what type of service we need to provide to readers in terms of what can be quantified and percentage. It's really the metal table. The stats are really to athletes and the different categories that's sort of like the bread and butter of covering the covering the Olympics, and it can be highly automated. But then the most interesting part my opinion, and I mentioned this to you before the meeting began, and also sharing social media might lead to lack of interest for sports in general. But I'm extremely interested in everything that surrounds sports. And the Olympics are always a reflection of a particular place and a particular time. So how the Olympics have changed historically, reflecting the times right I think about the the Munich Olympics, for example, in the 30s, right, yeah, that was a particular representation of at that particular moment in time. The social political, economical economic situation was somehow reflected the Olympics are both a reflection and also a product of their time. So how the current Olympics reflect the current times. For example, there is an increasing increase, increasing interest in environmental issues. There's increasing interest in equality, increasing interest in LGBTQ issues, right. And this is something that I think that is very a very much on the news these days, sometimes for very, very bad reasons. Because some big organizations have fueled a moral panic against LGBTQ people, particularly trans people, and I don't want to name names, but it will the New York Times, for example, very recently. So yeah, we will look anywhere right. So there is something to be said about representation in the Olympics. How do they reflect the present moment? And also the time and the place right to to how do they reflect the city or how is the city going to react, how the place where the Olympics are going to take place a or how they're going to be changed? Or how are they been changed? Is that changed positive? Is it negative? Has it been positive or negative in the past, providing that type of context, I believe is also quite important. Love
it. I'm going to ask you one more question and I'm going to go to questions is submitted via q&a and questions from the room downstairs. So bring on your questions. But last question from my side. Parents has been very loudly marketing the idea that these are going to be the most sustainable games ever. How would you use this team to use data visualization to either monitor those promises or tell the story of the delivery of those promises?
That's a really interesting question. I think there is like one thing I would look at like early would be what the legacy has been in different cities, while the promises were promises, made promises where they kept, you know. Yeah, and I think you've got some great examples like London, I think where I think actually a lot of that stuff is still very much in use, but in other cities where it's not. So there's something about that I think there's something about you could do like a carbon Olympics, can you like, you know, compare each game to how much carbon I'm sure there's like some way you can estimate that. One thing I was thinking about the kind of the, the youth angle, I mean, looking at a lot of the ways that the a lot of really good people use Tiktok around data is so interesting, where they'll have like a screenshot or something behind them and then talk to it like a human. And I think maybe there's something you could do with that. There's also I wonder if there's a way if you had the money, so you like the New York Times, right? And money's not an issue. You they did this piece around air pollution a few years ago where on your phone you could see particles floating in the air for different cities. What they could be in, like in front of you. It's very cool like this is California during the fires. This is Delhi today and very neat and something like that, which you might be able to monitor real time because you can monitor a QI and stuff. There's things that you can monitor the Olympic things. During the game you can you can monitor because they can be a ton of people just turning up the obese flights or people going there and it's like there is like a ton of ways I would love to know what measures the Paris Olympics team to define our sustainability and what are the what are they deciding counts? And then hold on to that.
Yeah, I think I think that basically didn't like that idea. Yeah,
he's he's not a fan of pollution.
The best I haven't smiled this much in the seminar ever before. I think it is my new official support. What is it called? The support animal and he needs
his nephew. He actually does have an Instagram account, which is mostly him on the beach.
