Transcript
Speaker: Welcome back to the Policy Viz Podcast. I'm your host, John Schwabisch. I hope you're well. Hope you're excited for another episode of the show. On this week's episode of the program, I am joined by Lauren Klein, the main author of the new book, Data by Design. It's a fabulous new book. Actually, the time I'm recording this, I don't have the actual physical copy of it. i just have the e-version, but it is beautiful. It is delightful. And that's nothing to save the content. Really great content in this book. taking a historical perspective to data and data visualization. Lauren and her co-authors, her collaborators, recreated a number of historical visualizations, taking into account the hierarchy, of the powers, the structures that led to the creation of these visuals in the first place. so it's a really interesting take on the history of data visualization and what you're going to hear in this week's episode of the show. is the conversation Lauren and I have about the structure of the book, the goals of the book, and the process of not only creating these visuals, but actually getting those historical artifacts. Because it's not as easy as just downloading something from Wikipedia and calling it a day. So if you've never had to do this before, you're going to learn something about trying to grab high-resolution historical images, particularly with a data visualization book. So I think you're really going to enjoy this week's episode of the show. so Here's my conversation with Lauren Klein, lead author of the new book, Data by Design.
Speaker: Hi, Lauren. Good to see you again. Hey, John. Yeah, it's been a minute. It has definite has definitely been a minute. It has definitely been a minute. um How is Georgia? How is the kickoff to the semester?
Speaker: Um, it's good. It's exciting. It's exhausting. You know all the things that happen in week one of classes, which is when we're talking, but yeah, no, so far so good. How many classes are you teaching now?
Speaker: I I'm teaching two. Okay. Um, and I, I have this funny appointment, so I'm half in a data science department and half in an English department. And so I joke that I live in a natural experiment, but it's totally true. So one of my classes is like, it's supposed to be the introduction to NLP natural language processing for data science students. And so that class is like, here's how AI works. And then in my English grad seminar, it's like, AI is terrible. Let's learn all the ways in which AI is destroying society. Oh, wow. Yeah, don't get them all one room together. That would be super good. Well, I do. I let my seniors, I let my data science undergrad seniors enroll in my grad seminar as a capstone seminar. And it's actually, it's really good to have students with technical knowledge
Speaker: in dialogue with really theoretically yeah rich grad students. like Usually the data science students are like, oh my gosh, they they they use so many words. They have so many things to say. and the grad students are like, oh my gosh, we just we said we wanted to like make this map. And then they just they did it.
Speaker: Love it. That's awesome. All right. Bringing people together, um which is kind of what you're doing in the new book, like bringing lots of people together. You've got a whole crew of co-authors and students and collaborators on this one. So it looks great. I'm waiting for my hard copy. I've got the digital version, which is not going to be I'm, you know, MIT does have a nice reputation of good, like solid.
Speaker: quality book. So I'm looking forward to holding onto it. So data by design, can you start us off? Like, what is the goal? What is the message? What do you want people like? What is that core thing you want people to know when they go to Amazon for for shopping?
Speaker: Oh my gosh. So this, you can tell this is the first time that I've ever talked about this book. Cause the answer is that I don't have one answer to the core thing that you want people to know. But I will say like, there's sort of, I was thinking about this and One of the reasons why it's hard to answer is that this is a really long project. I worked on it, I say a decade, but if I'm being honest, it's probably something like 12 years. It's as old as my oldest child who just turned 13, which I know because there's a picture of her in a baby Bjorn outside of the first archive that I ever went to on my first day of research for this project.
Speaker: um But I sort of, I put it aside and came back to it and put it aside and came back to it. kind of each time my goal changed, I think in a good way. So like the first time it was a pretty clear goal. um I just wanted to place the history of data visualization in its historical and philosophical context. I thought that that had not been done before. You know, we've got, and if you look at the histories of visualization, who writes them, it's generally statisticians, which is fine. Great. Like, I love that, but they're not trained historians. And so one of the things that I can do for my background in the humanities is broaden the context through which we understand the emergence of the field. So that was like, you know, goal number one. But then i wrote Data Feminism with Catherine and I put this book aside. And one of the unexpected pleasures of that book was that people actually read it. In humanities, you don't expect people to read your bill your books. you know um But it was really exciting to think about our ideas impacting practice. And so when I came back to it after Data Feminism, I really pushed myself to say, like how does this history change what we do now?
Speaker: And so if you look at the book, there's now at the end of each chapter like bullet point takeaways that we have a set for people who look at visualizations and interact with them and then a set for people who design them. And I really pushed myself to ask like, okay, how does learning about the 18th century or, you know, the weird life that William Playfair had, you know, how does this change how we like, you know, design time series charts today? Like really really, and sometimes it was obvious to me and sometimes It was tricky. So anyway, so that was the sort of the second time. And so I really did want to sort of write and I do want to change and enrich how people design visualizations today. And then the third thing, you know, when you're writing a book, a book book, you know, this actually it began as a website and will eventually again become a website. But one thing that happens when you write a print book is that you send it off to the press and then you don't see it for like a year because it's getting...
