Become a Creator today!Start creating today - Share your story with the world!
Start for free
00:00:00
00:00:01
Exploring the Evolution of Data Visualization with Moritz Stefaner
697 Plays6 days ago
In this week’s episode of the PolicyViz Podcast, I sit down with data visualization expert Moritz Stefaner to reflect on his journey in the field. We discuss Stefaner's work on the Data Stories Podcast, his shift from bespoke data visualization projects to scalable design systems, and his collaborations with organizations like the World Health Organization. Moritz shares insights on the evolution of data visualization trends, the importance of clear communication, and the challenges of building sustainable design frameworks.
Keywords: Data visualization, Moritz Stefaner, PolicyViz Podcast, Data Stories Podcast, Design systems, Data journalism, WHO design language, Data communication, AI transcription, Data storytellingmathematics, Al, machine learning
Subscribe to the PolicyViz Podcast wherever you get your podcasts.
Become a patron of the PolicyViz Podcast for as little as a buck a month
Explore all of Mortiz’s amazing work on his website
Follow me on Instagram, LinkedIn, Substack, Twitter, Website, YouTube
Email: jon@policyviz.com
Recommended
Transcript
Welcome and Introduction
00:00:12
Speaker
Welcome back to the Policy Biz Podcast. I'm your host, John Schwabisch. On this week's episode of the show, I welcome my old friend Moritz Steffener to the show. Moritz is the former co-host of the podcast Data Stories, which is one of the first podcasts dedicated to data visualization.
00:00:30
Speaker
He is a designer. He is a developer. He is a teacher.
Navigating Organizational Hierarchies
00:00:33
Speaker
Moritz is one of the first sort of data visualization professionals that inspired me to get more involved in the field and to try things and to experiment with different forms and shapes, ah different visualizations.
00:00:47
Speaker
And so we talk about a lot of the different work that Moritz has been doing over the last several months. And in particular, we talk about how he works with and within organizations. ah We talk about how he navigates between different levels of management, as it were, the hierarchy within different positions within organizations.
00:01:04
Speaker
So if you are working in the field of data or data visualization, and you are working in an organization, as many of us are, and trying to navigate how to sort of balance all the what are sometimes competing opinions or goals, I think this conversation is going to be for you because it's really interesting to hear how Moritz thinks about approaching these types of projects. So I'm not going to talk anymore. Just going get you right over to the interview. Here's my conversation with Moritz on this week's episode of the PolicyViz podcast.
00:01:37
Speaker
Back together again. Moritz, we finally get to see each other.
Life Updates and New Projects
00:01:42
Speaker
It's been too long. How are you, friend? Good to see ah it's been a while indeed. yeah great to see you, John. I'm doing fine.
00:01:49
Speaker
Still here my little like office is here in the countryside in Germany. German countryside. much has changed. Yeah, I'm going nowhere. ah Just kids getting older, more gray hair. Just exactly. exactly Yeah. there you will You've been busy. I've seen a lot of your your, lot of your stuff. Excited to talk about it.
00:02:06
Speaker
Maybe we could talk about this podcast project you did. Cause it's kind of like, you can kind of look back and then we can look forward. So like, so for folks who don't know you and Enrico Bertini for a long time, hosted the data stories podcast.
00:02:19
Speaker
And recently you pulled all the transcripts together and built a little tool with it. So, um, you want to talk about like, well, first, I guess the question is like, what inspired after all this time, like what inspired you to like i finally do that?
00:02:32
Speaker
Yeah, yeah, yeah. So yeah, it's true.
Transcription and AI Tools
00:02:34
Speaker
So we started the podcast in 2012 and I was always like, wow, it's such an archive of, of like great conversations and also reflective really of how the field has changed and how our thinking has changed. So got really curious, like if we can make that archive more accessible and,
00:02:51
Speaker
And we did play with transcriptions, but back in the day, we had them done manually. So we had to commission them and had to pay for them and then still had to correct them because there's a lot of peculiar names in database and special concepts like Voronoi maps and so on. So you can't people to know how to spell that, right? And so we sort of gave up on it again, we were always like, oh, it's such a pity because...
00:03:16
Speaker
like Audio is such a monolithic thing. It's hard to look inside and yeah cross-read it, basically. It's difficult with audio. In text, you can search and you can reassemble and count words and whatnot. And then...
00:03:30
Speaker
and then um there was this big AI wave and it was like, okay, cool. AI is everywhere. What can we do with it? And one thing was, oh, you can transcribe like audio to text now pretty well.
