Hand crafted data (with Stefanie Posavec)

Hi, everyone. Data stories as usual with Enrico and Moritz. Hi, Moritz, how are you? Hi, Enrico.

Doing great, thanks.

Data stories number ten.

Yes. Wow.

We made it up to ten. I'm really, really surprised and excited. It's true that it's been almost one month since our last episode.

Yeah. We need to apologize. I think we need to apologize.

Or maybe not. I mean, yeah, it's getting harder and harder. I think part of it is the summer effect and maybe, I don't know, the Olympic effect, the Olympics effect. Moritz, possibly. Yeah, possibly.

I've been working on a project that has been quite Olympic over the last few weeks, and it's for the games at the last three weeks. Basically I've only been working. It was crunch time.

Yeah, I could feel it from you.

And we had a premium deadline with the opening ceremony, so we couldn't really shift anything around. But I guess we will do a special one when the whole madness is over. Special one? A special podcast on the project, I think.

Yeah, it's one of those cases where you just have to ship it, right?

Yeah, yeah. And we had a few curveballs thrown our way and not everything went the way it should have gone. And so it was.

Come on. It's. There is always.

Yeah.

You don't see what's behind the scenes.

There are always some glitches here and there. So how is the feedback from the project?

Good. Good. I mean, although what we hoped we would make it sort of into the sports fan scene somehow, but maybe it's a bit too abstract for that. Not sure. So it's, as a couple of my projects, it stays more in this design visualization scene and doesn't really go out of that.

Do you briefly want to say, what is this Emoto? And give the link, the URL in case.

Yeah, we already post the link. Yeah. So it's called Emoto. And what we do is we track Twitter in real time for messages related to the Olympic Games. And then we analyze if they are positive or negative. And we try to visualize the mood around certain topics at any point in time, in real time. And also try to do some more long term analysis, like look at certain developments of individual topics and blog about that. So there's a little data journalistic angle on that side. And now we're working on a data sculpture. So it's a big piece, like individual plates, one plate for each day. And it will be milled and with an overlay projection will be exhibited in Preston beginning of September.

Wow, it looks like a huge project. How many people are working on that. I guess it's not only you, right?

No, no. So it's me and studio NAND working on the design side. So that's four people and the front end programming we have Gerrit Kaiser, who is a backend programmer and we have at future everything in Manchester, the whole organization and the PR and the management and that's another, let's say four people part time. So it's a big thing. Yeah, yeah.

And are you. One thing I wanted to ask you, are you. Do you have direct access to the Twitter fire horse or firehose or what?

No, we use the garden host. So it's the 1% streaming API that everybody gets. But that's plenty. That's plenty. So I have on my hard disk, as we speak, 10 million tweets related to the games. And these are just english ones and just the ones that have an emotional classification. So they are like positive or negative. They contain positive or negative terms. Right. So it's just a small sample of the tweets we could have.

Wow.

Yeah, yeah, yeah.

Good. And you follow some. A number of hashtags about the Olympics or what?

Yeah, basically we search for, first for very generic terms like London 2012 Olympics or Michael Phelps, you know, something like that.

So I guess this is totally hard coded, right?

Exactly. That's one query that we run against the Twitter API. And then within these results, we then try to split up by the different disciplines, the athletes or topics like traffic and weather and so on.

Cool.

And this is what we work on. Yeah, but as I said, we need to talk more in depth about that by times. Yeah.

Cool, cool, cool.

Second big news, you are moving to New York.

I'm moving, and it's getting totally hectic here and I'm really happy to give this announcement for the first time on data stories. So we already have one scoop today. Yeah, it's great. I'm actually moving to New York City and I become assistant professor at NYU Polytechnic there.

That's so cool.

And of course, as you can imagine, I'm really excited about it and it's a big next step. And this doesn't mean that I'm happy to leave from Germany and from Konstanz in particular, because I really love it here, especially now in summer. It's an amazing place and if you have kids, it's just wonderful. You can go around. There is a beautiful lake and it's fantastic.

But, yeah, you can do that when you're old. You have to move to New York now.

Yeah.

And actually moving from Konstanz to New York. I think it's going to be a very interesting experiment. I don't know if I'm more scared or excited or both things at the same time. And I don't know what to expect. But for sure it's not going to be boring.

I think so, too.

Yeah. And I mean professionally speaking, I think New York, I'm really excited because I think New York is going to be, I don't know, really, really special in terms of developing new projects and meeting people and visualization is so much developed in New York that I'm really looking forward to meeting lots of people there. I already exchanged a number of emails with people that I always wanted to meet and I think it's going to be fun.

Yeah, definitely.

You should come over, actually.

Sure. I come for every data store is recording. Every two weeks I fly over and we have a coffee.

We should start collecting some money in some way or another in order to organize these kind of things.

Exactly. So. And today, let's talk about today.

Exactly. So. And today, let's talk about today.

Yeah, this episode is not to talk about ourselves. We just wanted to give a brief update. But we have a great guest. And Moritz, I think it's better if you introduce her for the first time. I think it's great that finally we have a woman.

That's true.

Yeah.

We should have more.

Datastore is ten. We should have more. But yeah, we're working on it. So our guest today is Stephanie Posavec.

Hello.

Hi, Steph.

Hi, Stephanie.

How you doing?

I'm all right, I'm all right.

You said already you're a bit, bit ill, but sick.

A little bit of a cold in this second winter.

We're having the warmer winter.

Yeah.

Moritz, Moritz, finally, eventually we have a real sexy voice in our data store. In our.

Oh, that's exciting.

