Episode #110: Ben Welsh

By • • PolicyViz Podcast in Online

Recording

Listen on the original podcast site

Show the timestamped transcript
  1. Welcome back to the PolicyViz podcast. I'm your host, John Schwabisch. Now we're into 2018. It's
  2. probably a good time to look ahead, but also to look back a little bit on what happened last year
  3. with data visualization, what happened in the last several years, and of course, what's going
  4. to happen in 2018. And to help me do that, I'm very excited to have Ben Welsh, who is the editor
  5. of the data desk at the LA Times to chat about what's going on over there on the west coast.
  6. Ben, how are you? Happy New Year. Thanks for coming on the show. Happy New Year, John. Thanks
  7. for having me. I'm excited to chat about all things LA Times because some of the great stuff
  8. that you guys do that tends to be west coast doesn't always make it over here to the east
  9. coast. You know, it's sort of like the little story about the wildfires. And then I get to go
  10. to the LA Times website, and really see some incredible stuff about wildfires or agriculture
  11. work that doesn't sort of make the front page over here. So why don't we start with having you talk
  12. a little bit about yourself and how long you've been at the times and the work that you all do
  13. there, and how that data desk works in relation to the rest of the newsroom? Yeah, sure, no problem.
  14. So I've been doing the kind of nerdy thing for, you know, gosh, I guess I have to count now. I
  15. guess probably about 15 years. And about the last 10 of those have been at the LA Times. And when I
  16. started in 2007, I was working for la times.com, which was in an entirely different part of the
  17. building and wasn't even in the editorial quote unquote section of the newspaper. And in the 10
  18. years since that's changed. And we've kind of gradually integrated and we have one newsroom now.
  19. And within that one newsroom at actually kind of the center of it physically is where our team,
  20. the data desk sits. And we're kind of a small squad of reporters and computer programmers who
  21. use our technical skills to get stuff done, to find and tell stories to solve problems to do
  22. research to make things happen on the internet. And there's really a lot of different shapes and
  23. sizes that takes I don't think any of it would really surprise you. You know, one observation I
  24. would kind of make about our industry is that the types of things people are doing are really
  25. quite similar from place to place. But the way in which they're done sort of within the organization
  26. or who does it can vary quite a bit, you know, kind of the mix of different tasks that are done by
  27. the mix of different teams in newsrooms can vary a bit. And even though our team is called the data
  28. desk, we do a little more than that, just because of the way things have evolved. And so in addition
  29. to the kind of, you know, old school computer assisted reporting type stuff where you're
  30. generating nutgrass from big hairy databases. We also do a lot of systems administration for
  31. newsroom tools that help us publish custom stuff, and fancy stories and things on the web, we do
  32. the design for a lot of the long form and fancy packages that come through each year.
  33. And we do some data visualization and other work as well. And it really can kind of vary
  34. depending on the project and the people involved and what needs to get done. So I tend to describe
  35. our team as kind of the nerd Swiss army knife, where we're trying to use our computer skills to
  36. solve problems and make stuff happen. Right. I mean, I assume there's a big culture shift having,
  37. you know, move the two newsrooms together. What was the transition like from going from a newsroom
  38. that was your I would guess a traditional shoe leather newsroom and a separate sort of dot com
  39. newsroom and then pushing those together? Was there a big culture change, a big effort to try
  40. to merge those two different I would guess cultures and skill sets together? Yeah, I mean,
  41. so when I started the la times.com was in a different part of the building with a different
  42. reporting structure, I was hired there and didn't even work, you know, I guess for the editorial
  43. department. And now you know, I'm I guess an editor within it and we're all employees under
  44. that umbrella. So like organizationally, it was all kind of moved under, which was done
  45. in kind of a big reorganization that went along with some pretty large layoffs. A number of years
  46. ago, and that was kind of, you know, a big moment for the LA times when all that happened. But,
  47. you know, just one change like that doesn't sort everything out. These types of changes are really
  48. long and complicated and evolutionary and uneven depending on, you know, where you are or what
  49. you're doing. And it's kind of really a struggle that everyone in the industry is facing. And we
  50. are to just try to, you know, get to where we need to be. I don't think there was ever one moment
  51. where suddenly, you know, the switch went on and things were different. But there's been milestones
  52. throughout, you know, that was a big one when the merger happened, like another big one that's very
  53. procedural that I'm proud our team was involved in was having all of our copy being written in the
  54. web CMS first and then exported to a print publishing platform, rather than vice versa,
