Transcript
Speaker: Welcome back to the policy of this podcast. I'm your host, John Schwabisch on this intervening middle episode of the show. I'm talking about the first in a series of podcasts where we talk about my favorite graphs. Now, these are the graphs that help sort of shape my own perspectives. on data visualization, data communication, data storytelling. And I'm not doing them in any order. It's not from first to last. It's not alphabetical, not by time, just by what I feel like recording on that particular day. So that's the goal here. Hopefully this will provide you with some, you know, maybe historical
Speaker: framework for data visualization or maybe this is just a vanity project I don't really know but hopefully this is worth it to you and you find some value in exploring some of these projects that maybe you've never seen before or you forgot about depending and we are gonna start with The Guardians Project from December of 2017 busts out how America moves its homeless. Now, if you are watching this on YouTube, I'm going to split screen this so you can actually see the project. If you are listening on the podcast, on the audio, I will describe what I'm looking at so you can listen to this or watch this wherever you like. So, and I'm not gonna go through every detail, right? i'm not going go through everything. i'm just going to you the highlights, the stuff that I love most about these projects. So,
Speaker: We start with a looped video of a bust coming from the left side of the screen to the right side of the screen. Here is the headline. Bust Out, How America Moves Its Homeless. Each year, U.S. s cities give thousands of homeless people one-way bus tickets out of town. An 18-month nationwide investigation by The Guardian reveals for the first time what really happens at Journey's end. Now, this is a combination of data work from the Guardian team with a number of their journalists, with the development done by Nadi Bremer and Shirley Wu. And so what happened in this project is one of the first projects, at least that I can recall, that really did, I think, a superb version of combining journalism
Speaker: video with the people who are the subject of the news story with data visualization and in this case it's inter it's animated data visualization so some interactivity but it's mostly animated visualizations and I'm gonna start at the top here but I'm really gonna focus down as i get further down to the visualization that really is the one that that sticks in my head that I remember so the whole story starts with the story of a person named Quinn Raber who was Bust from San Francisco to Indianapolis so the whole thing starts with a short two paragraphs and then a big map of the United States and we see as we scroll so it is a scrolly telling the animation starts with a red line starting in San Francisco San Francisco and moving east and
Speaker: to Indianapolis and the tax scrolls on top and so today this is pretty standard right this is pretty standard parallax scrolling at the time it was new I mean today you could do this in a tool like flourish right you could just do this right out of the box but at the time this was a really new approach that was that was being tested out so we scroll through the first part we're getting into the new story it's setting the stakes for what we're going to see The map then changes to a choropleth map looking at the state level homeless rate ah per 100,000 people.
Speaker: As we get to a certain part here in the story, um we start to see some more animation. And as we scroll that animation continues. So several states are highlighted on the map in red dots with labels. And then we start to see the animation of these little, let's say, blue dots um as people move from city. to city and They are moving the bubbles are expanding as the population moves between city and city and then below that is a timeline on a vertical bar chart showing the number of bus relocation journeys So we have two visualizations working simultaneously in an animated way So there's nothing for us to click highlight, you know get more information on it is sort of the movie version of this new story. And of course, now today we see these, well, maybe not as much animation today, but at the time this was really new and this was really exciting. Remember, December of 2017. Okay, so we scroll down and now the visualization moves off the screen and now we're into text and we're into text with some pullout quotes. So it's not just a wall of text. We've got the pullout quotes. And then I think the part that really sort of
Speaker: for me did such a good job of mixing these methods of storytelling, data visualization, qualitative work, is we have a couple of videos with people who are subjects of the story. So we actually have the video, we can hear their voices, we can understand their story a little bit more.
Speaker: Okay, I'm going scroll down here because there's a couple of videos and a wall of text, um a couple of of of looping videos. And then we get to the part that I just remember. This is the part of the story that I remember. We get to a visualization reply about almost halfway through homeless relocations from New York City is the headline. Around 650 people were flown to foreign countries. And when we get to this part, as we scroll down, we get one blue line, an arc from New York City on the left to New Zealand on the far right.
Speaker: And so this is an arc chart, which I can't recall really seeing one that worked this well. So we see the one line and then a whole bunch of others fill in. So we have um New York City on the far left, New Zealand on the far right, and then between we have different countries, right? So now we're going to foreign countries. We have Mexico, Dominican Republic, France, France, Nigeria, India, and it's scaled by by location, by by journey length, which is also just really smart. and we've got a number of different attributes working together. So the height, the thickness, and the color saturation of the lines are doing a lot of work here to show us you know the longest journey. And then along the...