Yeah. It's not the usual exotic hero voices.

No, that's our usual stick. But Steph, in turn has a funny mixture of American and British accents. Yes, it's fun because you're originally from Colorado, I think.

Yes. Well done. Denver, Colorado.

Denver, exactly. And now Steph, I mean, probably all of our listeners know her work, but I still, I should mention she's now living in London, is working as a, I would say you have two jobs, right? Your graphic designer.

Yeah, I'm a graphic designer. That's what I tell everyone, but I, I work with books and data.

Exactly.

But I'm. But mostly I work with data from books, so there's a bit of crossover. I used to be a book cover designer and now I often design websites or books or book covers or information graphics or anything relating to data. It's a mix.

Exactly. But you started off, like, your professional career started at Penguin.

Yes.

Where you were, like, doing the book cover design.

And then I quit to go freelance almost two years ago, so.

Yeah. And probably, I think you became famous with the literary organism, at least in the data visualization world.

Yes. Famous in the data visualization world. That makes. That's quite a special achievement. It's basically like being.

Yeah, no, but I would say at least, I mean, you were noticed then. This was, I think, your master's project.

Right.

And what you did there, if I can summarize that, is go through the whole of Jack Kerouac's book on the road.

Yeah.

And hand annotate it, which is, like, crazy. And you have these beautiful shots on your website with all this text marker, you know, annotations of the book.

Yes.

And you would classify sentences according to, I think, their content.

Yes, yes.

Or the theme of the book. Like, there were a couple of key themes you identified, and then you would look. Which sentence would support which theme.

Yes.

And then make these beautiful hand. Let's say. I'm almost saying drawn. I mean, it's done in illustrator, but it has a strong handmade touch to it. These different diagrams that. That illustrate the structure of the book and highlight different properties, like the sentence lengths of, you know, how far the sentences follow each other or the chapter structure and stuff like that.

Yeah, got it.

Is that accurate?

Yes, it's very accurate.

Very good. Yeah. And it's a beautiful project. And the funny thing is, people are always like, wow, how did she program that? I mean, it looks so organic and nice. And then you say, oh, I didn't.

Yeah, no.

Oh, my God. And that's, I think, the unique thing about your work that once you know how it's done, you get a huge respect for all the little mouse clicks that must have gone into it.

Yeah, I know. And I should learn how to code. I've been saying that for about two years now.

It still didn't happen.

No, no. I've started a little, anyway.

Yeah. But it's good. I think that's. That's what sets your work. Your work apart, right?

Yeah. But it's good. I think that's. That's what sets your work. Your work apart, right?

I mean, yes, that's. I'm going to use that as the base to do personal projects. I think the handmade aspect.

Exactly. Yeah.

Yeah.

There is one thing that I wanted to ask to Stephanie regarding handmade visualization. Did you, did you ever give a look to. So actually, visualization started as a handmade craft. I was wondering if you take any inspiration from, from people who, I don't know, in the 19th century used to draw their visualizations by hand or it's just something that you started doing. Regardless what other people have done in.

The past, I think I found a lot of kind of those, the 19th century visualizations. After I became interested in this way of working, I think I just came across it because I came from a print background and so I didn't really know what was possible. So I was just working with what I had, so. And what I had was just kind of my brain in my hands.

I didn't know.

So it was just like lots of, and, you know, lots of time on a masters. So I think it just came from not really knowing what was out there, like why I ended up working.

So you did not set out to say, well, everybody's like, doing these fancy information graphics, and now my twitter will be, I do it by hand. Right?

Yeah, I think because when I originally became interested in this, I sort of assumed that, oh, well, when I start my design career, I'm just going to work at a really cool small design agency and design logos and brochures and brands. And I never really, I didn't realize that there was a place for the type of work that I was doing.

Ah, okay. Yeah.

So it came as a surprise to me to realize that people would actually pay me for that. I thought it was just gonna be a project I'd sit on and that was it.

So, yeah, it's interesting because, I mean, you're not directly an artist, at least in the art sense of the word. You know, like museums and galleries, although you have been exhibited, for instance, at MoMA and so on. But, but I think it has more touch of artisanship or like craftsmanship.

Yeah, I think.

But then it's a very freely developed work, too. So it's so interesting, all these different markets you have, all these different niches where you can work. Actually, it's true.

Yeah. I think craftsmanship is probably a better way of describing it, particularly because people will ask me about my practice, and that seems like a very art word and I don't feel very comfortable with using that. So I think designer and craft feels a lot better for me.

Yeah, yeah, yeah. But still, like, the projects you're known for, they all, or many of them are self initiated. So there is this sort of, you know, your authorship is quite clear there.

Yes, yes. Now, the only problem, problem is trying to do that self initiated work. You know, while I'm working, which I still haven't quite figured out. Yes, I know.

Yeah. That's interesting. Yeah. And now you call yourself a data illustrator, right?

Yeah. That's interesting. Yeah. And now you call yourself a data illustrator, right?

Yes, that's why I say I call myself, though I'm still not totally confident enough to use it when I'm trying to get work. But I always put it in there next to information designer, so they get the gist. But do you want me to define the term or.

Yeah, I was actually going to ask you, how do you define data illustrator?