00:03:41
Speaker
there's There's a bunch of tools you can use. um yeah And we used assembly.ai. So it's a... special service just for audio transcription. And so, yeah, I played with it. And then I was like, oh, this is amazing. And um it does chapters, it does speaker recognition, because you also want to know who said what, right? Yeah. And it does summarization and whatnot. and yeah So you popped them all in there.
00:04:09
Speaker
yeahp you had you had You had transcripts during the show though, right? Like when you were running the show for a yeah you had. Yeah, exactly. So first we had like manual transcripts and towards the end, we also had automatic transcripts using Whisper, but it wasn't that systematic. And this is the first time we went through the whole archive, okay transcribed everything. And now we have a full, full text archive. It's like 1.5 million words of data stories. You know, it's like a big heap. 1.2 million are Enrico, right?
00:04:38
Speaker
Yeah. Actually not. Actually not. I ran the numbers and I actually talk a bit more. yeah but The stereotypes are not are not true.
00:04:48
Speaker
so Good to know. Good to know. yeah All right. yeah so So with this now new database, you built like a front end visualization tool.
00:04:59
Speaker
Yeah, right. So um all the transcription, by the way, was done by Misker Knapik. He's like a database freelancer. So if you have something civil in mind or any like open data project, he's great. And I know he's always looking for commissions, so I can recommend working with him.
00:05:14
Speaker
And... um Yeah, he helped me with that process and ran all the transcriptions and invested a lot of time in fixing the last 5%, which are exactly that arcane, you know, and special right names and concepts that you want to get right because it's a database podcast. Yeah.
00:05:31
Speaker
all the domain specific stuff needs to be writing. Yeah. Lots of manual replacement transcriptions. And wow yeah, but now we have this big collection of um timestamp speaker annotator text, and we can, let's say you can search for
Evolution in Data Visualization
00:05:46
Speaker
a text.
00:05:46
Speaker
When do we talk about big data or something like this, and then jump to all the audio files immediately. Right. Or the audio snippet actually where was mentioned. And You can also do like a trend analysis, see like when ai or maps or crafting or whatever was mentioned most. So you can really do a bit of archaeology almost on the field. so when you when you think back on that you know basically decade...
00:06:16
Speaker
time span I mean, a lot of when you look at those trends, a lot of it comes out of what's happening in the field, but also who you and Enrico wanted. And there was other people I remember at one point, right?
00:06:27
Speaker
Like who you could get on the show, who you wanted to talk with. So like reflecting back, how do you sort of see the show and in retrospect? Yeah, yeah. I think that was a great opportunity to think a bit about all that. And so we started 2012. And think back then,
00:06:43
Speaker
and think back then um everything was much more open field and much more experimental. I think a lot of people like I were just working on, hey, what are novel ways we can represent data?
00:06:56
Speaker
Can we invent new chart forms or narrative formats, you know? And... Also, this connection between research and practice wasn't that established. So it was kind of unique that I, as a designer and and practitioner, would talk to a researcher and would realize, oh, say the same words, but we don't mean the same things. That's crazy. will say that that bridge has still not been built. Yeah, exactly. Yeah, but, ah and you know and And so I think 2012, 13, 14...
00:07:28
Speaker
When we started, this was very much a golden age of data journalism. and like And I think a lot was driven by New York Times, Guardian, Washington Post, you know who did really amazing new narrative formats, applications, really pushed push the field there a lot. And also, again, the the solo practitioners or the small studios who who built really good like yeah new ways to look at data. Right.
00:07:53
Speaker
Yeah. And then maybe 2015, 16, 17 is maybe more the age of back to analog and crafted data. i think Deer data came out 2015-ish or so. And yeah I think there was a big wave of...
00:08:08
Speaker
going analog away from the screen, um, crafting and, and haptic and texture and, yeah um, also more between the lines and, you know, playing with, with the rhetorics and all the associations and maybe a bit of an art push.
00:08:25
Speaker
Right. Also back at that time, I did the data cuisine project a lot, like where we right cook data representations or prepare data dishes, you know, really stretch the field or the envelope a bit. Yeah.
00:08:39
Speaker
And then maybe 17, 18, 19 is maybe then sort of the pendulum swings back again. People were more looking at functional aspects. Does this actually work?
00:08:51
Speaker
You know, can we scale it? yeah How, what's the ROI? You know, I think things became a bit more utilitarian maybe in these years.
00:09:03
Speaker
Mm-hmm.