Speaker: horizontal axis where the city set the size of the bubble corresponds to that to the number of people and as we scroll continues well we then get from ah New York City they show New York City to Puerto Rico so they zoom in on to the most common destination so here we have Canada Dominican Republic Puerto Rico and a few more. And then as we scroll down, we get to the most popular destinations in the United States. And now again, this animation kicks back in with the same structure set up with New York City on the left, the US cities stretched out along the horizontal axis, San Francisco on the far right. So it's using the same horizontal dimension of distance the same coloring on the arcs and then the same size along the bubbles. But we have three visualizations going from the global
Speaker: to I guess I would say maybe like the Western hemisphere and then zooming in to the United States itself. i man I just love this particular visualization. I thought this was, and still do, when I say thought, I just mean at the time, but still do think this was such a great piece, such a great use of the arc chart that we can show distance here and we can use the combination of color and saturation and size of the bubbles to show all these different pieces of of the data. So that brings us to about halfway through the story. That's the part that I remember. That's what sings to me. um I'm just going to a little bit further here just to show they have more visualizations.
Speaker: So there's kind of Sankey diagram a little bit further down that has um where recipients go based on the median income comparing their origin city to their final city. So we have the the sort of normalizing the data of where they started relative to the median income of where they ended up. And so the the headline here is that 12% of people were relocated to cities where the average income is higher than where they started and the remainder 88% were moved to cities that were lower. and next to that what is a sankey diagram is like a little histogram a histogram of showing the distribution of that median income just really smart really good ways um to uh combine these different data types and as i as i hover and click there's nothing to hover and click on right it's not it's not interactive which
Speaker: You know, I think there was a period of time where everything was interactive. This was just smart because it's still telling you the story and you don't have to click and hover and get all the details because, you know, I think there was a time where everything was clickable. Everything could hover and get more details. And at least for this story, this piece, um there was maybe a realization or recognition that, yeah, not everybody's gonna click. Not everyone wants to click. Let's just get through the story. Cause only halfway through. So this is a big investigation that they did. Okay, so we keep going down. We've got some more video. We've got more texts, more ah photographs, more videos.
Speaker: um Telling people's stories we get a little bit further down and then we look at the relationships between people who've got sort of ah a bubble scatter plot and again more videos and then the end of the story at the very end is another bit of a scrolly telling this one shows animations of ah people represented as bubbles moving um from city to city. Sort of overall what that looks like as people journey between those different cities. And we see a couple more animations before we get to the end of the story um where we have to finish up the video of Quinn Raber who was the subject of the new story in the very first paragraph, in the very first piece. So in my...
Speaker: opinion, my perspective, this is just an expert way to tell a story. This is actually telling a story, right? We throw around the words data and story a lot. This one, I think, just does such a great job. And for me, the arc chart, the animated arc chart is the thing that just sticks in my head. It's just a piece that I have really loved since it came out back in December 2017.
Speaker: So that's my perspective on this first of my favorite graphs. And I have a special treat for you. I asked Nadi and Shirley to send in their recollections on this project. And they sent me a fun little audio snippet. So I'm going to play that for you. Take a listen to Nadi and Shirley's ah recollections on this great project from The Guardian bust out. And I hope you enjoyed this episode.
Speaker: i'm just to let this play. And I will see you next week once again for an episode of the PolicyViz podcast. First question about Bust Out. How did it come together, Nani?
Speaker: Yeah, so I was approached by two journalists from The Guardian US about how they had been working on um like an entire news section on their website about homelessness in America.
Speaker: And that they'd been gathering data about how homeless people were being bussed around the country and and data about like the bus tickets given out by shelters in cities.
Speaker: And they wanted to tell the story with this. They knew it was like a very, like they were sitting on a treasure trove of information, but they needed somebody to bring that information to life through data visualization. So um they they asked me if I was interested and I naturally said yes, but I also could see that this was bigger than I could take on a loan. So that's when I thought, well, immediately I want to want to work together with Shirley on this. So that's ah that's how we teamed up.
Speaker: And then... um With my background as ah as a data analyst, I took more of the data cleaning and the data analyzing on me. And surely, we can talk about it later, but you took more of the coding on you.
Speaker: um And then there was... by luck we were all in the same city san francisco together for two days like the journalists and you and me and then it's that's at that point that i did all of the data analysis and we talked on about how the whole story should kind of work which the different subsections should be and then we started brainstorming about the different like visual designs that could go with that um yes that's kind of how how the start of it came about
Speaker: Yeah, and actually... I think you had a lot... m Oh, yeah, yeah. I was like, you could talk about the challenges you've had because I think you had the bulk of those.