Well, I think I need to label myself a data illustrator because, and I think this is actually just a step along. You know, I give talks and probably what I say I do evolves with every single talk I give at a conference. So I'll probably change my mind next talk. But it's just because when I would present my work, people would kind of hold my work to kind of the rigorous standards of data vision, like more traditional data visualization or information design. And I felt like there were a lot of ways that I used data that didn't really seem to, that sort of meant that my work didn't really fall specifically into those two categories. So I think there needs to be another, there needs to be something called data illustration, where you're using data just in, possibly in a way to communicate a more subjective emotion, or you use data to fulfill a design brief where the data is secondary to another more important message. I'm just trying to find a way of explaining how I, as a graphic designer, like to use data to sort of arrive at a design outcome. And sometimes it is different than being a strict data visualization or a strict information design thing.

So what do you say? Go ahead.

Sorry, Moritz. No, I just wanted to say that in a way, I really like the term because in a way it describes much, much better a lot of, I think, a number of works that don't really fit into the label of data visualization. There are lots of things that I wouldn't directly call data visualization if they have the alternative of calling these things data illustration. I think it's really, really nice.

I like it because I don't think it's. I know there's data art, but it's just, for me, I, you know, it's, this is how I arrive at the kind of the very mundane thing of designing a CD or designing a poster or, I don't know, arriving at a solution for a graphic design commission or something. It's, it's not, I like that. It's not lofty, it's not meant to be put on a wall. In every case.

There's something to the term definitely that captures quite well what you and maybe a few other people are doing. And probably some aspects of my work are the same that I say, okay, the data is the one thing, but it's also important what happens expression wise in the originalization or what happens between the lines and what is expressed beyond the pure data communication. Right.

Yes. Yeah. So it's, what's beyond that, I think is really interesting.

Yeah. So maybe traditional data visualization is more like photography, like where you try to, and not the art artistic type of photography, but maybe trying to find like photojournalism, really one to one representation of reality, something like that.

Yeah.

And if you're an illustrator, you try to bring, or you will often bring in more of your personality. You take reality as sort of the starting point, but you might get carried somewhere else. Right.

Yeah. Yeah. And it's not style because I think an illustrator probably would be offended if I said that. But you're right, it's their personal, there's a subtlety to it where you're communicating things, but they're not as not so easy to define what you're using to communicate that message.

That's interesting. But is it also, I mean, I can see how you use the term to change the mindset of people judging your work in a way, you know, or the perspective, so they don't judge it with the same criteria. But is that maybe also a bit defensive in a sense that you say it's just illustration? I mean, that's a provocative question. Maybe.

Yes. No, no, it's not provocative. It's, I think it is a defense just because I'm, you know, like I've discussed before, I sometimes I do feel quite intimidated by the, I think people can be quite harsh critics when they're dealing with data visualization and so on. And so kind of in order to kind of, I don't know, to let, to ask people to let me be or like to prepare for it, then it's a valid. Then I say, well, I'll just make my own definition of what I'm doing.

So have you heard that argument quite often that, well, that's all very nice and colorful, but I cannot read anything from it, or I think your data encoding is not that accurate. I like have people, I think, well.

It's more like, is this functional?

Yeah.

How does, how is this functional? How does this add insight? Yeah, how, yeah, is this functional? How does this add insight? How is this better than reading the book, you know, and, but in a.

Way, I mean, let me, let me, let me say something. I think honestly, I mean, I have had huge debates about this thing on functional aesthetics and all this stuff and whether the visualization community, if we can really define one, is too critical or if it's too hard to enter into this world and so on. And from the one end, I fully agree that there is a tendency to overly criticized works. But at the same time, I think what we have here, it's a nice story about, because of this criticism, you've been pushed to define your work in a different way and maybe come up with a label that is going to be useful for some other people. And I really like the fact that you are not pretending to frame your work into the same frame other people use. And just say from the very beginning, look, what I'm doing here, it's something I call differently. So if you want to look at my work, you have to put another pair of glasses. Is it correct what I'm saying?

Yes. And so, I mean, I think, yeah, that is a good thing to come out of it, though. In the beginning, when I was, when people were first asking me about my work, I think I kept trying to put it in a. Yeah. Data visualization point of view, but it has, it has evolved. But I think now, yeah, I suppose the criticism has been very useful. And I think now I just realize that, you know, everyone is who they are. And so I, you know, I think what I'm interested in is going to be very different than what a lot of other people that work with data are interested in, and that's just the way it's going to be. I suppose so.

Yeah. Yeah. I mean, in other words, I wanted to say that I don't have a definite position on whether this level of criticism is good or bad according to the angle I used to look at it. Sometimes I think it's good and sometimes I think it's bad. I think in a way, I wouldn't like to have a situation where there's no criticism because criticism, even if it can be painful from time to time, it also helps developing this kind of new things. So I really don't know whether it's better to have what is defined. I mean, what is the right point there, the middle point there? I mean, I think, I wouldn't say that having criticism or even harsh criticism is necessarily bad in a way. It can also. It has a function itself.

Yeah, I agree with you. I think, like I just, earlier this week, a collaborator and I showcased some work, and we were, we got some very tough criticism, but they were, it was well intentioned, so it was really useful. I'm happy we got it. I think the difference is when people are harsh critics without meaning, well, you know, or maybe they mean well, but they're, I think as long as, you know, somebody's on your side when they give you that harsh criticism, it's okay. But if you feel like they are against you, then it's, I don't know. And when it's more confrontational, I mean.

For a certain time, it certainly felt like the traditional visualization scene was offended by how people use visualization.

Yes.

And I just recall, you know, Robert Kosara talking about lines in the sand and, you know, which side are you on? A few years ago, it was a bit aggressive towards non standard uses of shapes.

Yeah.