From Bespoke to Systemic Projects
00:09:04
Speaker
um And also, DataVis itself got a bit more critical of itself. Like I recall there were a lot of like discussions around, is big data analysis a good idea in general, you know, or what are the limits of what you can do with crunching numbers and yeah also what are limits of fact-based communication, you know, maybe we're not getting through regardless how beautiful we make the charts. Maybe there's still ah big gap, you know, and you're never really able to persuade people who don't want to be persuaded and so on. And I think...
00:09:40
Speaker
That was a big, like maybe first soul searching phase in a sense. I think the same thing, not that I'm in this area, but I think the similar thing happened on the academic side too, where it was like, do we really know what we think we know? like I think of like the replication crisis also in psychology and write and like asking a lot of questions suddenly. Right, right. Like how do we read a pie chart? Like we don't, nobody knows. like Yeah.
00:10:08
Speaker
We all think they're terrible, but nobody knows. Yeah. yeah So I think we lost a bit of naivete, but also, i mean, that's the only way to really make progress. Well, yeah yeah. I mean, it's interesting over that time span too, it was like at the beginning, 2012, first era was like,
00:10:24
Speaker
Everything was interactive and you could click and hover and everything. And then by the end of that period, it was like, we don't need to make every bar chart. Like, let's just make a graph and just, yeah.
00:10:38
Speaker
Yeah. Yeah, also more devices, more um yeah multi-channel, you know, and then people realize, oh, a little square image is actually something that works everywhere, you know, or a movie, you you know, video clip, like 30 seconds. People just do it in the subway and so on.
00:10:54
Speaker
So, and i think that that all really... Yeah, came to full effect like around these years probably. and Right. Also, I sort of moved away from more, let's say, experimental and communication focused projects, more it's towards tools and like long-term projects and building systems and like maybe going from artisan more to architect, you know, some to some degree. Yeah.
Design Systems for WHO
00:11:20
Speaker
Also because I thought like, huh, I mean, it's kind of nice that you can always come up with new engaging visuals, but is that all that's there is to it, right? Or how do we create lasting impact or give people actually something they can work with and and yeah do something better?
00:11:37
Speaker
I still remember, maybe this was towards the end of the period, you did um you did a piece for Scientific American on the on the bees, right? And it was like um this honeycomb graphic. Yeah, yeah,
00:11:48
Speaker
Like this static piece. And i feel like I haven't seen that sort of stuff from you as often over the past four or five years, I guess. Yeah, absolutely. Yeah, yeah, yeah. So, yeah.
00:12:01
Speaker
So what was, i guess, what was your transition in your work? And what did you transition kind of from that sort of bespoke work in doing now?
00:12:13
Speaker
Yeah, it's a good question, but I think i what I sort of felt, maybe I played through to some degree is doing these single one-off communication pieces.
00:12:24
Speaker
And then then I started to think more about how do you create systems or how do you enable other people also to create good graphics, right? Or how you establish a certain culture or a certain language?
00:12:38
Speaker
um Basically, moving from the particular one thing that you craft really well one level up and think about, okay, what's a class of things, you know, that I've crafted well now.
00:12:49
Speaker
In a way, it's it's a trap also because once you move that level up, you also lose a lot of control and yeah a lot of the quality comes from, like, taking every single thing super serious and thinking about it. and and like So I sort of, yeah, I want to go back to the crafting part as well. um On the other hand, I think, well, it's...
00:13:09
Speaker
Yeah, if we really want to have effect at scale and and like lasting effect, maybe you also need to think about these these larger systems that things are embedded in or yeah where where things happen.
00:13:20
Speaker
no Yeah, i mean, if you think about large organizations and they're just trying to get... yeah right they're just trying to build their own internal, even just their internal metrics, their own internal KPIs. Like they just need to get stuff done and out the door.
00:13:32
Speaker
So you've done a bunch of like big projects. You did a thing with the World Health Organization on like a design language. And you did a, no most recently, like this climate conflict index thing.
00:13:47
Speaker
um You want to talk about those projects and like what it was like sort of building like at scale rather than like, a single bespoke. Yeah. yeah Yeah. I think the transition piece there was first that COVID dashboard. So I built together with studio NAND, the official German like COVID vaccination dashboard.
00:14:06
Speaker
And ah was really exciting, like to be able to do that. And It was like a one-off piece because we had to build it really quick. you know because Suddenly it was like, oh, we're vaccinating now. We're collecting numbers. We need to show them and right we need to go online in like three weeks.