Speaker: Well, I want to actually also say first how fun it was to get to work on this with you, Nari, because I think this is our, outside of data sketches, I think this is our biggest collaboration. um And also how fun those two days were.
Speaker: i think that when we work remotely with clients, we kind of forget how fun it is to be able to ideate in person. And then, like, energy of, like, the four of us. Yes. Like where the journalists knew the story and then you knew the data and then two of us were like pitching the designs.
Speaker: And that that was ah such a fun two days. I'm going to start there. Like Nadi said. um ah like naty said I was asked to pull the ah the coding part of the side of it together. So it it was because this one this project was not only data visualizations, but they also wanted media. They wanted photos and they wanted videos, looping videos, um and they wanted all of that to work um seamlessly while the scroll was happening. um So i was in charge of pulling that side together.
Speaker: And I think that's where I ran into a lot of the challenges. um not Not as much to do with the actual web implementation, but this was the first time and maybe only time that I've had to implement things and not only for the what the browser, but for a client's mobile app and the mobile app ah not only on the mobile app on both iOS and Android. And then...
Speaker: having to debug a web view within iOS and Android. And I think that that by the level of pain, um I think that that might have been the biggest challenge is not being an iOS or Android developer and still trying to figure out what the bugs were inside those apps um with no clear understanding of the debugging tools for those apps.
Speaker: Yeah, I would say that was the biggest challenge. I think that I, ah for about two weeks, was not in a good place. but I remember that. i Oh man, yes. Yeah, because It's kind of... Because I think we didn't face a lot of challenges like in ah in a non-coding sense, because all the collaboration between the journalists and us together, it was all very smooth and we were all very much aligned. So then, yes, those coding challenges did did grind you down for too long.
Speaker: Yeah, and I guess this was pre-being able to just ask an AI, like, why is my... app not working. um So it was just a lot of banging my head ah on ah on the laptop and then asking for help from Guardian developers in Australia um because they understood the apps the best. it was It was honestly really cool to see all of the like help we were able to get from people within like the Guardian organization. um
Speaker: But yeah, so I guess then um the question is like, how do we look on it now? You know what I realized? I think it is now eight or nine, I think nine years since this piece published.
Speaker: Yes. Yeah, it was 2017. Yeah, man, it's been that long. um I think it's still one of my like definitely like top ten favorites. projects and I still get sometimes people telling me that they read it back at the time or it made an impact for them and also believe it it it generally made an impact on on this issue where I think the journalist got on like national television few channels and was able to tell the stories and that's good um but honestly I just my my personal memory is just that it was
Speaker: like how it was working together together with you for the first time and how fun it was like ah like you also mentioned to be able to kind of brainstorm ideas about the same thing like with data sketches it was both of us working separately but giving other feedback which is very different than working on the same story um and working together and making it making it happen um so that was that's also why to me it's like a one of my precious precious projects uh in that sense You're right. That's such a good point. I totally forgot about that. yeah, to echo what you were saying, that
Speaker: Actually, yeah, I don't have any more to add on that. It was a really fun one. And I think it might be the only one where we got to work on the same code base, on the same project, and be able to brainstorm together. um And also, um yeah the same for me is like, I think it is one of the projects that people remember me by. It's always either Hamilton or this.
Speaker: And then like a few other, then there's like a few other projects that people... will be like, oh, I saw this. um I guess terms of yeah personal impact, um I think that that was an extremely empowering project for me because I think this was our, this was maybe my second or third big client project. And I think that before this client project, I had steered away from very serious topics. I just was like, oh you know, I'm,
Speaker: I don't have any domain expertise in, um, in any of these, uh, topics. So let me just kind of stay in my lane. Let me, uh, let me, let me make a dedication to Hamilton. Let me like cover things. Like, let me cover things in pop culture that nobody can get mad at me for covering.
Speaker: um, right and And then I think this was the first time where I got to work on something that like felt so serious and so important.
Speaker: And like Nadi said, that the journalists went on and talked to policymakers and um that it was that it sparked a conversation.
Speaker: um was an extremely empowering experience of like, oh, if I'm able to work with domain experts like that that is the role I want to have is impact like is enabling domain experts to tell the the important stories um so yeah still one of the most important projects I've gotten to work on
Speaker: nice do we have other memories worth sharing I think i think that's i um I don't know, but i feel like we're definitely over the five minute mark by now. Alright, thank you. John. Bye-bye.