I think things are a bit more relaxed by now, and I think everybody understands a bit better what the other side, if it is, is up to. But I can see where that comes from. Yeah. And I just realized the literary visualization, I mean, it's something that people haven't been doing before. Right. I mean, that's the other thing. Like the whole, the whole goal or the whole theme is so different from what you would traditionally visualize. I just recalled, of course, Boris Miller, who was my professor in Potsdam, who has been doing all these poetry visualizations, poetry on the road, and also Stefan Thiel, one of his students, who did the understanding Shakespeare project as a master's thesis. So these were all very, let's say, I think if you work with literature, you have to. Maybe you have to work a bit differently, too, you know?

Yeah. I mean, it's inherently, you know, when you're gathering the data from literature that you're already putting your own personal stamp on how you categorize that data anyway. So I. But it'll never be perfect. Or maybe it will, but.

Yeah, but it's by no means hard data.

No, no.

And we have the same now with the tweet analysis. So we do have automated analysis of the tweet content and the tone, but in the end, it's very fuzzy data. And maybe you have to work in a very fuzzy and personal way with that. To be more honest about the whole project, instead of visualizing the precise values, I'm fairly convinced that this is the way to go there.

Yeah.

Nice.

Yeah. But I like how you've, like how you found that definition now and that term. I think the term is great, too. Data illustrator. So let's hope it helps a lot of, like, talents, you know, to say, yeah, that's what I'm doing and this is why. It's why it's good. Yeah. It's important to have this a proper profession. Know what it is. Yeah.

Before we move to the next point, I'm wondering whether Stephanie has any suggestions for. For. I don't know. Let's imagine that among our listeners there are some novice data visualizers and maybe they are totally scared about the community. Do you have any suggestions for these people?

Talk to the two guys that run data stories. You guys all seem quite nice, actually.

That was prepared.

They paid me.

Yeah. Euros, easily earned. And now we get all the emails. Fantastic. I think Steph does great advice, too. Yeah.

What would I tell people? Well, it's not as scary as you think, I think. I don't know. What would I tell people? I think it's. Can you repeat your question? I need to think of an answer.

So let's imagine that. I'm sure that among our listeners there are people who are data visualization novices. They are just starting and maybe they are as well, totally scared about reaching out to the large public because they see that there is a lot of criticism as soon as they put something out. So since I think you went through this painful moment and you actually, it looks to me that you solved it by just coming up with a new term. And I think it was brilliant. Do you think there is anything you can suggest to these people how they should react to this kind of criticism?

Oh, God. When they get criticized and then. Or just how to prepare yourself for it? I think, um. Oh, I don't think it's very difficult. I think it's difficult for me to handle criticism online. Um, I think. I think the only thing I can tell people is that you have to let a lot of the online criticism go just because, like, it's. Half of it is, um, it's just because they're behind a computer and.

Yeah, but if somebody's wrong on the Internet, you cannot just ignore that. Yeah, you have to do something. No, but you're right. I mean, just, you know, there will always be, like, funny comments to anything on the Internet. So. Yeah, it's part of might as well be your work that gets a funny comment. And I totally agree. Don't get hung up on that. That's. Even though it's not healthy, I say.

This now, but there is definitely one person, one person out there who criticized my work in the beginning. I won't say who it was.

Yeah, we want. Now we want the name.

Oh, I'll tell you guys after the podcast. But it was that. The one person that I just. It's nobody you know, but it.

Is he into bar charts?

No, no, I don't think it was a data person, actually. I just found it. Yeah, exactly. But I don't know. I think. Yeah, you have to ignore. It's true. It's better to have your work being criticized than other people's because that means that your work is out there. So I think it's a reasonable trade off, though. It does make you. It does remind you to not just criticize people's work without even thinking it through. I mean, there's. Sometimes I've wanted to say something about somebody's work on twitter that might not be the most, like, you know, it might not have been good criticism or. I've actually just started to type it and then deleted it because I know, you know, it's actually not really nice what I was gonna do, so.

Yeah, sure.

Yeah.

I remember reading some time ago a book from. I don't remember exactly who he was, but I remember it was one famous blogger, and he basically called. It was nice because in this section, he was saying, look, every time you put something out on the web, you will always find some vampires, and you have to get rid of vampires. I mean, just. Just don't care about it. They are vampires, and they suck your blood just by writing catalyst sentences. And that's it.

Yeah. If only there was a button that you could press that would just make their post explode or something right in front of you, that would be useful.

Yeah, and hurt them a bit. No, it's a lesson to learn. I also got hung up on, like, really senseless help, not helpful discussions on Twitter and on blogs, and rarely something productive comes out of it. I mean, in some cases, it was also good to fight a few fights just to make us dance. But after five times or so, it's not funny anymore.

Yeah.

Yeah. Moritz, I don't know if you have another question ready. Yeah, I mean, otherwise, I have something I want to talk about.

Yeah. Moritz, I don't know if you have another question ready. Yeah, I mean, otherwise, I have something I want to talk about.

It's a general point, more like. And it's something you said, I think, at IO, maybe c conference. I've seen so many talks from your staff, I can't even separate them anymore.

You're my biggest fan.

Yeah, of course. I travel after you. Didn't you notice?

Yeah. You're outside my house now, aren't you? You.

Oh, sometimes, yeah. Anyways, so one thing you said, and I had to think about that a bit like that. So that the time you spend with the data, you know, that it's not just because, so you don't just press a button and there's your big tree graphic, but you have to construct the tree yourself, basically. And that this time you spend and the effort you put in sort of gives the graphic more weight, basically, or more, I don't know, that this invisible time you dedicate it might give it more value after all. And I found that really interesting because, I mean, there was this talk by J. Thorp last year about the weight of data, like that we should think about what is behind the data. And Jen Lao gave a similar talk at I o about data and memory. And so I was thinking about if you had any thoughts on this extra aspect that again, what goes beyond the pure graphic, that the time you spend with something is alone creating value. Does that make sense or am I like. Totally.