Speaker: you know, routed and they do legal clearance on the images and they copy edited and whatever. And so when it came back to me and copy edits, it was right in like the first or second month of the second Trump administration where higher education was just getting attacked right and a left and words were getting banned and professors were getting and fired. And then there was also the sort of undercurrent, which we would become a major tidal wave of AI, which was threatening from a different angle, kind of sometimes the same angle, um you know, what it is that we do in universities. And so it became really important in the end to really stake a claim for the value of research, um and in particular of humanities research, which is often perceived as abstruse or narrow or irrelevant. And certainly slow. I mean, like I just told you how long this book took. yeah um But I think it really matters. And so the last pass I did through the book, mostly in the intro and in the conclusion, or I guess the preface of the epilogue, there's like sort of end front matter and matter, right? Yeah. um Was to really be explicit about what I thought the political stakes were of the project and why this kind of work matters. And
Speaker: I believe it, you know, and now when people say like, why did you write this book? Now I say, well, it was to show why what we do is, it really matters in the world. It matters for the future. It matters if we want to have better ideas about how to get out of this mess. We need people spending 12 years writing books about the 18th century. You know, I mean, we also need activists and we also need organizing and we also need unions and policy and all these other things. But we do need people saying like, let's think really hard and deeply about how we got here.
Speaker: Yeah. So I want to come back to some of these, what people can learn from William Plafer or what people can learn from these graphics in the, I mean, really going back to the early 18th century, right? Even before that, in the preface, at least, you go back even further. So I want to come back to to some of that in a bit. But So let's talk about the, before we do that, let's talk about the book itself. So you have, I believe if I counted right, 11 collaborators slash coauthors. And I'm curious from an author perspective, like was that...
Speaker: harder or easier than like working with just Catherine? Like what what was, it sounds to me the way you just described it and the way it's sort of written up in the book that it sounds like this was kind of a book that grew out of a practice with lots of students.
Speaker: Yeah, yeah. I mean, like to answer the question, to it like it's both easier and harder, right? yeah, It is challenging in that most of my collaborators, as you just said, are students. And the i don't know like who's listening and how long ago they were a student, but the student lifestyle is very different than the lifestyle of a middle-aged mom academic yeah like me. you know And so like I've got my to-do list, I've got my schedule, I've got my calendar, and I...
Speaker: I'm very clear about what needs to get done. And I think when you're an undergrad, when you're 18 years old, you're like, oh, no, I have a midterm tomorrow. you know, like that is my all consuming priority or like, yeah oops, I stayed up all night and now I need to sleep all day. You know, and these are normal things that should be happening to college students. But one of the things I'm most proud of over the course of this project is I think that I learned how to be a really good project manager of students.
Speaker: You know, like I really learned what a realistic chunk of a project was to sort of delegate to different people depending on what their sort of constraints were and sort of where they were at that point in the semester. I feel like I figured out a good meeting cadence. We have good asynchronous communication. i really feel like, and that was a skill like, you know, like you don't learn that in grad school, you know? No. So I really, like, I'm actually, i feel like we, like, I figured out how to make it work. And I hope that the other
Speaker: authors on the project, the student authors would agree. Um, but sometimes it was tricky, right? Because sometimes it'd be like, look, I need to get a job when I graduate. So I'm just like out for the next three months. Cause I'm interviewing a New York city, right you know, and like, right you need to be like, okay, that is your prerogative. right yeah And then you're like, oh no, no one can prototype my visualization for me. Right. Yeah. That's right. Yeah.
Speaker: Um, so anyway, so like, so that part was a little bit harder, but the other thing, you know, and I think about this, um, you know, now, because I'm like, oh my gosh, like there are a lot of moving parts. Would it just be easier for me to just write the thing that came out of my brain, just me by myself? But it's very lonely.
Speaker: And sometimes you're like, I'm stuck here. I don't know what to do. And there's something really nice about having sort of like a built-in real-time feedback committee where you can be like, it does this make sense? Is this where we should go And people who are not just invested as readers and friends and colleagues who you can bounce ideas off of but people who are legitimately invested in the project because they're part of the project. Like that's that's really nice, you know? So I definitely feel like it will be a transition moving back to even smaller collaborations. The book is really interesting for lots of reasons. But one reason I found interesting just on the visualization side is you don't just
Speaker: take a set of graphics and talk about the historical, where they were in that moment in time and who created them and why and how and what the impact was. But you also take time to redesign some of those going back to the original data. I'm thinking of the slave trade graphics in the first chapter.
Speaker: And so I could imagine that conversation about what the design should look like is certainly easier with more people.