  55. which might not sound too oppressive for people who work on websites and haven't worked in a print
  56. newsroom. But the reality is is actually the reverse most places where you're writing in a print CMS
  57. and then at some boring bureaucratic process has to play out before it is, you know, exported to
  58. the website. And just even reversing that last year to me was in its own way kind of a major
  59. milestone in the LA Times history. Yeah, were there certain stories over that time span that
  60. maybe drew people in even more engaged them to say, Oh, this is something that I can use in my
  61. reporting to, you know, connect with or communicate better to the audience? Like, were there milestone
  62. projects that you remember over that time as well? Well, there's definitely been a lot of big projects
  63. that you know, I'm proud of on our team. I don't know if we can necessarily take credit to having,
  64. you know, done the big story that that changed the whole thing. I don't know. I don't know if you
  65. could say there's one story that really flip things here. But I would say, you know, there's
  66. projects we've been involved on that were that were entirely kind of untraditional, that didn't
  67. involve creating, like a traditional print story at all. Things like we had a project, you know,
  68. eight or nine years ago, when I first started called California's war dead, where we compiled
  69. all the obituaries that had been written about Californians who died or Iraq and Afghanistan,
  70. made it into a database, then made that database into a website that had a page for each person,
  71. and then also powered a lot of enterprise pieces. And then on a going forward basis, after that,
  72. we were publishing to that application before we were even researching and writing the obituaries
  73. and sort of incrementally updating them as the work continued. And I think that that because the
  74. obituary writing process was a project that so many people in the newsroom were involved in,
  75. in the writing and research for, I think it was a way for us to expose a lot of people to doing
  76. things kind of a different way, and kind of conceiving of what you're creating in a more
  77. structured but also, you know, less conventional storytelling format. And so I guess in my time
  78. here, I think that's been a big one. Yeah. Now that you have the newsrooms together, and we've
  79. seen a big change in technology and the way people are receiving their news, can you talk a little
  80. bit about the balance between desktop and mobile and how you all think about that as you're creating
  81. either the small projects or even these big narrative features? Sure. I mean, everything
  82. has got to work on the phone. If it doesn't work on the phone, it doesn't work, you know. And
  83. increasingly, you know, a majority of readers are often seeing stories on the small screen. And so
  84. I think, you know, from the design standpoint, if not start with it, it has to be paramount or equal
  85. to the desktop. And that's just, you know, now fundamental to kind of the custom web development
  86. our team does. And I think probably every team like ours does. I think, you know, one symptom of
  87. that or one outcome of that is, you know, this won't surprise you, I'm sure is, you know, kind
  88. of a move to more static graphics over interactive graphics, in many cases, because they're oftentimes
  89. just better suited for mobile and sort of the recalculation of the cost benefit analysis of,
  90. you know, how much work you put into something. The increasing weight on mobile as an input,
  91. you know, makes the investment on the interactivity kind of often seem less worth it, you know. And
  92. so I think you're definitely would see a lot more static graphics at the LA Times and pretty much
  93. everywhere across the industry. I think that's probably only going to increase, you know, I think
  94. that trend is going to continue. And I think, you know, you know, really boring, nerdy stuff,
  95. like you got to make sure the page isn't too heavy, you know, and, and you have to think about also
  96. how you can maybe output stories from data in an automated way or things that maybe serve that
  97. mobile audience that maybe wouldn't serve a print or sort of a long reading audience that's putting
  98. more attention in little bits and bytes of information. Like an example I would give is a
  99. few years ago, one of my colleagues developed a bot that automatically writes and publishes
  100. stories about earthquakes within seconds of them happening by converting the sort of pulse of
  101. information that comes out of the government seismographs into a very simple post, which then
  102. is sent to the copy desk automatically who review it before sending it live. And that allows us
  103. almost literally within seconds to have a post with a basic map and a basic summary of the
  104. earthquake ready to ship. And then if it all worked out publish, and that allows us to get that little
  105. bit of information that people more often in a mobile space are trying to find kind of like
  106. immediately, hey, was that a big earthquake here in LA just seconds ago, right? And conceiving of
  107. kind of the audience's needs, from kind of the mobile point of view can lead you to like put an
  108. emphasis on creating more things like that. Yeah, that's really interesting. Everyone says mobile