00:14:21
Speaker
um But then we also thought about, okay, how can we um address everybody in Germany, make sure everybody in Germany understands it you know that it's available in all languages, that's fully accessible, it's easily shareable.
00:14:35
Speaker
Right. ah It's presented in a way that's like redundant, like maybe some people prefer text, other people prefer graphics, other people prefer a metaphor, like we had a vaccination clock that would show you how many people are vaccinated per second.
00:14:49
Speaker
Other people want to have the full detailed statistics, otherwise they don't believe you. You know, it was also this time of lots of skepticism and discussions. Right. And so we were thinking really hard and discussing a lot about, okay, how do we make this as simple and accessible as possible?
00:15:04
Speaker
right And so really super solid, straightforward stuff. And through a few um detours, this led actually to that WHO data design language project, because at the same time, the World Health Organization was rethinking in a smaller team, okay, how do we actually communicate and store and share data, right? Because they also realized, oh man, it's everybody wants real-time information now on public health, you know, and that just wasn't the case before, it right? And so they had to rethink their processes quite a bit. And um Core is an agency in the UK and they brought me on board and I was able to also then put together a team
00:15:44
Speaker
um freelancers and an agency, Nine Elements in Germany, and a lot of external consultants like John, like you. and And so you helped a lot on the design system part, for instance. And so we were able to really get input from from great experts and really rethink from the ground up, okay, how should data communication look like, you know, in 2020 something for an organization like this. And right so we put a lot of emphasis on really like simple building blocks,
00:16:16
Speaker
um ah transparency, wide accessibility, um like really like a Lego system, you know, super solid and everything snaps into place. And that's really robust and and and can withstand in a lot of like, you know, context and situations. and Does that, ah for for someone who spent their career, like building these super bespoke different kinds of projects, right?
00:16:43
Speaker
Does that sort of like Lego, design process, does that like hurt your heart a little bit? Or do you just feel like most people like just need to... like they're They're not building honeycomb plots, right? like they're just they're they're They're doing research or doing data analysis and they need to get this information out to them public and and yeah and that's okay.
00:17:06
Speaker
but But does it like hurt you a little bit on the inside? No, it didn't hurt because we said, this is the way to go, you know, and and this is was, you know, we we knew this is the way to go and this is how we want to do it.
00:17:17
Speaker
and But you're right in the sense that the actual practice of building a system like this is very different than building a cool like data graphic that you just make right? yeah Because you really have to think in systems. And so we had all this like design tokens that would define all the colors and font sizes and...
00:17:35
Speaker
light mode and dark mode and high conscious mode and so on. And you always have to think in abstractions and within that framework of what you have already. And so it's like a big cathedral, right? It's, you can't just do something like, Oh, let's make this orange, but you have to think about, okay, what's the semantic meaning of this thing and which color, like which abstract color type is it? Right. And so yeah will that work?
00:18:01
Speaker
And yeah, I think it's cognitively very demanding and I have high respect of people who do that well because like you have to concentrate really hard. Yeah. And the other thing is, um yeah, you have to be careful that you build a system that's so simple that's still expressive and not limiting but always creates new opportunities. And it's never a rule book that suffocates you, you know? And I think that's the danger with all systems is like they grow, grow, grow until they become like heavy and
00:18:34
Speaker
obscure and alone like get an administration basically. And so you always have to think, it's like gardening. You always have to think about, okay, which part can we prune again? Or ah what which what could we now combine or name better? So it's a constant gardening activity.
00:18:49
Speaker
and But I think the most humbling part was that we realized, so we've been all had an expert team, great people, all really um seasoned database designers,
00:19:01
Speaker
But we had endless discussions how a line chart actually works, you know? Like, I don't know. For instance, you have years, right? Like you have five years in the line chart. So yeah you make five sections on the horizontal axis, right?
00:19:15
Speaker
But now does the line start in the middle of that first section or at the beginning, right? At the end. But either way you do it, you run into trouble like, this can't be true. Like, but let's see how how proper software does it.
00:19:29
Speaker
And then you're like, oh, they all do it slightly different. It's, you know. And you know they all give you options to change it. Yeah. Exactly. And then you're like, wow, I had no idea. like it really, it's humbling in a sense that you really think about the the fundamentals of, you know, of how charts work. And so, yeah.
00:19:46
Speaker
I love this metaphor of the of the gardening. How did you leave them? kind of empowered to, or with a plan, I guess, to do that gardening over time.