I was going to add by the way before I forget as long as we talk about pollution that brought back to memory, a project by Reuters a in from 2018 titled a window into Delhi's deadly pollution. That is essentially what they did was to position a camera on the same on a roof for a period of time. I don't remember whether it was a month or something like that. And they took a photograph every single hour, something like that. And you could see sort of like pollution changing over the day, how we didn't get how we small increases and decreases throughout the day. And they represented that graphically through line charts, but then they pair that through the for the photographs, and it's a striking project. I showcase it in my in my most recent book is super
love that project. And I think there is just a good reminder. You don't have to rely on other people for data. I know he said this at the beginning. Make your own data. You know it was there was nothing stop you you know like if you like a Paris local news outlet right now there is a ton of stuff that you could just make for yourself. You don't have to wait on the IOC or somebody's going to generously give you their their data you can just make your own
I am still waiting for questions to drop her nudge nudge wink wink. People downstairs and people online. Come on. Feed us please. But while we wait for them let's see what do I have in backup? Well, what's the most expired exciting sport at the Olympics to create a data visualization for
weight loss? I think it's always I do actually have a filter for about I think it's always the youth sports that people don't know. Right? So there's a lot of stuff this time. What is it this time there's like a boarding boarding is
breakdancing,
what have you because show how to do like a skateboarding move that somebody tells me about picks. You know, that that again is a human thing. And the great thing again about what skateboarding is a gaming phenomenon already maybe there's a way you could tap into that gaming feel? A little bit and then some like breakdancing. I think I'm certainly when we're thinking about on the trend site. We'll have pages that are about how people are searching for what I didn't mention that. So I work with Google translator is there's a site which is public and free to use called trends.google.com. You can go there and explore any sport and how people are searching for it real time. So that's real time data that's downloadable and accessible if you can look for another data source. And you can see what other questions that people are asking about breakdancing during during the Olympics and also the people that who was suddenly kind of coming up that you've not heard of that I knew and and and why wasn't people asking about them. So I just think there's so much potential there. Yeah.
God bless you just seem heard. Let us know what country you're from. Because we love to keep track. But to see him wants to know can you give me some great examples of data journalism, please and I know for a fact that Alberto has 16,000 tabs on his computer that I can show
ya as much as I'm not interested in sports, I'm extremely interested in the visual representation of sports. So I'm going to give everyone essentially a few places to go where you can see great examples of, of visualization. So if you don't mind I'm going to I'm going to share my screen and give you those those sources. The first place to go I think is a flowing data flowing data.com This is a website by Nathan EO who is also the author of a couple of books about data visualization. He has recently published the second edition of his first book visualize this which is really great. I strongly recommend that book but in any case, in his website, he has an Olympic step analytics category. And here he collects some comments on a graphics related to sports that he has seen a throughout the year. So this is a great source of inspiration. It's one of these collection of graphics that you can go to essentially borrow ideas from one of the things that I always tell my students is that one of the best ways to learn data visualization is to try to copy but not in the sense of plagiarizing people but trying to borrow ideas and mix them up somehow. So this is a great place to go to get to get inspiration inspiration
low, barely said all artists imitation and
you all are already seen in part imitation. Go beyond the invitation but another one is a data sketch. They have this website 2020 Olympic Games and they collect tons of a again, tons of examples of graphics. These are much more recent, obviously these Afghan 2020 There are there's a very good one by Bloomberg, this one here that are really like the sort of like cartogram battleground type of visualization. It's really well done. I, by the way, recommend people to go beyond the usual suspects. And by the by this I mean, we all love the New York Times and I love the New York Times I have many friends who work at the ground and former students who work at the graphics escapada your time to write we always talk about the New York Times we need to talk about other organizations. The Washington Post produces great work The Economist magazine, Bloomberg, Reuters and even local local organizations and you can see some examples in this in this page and also on this page. By the way, this refers to flourish flourish is a freemium data visualization tool and they have these articles that I haven't seen before. I noticed it before because I wrote the I read this page, how to visualize the Olympics and they have plenty of examples of how graphics simple graphics that can be done to visualize your Olympics and then there's plenty of articles about that about visualizing the Olympics. This one is from 2014, a bike night lab. So I don't know whether you want me to post all these links in the chat. We'd
love you to post those in the chat. Thank you. Thank you very much.
Save a flourish I mean, so it just obviously is just uh we were just so we're transparent we we actually funded flourish when it was starting is a fantastic fantastic tool. It's you can do a ton of stuff free and and it's just a great because it's built by data journalists is a great date journalism tool and I super recommend it for anybody who's just starting, or even like we use it all the time now. It's just this fantastic.
Ditch Kamkar says we've talked primarily about representing the Olympics from a global perspective. What about representing the Olympics from a particular country's target audience in mind, do you stick to following the home team? How do you create enough excitement about other countries? achievement? I think? Yeah.
Really good. That's a really good question. Yeah. Because there is a natural bias to just do stuff that we know and we'll know is like kind of old people are going to kind of want to do that. I think showing if this is where actually the alternative metals table can help because you can share the country's punching above its weight. I grenaded it in 2012. You know, like where you got, you can actually say, Well, if we take population into account, we're doing really, really well. I think that's fine to be cheerleading for the home team in that way yourself in context and showing how it fits in and what is what are other countries doing is different to what your country is doing. So she bringing the kind of a global perspective from it in a way that's relevant or understandable to people. That would be my my advice there for sure.