Speaker: And presumably, you can just say to the students, go build the thing, and we'll talk later. I mean, yes and no, right? I mean, I think it's super it's super interesting because i feel that oftentimes, I have a very clear sense of what I want a thing. Not as much what I want it to look like, sometimes I do, but like what I want it to do. and often And I'll sort of come to the students and be like, here's the data.
Speaker: here's sort of what I think we should do with the data or the kind of thing that I want to show. And then the question is like, well, how do we show it? And then the interesting conversation comes between like on the team, I have people who are getting their ph ds in visualization in a computer science department. Shout out to Xiao. I have people who are getting a PhD in English. Shout out to Margie. um you know And so like so Margie Adams, for example, is very familiar with sort of theories of the archive and how you deal with sort of the damaged evidence of the past, in particular with respect to slavery. And there's a huge body of work that is very theoretical about
Speaker: what you should shouldn't do with that information and who should shouldn't do things with it, right? And then you have someone like Xiao who's been instilled in sort of best practices in visualization design and the goal of clearly communicating a single, you know, like there's, you know, and there there's almost sort of conflicting disciplinary perspectives. yeah And everyone is bringing their own perspective to the table.
Speaker: And then my job is kind of to try to merge those perspectives. So we end up in a place where we all either see what we believe visualized or we're sort of converted to the other perspective. And I think it goes in both directions, right? yeah Where I think, you know, and for some people, especially dealing with a lot of this 18th century data, which sometimes is of things traded, but other times is of people who are traded, right? like There's very much ah pretty core principle held by a lot of people in the humanities that just is like, you don't touch that. That's irredeemable data, right? like There is nothing good that could come out of trying to re- and plot
Speaker: these numbers that were the first step in the process of turning people into saleable goods, right? Right. That's just a dead end. um And many people in the field, like that is a, like, no one will question you if you say that, but just so like, you're right, yes. And to to a large degree, like I i feel that very deeply. Mm-hmm.
Speaker: And then on the other hand, you have people like of which I'm also a part, which is like, you can always learn something about data. like Visualization is so cool. There's got to be something that we can do with this that is generative, that lets us understand this really troubling time with a little bit more nuance that makes us pause and think, that sort of inflects our own choices going forward.
Speaker: and so that was that was sort of what I was trying to do with that particular visualization was, and yeah. Yeah. yeah I mean, it is an interesting question to think about how, and maybe you didn't have this in your group, but how people might disagree about, I mean, not just yes or no, the binary question, but even as you get deeper into the design aspect,
Speaker: At what point? Because you talk a lot about power and hierarchy and, you know, who who sort of writes the history. And I can imagine within any subset of people working on a visualization like the ones that are in the book saying, we shouldn't do this or we should do it this particular way. And I mean, I can imagine tensions and emotions getting pretty high.
Speaker: Yeah, you know I mean, I will say like our group is a pretty good group. And I think part of the benefit of working on this project for a really long time is that we were almost never in a rush. And actually, I do feel this as someone who does a lot of of work across technical spaces. And i like I write for technical conferences. And then I also write books that take 12 years. like I do think one of the values in humanities scholarship in general is that there's never a conference deadline. You can always push things. I mean, it's both a a feature and a bug, right? But yeah yeah yeah yeah you know if you haven't yet come to consensus, you can be like, well, we haven't reached consensus. We're not done yet, right? And I will say that um we don I would say that the
Speaker: Not as much in that chapter. And there was that we did a chapter that has to do with indigenous mapping. And that was the one that I think was a little bit harder for the group because we had and no indigenous team members. And if there's like one thing that you know about indigenous data sovereignty, it's like you must do this in collaboration with and led by the people whose data you're working with.
Speaker: And it was that I left the chapter for last. It's the middle of the book, but it was last one I wrote. And um because I knew that if you were telling sort of a counter history of data visualization, there needed to be space held to tell the story of indigenous mapping practices because they are so robust and everywhere and are the obvious counterpoint to these sort of dominant accounts that we're getting. But you know the team was what it was. And, right you know, I really believe in taking people's perspectives and experience seriously.
Speaker: think and i think probably had I started the book from the beginning later, I might've been a little bit more intentional in who we had brought into the project. But because my collaborators were students who sort of showed up their first year being like, I heard about your research group on the internet. Do you have a job for me? You know, like I didn't, yeah you know, I mean, I didn't go choose people. that I didn't go choose people, right? Like they, you know, it's ah it's an educational context. They chose me.
