Become a Creator today!Start creating today - Share your story with the world!
Start for free
00:00:00
00:00:01
Episode #75: David Bauer
173 Plays8 years ago
Welcome back to the PolicyViz Podcast! As I’m sure you’re aware, data visualization tools are a regular source of discussion: Which tools to use? Which one is best? Which one allows me to do this or that or the other?...
The post Episode #75: David Bauer appeared first on PolicyViz.
Recommended
Transcript
Introduction and Sponsorship
00:00:00
Speaker
This episode of the PolicyViz podcast is brought to you by JMP, Statistical Discovery Software from SAS. JMP, spelled J-M-P, is an easy to use tool that connects powerful analytics with interactive graphics. The drag and drop interface of JMP enables quick exploration of data to identify patterns, interactions, and outliers.
00:00:19
Speaker
JUMP has a scripting language for reproducibility and interfacing with R. Click on this episode's sponsored link to receive a free info kit that includes an interview with DataVis experts Kaiser Fung and Alberto Cairo. In the interview, they discuss information gathering, analysis, and communicating results.
Welcome and Guest Introduction
00:00:49
Speaker
Welcome back to the PolicyViz podcast. I'm your host, John Schwabisch. On this week's episode, I'm excited to have with me David Bauer from NZZ. And David, where is NZZ exactly?
00:01:01
Speaker
We're in Zurich in Switzerland. Zurich in Switzerland. Great. Well David, thanks for coming on the show. I'm glad we can catch up here on a late on a Friday afternoon. How are you? Excellent. I'm just ready to go into the weekend and glad to be on the show now.
00:01:19
Speaker
And only a few more moments until Friday night beer, so that's good. Well, I'm glad we were able to connect.
Role and Team at NZZ
00:01:27
Speaker
You recently had a post, a new tool that you have all developed and have introduced to the newsroom called Q, which I want to talk about in some depth and you have a great medium post about it. But why don't we start by having you introduce yourself and talking a little bit about your role over there at NZZ.
00:01:45
Speaker
Sure. Well, I'm David. I'm the head of storytelling at Noitsu Ishitsaitung. The storytelling team is what other news organizations might know as a graphics team or maybe an interactive team. So it's pretty much the same that we do. We are a team of 12 people. That includes data journalists, graphic designers, developers, of course. And we support and work for and with the newsroom
00:02:15
Speaker
to make sure that the stories that we want to tell are told in the best possible way. And so that means we obviously create a lot of graphics both for the print edition and online. We make interactive graphics. We tell our own stories, mostly visual data-driven stories. And of course, we build templates and tools for editors and reporters to work with.
00:02:42
Speaker
Now, you mentioned stories a few times in that description. How do you identify yourself? Do you identify as storytellers, as journalists, as developers? How do you sort of, inside, how do you identify? Well, all of the above, basically. I mean, we call the storytelling team, so people like to call us storytellers in the newsroom. But I mean, we have different backgrounds. I have a background in journalism, maybe. So I see myself mainly as a journalist.
00:03:11
Speaker
others see themselves as mainly graphic designers or developers. And the goal basically is always the same, improve the experience that readers have when they come to read the news or in-depth reporting on our site. So it doesn't really matter how we call it again.
00:03:30
Speaker
Great. Well, let's talk
Development of Q Toolbox
00:03:32
Speaker
about Q. So you had a post on Medium about building this new tool. And so maybe you can sort of give a little bit of background about the development of the tool and what it accomplishes there in the newsroom for folks who may not have read the post yet. Yeah, sure. So
00:03:49
Speaker
I'm going to go back maybe one and a half years when I joined Nsetset and when basically this new team was formed. And the situation back then was that we had, I guess, a lot of traditional news organizations, a print graphics team that was well established and produced graphics for print. And then we have the interactive team that worked mainly on bigger projects on the digital side.