Yeah, good shout. Alexander, Malia Rinko, along the same lines, as the previous question, but while we're on the sharing links portion, can I just flagged you, Alberto that the links you shared only went to the panelists and we just want to put Oh, everyone's attention if you can change. Yeah. And while we're on that, Alexandre Malaya and Coco's, what are the most accessible and preferably free online visualization platforms that you would recommend to another journalist for creating sports infographics for the Olympics.
We free free completely free and open source I would say your role graphs, I will actually type in the answer to that question. And good post it here. But a row graphs a composer link to row graphs is an excellent, a free and open source tool for data visualization was created by a team at Politecnico di Milano, the university flourish which we have just mentioned, a file is freemium, meaning that you can use it for free is only the flourish logo will appear and in your graphics and the graphics will not be hosted. In your in your server. They will be hosted in flourishes servers. If you want to use the full version, then you need to pay for it. There is another tool called Data wrapper which is excellent, great, great, super simple a to use data visualization tool is not completely free. But then I mean, there will be a moment in which after you have used these tools, whether these are tools that I use in my classes, Introduction to Data Visualization, so we use anything from data wrapper to flourish to regrets Adobe Illustrator which is designed to then there is a point in which you need to further customize your graphics and then you need to start learning a little bit of code. So if you want to produce interactive visualizations for the web, you need to learn HTML, CSS, JavaScript, and in particular, a library called d3. But I'm a huge fan of the R programming language. It's great language for visualization and data analysis. And there are tons of free resources online. I actually mentioned in my response, they are a book written by Hadley Wickham, who is also responsible for many of the developments in the art programming language and the book is titled IR for data science. That's the best intro to AR that I have seen so
much. There's so much visualization you can do in our it's amazing.
You can create interactive visualizations with are they have like extensions to add JavaScript to it. It's it's a rabbit hole. Once you start learning art. It's like it's absolutely incredible.
How do you feel about Canva?
Well, Canva owns flourish now. Yeah. So so obviously, they've got flourish was a fantastic tool, so
and kind of teeing up for the next question, how do you feel about mid journey ish generative products for use in this kind of venture?
I think so. I thought about this a lot, partly because obviously, I work at a company that does a lot of generative AI work. It's a huge part of a huge part of what Google does. And I My advice would be to focus on what's you know, like, I think I think the stronger you get the strongest results out of out of out of AI when you're asking for opinions on things, not for like, yeah, who was the first person to x? You know, I think I would, I'd think you can work out yourself. What are you gonna get out of it? And maybe there's something you do read around the text generator and the context. Like, you know, tell me about skateboarding while skateboarding is, you know, when you've got so many sports, there's, there's things you can do that in terms of the visual side of AI I don't know, I think you could do. You could I think that's maybe where you'd have fun with it, you could say at this stage, because we're still pretty early on in this journey. So could it be like generate, I want to look like what what training would I have to do to look like Michael Phelps or what what, you know, how would you create the perfect swimmer, the perfect gymnasts? You know, maybe there's stuff you can do that. There was a piece that ProPublica did a few years ago about workers This is rather than problems workers compensation, where should how much he got paid for different parts of your body, and you could like you could, you can play around with the numbers and stuff. And that change the visual so maybe there's stuff like that we can bring that AI into something else you're doing so it's complimentary. Usually, it's complimentary is to me to my mind smart as way and then there's using AI to just like process a ton of documents. So you've got document, Tom. Now there's a Google tool called pinpoint, which is kind of amazing. It is going through tons of documents and helping you analyze them. Thanks. That's the more the more traditional data journalism approach. But yeah, I
think that answers your question Francisco betregal, but
I will add a little bit. I will refer to the old saying that you're noticing is a discipline of verification. So essentially, it's like I see generative AI not that differently to the way I see older technologies such as Photoshop, for example, when Photoshop here like 30 years ago, or something like that, there was all these talk about there's always going to be a need for photo editors anymore, because now everybody's going to do these, whatever. Yeah, sure. I mean, many photo editors are not working anymore. In the news industry, but the tool doesn't substitute person. The tool is an accelerator of what what a person can a person can do. So I see generative AI as an extension of our cognition the same way that it see any other tool I would discourage people to use it right now to generate images a, you can get inspiration from things that you can get for generative AI but that doesn't substitute the need for verification and your own hand on your own brain applying some critical work to the to the creation of those of those images. The models right now, are too crude to be trusted, both in the on the side of generate generating documents or data and on the side of generating images. It's great again to accelerate the process I have used a I choose to generate code that accelerates my workflow. But at the end, I need to read the code and test the code and make sure that it does what I want to what I want to happen.