Speaker: yeah um And so for so for a really long time, you know, we were like, Is this, what do we do with this chapter? Should we, you know, like there was data, there there is geo data, you know, like we we all know how to make a map, you like we could do it. And some people thought, you know, like, should we try it? Is there a way to, maybe we can sort of reclaim it in some way. And then other people were like, absolutely not hard. No, this is my red line and we should not cross it And um you know we even, we we sketched a lot of things. I should say we, Tanvi, the sort of the main art designer who went to art school on our project, who begins her processes a lot just by sketching. Like she would sort of sketch things out and then bring them to the group and be like, I don't want, like, and we'd be like, these are beautiful. And she'd be like, I don't think, I don't want to use these. Like, no, we're not going to use this design because this doesn't feel right to me. right um
Speaker: And so we really had to work through what ah an ethical path through that chapter would be where all of us could feel comfortable signing our names to the end product. And I feel like that's the, like for me, that's always the question. It's like, when this is out in the world, If you're on a podcast, you're standing up in front of a room and someone says, why did you do this? Yeah. Like you need to have an answer. you You need to be able to defend it. And the the defense could be like, I'm not sure that this is the best thing that we could have done, or maybe I would have done it differently, but you can't just stand up there and be like, oh, good question. I never thought about that. Right.
Speaker: I mean, I don't know how you how you feel about this, but I would guess that you can't get up there and say, well, I actually think we did the wrong thing here. Like, yes and no. as a collaborator, as a coauthor, you kind of have ownership of it. And I get that there are disagreements, but you know, before the thing goes out in the world, I think you sort of either implicitly or explicitly say, this is what we've agreed on.
Speaker: Let's not let everybody behind the curtain to all of our disagreements on that Yeah, that's exactly it Right. It's like, you never want to be in a situation where you need to throw someone else on your team under the bus. Yeah. Yeah. yeah Like, i think it's like, you know, like, it's like, if my name is attached to this, can I describe this collectively and be like, we discussed this, we thought, we thought about this or the other thing. And in the end, this is what we decided. And it's like, Catherine and I did this too. Like the last thing that we did before we sent data feminism to the printer.
Speaker: was that there were a couple of sentences where we've gone back and forth and the two of us had different ideas. And like the last thing we did was to go through them one by one and be like, if my name is attached to this sentence, you know is this OK? And if it's not, then we we had to dial it back or say something different. But right you know if it's like, well, like it maybe I wouldn't have quite used those exact words, but it's fine. Then that's kind of, you know.
Speaker: And I think that's really important. Yeah, yeah. And also to your, I mean, I think to your original point, like someone asked you the question and say well, this may not be the best solution, but this is what we thought at the time or even now is, i mean, is there ever a best? I mean, especially with some of the content that you're talking about, right? Like, I don't know if there is ever a best solution to using data about selling people.
Speaker: It's a solution to a bad solution. bad data set, I guess, might be the way to put it. i don't know. Yeah, but I mean, I guess in a way that's sort of, I almost feel like want to one of the main contributions that I wanted to make, which is another thing that comes from the humanities, which is the fact that like in the humanities, if you study history or if you study almost anything,
Speaker: A lot of it is bad, you know, and just because things were bad in the past doesn't mean that you can't learn from it or that you learning more about it makes you bad. Right. And I almost feel like in certain tech spaces, when people are confronted with evidence of a thing that is bad,
Speaker: there's sometimes an impulse to be like, oh, I should just walk away, right? Like that's, it's untouchable. Like that that's such a terrible thing. It's so separate from me and where I am either because of,
Speaker: time or like who I am or what job I have or whatever, like, let's just leave that over there. And you don't really have that choice in the humanities, because if you want to be able to learn something new about the past, you need to, you need to sit with like, I mean, this is what Donna Haramay calls staying with the trouble, but like, you need to sit with some really heavy stuff and then ask yourself what you're going to do about it. And, yeah you know, both in terms of technology and just in terms of my sort of humanistic scholarly practice, I am really someone who likes to push myself to ask, like, how do I go through? Like, how do I move forward?
Speaker: um And I feel like in like I'm having a lot of soul searching about this with respect to AI right now, because the people like the on the people on the side of refusal, like full on refusal, i I understand that. Like I, I understand and respect that stance.
Speaker: And I'm asking myself, like, is AI the limit case for my general ethos, which is like, there's gotta to be a path. Like, even if it's narrow, there's gotta be a path through.
Speaker: And i I don't, I mean, at least once a day with some new ridiculous headline, I'm like, oh yeah like maybe not, you know? Right, right. um Okay, so I want to ask about the visualizations in the book. And I want to ask two questions of you so you can sort of pick. But I also wanted to say that I think your book and my Better Data Visualizations book, as far as I know, are the only data viz books when they include the Menard Napoleon March graph also include the top panel on Hannibal's March to the Alps. So I was very satisfied to see that because I don't think most people know that that Menard graphic is like the way Tufti shows it's sliced off. It's like sliced half that page is lost. So I was really happy to see that. um Okay, so
Speaker: I want to ask about like the conceptual thing and then the and then the practical piece, which is how did you go about choosing the visualizations you did choose to include and redesign in the book? And then also, because I'm actually in the process of doing this now, like what was the process of getting the high resolution images to include in the book? Because that is not as always as easy as it sounds.
