From the course: Data Visualization: A Lesson and Listen Series

Listen: Julia Krolik

(upbeat music) - Joining me now is Julia Krolik. An information designer, data scientist, artist and entrepreneur. She's a creator of data art through her studio, Pixels and Plans. Where she and her partner Owen, design, code and strategize to develop innovative products incorporating both art and science into their methods. Julia, welcome and thank you so much for being here today. - Thank you so much for having me Bill. It's really exciting, glad to be here. - So, Julia, you're a data person, I'm a data person. Let's just jump over everything and start with the toughest question of all, okay. So, as data people, why should we care about touchy, feely art stuff, right? We care about cold, hard facts. Numbers, evidence, data. Make the case for art first. - I mean, that's a really, really great starting point if we're talking about data art. So, I mean, if we go back to kind of the old original definition, if you will, of art. Art is kind of this idea of form and aesthetic and then, design, which I think where data visualization sits. It's this idea of, you know, form and function, if you will. And I think that, the biggest thing for me at least, as I walk these two kind of paths is that, art allows that complete creative expression. But the other thing is you can reach new audiences with it in like, unconventional ways. So, and also puts things in a new light and yeah, I think that it's so important to enable that kind of creative control and creative expression especially for folks that that want to go that route. And we also know, especially, you know, how hard it is to push facts at people all the time. And I think art is kind of that, maybe almost an antidote. - I guess I like to think of data art as being data visualization, where accuracy and precision of the data is maybe less important and the aesthetics and the meanings behind the data are maybe more important. And that's a pretty fuzzy line of course, because you could make the case that, you know, some data journalism and even business data visualizations might weight aesthetics a little bit over accuracy at times, but certainly you would never call it data art. So, how would you define it? Data art. - I mean, this is going to be a personal definition of data art, I think, in terms of, I guess self-reflection in my own practice in data art. The biggest thing is that you use it in the methods, right? It's kind of part of that creation process. So, whether it's using a data set or it's using data science and some maybe part of that analytics from it, maybe you're drawing on even, you know, machine learning or some kind of a statistical method that you put into your artwork or some of the more, I think, common ways that we see data art, where it really is kind of combining visualization as a practice with that. Or maybe it's using elements of it, so like, elements of visualizing. I think that the best way to kind of really define it is maybe give an example. So, if that's okay, I'm going to screen share and do that now and kind of talk about what specifically, at least is kind of data art and what isn't and what makes it kind of art. So, let me just do that now. So, this is an example of a project that Owen and I worked on called community flow. And I'll give a little, like, quick background on it and then, get into this definition of data art. So, what ended up happening is a non-profit called Waterlution ended up having different projects in different communities on the great lakes in Canada. And so, what each artist was tasked with was having people come and they had a workshop where individuals created different little art works. And then, those artworks were taken by the artist and then, they created a bigger artwork piece. And then, Owen and I were tasked with creating I guess, you can call it data art to kind of combine all that in a digital way, so that folks could explore, you know, all of these initiatives at once. And so, what's interesting about it is, so here's a network in Thunder Bay, for instance. And if you click on it, you know, you will see the overall, you know, work that was created. So, for instance, this quilt here, which was what was made in Thunder Bay. But what's interesting about this is that you can click on individual pieces and have a look and see, you know, what was done individually as inspiration. And you can actually go through and, you know, reset it and there's a couple of other features and you can go to, you know, another place where this artist, you know, in Sound had people come, kids, everyone come and just do different drawings, and then, they ended up putting them in these light sculptures. Now, the reason why this is data art isn't because it's actually data visualization per se, the data here are images, right? But it is using part of data visualization which is a D3 circle packing algorithm. And that's something that we're quite familiar with, but maybe not seen in this kind of way, right? And I wanted to show this particular example, because I think conventionally, when we think of data art, we think that it does end up looking like a data visualization or has elements of it a lot. And this is also the case here, but sometimes, you know, it could be physical manifestation of data where, you know, somebody builds a bar graph and it puts it in a gallery and that kind of thing as well. So, those are kind of really broad ways of thinking about data art, but I wanted to share at least one example. - Okay, now, why data art, as opposed to data visualization more generally. Explain where these two disciplines maybe intersect and why it matters. - So, it's a really great question. I think that one really important thing about the two disciplines is to kind of really hone in on the audience. I mean, it's really about what the intent is, right? You have clients that require certain, you know, specific ways of reaching your audience with data base if you're working in that space and that has those requirements, but data art allows you to take it, like, I think to a whole other level, right? And it could be that, you know, I've seen client work turn into data art as that next step, because they know that, you know, maybe where they have taken it to kind of how that again, going back to that form and function, that functional aspect of communicating research, or results or whatever through the data viz, than maybe when it's time to put a public face on it or