00:19:57
Speaker
Yeah. So I think I was just watching out for that, like in a sense that I took that role. sort of Yeah. Pruning, pruning it also. Yeah. And just building,
00:20:08
Speaker
the simplest stuff that's absolutely needed. like so so there's this yakni principle, you ain't going to need it. like you know like If you come up with something, probably you're not going to need it. So you know really, yeah unless something's absolutely needed needed, you just don't build it. And so we just had six chart types in the beginning or something.
00:20:26
Speaker
And it was vehemently defending those because they were not... but we need stacked bars, or we need a scatterplot, we need a stack. You can also do two line charts next to each other and then coordinate them. Or, you know, it's like always looking for easy ways how things can be solved within that simple framework.
00:20:43
Speaker
Yes. And the last part was really building everything out with code so that... We had a system where if we were to change a tone of blue in Figma, we were able to save and export and then rebuild, and it would change everywhere.
00:21:00
Speaker
so So there is no cost to adjusting things afterwards. so we we yeah And so we built a smart system there um that that would enable that and that would allow us to really...
00:21:14
Speaker
change things as we see the project needs, you know? and Right. I think that worked well. What was the adoption part like? So I can imagine you have this team of people, you've got the core folks at WHO, you've got your team, and then you sort of like release this thing out to...
00:21:31
Speaker
I mean, i don't know how many people work at WHO, tens of thousands, hundreds of thousands. I don't know. Yeah, lots of teams and across the world and so on. So it's really hard. yeah Did you get to talk with them about like that piece of it?
00:21:43
Speaker
Yeah, so I'm sort of out of the project by now, so I can't really say exactly how that goes. But the strategy around it was to build first a few individual products that are important also to the organization and build them in this small team and then sort of make bigger circles and... and either build custom solutions for other teams or have them sort of adopted fully.
00:22:05
Speaker
um Yeah, but this process takes a while because of course there's like content management system and wider considerations and
Climate Change and Political Stability
00:22:15
Speaker
and so on. And so, We basically built this core of these are the principles. These are the building blocks. These are really good examples of how it should work.
00:22:23
Speaker
And yeah there's also the software part of it that's really clean and and nicely done. And yeah now they have to sort of run with it. Yeah. Yeah, it is. It is interesting. i mean, I can't tell you how many, how many groups I've worked with where it's like, you know, I'll be teaching something and I'll say, so do you all have like a style guide or a color palette? And you'll see like one person on this side will be nodding. And the other person, these people on this other room, like, no, we don't have anything like that. Cause you could build these systems, but if people don't know about them, yeah then, you know, they don't really accomplish the goal. Um,
00:22:57
Speaker
so So that's a really cool project. But then more recently, I mean, I think like really recently, right? the You have this climate conflict index project that just came out. You want to talk a little bit about that one?
00:23:08
Speaker
That's kind of like more core, how I think about like Moritz's like work out in the world. Probably in a sense that it has also that system parted, but it has a real content, like in a sense that, oh, we have a new data set that is created extra, like specifically specifically for that project. And yeah now it's our role to communicate that and demonstrate how used. So yeah i think I enjoy that, that it's,
00:23:32
Speaker
yeah, there's a concrete content that we can actually work with. And so, yeah, it came out of, so the German foreign office, um they want to understand better how climate change will,
00:23:46
Speaker
overlap with, or like hazards from climate change um will overlap with violent conflicts and other like pressures on on a region, right? So, you know, today like different regions of the world get these types of stress from all kinds of different factors. And um yeah, we believe like unforeseen climate disasters can, for instance, like cause migration, which causes then, you know, conflicts other elsewhere and and they have that yeah on the on the radar. basically know what to do about it.
00:24:17
Speaker
right so um we looked into existing data sources and built like a first quick prototype based on that. But then we realized to make it actually actionable in-house and also for others, it's an open source project.
00:24:30
Speaker
um There should be like a harmonized set of data layers that you can just put on top of each other and and combine them yourself. And that are all on the same spatial grid, the same time grid.
00:24:42
Speaker
So you can see where will, or where have been floods, where have been heat waves, where have been armed fights um and also how, how well is the region equipped on ah on a political administrative level to respond to to problems, right?
00:24:59
Speaker
yeah And um yeah, so they um created a whole research project. So we have the PIC, which is the climate or an institute that investigates climate impacts.
00:25:12
Speaker
And we have the University of the Armed Forces, which are obviously the security experts, and they have researchers now working on that. And we work with them, small team of database experts, to put the data to use in-house, but also explain it to the outside and make it accessible. And so, yeah yeah, it has a lot of different facets. So we need to sort of understand how how the data side works, how the application side works, do a lot of user-centered design also.