It does make it make sense. I just need to think about how to.

Very high level.

Yeah, yeah, we're getting philosophic here.

Smoked so much weed before the podcast. I'm sorry.

I think there is something about the weight of data and then the process of engaging with it that like, the reason I like to do things by hand, or I tend to go towards projects where I have to engage with the data on a really intensive level. I think data is really heavy. But in. Sometimes I think it's a shame when you don't know when you look at a graphic and it was just a button was pressed and then it just popped out. And I think it's a shame when you don't realize how much like blood, sweat and tears went into kind of collating it and kind of put bringing it all together. I think it's a cr. I don't know, handcrafted data. I'm not really making any sense now, but I remember when I talked to IO, I talked about handmade versus computer made. It was like a hand knitted jumper versus a machine knitted jumper. Like, you know, they're different. But, you know, I mean, there's something in tri. There's a difference between the two, but we always lend more value to the handcrafted thing in many cases. So.

Yeah, yeah. Because it's personal, because of the person behind it. Exactly.

Yeah.

There's something attached to it. Right. It's not just the thing, but it's more.

And perhaps that also goes back to what you said earlier about how, you know, it's kind of projects that work with data that go beyond the data. So perhaps that bit of dealing with data is another bit of the beyond. It feeds into that. Beyond that, I like to look at, I guess.

Yeah, it's very interesting. And I think if you're very much into code and statistics and numbers, you might forget occasionally about what stands behind the data. And if you actually take your time and think about each number and how that number came about, I think you might develop a different relation to the truth behind the data. Actually. One more question. When you work, like, how many decisions do you make on the fly? Like, do you have, let's say, an idea for graphic in mind before you start, and then you just execute it more or less like a robot, or do you make the graphic up as you go along?

You mean if I'm kind of doing it all myself?

Yeah, let's say the Kraftwerk poster.

Oh, that one. Hmm. I had an idea, and then this Kraftwerk poster, just for anyone, it represents the. To scale the amount of cassette tape needed to record Kraftwerk's Computer World. So for that, that one, I knew I wanted to represent the entire amount of cassette tape needed, but I wanted to make it look interesting, so I made it spell out computer world in letters. This never ending length of cassette tape. It's like a labyrinth that if you draw a pencil upon or trace your finger upon it, you'll. You'll trace your finger along the entire length of tape needed to record that album. And then I'm a real type.

But how do you do that by hand? Like, match the length and have a little typeface, you know, it's like I.

Started to work through it and illustrate. How did I do that? I don't remember. Is that weird?

Probably. It took you two weeks without sleep, and this is why you can't remember.

Oh, oh, okay. No, no. I use. You can determine length in illustrator of lines. So I kept creating them, and then I would, like, kind of trace through it with a pen line in illustrator, and then I would figure out the value, and then I would make all of my letters and then add it all up. And then if it was too high or too low, I would begin to add more, like, loops and turns in the library.

So you first did the outside, and then you noticed. Okay, I only spent, like, 20% of the tape, so I can do a few more loops inside.

Yes. Yeah.

Good thinking.

Yeah.

And then the way that I added all of the. Because you can also see by every minute, is marked out on this labyrinth where you are in the recording. And I did that by, again, making a tool that had one inch or 25 millimeter dots set 25 mm apart. So I use that to measure and use. I'm not making any sense, but basically just lots of kind of really wonky ways of using illustrator.

That's what I figure. No, but it's amazing. But you have, like, let's say you have a rush visual idea, but you still develop the details as you go along, right?

That's what I figure. No, but it's amazing. But you have, like, let's say you have a rush visual idea, but you still develop the details as you go along, right?

Yeah. Or if someone, I mean, if I ever work with anyone else, if it's something that I can't, if I need someone to develop it for me, then I usually end up sketching or creating, like, the graphic rules, the parameters of how you visualize that data. I'll do that in illustrator, and maybe I'll do it with a little bit of test data. So I'll begin to understand where the high point and the low point, low points within the data fall and sort of what distribution and what the data looks like. And then, yeah, create the designs based upon sort of how the data looks and feels and then test it and then give it to someone to code. So I'll send them a series of PDF's that are ridiculously instructive, telling them percentages of shapes and colors. And if this, then that, if this and that. So it's like writing a program, but it's just all done in PDF form in a really annoying way for a developer, I'm sure.

So, Stephanie, I'm wondering, is there, are you able to describe to us whether you have some sort of predefined process that you normally follow? And I'm sure that our listeners would also like to have more details about the tools you use.

Oh, okay. Well, if I need to visualize something, what will I do? I'll get whatever data I'm working with, and I'll try to, even if people don't have all the data at that point, I'll try to ask questions about, like I said before, you know, what I figure out, what are all the points that I need to, what are all the different types of data I'll need to represent? And then I'll ask who I'm working with, who's giving me that data, where the high and the low points are, what the range is for a particular type of data. And then I'll start to think about, is there any way that I can use color? I mean, is there a metaphor for this data? Is there a way of converting it to form where the form, I don't know, adds more insight to its message? So, like, from this. Ok, go. Album artwork that I've done because I was dealing with text that dealt with rainbows and the spectrum of light and diamonds and mountains and so diamonds and rainbows and circles and prisms and those were the shapes I was using in the ideas that I was coming up with. So, and then I'll start to sketch on paper with pencil how I think that data could be represented. And then when I'm happy with some pencil sketches, I'll start working them up in illustrator. And then at that point, I'll probably show a client or whoever I'm working with and then choose colors and figure out what data I can attach to those colors.