  109. first, you just said mobile first, are you literally thinking about Oh, we're going to do
  110. this new story on, I don't know, wildfires, is it the development is literally mobile first? Or is
  111. it you know, we have an idea for how we want to tell this story. And now let's make sure that it
  112. works on mobile. So my question is really whether it's thinking initially in a small screen versus
  113. thinking sort of broader and then modifying elements of the story so that they work on a
  114. mobile screen. I mean, in the last couple years, our narrow team, the data desk has kind of evolved
  115. more to projects work, off breaking news work. And so on that particular example you gave hasn't
  116. come up for us. But you know, our graphics desk, which is, you know, just increasing in strength
  117. and does a great job on most of the kind of the breaking news work like that, they definitely are
  118. thinking that way and are taking that approach. When it comes to something more like a long term
  119. project or something that you that you want to have kind of be special. I don't know if we think
  120. mobile first, but we're kind of thinking like what you know, at its bare essence, what is the story
  121. we're trying to tell, right? And then what is the most powerful way to reach, you know, our audience
  122. as we conceive them, you know, with the tools we have. And I think, you know, as part of like that
  123. conception and that idea development process, you have to be thinking about people's phones,
  124. because that is, you know, this gigantic percentage of your audience. And so for me,
  125. when it comes to like the projects work, I think it really begins with what's the story and how do
  126. we best tell it? That's really interesting. You sent out this tweet in December about 30 of the
  127. most viewed pages in 2017 and nine were by apps or designers. Can you talk a little bit more about
  128. that expand about that? You know, what does that I guess really what does it mean to you as someone
  129. who has seen this transition to have about a third of the biggest hits be from your team or from the
  130. development team? To me, it's just a sign that what we're doing is connecting with our audience
  131. and is kind of at the core of what our newsroom is doing to try to change and survive like in the
  132. digital space. One of my sort of maybe secret motivations for writing a post like that is that
  133. one of my kind of pet peeves about particularly about data analysis projects or about sort of the
  134. card investigative work is sometimes there's an idea out there that you know, they don't draw a
  135. large audience, you know, or that they're sort of part of an outdated sort of prize winning process
  136. that happens in newsrooms or something. And I just think that that is at least in our case 100%
  137. untrue and that our team is contributing to and trying to really move the needle with reaching a
  138. digital audience and doing new things. And I think that I think that's almost certainly true,
  139. probably most other teams like ours. And I think it's just an idea that I want to kind of reinforce
  140. and spread because I don't often see it hammered as hard as I think it should because there's
  141. really good reasons to invest in the type of work we do that are, you know, to be frank, you know,
  142. commercial. And I think that's fine with me. You know what I mean? And so that, you know, that's,
  143. that's one reason I think that that's an important story to tell it to is I'm just really proud of
  144. the people who work on our team, because I feel like that's a sign of their hard work paying off.
  145. You know, if you were to look at those nine things, I didn't really elaborate on what they
  146. were, but you would find kind of a broader array of things than you might think of with our team
  147. being called the data desk. They're not all just like data stories. The number one most viewed
  148. thing at the LA Times, at least so far, this year was a series of editorials that our opinion page
  149. put out early in the year about President Donald Trump, and some problems they had with him. And
  150. in an effort to make a bigger splash out of the series of a half dozen editorials,
  151. Andrea Roberson, who's on our data desk team, designed using our data desk custom publishing
  152. tools, a series of pages outside the content management system that really kind of blew these
  153. up and made them look special, you know, and improve their presentation. And you know,
  154. I think obviously, you know, the primary reason that those editorials succeeded is the content
  155. and the writing and the people not on our team who, you know, did all that. But I do think a
  156. significant component of it was the excellent work she did in the design of it. And so that's why I
  157. think, you know, that that would be a case where our team kind of had a hand in in something that
  158. you might not think of as the narrow data work, but was really important. Right. You know, I've
  159. noticed over at the New York Times that they've started adding the developer names to the bylines
  160. of news articles. Have you all started to do that? And if not, is that something that you think,
  161. I think going in line with what you were just talking about of your team being a crucial part
  162. and a central part, is that something that your group is looking to do in 2018?