00:25:38
Speaker
talk So how does that, how they use for you, it sounds like there are a lot of, um elements in that that I would guess you probably don't have like a lot of experience working with like military data, military simulations.
00:25:52
Speaker
Okay. So correct me if I'm wrong, when you're working on a project, um you're doing at least some basic kind of gut checks on the data, right? Like this yeah makes sense. This feels right to me.
00:26:04
Speaker
When you're doing a project like this, where there's so many different data sources from so many different places where, i mean, no individual could be expected to be able to have a gut check on military whatever. um ah How do you think through the data? How do you approach that? Do you do you just trust the team? like what is What happens for you?
00:26:25
Speaker
yeah. yeah So we have a lot this like connecting role between we have the domain experts and the analysts and the policymakers in the office um and desk researchers and whatnot.
00:26:37
Speaker
And they know what they need, right? And they also know what they usually use and they know what types of formats work in-house. um Yeah. And so, for instance, they know, okay often it needs to be printable still or like easy to circulate, you know, nothing complicated.
00:26:56
Speaker
know High contrast, static, well annotated. um But also what we learned is in-house, there's almost like a social media culture in the sense that graphics work really well, maps work really well, and especially annotated graphics and maps, like one slide basically that has a clear message.
00:27:14
Speaker
has a few detailed bullets and then annotate annotated spots on on a map or something that's perfect. right and so And I think that's a lot what people now realize, oh, that works well in the web, also works in these work contexts, which is yeah interesting.
00:27:27
Speaker
And um so we learn about that. And then we have the researchers and they say, oh, this is actually something we can provide data-wise, or we have these different alternatives of how to calculate something.
00:27:40
Speaker
um And of course, there's a mismatch between what people say they want and what they actually want. And then there's also a mismatch between what they actually want and what can be done. Right. And so yeah yeah you constantly have to sort of think on your feet and like translate and negotiate a bit between these different layers. Yeah.
00:28:00
Speaker
Yeah. Forces in the project. And I think that's, that's a lot of our role, like, like this, um, you know a sort of node between these two kind of separate yeah and of course we cannot tell the researchers exactly what to do because they're researchers and they need to have their freedom also and we respect that at the same time they also wanted it to be effective and to be used and so um yeah um we we sort of try to you know Just translate between these different Translate and bridge all the all the different skill sets yeah and the and the content areas.
00:28:35
Speaker
yeah yeah And actually the graphics or like any interactive prototypes we do, they play a huge role there in um just enabling that discussion. Because you often discuss in theory, oh it would be great to have a map that shows all the hotspots and blah, blah, blah.
00:28:52
Speaker
And then you see it and then the real discussion starts. Oh, right actually, that's quite seasonal. Should we do it yearly or should we do it seasonally? you know Or like is it actually better to show all the floods or shouldn't we show them only where actually people live? you know There's huge floods in Siberia, but it's not a big damage. Yeah, maybe we should. But then in other contexts, you want to see the raw data, right? ah Or right how much do you normalize the data? Or do you show raw values that have a unit, you know which are easier to understand? But then...
00:29:24
Speaker
maybe sometimes it's better to do something more yeah elaborate, but then it's just the score between zero and one. and And so, you know, and all this just happens when you build quickly, build data representations and have people play with it.
00:29:36
Speaker
Yeah. And do you feel like those deliberations always end up in a better final product or or have or have you found in some cases that like you have these deliberations and you end up on this curve where you have to like pull people, pull people back because in like my experience of working, especially with like researchers and scientists, right. It is, um,
00:30:02
Speaker
um I guess it's even when you're saying, you know, we're building this for the, for a broader public, it's easy for people to lean back into what they're comfortable with. Like, let's use all of our jargon. Let's, this is what I'm interested in, but it's not going to be great for the general public, but yeah. Like, do you find that, that bend in people's desires? Then you have to kind of reign them back in.
00:30:25
Speaker
um Yeah, it can go all kinds of ways, of course. so yeah Sometimes you have this faster horses thing, so I want exactly what I'm doing now, but but somebody else should do it for me or it's already done. yeah yeah But then people don't even question if what they do is fundamentally the right thing.