00:04:17
Speaker
But what was missing was basically means or tool for editors to create daily graphics for online that are easy to produce, that are not overly sophisticated. And the problem was simply that there was no one in either team that was able or had the time to produce so many graphics for daily use. And so the obvious
00:04:45
Speaker
The first thing I initiated when I started at NZZ was to say, okay, we're going to build a toolbox, not just a set of tools, but actually a toolbox that is kind of a one-stop shop for editors.
00:05:02
Speaker
And so pretty much after some projects that were ongoing had finished maybe two months into the creation of the team, we started building the toolbox and then maybe another two months later we were ready to launch the first version.
00:05:18
Speaker
Well, that's fast from getting that out. How do you tackle the issue of journalists who may not have familiarity with data or with curating charts? Do you
Training Journalists on Q
00:05:31
Speaker
sit down with them? Do you do trainings? How do you sort of move journalists from all the way, one might say, on the journalism spectrum with not a lot of data experience to move them a little bit closer to being able to create responsible, clear graphics?
00:05:47
Speaker
Yeah, so there's kind of different approaches that need to work together to achieve something and we're still struggling with all of them. One obviously is like changing mindsets that needs a lot of just talking to people and trying to convince them why it's a good thing that actually it's not just additional work, but something that improves their stories in the end.
00:06:14
Speaker
then there's the technological component. So we try to keep really as simple as possible. We resist all kinds of additional options that might add a bit more flexibility to some people, but would leave others out where it's no longer simple enough. So we try to keep it as simple and possible.
00:06:39
Speaker
and actually made it a rule that it should take you no longer than five minutes from logging in to having the finished map or chart or whatever in your article in the CMS. So that's what we try to follow. And then, of course, it's a lot of training, one-on-one training, having groups sit together where we show new tools, where we show all the tools again, where we have added new features and so on. So it's a combination of all of those,
00:07:09
Speaker
And then maybe as a little extra, we have created a Slack channel that alerts our team whenever somebody creates a new chart or a new map or whatever. So we can quickly go into check if that's okay, if everything's correct or also see where people might be struggling. Sometimes we see drafts that are not finished and we can get in touch with those people, ask them why they couldn't finish it and tell them or show them.
00:07:38
Speaker
So that has proven quite effective for us.
00:07:42
Speaker
Yeah. Prior to launching Q, what were the main tools that people were building? And what was the major pain point in the use of those tools? There weren't many. I mean, there have been some tools that were used in the interactive team that existed back then. But the only tool that I'm aware of where there was an effort to roll it out in
00:08:10
Speaker
in the newsroom in broader terms was data wrapper, which I guess most of the listeners know is a great tool for creating charts and also maps. But also it's something that we feel is a little, if not a lot too complex for just most of people who are just used to writing great stories. And
00:08:37
Speaker
So
Q vs Existing Tools
00:08:38
Speaker
yeah, the experience was that some people in the newsroom actually used data wrapper, maybe 10 or 15 people, but it was just not the kind of tool that would allow the whole or large parts of the newsroom to work with these kinds of tools. And also it's built for basically one use case, which is creating great charts of whatever variety.
00:09:03
Speaker
And what we had in mind is a whole set of tools that go beyond charts like pointer maps or quizzes or for elections or for referendums, which we have a lot in Switzerland. And there is not one tool that covers all of this, obviously. So if you rely on
00:09:25
Speaker
different tools for each of those cases, you would have five different tools that have five different URLs, five different logins, five different interfaces, and that just confuses people too much so that it will be productive in everyday use. Sure. Yeah, that makes sense. Can you talk a little bit about the technology behind the tool? I would suspect a lot of people listening are curious about
00:09:52
Speaker
actually how you went about and built with the programming language behind it. What's the techniques or the approach you took to actually building a custom tool like this? Yeah, sure. I mean, that's kind of difficult territory for me because I haven't built it myself. I have an excellent developer called Benny Bues who did basically all the work.
00:10:12
Speaker
It's browser-based. It's mainly built on Aurelia framework, so a JavaScript framework. And then this is kind of what the toolbox is made of. And for the individual tools, we then rely on different technologies that are out there. The charting tool right now that we have is built also in a library called Chartist.