Speaker: Just like go download the Playfair thing and you're good to go. So maybe you can start with like why these particular graphs and then we can talk about the the practical piece of it. Yeah, so it's such a good question. So each chapter has sort of a core image. And if you think of, and the reason why that is true is because in the humanities, when you write a book, the structure of the book is that there are four or five chapters and each of the chapters has a central text or image or artifact or whatever. And when I conceived of this book,
Speaker: i was only I was in my second year of being an assistant professor. like i I didn't know that there were other ways to do things. And I had actually thought this would be the first book I would ever write. It would be the book that would get me tenure. And pretty shortly after that, it became clear that there was no way I was going to finish that in time. So I had to so so i think you know one of the interesting things is i wrote the i rewrote the introduction towards the end. And I rewrote the introduction as a more mature scholar.
Speaker: And as I was writing it, I was like, oh, I wish that I could do this all again and make the chapters all like this. um But on the other hand, I do think that there is value and there's especially value for people who are not familiar with that form of research to see what you can do with one thing, right?
Speaker: Like you can get a lot of meaning out of a single image. And so I actually, so i go back and forth, um but that it is what it is. But like how I picked them, you know, some of them were images that I knew i wanted to talk about. you know i knew i I sort of felt like I had to talk about Playfair because that's where I started.
Speaker: And that started as the first chapter and it moved to the second one because i at a certain point, I didn't want the book to be so connected to this dominant history. But I just felt like there was so much more to say about those charts than people had said. yeah um Same thing with the Elizabeth Palmer Peabody charts. Those are probably the first charts that I saw that like blew my mind that I'd never seen before. And I thought they were so weird and beautiful and captivating and I had no idea what they meant.
Speaker: um And so I knew I wanted to talk about those. um You know, i the the other ones, you know, like you sort of,
Speaker: you do some thinking about what you think the important strains of the argument are and what the images are that can help you tell that story the best. And for me, they have to be images that are visually striking, um that can sort of stand on their own as images.
Speaker: They need to be able to sort of expand when you place them in the context of their time. There needs to be sort new things that you learn when you say like, who was making this, what was going on around them? How does this connect to the broader political forces that were playing out at the time? Like that needs to, you you sort of need to either see that in the image or see how the image changes when you know that.
Speaker: um And then I also thought a lot about who was making the images and what perspectives those people might have brought to their own practice. Like, you know, I'm not going to lie you know, if you look at the heroes of data visualization, it's all mostly British, Scottish, and some French dudes, you know? Right, right.
Speaker: you know And like there's a lot more people in the world making visualizations all the time yeah um who are not you know like from that pretty small, isolated geographic area. In this project, you know again, it could have been, it never was going to be a visual history of like everyone who ever made a visualization in all places of the world, because I believe in disciplinary expertise. And my own expertise is the sort of Anglo-European 17, 18, 19th century. So i like I know a lot about that.
Speaker: But even within that space, I was like, I want to be very intentional about sort of shifting the spotlight onto these different types of people who are representing different constituencies who are making visualizations at that time. That's why there is the chapter about indigenous map making where I focus on a Baothic woman named Shana Dithit who made these really interesting and fraught series of maps in and the late eighteen twenty s It's why I have a chapter on Du Bois, but I actually don't talk about Du Bois as much as I talk about his students, because I feel like that's really the untold story of those charts. In a way, i actually feel like it's a little bit tricky. like It's convenient that we now know about Du Bois' charts, because he's another great man who just sort of gets slotted into the hero narrative. And it's like, no, he like he had a whole bunch of students. A whole team, right? A whole team, you know? Yeah. um And likely, the only reason why he was able to make all those charts in time
Speaker: was because he had a whole team of people working on it together. As you were sort of sketching out or outlining the argument and the structure of the book, um did you find that you had um an idea of the visualization or the or the creator that would be the best representative for each one? Or were there ones where you had like Du Bois, Playfair, the slave map, like all those are kind of like,
Speaker: you know, kind of like in the back of, I think your head, my head, like, you know, if you're thinking about these things, these are the ones that kind of pop up, right? Were there ones that you ended up using in the book that you, you know, kind of discovered in the process of of writing and and developing the project?