00:30:42
Speaker
yeah And then it's hard to convince them otherwise if you can't demonstrate it. a complete solution you know that that replaces that and so of course there's some inertia there um and or as you say you can just like in a group you can dig the deepest rabbit holes of course if you're not careful and but i think to some degree yeah you just need good like project management and i think design leadership in a way like that so my theory is if a project is I might be biased because I'm a designer, but if you have a strong like design leadership in a project, yeah it it doesn't really happen because then there is a clear vision of yeah which phase you're in and which type of input you're looking for.
00:31:25
Speaker
And if you're expanding possibilities currently, right? Or if you're narrowing down and that's something you would just... um And now it's also like, oh, first we you know really go blue sky, you know open field, and and what could we do? And then you say, right now we need to get realistic and pick one of those solutions, actually build it out and test it and yeah not keep changing things. And I think, yeah, if this is well prepared and and clearly communicated, then it can work.
00:31:54
Speaker
But it it takes a bit of time and um it's not always the same in every group. Yeah. yeah Yeah, I think that this is where design practice can can really bring a lot of value. And I can e immediately spot a project where there is no design leadership because it can have exactly these problems. It's just yeah endless like discussions or endless bug fixing, but no actual like new improvements or all these things. So yeah.
00:32:21
Speaker
Yeah. one no when I want to represent it. Yeah, no, so so it's funny. So if if you were... um If you are sitting right now in front of a room of, you know, students about to come out into the world with their, know, they want to be data visualization professionals.
Beyond Technical Skills
00:32:39
Speaker
specialists, i don't know, data visualizers, whatever the whatever the term is, right? yeah yeah What would you recommend they a skill scal or skills they develop aside from you know the technical part of you know being, let's just say, a good designer or a good developer? And what I'm thinking here is like,
00:32:59
Speaker
the person to person skills that you just mentioned, like, you know, would you say like a project management skill, like interpersonal skills, like, you know, writing skills, like what would be the skills that you would say? Like, this is what you really like to do the job.
00:33:14
Speaker
You've already got the design thing handled, but like to do the actual job, like what would those other skills be that you would use? You think people should have? That's a great question. So I think the communication part is huge and and people tend to underestimate that. yeah that And yeah it's never as easy as you just do the work and then the work speaks for itself, you know? and Right.
00:33:37
Speaker
And if you you dream of that, it's like, oh, couldn't I just do the the work? And then people say, oh, it's great. Yeah. and But if it were like that, you would also miss so much interesting, you know, discussions and learnings and and, you know, like advances you can get from like and first being irritated and then understanding somebody's position or how people can totally see the same thing totally differently. um Yeah.
00:34:02
Speaker
Yeah, I think that's a big part. So I think generally it's good if. If you have some environment where you can quickly produce a lot of different visual data representations, so it can be sketching, it can be code. um I'm having so much fun right now with observable plot and observable framework. yeah I think um like Mike and Phil, Philippe Riviere, Mike Bostock, they're doing amazing work on the product.
00:34:30
Speaker
And for us, it has been really a game changer in terms of sketching with data real quick. So big shout out here. um But it can be anything. you know so It just needs to be one environment where you can quickly load data that's reasonably like large and get a lot of different perspectives on it. Or if you have an idea, quickly be able to demonstrate and Because working with these concrete artifacts...
00:34:56
Speaker
That's the game changer. And that's the real quality we can bring, because this is often what people are not data visualizers who are more technical or more conceptual, they don't have these artifacts, right? And so conversation stops at some point and you're like, yeah, but then we'll have to see. Or we just decide for one of the options, but we don't actually know how they will play out. And I think we have that gift, basically, that we can make these things all concrete and...
00:35:21
Speaker
and spur new conversations. And I think, so being able to create these artifacts is the one thing, but then also framing them, presenting them in the right way, thinking about how to present ideas, how to elicit the right type of feedback at the right point in time.
00:35:40
Speaker
I know John, you've done a lot on on presentations as well, right? So it's a lot also this storytelling skill, you know, there about what's the story of our project and where are we in that story, right? And sort of, yeah um but also not prescribing that, but also shaping that together and and being open there. um I think that's yeahs sort of key.
00:36:01
Speaker
Yeah. I want to finish up. You mentioned a blue sky earlier, and I wanted to ask you what I asked Enrico a few weeks ago.