00:10:35
Speaker
We're going to rebuild that now, probably switching to VEGA and for maps. We use, for example, the OpenStreetMap layers and bit of a map box. So we have the flexibility to use for each tool the technology that is
00:10:52
Speaker
Best fit for for what we want to do right right, but I recommend like reading the the medium article where any describes with a little more Expertise and in-depth was what the technical details of Q. Yeah. Yeah, absolutely and I'll link it on the show page
00:11:09
Speaker
I'm curious whether you think this is the future of data visualization tools for organizations and newsrooms. You mentioned you could have a data wrapper or a plotly or a Vega or all these sort of different tools, but you don't want to maybe have people need to go to five different places to do something.
00:11:31
Speaker
So do you think that the future is these custom sort of tools that newsrooms or organizations can build internally to meet all their different needs and branding and styling guidelines. It depends I would say I mean we're in quite a privileged position that I have.
00:11:47
Speaker
quite a large team compared to most newsrooms out there. I have developers that can take care of something like this. I have designers that can add design and so on. So I think for most newsrooms that are lucky if they have one developer, some don't even have developers, smaller newsrooms, they
00:12:10
Speaker
would always kind of need to use tools that are out there. And luckily there are quite a lot of tools that are good that are out there. And we kind of have the privilege to work on a custom built solution that is excellent for what we need. So I think for newsrooms of our size, we have around 300.
00:12:36
Speaker
people working in the newsroom and bigger newsrooms, I think actually this could be some sort of future. I wouldn't say this is the future, but certainly one way too flexible enough to include tools from others that are open source so that not every individual newsroom builds something from scratch.
00:12:57
Speaker
Yeah, I see potential in open sourcing tools and like these things coming a little closer together. But then in the end, each newsroom has their specific needs, their specific technology stack, their CMS and so on. So I don't think there can be that one super tool that works for most newsrooms.
00:13:19
Speaker
And it's interesting you mentioned the open source because I sort of see this,
Open-Sourcing Q
00:13:22
Speaker
there's obviously a tension I think here because on the one hand, news organizations are competitors. The New York Times and the Washington Post are competing for subscribers. But on the other hand, we've seen especially over the last, at least in the United States over the last year or so, especially a lot of news organizations I think working more closely together. ProPublica has this
00:13:49
Speaker
I think they are the leader, the hoster of an organization that sort of brings different newsrooms together to pair up and combine efforts. So I wonder whether you think that open source piece of these tools is going to grow so that, you know, the smaller newsroom in somewhere else in Switzerland or in, you know, Omaha, Nebraska, or whatever it is, has access to these technologies where they may not be able to build them
00:14:16
Speaker
on their own but they can take advantage of things being produced by larger newsrooms. How do you sort of see these two sides of the media landscape?
00:14:24
Speaker
I mean, there's no doubt that the competition aspect is something that you have or must have at the back of your mind when you open source something. But I really try not to make that the key driver of decisions. I mean, in the end, we're also standing on the shoulder of giants. We use a lot of open source
00:14:47
Speaker
software so we're benefiting as well so it feels like on one hand it's giving something back to a larger community not just media but like every everyone who builds something for the web
00:15:03
Speaker
But also we see quite a lot of potential in benefiting directly ourselves when we open source this when we see other people use it improve it maybe add new tools that we in return can use so this is kind of the more positive and not fear or competition driven view on on software and this is the kind of.
00:15:25
Speaker
the view we want to focus on. Yeah. Yeah. Can you maybe share with us a story or two of how you've worked with journalists in the newsroom there? Either they've had successful interactions with Q and the tool and your team, or they've hit a wall even there. Are there some experiences sort of jump out to your mind as you've sort of gone through the last year or so with the tool out there in the newsroom?
00:15:51
Speaker
Actually, I can't think of the one single story that would illustrate all of our struggles and successes we had with Q, but the typical story is it takes a little time of convincing, why should I do this? Why should that be part of my job? Then you get them to understand that this could actually be useful.