connect different, you know, audience segment, you can then do something that's a little bit more creative and building a narrative where you see more a presentation in that. And another reason which is a big one is, just having fresh eyes on it. So, if you have a creative look at, you know the very same thing that you've looked at from, you know, a different lens, it's almost like, becomes exploratory in nature, right? When you have a different lens that you put on to look at the same project. So, I think it could create fresh ideas and might even help inform the actual like data visualization aspect, right. I think a lot, in terms of kind of creativity and that direction, just because, I think having a message and the audience is such a huge part of our creation that I think that if you kind of put that into the data viz practice, then I think it'll help inform it a lot. - So, you moderated a fireside chat about data art for the data visualization society. And full disclosure, we're both on the data visualization society board. And I really liked your story about how you were working on a project that was using a government satellite photography dataset and you wanted to use it for an art project and you had to have multiple phone calls and go through five different people to get the right person, to get permission, to use that data for an art project. No one understood what you were trying to do, maybe even why you wanted to do it. So, the question is, when you think about this audience for this show, my audience. This is data analysts, data scientists, yes, data visualization practitioners, but it's also HR people, marketing people, you know, people who work with and manage data sets like you described. Many of them are not artists, maybe not coders necessarily, but if you had one minute to explain to them why data art matters, why this is an important and worthwhile effort, something they should be aware of and maybe help to enable, what would it be? - I think looking at that example that you mentioned that I talk about in that fireside chat is a really great kind of case building, you know, way of doing this, if you will. So, intersection was a project that involved government data as you mentioned, and the outcome of two artists, myself and Owen, taking a dataset that was really invested in by the province and, you know, it's really, really great ortho photography dataset, but it is locked up and we did have to get access to it. But in return and putting it in a new light, through creative coding and making it look interesting and submitting it to a film festival and then, submitting it as an artwork to various places, one of the ways that it was presented was on the subway screens in Toronto, which is, you know, a pretty big Canadian city. And so, you know, you had over a million people over the course of a month long exhibit for like, a big exhibit called the Scotiabank photography festival, see that. And so, and that just brings that awareness to that dataset. And if you have folks that are more interested and are just in general aware of that data they might be inspired to, you know, if there's a kid that is standing at the subway and is looking at this work and asks, you know, their parents about what that is and they read the context and explain that it's geography and then, they get interested in geography. Then, that is the case for data art, because it does come out of this, you know, maybe conventional way and into something that is completely different and perhaps inspiring or informative or communicative. So, I think that if you are in the position of having these conversations or even thinking about it then opening up data sets is a big one, right? Maybe having challenges, creative challenges, if you're in the government, you know. Thinking about like, ways that you can potentially inspire to get your data out, because the other big thing is that, by showing, especially for government, you know, we showed this data set to the public and then, that builds public trust and maybe potential investment in future initiatives, right. So, you don't even think about it initially, but that's the power that it could have, because it's now seen the light of day as opposed to being, you know, locked up behind a pay wall and some kind of like government request, web forum, right? - So, this of course, overlaps with your studio's work. Tell us more about the type of work that you do and how you collaborate with clients, bringing research and science together with visualization. - Definitely, yeah. So, I have a company with my partner, it's called Pixels and Plans. And I often describe it as kind of like mad men for research. You know, a lot of the time it's. So, we work mainly with NGOs and government clients. And I love working with our clients, because they have such different needs. And I love learning about all the new, different various initiatives that they need help with getting out into the various audiences. The main thing is, you know, I mean, sometimes it is creative work, like community flow where we do get hired more as creatives. And we do try to reserve at least, you know, we try to do one creative project every maybe year or two, because it is a big undertaking when all of our ideas are, you know, are everywhere and we want to do so much. And then, on the other side, you know, we definitely build, you know, traditional dashboards in D3 and of course, doing the more traditional client data viz work, but always having that kind of creative angle and definitely always thinking about, you know, educating our clients about their audiences and where their audiences are. And sometimes I find, I don't know, I think I'm not the only one when I say this, but oftentimes it's explaining what's possible before you can talk about what the work will be, because they're not always, you know, keep up with latest trends and then, thinking about longevity within the company too, like there's this, you know, are they all on Power BI or do they have, you know, a team that could support D3 development and that kind of thing? I think it's also important to have those conversations as well. - So, given that our topic today is data art. How does that really weave its way into your work? Do your clients ever request more work on like, the art end of the spectrum or is it always that balance between the two? - They're different clients, I think, you know, various organizations that have kind of creativity in their mandate, kind of like the