  163. Oh, totally. Yeah. I mean, it's something we've done for years. Okay, honest. It's something
  164. we've done for I can't even remember, but at least five years or more long time. You know,
  165. there's different ways it gets kind of handled depending on the context. And you could I'm sure
  166. finding consistencies and how it's been applied with zero exceptions, you know, if someone has
  167. done custom web development work that even if it's strictly cosmetic, they receive a production
  168. credit at the bottom of the page. If you know that production work is for like sort of a data
  169. graphic, like say election results, or some sort of standalone visualization, we do traditional
  170. bylines at the top of the page. And then, you know, in cases where it's like a long form narrative
  171. where we haven't done just cosmetic design, but we've also contributed data analysis. There's
  172. sometimes kind of a mix between whether you receive kind of the traditional byline alongside
  173. the writer, which in many cases you do, or, you know, this sort of print tradition of a
  174. contribution line at the bottom of the writing, which is based on some judgment call about, you
  175. know, the significance of the work that's been contributed. And, you know, there's a lot of,
  176. you know, messiness and give and take and when that happens and how, but, you know, to me,
  177. I think it's important that people receive their credit. I, you know, it might be better if we had
  178. kind of stricter rules in a system that everyone was on the same page with. But, you know, these
  179. things are evolving all the time. And to me, it's just something we're consistently trying to sort
  180. out. Yeah. Yeah. I mean, it seems like continued evolution. Oh, yeah. I want to get your thoughts
  181. on what were your favorite stories from 2017? What pieces you all worked on that you thought
  182. were highlights, not necessarily that may not have been the biggest, you know, page views,
  183. but which one sort of struck you as both the editor and as a reader? Sure. I mean, so here
  184. there's a few local or California stories that might not have got all around the world that
  185. our team participated in that I think were real good. Let's see. So our first example might be
  186. Ryan Menezes on our team, who's sort of a actually a statistics person who has become
  187. a journalist on our team partnered with Ivan Penn on our business desk, who's sort of a long time
  188. energy reporter to use data to help Ivan sort of raises ambitions and write a bigger series of
  189. stories about some phenomena. Did I get that right? I never know. Yeah, I'll have to look that
  190. one up. They probably both were wrong. But some stuff. Yeah, so Ivan's an energy reporter on the
  191. beat, right? Right. He knows lots of stuff. But sometimes in just covering things that happen
  192. out there in the world, you can't take a step back and kind of get at the bigger issues you want to
  193. get at with the kind of good enterprise. And that's where data can often help. And in this case,
  194. I think it did where he had sort of been hearing from people and kind of sort of developing the
  195. idea that, Hey, maybe California just has too many power plants just has too much power. And
  196. this is not something that was in any story anywhere or that had really been written about
  197. at all. It was just kind of this thesis that some people on the fringes were kind of putting out
  198. there and he wanted to find out if was true and kind of evaluate that. And so we sort of teamed
  199. Ryan up with them and using a lot of data, some of it publicly available, some of it gathered,
  200. you know, through records requests and other means, were able to like really dig in and go at
  201. that question. And the answer they found was that it's true that California actually has a gigantic
  202. and growing surplus of energy being generated by the power plants out there across the state.
  203. And we continue to build more. And that's actually one of the reasons why price continues to go up.
  204. And so we have this kind of broken supply and demand curve of demand is actually decreasing
  205. due to like energy efficiency, green alternatives, yada, yada, yada. So demand is flat and supply is
  206. increasing, which, you know, if you remember high school economics, that should result in lower
  207. prices, but it's not prices are going up. And the system in some kind of fundamental way is broken.
  208. And they were able to really dig into that through data. We wrote kind of three big kind of stories
  209. through the year, which we tried to bring to life and take make this boring, complicated concept
  210. relatable through illustrations, and sort of a mobile first sort of like slider kind of concept
  211. that was separate from the main stories. And you know, I thought that stuff turned out real good.