00:16:15
Speaker
Then they see it for the first time, check it out. I think it's interesting. Then they forget about it again. And then there's maybe the first story where they really see that additional map or a simple chart could help.
00:16:32
Speaker
And usually those people then create that thing on their own and we kind of just pitch in and help do the polishing in the end. And you know, it's quite a good experience to see somebody happy about how such a simple thing as a map has improved their story. That sounds like the fairy tale version of all of it. We obviously have
00:16:56
Speaker
other stories as well where people try to use it and then hit a wall because that one feature that they think is the make or break feature for them to use it does not exist or does not yet exist. That happens too. I always like interactions with correspondents that are all across the world basically.
00:17:21
Speaker
which
Journalists' Experience with Q
00:17:22
Speaker
mostly have never seen me neither have I seen them of course so we interact via Slack basically and we have quite a lot of correspondence now that really like the the opportunity that they've been given to do more with their stories than just write which is for the most
00:17:43
Speaker
part of their career has been what they've been doing. Yeah. Have you seen in those interactions, have you seen a journalist who has come to the tool, learned to create something, you know, maybe a simple map or simple graph, and then they've wanted to take it a step further, you know, maybe not build something themselves, but build something or have something built or work with your team that's a little more complicated
00:18:07
Speaker
little more involved, maybe more long form, what we might call an immersive narrative with multiple graphs and you know, sort of different things. Have you have you seen people try to or want to take that next step because now they see what you can do and they and they see sort of the, you know, what's the metaphor going up the hill or something to sort of, you know, a bigger grander thing. So what we what we always see is journalists kind of hacking their way through
00:18:34
Speaker
non-existent features in Q so they try to trick it which is always interesting for us to learn like some weaknesses of Q but also what kind of things you could do with Q that we didn't imagine like this so this is always interesting and of course often when when people use Q regularly and become aware of its limitations then there's usually conversation with us to see what other
00:19:04
Speaker
options there are to visualize certain data sets or so on. So usually those people who use Q regularly are also the ones that use a lot of data in their reporting and think in more visual or graphic terms about stories than other journalists that might not use Q as much.
00:19:26
Speaker
Yeah, that is really interesting. Let me just ask this last question. Where do you see Q going over the next year, six months, year, two
Future Plans for Q
00:19:36
Speaker
years? I know you've written a little bit about this in the Medium post, but where do you see development going and how it's going to impact the newsroom there over the next year or so? So we're currently refactoring Q to basically turn it into a piece of software rather than just a working prototype, which is now.
00:19:55
Speaker
Then we'll add some more tools and one of the key drivers of this refactoring is that we want to use Q for different platforms within the end-to-end media group. So there are other newsrooms and other websites that we can't deliver to.
00:20:16
Speaker
right now but we want to do that and also we want to get a closer integration of how charts that are now produced in Q4 Online find their way into the paper which is now mostly a manual process and we hope to improve that a little bit.
00:20:35
Speaker
Really interesting. Well, good luck with all that. I think working with journalists and teaching them how to use a new tool must be an interesting but rewarding process over there. So David, thanks for coming on the show. It's been really great chatting with you. Cool. Thanks for having me. And thanks to everyone for tuning in to this week's episode. Until next time, this has been the policy of this podcast. Thanks for listening.
00:21:08
Speaker
This episode of the PolicyViz podcast is brought to you by JMP, Statistical Discovery Software from SAS. JMP, spelled J-M-P, is an easy to use tool that connects powerful analytics with interactive graphics. The drag and drop interface of JMP enables quick exploration of data to identify patterns, interactions, and outliers.
00:21:27
Speaker
JUMP has a scripting language for reproducibility and interfacing with R. Click on this episode's sponsored link to receive a free info kit that includes an interview with DataVis experts Kaiser Fung and Alberto Cairo. In the interview, they discuss information gathering, analysis, and communicating results.