Thanks. So let's go downstairs to Siena from Denmark who has questions. Oh, zoom in and unmute. Here we go. Ask you There you go. Go for it. Hi,
thank you for the interesting talk. I was just curious. About since the Olympics as an international event. Do you sometimes collaborate between media outlets and big projects? And yeah,
how does that go on the scene? Any collaboration
not me. I mean, I've a again, I've been a professor for quite a while. So I have not participated in Olympics of really projects recently other than projects. I've done a with a with Google, visualizing Google Trends data. I'm not aware of collaborations, whatever you send them
Yeah. And I think there is a ton of potential there you can really see where you've got, you know, a lot of small newsrooms. Maybe they only have one person working on day stories. Maybe they you know, they can kind of pull together and I would be surprised and stuff like this isn't happening in Paris just because I think the European journalists are very good at finding projects to work on data journalist says there's a kind of collegiality to take journalism, which I think doesn't exist and other forms of journalism partly because often people are just working on their own. So they they tend to know that their competition really well. So I think ton of potential there I don't I haven't seen a lot of it, I think because it's like, it's like elections. Yeah, election results. People are kind of like these are big projects running us. And so that equity kind of gets into it and just gets on with it. But then you got the stuff around, say elections or around the Olympics, which could be really interesting. Like, I think, yeah, looking at the IOC or going back over financial records, things like that, where it would make sense to collaborate collectively. Yeah,
the potential for an Olympic bass Panama Papers event happening in the pressroom is Yeah,
absolutely. There may be who knows already. So
more power to you, wherever you are. Charlie, can we go to you for a question from the UK?
And hi, thanks so much. I sent a question about maybe some of the more difficult projects you have both worked on due to kind of constraints with data. Just because I'm thinking of my own project, perhaps where there isn't much to work with. And I'm curious to know like, if you guys have any examples where you'd have to sort of battle through I mean,
I have one is a bit old out but um, when I was at The Guardian, I was there during the riots in 2011. And then what we want to do was we wanted to get the court cases to find out who was who was going to court for those for those offenses, like where people kind of get referred and where people came from where there's like any kind of socio economic factors. And even though courts date was supposed to be public, you know, they were they were asked they were going to charge us basically, to use the data. So those they will charge you 15 pounds a name. This is hilarious. So we went to the Ministry of Justice, I just said Ken, you're supposed to be all about open data. And you're not just putting a structure around to the courts to just like release the stuff at the end of the day. So they did and then we got swamped with everything. So every case, not just cases related, we had to kind of get a team of journalism students together to kind of pull through them. So it's like, I would and there's lots of kind of examples of that in dangers and what I mean by would say like, and maybe they'll project and if you know if you want to mail me, or whatever, I'm happy to see this new advice I can give well, I might know somebody who probably knows more than me, but with every with every one like that, I found it's better that you've got one step at a time you you kind of start with his mountainous thing that it's gonna be hard to get. But you've got to kind of like do it one bit at a time. Don't think about the whole mountain think about what's the next thing that I can do the next little bit that I can get to bring it forward and there is a ton of you know, there are a ton of levers that you can pull from freedom of information to, you know, to just even just pressuring people on the phone. Now emails never get anywhere but as you people hyping run up and around. So you know, there's a lot you can do there and I think there's a lot that is just basically is like the journalism side of it. And then and then you've got the data analysis side of it. But when you've got the data, then there's a ton of tools you can use to to go through it whatever it is so helpful, sorry,
something that I was going to add before and they didn't me maybe relevant to the answer to this question, which is sometimes we journalists tend to sort of like looking to get into it always from the same sources like government or sources. In this case, the Olympics. I mean, we got we can expand the scope. What about academics What about researchers? About about sports, right? We get one thing that I would add to war is using surgery's not only our sources of information, but also as collaborators in projects right maybe possible to create a collaboration between your news organization and an academic institution and then and then creating products together right that that can be done and I know that there are certain academics you know, I work with people for example, in a we do we study weather and climate change, and not all of them, but there are many people who are very willing to put the word out there of what is everything that is happening related to the environment, and they're willing to not only give you data, but also to help you understand the data better. So perhaps that is also true about sports. Char