  212. It's resulted in a power plant being canceled, which is pretty cool, you know, as far as impact
  213. goes. And was to me just a great case of how statistics and data can help you, you know,
  214. kind of find and tell and prove out stories that otherwise would not be told. And to me,
  215. that's one of the things I get a real kick out of in data journalism. Yeah, absolutely. I mean,
  216. having a huge impact. Were there any others that struck a chord with you this year?
  217. So Ivan's on our business desk, and I am kind of proud this year that we partnered with our
  218. business desk more than we have in past years. And this is kind of, you know, part of that
  219. evolutionary process is you have sort of allies that you gradually develop within the organization.
  220. And then you seek to expand that network, you know, and draw in new people and new sections of the
  221. paper to kind of, you know, the approach we're trying to take. And so this year, we did a bunch
  222. of stories with business that I thought turned out okay, there was the ones with Ivan. And then
  223. there was another set we did with two reporters, Natalie Kitroff and Jeff Mohan, about California's
  224. agricultural industry and how that has really changed in the last couple years, and how some of
  225. sort of the received wisdom and potted history of sort of how that industry works is outdated.
  226. And a lot of things are changing. So like, for instance, due to increased border enforcement,
  227. there's actually tons and tons of people coming in through legal visa programs for temporary farm
  228. workers. And this is something that's just just dramatically exploded in the last couple years.
  229. And we were able to write about that and kind of dig into that issue, problems with housing for the
  230. workers we were able to get at, I thought real well. And then also kind of an interesting economic
  231. story about another result of increased border enforcement, which is there are in fact wages
  232. are going up for people who work in some industries like grape picking. But that even though the wages
  233. are going up, they still aren't unable to attract Americans to take the jobs. And so, you know,
  234. this was another case where using data and kind of traditional on the ground reporting, we're able
  235. to do kind of a much stronger story than we could have done without it. Right, right. Well, it sounds
  236. great. I'm a huge fan. I actually have a couple of what are now probably older graphs that your team
  237. did that I use in every class I teach, because I think one of the things that I love most about
  238. what your team produces is the annotation on the graphs, they're all just, I think, uniquely and
  239. expertly done to explain, you know, concepts that people may not have any idea about, or even graphs
  240. that they may not be familiar with. And I think your team just does a, you know, extraordinary
  241. job of that. So thanks so much for that, for giving me enough material to teach. So anything
  242. you're looking forward to in 2018, I know we got a lot of news coming up, but any things in general
  243. that you see in the newsroom that you're looking forward to? Well, this is again, kind of, maybe
  244. boring to people who don't work at the LA Times, but we're changing over to a new content management
  245. system, the Washington Post's sort of arc system. And that's going to be sort of a kind of big
  246. behind the scenes challenge to be pulled off, you know, primarily by the product people who'll be
  247. responsible for, you know, 99% of the content that will be published out that system. But then it's
  248. also going to be new opportunities for our team that does special work to try to better integrate
  249. things with our homepage and section fronts to try to more closely integrate how we publish custom
  250. outside the CMS work with the CMS, which is kind of one of these challenges I think kind of everybody
  251. faces, which is, you know, you don't want stuff so far outside that copy editors, regular editors,
  252. junior developers can't really participate without like, you know, really ramping up their
  253. technical skills. But you don't want stuff so inside the CMS that it really, you know, limits
  254. what you can do. And so I this new system is going to allow an opportunity for us to kind of
  255. try to rebalance how that workflow works and how those tools work. And that's going to be kind of
  256. a challenge for us to sort out in the coming years, which I think there's definitely some good
  257. potential for to improve the status quo. Yeah, absolutely. Well, Ben, thanks so much
  258. for coming on the show and great work in 2017. And I'm looking forward to see what you all come
  259. up with in 2018. Of course, thank you for having me. And thanks to everyone for tuning into this
  260. week's episode, I will of course post links to all the stories that Ben talked about. I hope you'll
  261. take some time to check out the LA Times stories. If you're on the East Coast, we tend to sometimes
  262. neglect those folks working all the way out three hours behind us. But I'm a big fan. And I'm sure
  263. you will be too if you take a look at it. So do get in touch if you have comments or questions
  264. or suggestions for other guests. So until next time, this has been the policy of his podcast.
  265. Thanks so much for listening.

Downloads

Recording audio · Timestamped transcript