Speaker: Yeah, there's, um there's a whole bunch, like I ask myself that question a lot. And in many, you know, in many ways, my focus on sort of the people right next to the founding fathers is kind of how I was trained to do my own scholarship, you know, and think it's, I'm a artifact of my own time, you know, in, I went to grad school in like the two thousands, the first decade of the two thousands. And that was like the first wave of the corrective to dominant accounts. And so what we were taught to do was like, look in the room of the famous people and see who else is in the room. Right. And now multiple generations later, people are like, we don't even care about that room. Like that's not where the decisions were. Like, you know, we're, we're all interested in this other place. Right. But so I feel like my impulse, my personal,
Speaker: impulse as to that's sort of how I approach things. But there were these people who I encountered over the course of my research where i was like, oh If I were starting from scratch, would I have focused on them? And one of them, for example, is Francisco de Caldas, who's this Colombian naturalist who Alexander von Humboldt essentially stole the idea for the Chimborazo mountain diagram with the elevations. like Literally stole it from. like They were roommates um when they were both doing their naturalist voyages. And Caldas has these sketches in his notebook that predate Humboldt that are exactly the same.
Speaker: And Caldas, they're in pencil with watercolor. They're incredibly beautiful. um They're held both in Colombia and also in Madrid. And like no one knows about this guy. And I'm like, if we're going to valorize, they're like Alexander von Humble as this person who fused you know naturalism and visualization, biological data. Like, like he literally stole this idea from this other guy, you know, like he literally, like the, they at like the, we have the receipts as they say, right? Yeah.
Speaker: Yeah. And so like there could have been a chapter on that, you know, I really feel like the one missed opportunity that even would have been, um, in line with my sort of my general view. I don't have a chapter on Florence Nightingale. I feel like that would have been a really good way to get into the close connection between statisticians and the eugenics movement. And she sat uncomfortably close to that. And I feel like that story has not been told. And then there's also like I, you know, most of the archival research I did was in the United States on the East Coast and archives that I knew from my own training. And so they just have all these obscure 19th century
Speaker: mid nineteenth century big giant charts that are so weird, you know, and most of them, like most of them are about like our great country. Like a lot of them are about the founding, about the presidents, about wars. And, you know, there's like totally bizarro and illogical and also fascinating, like vines and snakes and, you know, whatever. um And I wish that I had pushed myself a little bit more to uplift some of that. And for a while, actually, of the million times I rewrote the intro, one of the ways that I so tried to start it was just to pick three of my favorite totally weirdo charts and just be like, you could start with this one. Here's a reading of this. You can start with this one. Here's a reading of this. You can start with this one. But then I felt like it was almost too arbitrary, right? And I did because sort of anchoring, like,
Speaker: really trying to say there's an established history here. Like major geopolitical events were happening at the time that data visualization or a sort of modern European centered data visualization was emerging that were totally the running in parallel courses with what we only have taken to be just the charts themselves. Like it was so important to me that I'd be able to present sort of like overwhelming evidence that this was the case that yeah I found myself reverting to the images that had the richest, most powerful evidence attached to them, which tended to be people who are already to some degree like central to these stories. So a person like Playfair, a person like William Clarkson.
Speaker: Well, it just leads to the the to the sequel to the book. That's all. Right, I know. Another 12 years to write the write the next one. oh my but um So I do want to ask about getting the images. Maybe no one else really cares, but I'm always curious about about this part. So what was that process like? I mean, I'm guessing some of these are relatively easy to get from online library catalogs, you know, at least, you know, a lot of the stuff that I guess Library of Congress, they have a lot of the stuff, but like,
Speaker: How many did you have to go to like the original source and find it and scan it or photograph it? Like what was that process? Yeah. I mean, so this project is so old that my first images from my first archival research trip, I took with a digital camera because I don't think I had a smartphone. And so I actually have these images and I had a Zotero database. And so when I first went, I would take these pretty rough images of everything that I saw and attach them to the
Speaker: metadata entry from the light from the library, the archive, and then take notes. And if there was a PDF version, if there was a book. i was So I had this big personal database. um And so with some of those archives, you just need to go back to them and say,
Speaker: Like, hey, here's the catalog number. Could you scan this for me? And it does take some weeks or months, and you usually need to pay them. I mean, the thing is, I i knew from having done an old school monograph, like this is the process that all historians use. yeah um And I knew it would take time. And I, for the most part, accounted for that. But there were definitely a couple that fell through the wayside where we thought we had high res versions. Or a lot of things that would happen. There are so many different versions of a lot of these charts. And yeah we'd sort of have this placeholder image and I'd sort of forgotten that it was a placeholder. And then I'd be like, oh, we can't use the third edition because I'm actually talking about the first edition. And then, or like the other thing is because a lot of these are hand tinted or um you know even like early lithography the colors change and in some cases or they just fade and in some cases I was like oh I really want the version of the chart that this archive has like the Peabody charts for example i ended up using the internet archive version of them even though I have incredibly high res scans from the library company in Pennsylvania just because the Internet Archive colors are so much better, you know, at least like in the year 2026.