Do you want to explain a little bit about the data set you're working with? Maybe Alberto and Simon could troubleshoots a little bit with you? Yeah, I
think it's more and this is partially done for like me not having time to really get into the nitty gritty of it quite yet, but it's more that I'm I'm struggling with how to like to pull the data if that makes sense. So So essentially, I'm working on a project which is looking at missing people UK and the types of certain demographics of missing people receive versus some other demographics. There's a lack of data in general when it comes to like race and gender of missing people. There has been some like academic work that's been done around it, but nothing that I have been able to find about, like, like news coverage, like media coverage, and I had this grand plan that I was gonna like, I don't know, like scrape data from the websites and like, create a framework which was based on like, all these different demographic factors, but I am not smart enough to work out how to find myself. But yeah, to find that somehow, and I also need to figure out how to use the data I do have access to, like, in the best possible
way, um, recently that I had to jump in on it by because this is quite a complicated thing. So I would say, really happy for the organizers to connect us and as well, there are a ton of scraping tools. It might be you don't need to, but but maybe share some examples of what you're looking at.
How many people are we talking about in the data set?
I wouldn't even know tell you at this stage. But I guess it's maybe not how many people but the length of time would be the relevant thing here. I think so like how many years back we would be looking at?
We know that 170,000 People were missing each year. Yeah.
I'm asking these because I have seen in the bone Simon needs to jump out.
Sorry, guys. Thank you.
Have a wonderful day. Thank you for your time.
I can see a little bit longer. So IDM. So you were mentioning you know that they needed missing data or you haven't answered people but perhaps you don't have the agenda or whatever. Depending on how big the data set is, sometimes you need to fill in the gaps manually. So there is a project that I showcase in my most recent book by our data visualization, designer and data genres from India Gorman Bhatia, and she did a project about a the songs in Bollywood movies, the singers who sang those songs, and and obviously there's not a public database about that. And she put the database together by hand. Essentially she wrote it she spent months pulling the database together herself writing it down, writing the names, then trying to identify the proper gender of each one of the singers, which was much more difficult than she thought it would be. So she had to make lots of phone calls. Talk to historians of Bollywood movies and things like that. It's like sometimes, particularly when, when it comes to beginners in this field sometimes people expect that data is going to be out there ready to be used and clean ready to be visualized. And that is not the case. Like I I always refer students to watch the spotlight movie. I don't know probably watch that movie, but there's a scene in which you can see the reporters from the Boston Globe, essentially writing back on the data from books copying data from from printed records onto a spreadsheet and spending months doing that. That's the it tells students that's how it works. Scraping data is perhaps not the exception, but it is not as common as you think. And even if you can scrape the data, the scraping the results will come with mistakes that you need to correct you need to double check ins verify the data. So again, depending on how complex and big the data set is due mainly to a lot of hard work
or the end and just seem says Tell us again the name of that book where the infographic is described.
Yeah well the book is my latest one is the art of insight but the the the the art of insight but the the the the designer is Gorman it sorry. Let me just send you for our website
coming soon in the chat. Summary if you are approaching data visualization the Olympic Games this year, look for the humanity of the games. Put the reader at the center of the visualizations you create, contextualize and compare and use historical data. Don't be afraid of handcrafted and hand drawn there is room for emotion and data visualization. Create a snapshot of this moment in history. The social, economic, political stories behind the Sports. Is there room for cross border collaboration or cross sector collaboration. Don't be afraid to create clean and patch your own datasets. And my favorite takeaway from today. What would be fun that's brilliant. Thank you, Alberto. And thank you.
Thank you for
it's been fantastic. We are back home for one last global journalism seminar next week with Alex burns from Politico. He'll be chatting to us about the Olympic Games and Paralympic Games feels like it. He'll be talking to us about the US presidential election. And we look forward to seeing you there. Thank you very much for your time and go Well