Community Engagement in Data Visualization
00:36:10
Speaker
where where's your uh data viz social media world at these days yeah yeah so i used to be super active on twitter but yeah you know it's yeah that has changed so that has changed and so i was like oh yeah i tried mastodon for a while but i i'm now pretty optimistic with blue sky it's kind of fun uh at the moment at least yeah and i also enjoy instagram so um
00:36:36
Speaker
um it's more visual and generally a pretty positive vibe you know yeah you wouldn't discuss big things on on instagram but it's good right entertainment yeah yeah you see what everybody's up to yeah yeah but i i feel yeah the online community is for me a bit more obscure now i don't even know where everybody is so yeah either i either i'm not part of the the cool uh yeah we're the old we're the old guy where is everybody where is everybody so Yeah, it it has shrunk a bit, think, my my world in that respect.
00:37:09
Speaker
But on the other hand, then you maybe concentrate more on your work. Right. So I was going to ask, like, do you... Yeah, I was going ask, like, do you... I go back and forth on this because I miss some of the discussions and the closeness and and that, but also, like, I don't... i've I've deleted all the apps off my phone. Like, I don't... read And so there's freedom to, like...
00:37:31
Speaker
read and like do other things. Right. Like, um, so, so how do you feel it like, I just wonder how you feel about that. Like the community is more dispersed now, but you know, I mean, it's clearly okay. Like the world is what the world is, but, um,
00:37:48
Speaker
yeah Yeah, I think that there is yeah more happening offline, more happening on other channels. It's kind of nice. Also good formats. I enjoy a lot of newsletters. you know like yeah It's a good format and it's also less...
00:38:04
Speaker
fragmenteds You know, it's more like a traditional, almost like a magazine or newspaper that thing, you know, it's a big thing with little parts and not a lot of chunks flying around. So I think that actually helps with the concentration, the depth. yeah um But I sort of miss this sort of marketplace thing, bazaar thing, you know, where you just go in and see what's happening. It's like, woo! oh, these two are fighting and what what's happening to go over there? you know And like du to have this feeling of there's a community you can always dip into. And I think data with society is it to some degree, the Slack channel there, but it's also just a very specific wedge of the whole field that that's yeah present there.
00:38:46
Speaker
And also we need more conferences, I think. yeah So I think a lot of the the really, really good data with conferences went away because they've been around for a long time. There was COVID and so on. I think all totally understandable, but I think we need a few good replacements now.
00:39:04
Speaker
Yeah. Yeah. I've been missing the missing the conference circuit. It feels like it's down to kind of like outlier from database society and some of the academic conferences, which are a very different vibe.
00:39:16
Speaker
Yeah. Yeah. um but Yeah, there's there's a few here and there, but I'm i'm also sort of missing out that thing. And it's going to be interesting to see how it how it develops. Like if the database community, you know, will find back together into one thing, or maybe if we really split up in different subgroups and right then maybe again, more adjacent to the, because we're sort of squeezed into fields,
00:39:42
Speaker
into the middle of other fields. Yeah, in the middle of a lot fields. Everybody goes to their other field that they are maybe most at home with. yeah i think that would be kind of sad. I think data viz is unique. and Also, I gave a few talks about what I think is a unique data viz mindset. Maybe we could link one of them.
00:39:59
Speaker
Because they I think there is something that unifies us all in terms of how we look at data and what extra value we can bring. and I think we should foster that and like, yeah.
00:40:13
Speaker
Well, mean, just like people have one foot in the data viz world and another foot in there. economics or design or anthropology or astronomy world, like yeah you know you can still kind of do both of those.
00:40:26
Speaker
Yeah, yeah. But I think we should have, or make sure we have like a good, vibrant database community that meets up regularly and we're allowed to- I mean, you just need more at Sconf, just like out on the farm there. TEDx my house. Yeah.
00:40:43
Speaker
I love it. Backpacks for everybody. Maybe camping. If camping is an option, then I could offer it. I could offer that.
00:40:53
Speaker
um Well, always good to see you. It's been a long time. Thanks so much for coming on the show. I'll share this these project with folks and hopeful hope they'll check them out. And they should definitely check out the Data Stories podcast transcript because was a lot of fun to like dig in.
00:41:08
Speaker
Search for your favorite words and maybe rediscover some old episodes. That's right. That's right. Thanks, Boris. appreciate it. Thank you, John. Thanks for tuning in, everyone. Hope you enjoyed that. I hope you will check out of Mort's project. I put links in the show notes.
00:41:22
Speaker
If you have any comments or questions about the show, if you have guests you want to see or listen to, please do reach out. You can get me on most of the social networks. You can reach out to me on Substack, where I host my newsletter, or you can connect with me directly through the PolicyViz website.
00:41:38
Speaker
So until next time, this has been the PolicyViz podcast. Thanks so much for listening.