So do you show the client alternatives or is it more like, okay, this is what we're gonna do?

No, I don't have that level of confidence. So I'd probably show them four or five options.

Okay. And these are really different options.

Like, yeah, well, hopefully sometimes maybe three main options and then a couple more that I'm, I can't bear to throw away.

Yeah.

So.

So does it mean that in the intermediary step of your process, you're. Can you hear me?

Yes.

Definitely, yes. Oh, sorry, I thought I had a problem. I was wondering, does it mean that during the intermediary steps of this process, you are never representing the original data with your visuals?

Yes, that's right.

Yeah, I think it's interesting. And then what? I guess sometime when you arrive at the final step, when you get to the final step where you visualize the real data, you might find that there is a mismatch or something. Or not.

Yes. Oh, yes. So then at that point, in an ideal world, because sometimes I don't think there's time, but if ideally, then I want to see what that data looks like and then work with the developer to possibly change colors or shapes or the scale of various aspects. So then it's right. So then it's correct. But usually, and then, like I said before, you know, before I do that, I hand this PDF document where I just sort of pick apart my graphic and I basically, visually and with loads of text, show them every single visual rule and what type of data is attached to it. So I do basically, yeah, I write them a set of rules. It's just that because I can't translate that into code, it's kind of, I need that person to do that for me.

And do you ever try to, so once you have the data in your hands, do you ever try to first give a look to this data in a way that I don't know, with simple diagrams using software like Tableau or anything similar, just to make sense of what are the distributions there or whether there is anything that you, you might want to put in focus on stages of your project.

Yeah, yeah, I will. I will do that. But then sometimes, you know, sometimes people ask you to do things before they're giving you any real data.

Okay. Yeah, sure.

So. Which I feel like I tend to do more projects like that then the other way around than having all the data up front. But yeah, no, I'll try to find tools online or even just illustrator, just dropping things into spreadsheet or Google just to understand what I'm dealing with.

Yeah. And sorry, Moritz, can I ask another question? Or you have something urgent that you want to ask? It's getting more and more interesting. I'm really curious about that. So you initially mentioned that you would like to learn how to code. And why do you want to do that? I mean, couldn't you just continue doing your work that is really brilliant without coding? What do you think would be the advantage of being a coder?

It would just be good to automate a few things.

Less tedious, I could imagine.

Yeah. And then also because just to understand what is possible with code. So maybe I don't need to code, but I would. What I'm trying to understand is what is possible? What can someone do for me? You know, what can they pull off the Internet? How can they grab all these? Where can just. How you do it, what's possible, really? So. Or just little tools to speed me along, I think, you know, counting words gets old after a while in some cases, but, yeah, makes sense.

Makes sense, yeah. And did you already choose any specific, I don't know, programming language or environments or anything like that?

Or what I would be interested in using? I've tried processing and. But I mean, I do work like my brother in law, who I've collaborated with, and he's used r and processing to create the visuals that we've both collaborated. Collaborated on. So, yeah, yeah, this is exactly what.

I would suggest, but I think Moritz doesn't agree with me.

Oh, processing is good. I mean, I actually suggest, I think.

The combination r and process is just perfect. I love it.

I mean, I would rather go for, I don't know, python and Python Tableau and D3 or so. That's probably my magic. See, no, but what I suggested actually is scriptographer. And so it's an illustrator plugin that allows you to write JavaScript, but referring to illustrator graphics.

Yeah. I downloaded that and it makes sense now. As a graphic designer I understand how people are making certain, certain things, certain graphic elements now. No, it looks good, I just need to learn how to write JavaScript.

No, it's true. Some of the demos in scriptographer, you see them and you're like, ah, I got it. How this poster came about, or how this, you know.

Yes, exactly, exactly.

No, and it's, it's, it's nice and simple and you can just, I think it's a nice, I think there's so much power in this combination of automation and manual work, you know, to do it or to find a smart combination there and scriptograph. It could be something where first you do something by hand and then you have the program generate something in addition and then you tweak it again by hand or something. So these workflows.

Yeah, exactly.

They haven't tried it, so then you.

Trace it all out. It could work with a pencil. Yeah. Turn out, convert it all to pencil.

Exactly. Spray something like that, have the machine sorted again and. Yeah, yeah, it's interesting, this, this whole handcrafted thing is, it's very fascinating and, and I think a very powerful paradigm. At first it sounds so tedious and so.

Not, so, it's a little bit tedious.

When you see the outcomes you realize, I think, I mean, if you have an eye for that, then you realize that there's this quality which is very hard to achieve by programming alone. Once you, when somebody has laid out something by hand.

Yeah, yeah.

Yeah, and I agree. What is even more interesting is this mix between having a tighter collaboration between the machine and the human. I think we are still very far from having optimal tools that help people find the right balance between asking the machine to do one part of the game and the other part is more manual and make it flexible enough to, to have some sort of dialogue with the machine. I think this is probably one of the next big steps.

Mission cyborg, you mean?

Yeah, yeah, exactly. No, but it's true, it's true, I see it all the times. I mean, the limit of machines is that they are in a way or another, they are not flexible and they cannot really capture subtle things that for us are obvious. Right. I think there's no way to let machine, I mean, I don't want to open can of worm, but I think machines will never do some of the things we, we are able to do. So the best way is to find the right, the right way to have a dialogue with the machine. Right.