Speaker: I don't know like which library they got them from, but their preservation methods, either before the book got into their control or after, just meant that like I liked it better. And there's an image on the back cover of the book um where we were going to swap it out with the higher resversion. And then in the end, I was like, I don't i don't want to. I like the other ones. I like them better. No, but I mean, I guess like, but it's a long way of saying that and I actually think it's important for people who don't deal with old stuff to know. is like You feel like everything is out there on the internet and you feel like the image that you download is going to be fine, but yeah there's so that's not true, actually. and There are a lot of situations in which you need to go to a real expert who's
Speaker: you know, an institution even that has devoted its work to the preservation of these and of these original artifacts. And actually, in one of the, um I think in the Playfair chapter, we actually include an image of the high-res scan that we got from the library company of the specific version of the Playfair chart, including, like, the color bar and the, like, georectifying edges on the side, because I sort of wanted to make the point that, like,
Speaker: It's not like we just press download from Wikipedia. you know like right There was a person yeah who we who carefully put this in their very high-tech scanner and sent us something. With the white gloves and the whole thing. Well, no, you're not supposed to use white gloves, actually, because apparently makes your fingers too clumsy. And they've done studies, and they discover that If you balance out like human oils as long as your hand are clean versus the clumsiness that happens if you try to touch books and gloves, it's actually better to touch it with your oily fingers than a clumsy white glove because you're less likely to crack the pages.
Speaker: Okay, so now i I need to go back an episode in in Widow's Bay and tell the ah the librarian in that in the historical society not to use the Oh, that's on my watch list. I'm only just catching up on my shows now that the book is at the printer. Right is that right as the semester starts, you get to catch up on shows. I sometimes watch them on the treadmill.
Speaker: Oh, there you go. Okay, there you go. Okay, so I have um a couple of sentences I pulled from the book, and I want to read them to you, and I just want to get your, you know, more insight onto these, because these are the ones of many that that just struck me that I was highlighting as I was i was reading.
Speaker: All right, so here we go. So here's the first one. By placing each map or chart that we encounter in its particular context, we absorb a fundamental truth, that visualization is only as powerful as how and by whom it is put to use.
Speaker: So i just want to get your expanded thoughts on that, because I think that's a good, for me, a a really good encapsulation of the book. Yay. That's my, that's like my, the like the nut, like that's like the sentence that I, that was from the intro. And when I wrote that, I was like, that's it. That's what I want to say. So I, it makes my heart fill with, uh, I don't know, relief, pleasure and relief. Yeah. Yeah. no okay good that I mean, but, but really, I mean, that's like, that's really, I believe that so strongly. And I hope that people who read the book will,
Speaker: come to see how this is true, right? It's like anyone can design a visualization of anything for any purpose, but, you know, it is like a fundamental truth, I think, that you only activate the power of visualization when you ask yourself, like,
Speaker: Why am I doing this? What are my goals here? Who am I designing this for? And if you do answer those goals, you could be incredibly persuasive and you can make an incredibly captivating image. And if you don't, like you're missing out. Or worse, you might convince the wrong people or convince people of the wrong thing. Right? Right. And so, yeah. So that, thank you. Thank you for that. That's my favorite. Well, that's my favorite sentence in the book. All right. Awesome. Great. Let's start that one. OK, here we go. Number two.
Speaker: This is maybe a little out of context it starts with a word, but, but that's okay. Okay. But from the charts in this book will emerge evidence of the opposite claim that complete objectivity is an illusion that context and power always enter in.
Speaker: Ooh, you're a good reader because this is a sentence that I tweaked in page proofs, which they hate you to do because the type is already laid out. Yeah. Don't add a word. You gotta worry about reflow. i i know, I know, but I actually, i I struggled over this because what I was trying to say, what I believe is to be true is like, nothing is truly objective because things are all created by people in a context and you need to take that context into account, right? You can be looking at data, but someone scoped the data collection. You know, someone decided what you're counting and what you're not.
Speaker: You can be looking at a chart that is trying to show you only what matters, but someone decided what they thought matters or like when the You know, X axis should start in. And like, those are the things that I mean, you know, like you're never looking at something that is just outside of the conditions that created it.
Speaker: However, I'm very sensitive to the fact that we live in a world in which people don't believe in science and knowledge and expertise.
Speaker: And the word that I added at the end was complete objectivity. Because initially, and I think rhetorically, it sounds better to say that objectivity is an illusion.
Speaker: um But I really didn't want this to be misinterpreted as like anti-science or somehow a view that it was pointless to not pursue objectivity. And this is something that comes from like feminist science studies. But this is the idea that we, of course, are always pursuing the goal of objectivity. Right. But it's almost like an asymptotic goal, right? Like we're never going to everything, but we can always learn more, right? And sort of the more you know, the more you can either you can assess the objectivity of what you know, and the more you can sort of create conditions of sort of composite objectivity.
Speaker: Because you're saying like, oh now that I understand about the person who was collecting the data, then I understand like why they chose this and not that. And then I can go find that other thing that was missing and bring it in. And then I have more a more objective perspective, you know?