community flow work and a couple of other works we have done. They, you know, we've done one work that hasn't been released yet, that's actually a physical work. It's a data visualization that's done with led lights and it's a tree and it's going into a science center, but it is based on actual data from a paper on forest fires. And so, you know, to that vein about our work, it's not always digital, which is really fascinating too, right? When you're in that creative space, you really think about the audience in this case. You've got people coming to a science center, well, back in, you know, when you could have visitors come to a science center. And so, they could turn dials and it's a very physical experience to, you know to manipulate data that way. And so, you know, so, those are clients that have those artistic needs. And then, the more traditional ones, I think where the creative really comes in is to just giving them, you know, ideas and we do see that often at meetings, you know. Sometimes it will come up, Oh, you want to reach this particular segment of folks, you know, that are the public. Maybe you should consider this creative kind of campaign. So, I think having that intersection really helps to have like a broader view of audience engagement. - And, you know, I've been curious to know what types of data lend themselves well to creating data art. Are there particular attributes that you look for? Do you need like larger data sets? Do they have to be in certain categories of information? What do you look for and what have you seen worked well or less well in that area? - It's a really great question. I mean, I think there's so many different types of data. There's of course the classic kind of personal data collection, right? Where you yourself turn the mirror on you as a creative and then you capture something about yourself and you reveal it in your work. I do love the journalistic approach as well. You know, I think sometimes it's just a matter of what data is available, right? Especially folks who are just starting out, they're going to go look for open data sets. So, you know, as a collective cultural narrative, if we think about data art, if the data sets that are being kind of you know, transformed into data art are representative of what data is available. Maybe that gets people thinking about, well what data is missing and what data is hidden and whatnot, maybe getting that out is of importance. But then, there comes the play of getting that data, right. If someone spends the time to look for it, to ask, to bother someone and to actually get something out, they also find that that really helps. In terms of kind of, you know, a lot of data or a little bit of data. I think that's all kind of a fair game. It really depends, I think on the platform, you know, and the intent on how you want to disseminate and reach the audience. So, it could be that you're planning something in an art gallery and you're going to create something that's physical or maybe, you know, you're going to pull, like we have one project called Depth to Water, that pulls a well water data set. And I think that's, you know, half a million data points. The reason why I like the data set is because, if you map all of the Wells in the province of Ontario, you get to see the geography of Ontario just by the holes that are in it, you know and you recognize all the features, you know, and that just is a Testament to how much access and drilling there has been to the aquifer. And so, without saying a thing you just see that and you say, wow, okay. You know, sometimes that just speaks for itself. And I think that that could be really powerful. And I think passion is a huge part of this. I think that, you know, it really will be the kind of driving factor in what data set will be selected at the end of the day. You know, is it interesting to you as a creator that wants to work with this for the next two months, three months, a year. And then, if that's the case likely, you know, it will be also of interest to others. And there are a lot of resources online for finding open data and looking at data, but then also, you know, I think that even journalists, for like doing personal data collection and whatnot. - Unfortunately we're running out of time, Julia. I want to thank you so much for being here today. It's always fun talking to you. And I really like this topic of data art. It's something that's sort of a little bit further a field for some of us, but I think it really helps inform people and help them make better decisions, even with their more businessy end of the spectrum, the type of work that they do. I know that you do all kinds of interesting things and I'd love to hear any closing thoughts you have about data art and some of the work that you're working on coming up. - Definitely. I think, you know, first of all, thank you for having me. I think it's such a huge thing to be able to talk about data art, right? Just the fact that we're having this discussion and there's more kind of knowledge being shared about the practice, I think is incredible. So, that is my plug for like, you know, thank you for shedding light on it, because I do think that the most important thing is the more people know about it and feel comfortable. And also I think realizing that it doesn't actually devalue your data viz or data science practice to be a creative practitioner within data. I have not found that in my practice, I think it hasn't enriched it. And I think it's important to maybe mention that, in case someone is afraid that somehow, you know, they won't be taken seriously, if they try a creative project. It's very clear, you know, if you position it that way. So, if you're kind of on the fence, don't be afraid to foray into that. And then, as far as, you know, our next thing Owen and I are working on a Slack app actually, for the data visualization societies where it started, because there's a kind of a need to visualize data viz channels or just channels in the Slack workspace in general. So, we created an app called Paulin, which at the time of this, I imagine is now released and you can try it, but basically it allows you to get a landscape of all of the channels in a workspace and then, right from where you're visualizing them you can actually join the ones that you have not joined yet. - Julia, thank you so much for being here today. Really appreciate it. And I look forward to seeing your work and what you come up with next. - Thank you so much Bill.

Contents