Everybody does what he or she does best and we work together and so on. Exactly.

And I think you mentioned something similar in your I o talking. I think you. Not exactly that, but you mentioned that in one of your projects. I think this one. It was the one from Max Planck. There are things that you had to do manually because you weren't satisfied with the look and feel.

Oh, yeah, no, the World Economic Forum visualizations. Yeah. There. We actually tried to do something like that that would give the experts a tool to work with an automated layout, and they could explore all the possibilities that this automated layout gives them based on the constraints of the data, but they could still pick the one that would fit their mental model best. And then I would take that and make a manual layout based on that automated one, which is much more, you know, much more readable and looks much more balanced and nice. So, yeah, I think there's a big potential in this ping pong between, you know, doing automation and then something handmade and back to automation and so on.

And I think it's the same. The same idea behind what is called this memorial thing done by year. Torp, I think you mentioned.

Yeah. The layout of the names for the 911 memorial in New York was done in a very similar way. Yeah. So the first, they had these. They solved like, a big chunk of the problem already. So the idea was to lay out the names so that people who knew each other would be next to each other or, you know, who work together. And so it's an optimization problem. And at the same time, they felt the last 10% or so, you know, cannot be done by a machine, but have to be solved by hand. So they developed that tool that would first do an automatic layout, and then all the problematic rest would be solved with human thinking. And I think that's a strong approach. Yeah, yeah, yeah. Cool. Yeah, it's exciting. Computers, humans, everything. Yeah. We could close with a few more examples of handcrafted data visualizations we liked, or. Is there anything, Steph, anything you want to talk about? Any shout outs? Any.

Yeah. The layout of the names for the 911 memorial in New York was done in a very similar way. Yeah. So the first, they had these. They solved like, a big chunk of the problem already. So the idea was to lay out the names so that people who knew each other would be next to each other or, you know, who work together. And so it's an optimization problem. And at the same time, they felt the last 10% or so, you know, cannot be done by a machine, but have to be solved by hand. So they developed that tool that would first do an automatic layout, and then all the problematic rest would be solved with human thinking. And I think that's a strong approach. Yeah, yeah, yeah. Cool. Yeah, it's exciting. Computers, humans, everything. Yeah. We could close with a few more examples of handcrafted data visualizations we liked, or. Is there anything, Steph, anything you want to talk about? Any shout outs? Any.

Yeah. The layout of the names for the 911 memorial in New York was done in a very similar way. Yeah. So the first, they had these. They solved like, a big chunk of the problem already. So the idea was to lay out the names so that people who knew each other would be next to each other or, you know, who work together. And so it's an optimization problem. And at the same time, they felt the last 10% or so, you know, cannot be done by a machine, but have to be solved by hand. So they developed that tool that would first do an automatic layout, and then all the problematic rest would be solved with human thinking. And I think that's a strong approach. Yeah, yeah, yeah. Cool. Yeah, it's exciting. Computers, humans, everything. Yeah. We could close with a few more examples of handcrafted data visualizations we liked, or. Is there anything, Steph, anything you want to talk about? Any shout outs? Any.

Any shadows? Hi, mom. I wonder she'll hear this. I'm sure it is, yeah. Hi, mom. She'll find it. That's all. That's my only shout out. Yeah, that's it. No, I'm working on a handmade project, and it's taking a very long time, so I will let you guys know. You'll be the first to know when I haven't finished.

What is it about, roughly?

It's about words in sentences and language and grammar.

Arbitrary sentences.

It's a very specific sentence. I'm not ready yet. I'm not ready.

So you're going to handle it? Tag clouds, basically.

No, no, no. Something a little bit better than that.

Don't say tag clouds. Some people.

I wonder if anybody does that anymore. Probably somewhere. It'll be an ironic thing in a presentation right now. Yeah, yeah, yeah. No. A few projects I liked. So Brady and Nash, she did something, I think, also for herself. It was a study project, I think. But what she did was she took her diary and analyzed how often certain words occurred. So that reminded me also of how Steph works. So she went through a very personal text and analyzed it by hand. And then she made paper based diagrams, but out of cardboard, and glued them together and made these little sculptures, and then took, again, photos of that and made a book out of that. So that's sort of a twisted way to come to a data representation. But the results, you know, they have this really this warm and nice feel and this unique style. And so I was always really impressed with that project. So we'll put the link in the show notes so you can take a look at that. And Steph also suggested somebody we also met at I o. Have you talked to her, Steph?

No, I didn't get a chance to see her talk either. Look at it online.

Natalie Miebach. And she does data sculpture, you could say. And it's based on weather data often. But that's my impression, at least. It's about climate and weather, and it's like basket weaving techniques and pearls and beads. And it's crazy.

But I like it because she actually has to touch the data she's using, spending so much time making her sculptures. I think it's this nice when you get to touch people, touch and directly work with the data that they're looking at.

No. And she's really like. I mean, you do it in illustrator, so that's fairly mediated. But she's really, like, making sort of a data basket and, like, has it in her hands. And probably that's the sculptural aspect. That's the furthest you can go there with the handmade. That's absolutely right. Yeah, yeah. Her talk is great, too. I saw it on Jimmy.

I keep losing you.

But I can only recommend that we will link it.

Hello?

Still there?

I am here more.

It's Stephanie. The line is getting very nice.

That's bad. But then again, we can also.

You don't hear me because I'm. I cannot hear you.