Speaker: um So anyway, so yeah, so that that like it's a tricky line. i think that it is true, but i i worried about saying that because of the time that we are in.
Speaker: I see. Okay. I see that, but I still, to me, it rings true. I get the point about where we are in this moment in history. Okay, one more for you. um And this one, I think, is from chapter one.
Speaker: This one rang true to me, especially because in the book that I have coming out later in the year, we really focus on the people behind the data. And this really... Had your book been out a few months ago, I would have quoted this in my book, but um ah my proofs are basically done. So, okay. All right. ah More basically and more profoundly, these diagrams, which which I think are referring to the slave ship diagram.
Speaker: ah These diagrams remind us that before there is data, there are people. People who contribute to the production of data and people who themselves become data, not always with consent. Yeah, I mean, this is another, you now like every time you read over your book, you're like, oh God, this is the thing that people can read and my name is on it, you know? And like, I'm simply at that moment. So this is, it's making me feel good because this is another line that I like. I mean, so this is a line where I was trying to, so the the example here is both historical visualizations of the data of the slave trade that were designed in their own time um in order to try to convince people to get on the side of abolition, but then also um our project team, as we were discussing earlier, decided to take that same data set or related data set and see what we could do with it in the present. um
Speaker: But I think what I was trying to do here is to say, you know, this lesson that there are people behind the data and you need to think about what they contributed to the data, whether they wanted to contribute to that data. And I mean, both like people who collected the data, but also who themselves might've just been counted. Right. Like it's not just a lesson that applies to the worst of historical circumstances. Like what I did, what I was hoping was that, and what I try to do throughout the book is to say, we're looking at these incredibly resonant examples because they're very powerful examples. But the lessons apply to everything, right? Like even when you're just like making a, like, you know, plotting the, like what biz did I make today? Like my Strava data recorded my jog today in the park, right? You know, like there's always like any chart that you're looking at always has people behind it. yeah And you can always learn something when you ask yourself, like,
Speaker: who made this, who's in it, who's being counted, who's not being counted. you know like i you know i think a lot about like like the Strava heat maps, you know which I use all the time, but it's like, those are made by people just like me.
Speaker: Yeah, yeah right right. Well, now AI. Actually, they replaced them with AI, which I'm really grumpy about because they're not they're no longer made by people just like me. But yeah no but i mean, that's the point, though, is it's like, You know, you the the the question is prompted by an example where you can't not ask that question, right? that's why i began with that one. But the point of the book is to say, just because it's not necessarily an obvious question to ask of any chart,
Speaker: doesn't mean that you shouldn't ask it. And actually, I think that it gives you better ideas about how to design visualizations differently and more truthfully and more creatively if you're thinking about how your data came to be.
Speaker: So yeah. yeah Terrific. Okay. I love it. I think that's a great place to end on. I wanted to ask, I mean, obviously people can get the book wherever they get their books, um but there is an online companion.
Speaker: Right. Or there's an online version of the book, I think. yeah there's an online transitions Yeah. Yeah. It's data X design.io and we're in the process of updating it for what I hope will be the final time it contains. like a condensed version of the text that's in the book. It's probably like 30% shorter. And so it is a little bit faster paced. There are some details that I ended up cutting just because no one likes to read a long thing on the internet. But very crucially, it has interactive versions of all the visualizations. the book is mostly screenshots or things were redesigned from the web to print but this was a native web project And so some of the coolest things, like we do a lot of scrollytales and like zooms and simultaneous text and image, especially in a lot of the process parts. So people can check it out. It's free. And free. And free.
Speaker: And it's teachable. That's what I wanted that. I wanted that one. Yeah, that's great. yeah um And where if people want to get you on their podcasts, or if they want to join your lab for the next 12 years for the next book, what's the what's the what's the best way to get in touch? Okay, so be careful what you wish for, speak coming from someone who took like nine years to get a PhD.
Speaker: um i'm I'm online, ah not not very, but like sort of. um lkline.com uh i work at emory university my email is there and of course like the last thing to ever be updated is my lab group's website it exists from like 2015 um but it does have probably hopefully like a working email address so okay awesome well i'll put those on the show notes people can check it out uh lauren congrats on the book it looks great um and uh i'm sure it's gonna do great and uh yeah thanks for coming on the show
Speaker: Thanks for having me. I'm so excited for your book too. I'm going to go your book now. Thanks a lot. Thanks for tuning in, everybody. I hope you enjoyed this episode of the Policy Biz podcast. Please leave me a note. Let me know what you think. Let me know if there are other guests you'd like me to talk with.
Speaker: And you should definitely go check out Lauren's lab at Emory University. And of course, you should check out the new book, Data by Design. You can get it wherever you purchase your books. So until next time, this has been the Policy Biz podcast.
Speaker: Thanks so much for listening.