Enrico cannot hear as well, so probably we can also wrap it up. I mean, one last thing I would like to mention is data sculpting wise. We had this episode on food and data visualization, and I was mentioning this one workshop idea where people would actually cook data presentations or produce dishes based on data. And we actually made that happen. And so in September, there will be, in the context of the open knowledge festival, there will be a data cooking workshop. And so you can register on the site, and it would be great if a few people would register. They're still open spaces. So unfortunately, Skype is now breaking up a bit. So maybe we try recall or we just wrap it up here. I'll see if I can reach the other guys for a second call. 1 second.

Enrico cannot hear as well, so probably we can also wrap it up. I mean, one last thing I would like to mention is data sculpting wise. We had this episode on food and data visualization, and I was mentioning this one workshop idea where people would actually cook data presentations or produce dishes based on data. And we actually made that happen. And so in September, there will be, in the context of the open knowledge festival, there will be a data cooking workshop. And so you can register on the site, and it would be great if a few people would register. They're still open spaces. So unfortunately, Skype is now breaking up a bit. So maybe we try recall or we just wrap it up here. I'll see if I can reach the other guys for a second call. 1 second.

Hello?

Getting better. Let's wait for a moment. Can you hear me now?

Yes, I can hear you now.

I can hear you too. So.

Oh, now I can hear you.

Yeah, Skype is getting worse. And it's also during daytime. Often I, you know, in the evening.

Maybe we should try with the hangout.

Yeah, maybe next time. Hang out. Yeah, next time.

So I don't know how to recover from that.

I can just cut it out. I hope I can cut it out. I mean, I, in the meantime, while I was offline, recorded this open data cooking advertisement for the workshop. So that should be fine. And then we can, I think we can just rough it up. Yeah, I think we can cut out the part where we sound like robots. I mean, we don't hear that because. Yeah, hopefully, I mean, it shouldn't be in the recording.

So I think you should leave it all in there. Like just what everyone say.

Hello.

Hello, hello.

Maybe in the out section.

Yeah.

Yeah. No, that was great.

Let's wrap it up. Yeah, actually, I had one last question for Stephanie, and it's whether I have a whole new apartment to furnish and I wanted to hang some beautiful illustrations around. I was thinking about taking something from your work, and I think I can print it somewhere, right?

Let's wrap it up. Yeah, actually, I had one last question for Stephanie, and it's whether I have a whole new apartment to furnish and I wanted to hang some beautiful illustrations around. I was thinking about taking something from your work, and I think I can print it somewhere, right?

Yes. Yes. You can print it on image kind, but yeah, on the imagekind.com website. But I can send you a link.

Okay.

And I probably have a more up to date. It depends on what you want. I have a more up to date version of some of the graphics, so.

Okay. Do you have a discount for friends for an interview?

Okay. Do you have a discount for friends for an interview?

Well, let's talk over email. I don't. We'll do email, but I don't. If I get, if I get thousands of people emailing me after they hear data stories asking for a discount, I'm gonna be very load of new friends.

Yeah, that was a trap. No, but really, I'm serious. I'm really thinking about. I would really like, like to start hanging visualizations around and I think I would start from some of your drawings. I really like them.

Oh yeah, definitely email me. I've already got a thousand emails in my inbox asking for a discount now. Already.

Do you guys have any of these visualizations on your apartments?

No.

No. Moritz, do you have any?

So behind me there's a big poster by lust. You know, lust in Holland. They are crazy.

No, I actually don't.

I mean, they are also in between generative design and data visualization. So it's a borderline activity again. But they made this huge, it's really big poster of the North Sea and it's like small multiples all overlaid all different kinds of information about the North Sea. So that's behind me. And rest is my own work mostly, I have to admit, it's a little self.

A little self exultant.

Exactly.

I like to be surrounded by great narcissist.

I don't have any. Only because our flat's too small. But we're trying to move. So next place, hopefully, so there will.

Be more space for base.

We should all have much more posters. Yeah, it's true.

Yes, yes.

It's image kind. Can you recommend that?

It does? Okay. I think it's the best kind of on demand art print site.

Exactly.

Out there because I don't need to hide infinite posters under my bed like I'm doing now with another print. So. And I mean, I haven't had any complaints. And you can print on high quality paper.

So I also looked like a year or two ago for like on demand prints and you know how that would work. And I also had the impression image kind was the best service. I mean, the downside for Europeans is that it's based in the US.

Yes. It's very expensive these days, isn't it?

But now we have a us connection. Right, Enrico?

Yeah, yeah, yeah, yeah. You know, I get a fee for.

Exactly. So if you fly back and forth.

Yeah.

To bring loads of posters. Very good. That was great.

Good.

It was great having you.

Great.

A worthy guest for our 10th episode, I would say.

Thank you.

I really liked it. Thanks a lot. And I hope to meet you all guys very soon somewhere around the world. And if you want to come to New York, please do.

Oh, definitely.

First we have a couple of seminar rooms we can use for giving talks.

Yeah, sounds great. Yeah. You have to keep us updated on how it goes. Maybe you should start. Who goes to New York? Blog or something like that. Or a tumblr at least.

Yeah, actually, yeah, I should mention that I plan to revive fell in love with data at some point. And, yeah, the main reason why I was so silent is basically because I was working on this old thing and it was just crazy. And it's still totally hectic. But yeah, I'll be back.

Once you live in New York, you will have plenty of time like everybody in New York. Of course. I'm not worried there.

Yeah.

Okie doke. Was great to talk to you.

Okay.

Thanks, Stefan.

Thanks to you both. Bye.