A conversation with Ben Welsh

By • IRE Radio in SoundCloud

Recording

Listen on the original podcast site

Show the timestamped transcript
  1. Data journalism is a collaborative field with journalists growing into better reporters
  2. with the help of practice and partnership.
  3. In this episode of the IRE radio podcast, veteran reporter, editor, and computer programmer
  4. Ben Welsh joins to talk about his journalism career and how he got to where he is today.
  5. As the founder of Reuters News Applications Desk, where he is currently an editor, Welsh
  6. has gained a multitude of tips and tricks for data journalism that he will share with
  7. you throughout this episode.
  8. Additionally, we discuss his personal site that archives his various data in investigative
  9. journalism panels and resources.
  10. This is Nakyla Carter and you're listening to IRE radio.
  11. So the first thing I want to say is just when did you know journalism was your thing and
  12. then how did you nurture it to be the person you are today?
  13. Oh man, we can go way back with this, I'm getting old.
  14. I never set out to be a journalist.
  15. It wasn't something I did much in high school or a career goal I went into college with.
  16. I was an undergraduate at DePaul University in Chicago, Illinois after growing up in Eastern
  17. Iowa and kind of stumbled into journalism accidentally.
  18. I had a job answering phones in the communication department as part of kind of one of these
  19. work study programs to make a little money while I was in school and that kind of thing.
  20. In that job I was entrusted with some very important responsibilities.
  21. One of them was pinning up flyers on the bulletin boards around campus about upcoming events
  22. that the department was running.
  23. So I was often the first person to see the notice of new things happening.
  24. One day I was pinning up the flyer and I noticed that it was an advertisement for an internship
  25. with two journalists who were taking up residency at the university, Carol Marine and Don Mosley.
  26. They were looking for a student to join them.
  27. I was pinning in, I thought, "Hey, that sounds cool."
  28. It sounds a lot cooler than this job pinning up flyers on bulletin boards.
  29. Because I was the first person to see the flyer, I suspect I was the first person to
  30. apply.
  31. I never asked how many other people applied, but I ended up getting that internship despite
  32. being ridiculously unqualified and began my journalism career, as it was, as the grunt
  33. assistant to this producer and correspondent team who were doing weekly television pieces
  34. and newspaper columns and some long-form documentaries.
  35. I got into journalism as an accident and pretty quickly saw it was a lot of fun.
  36. It was a great way to engage with the world and learn about it and was also a challenge
  37. and maybe something I could do.
  38. I think that was amazing because I think it's important to understand that everybody's
  39. plight into journalism is completely different.
  40. It's always interesting to hear how did you get started there.
  41. You talk about how you kind of got into the journalism realm.
  42. How did you start doing more so data type of things?
  43. I was, as I said, pretty unqualified to work in this journalism office I was working in,
  44. but data and the work that I do now today quickly found me in a public records request.
  45. We got a tip that a local suburb of the city had.
  46. The tip was that a local alderman had been bid-rigging some legal contracts in this city
  47. to rack up some money for himself and his allies.
  48. To check it out, we filed a public records request asking for a list of every legal contract
  49. or fee that was paid out to the city, to this particular law firm.
  50. It was my first FOIA ever.
  51. After a few weeks or months, I can't remember exactly, finally in the mail came a big long
  52. printout, a computer printout.
  53. This is like a 1990s or '80s style one where it's an accordion sheet of paper, dozens of
  54. pages connected by the little tear strips on the outside.
  55. It was all the legal fees, but they were spread out in this dot matrix style across a lot
  56. of paper.
  57. We initially had the question, "Well, how the heck are we going to make sense of this?"
  58. I honestly don't remember or don't know where I got the idea, but I thought, "Hey, what
  59. if I type it all into a spreadsheet?"
  60. It kind of came to me.
  61. I sat down there.
  62. It took me a day or two to go through it all and double and triple check it.
  63. We arrived at kind of a pretty big number, a big enough number to justify the story being
  64. on the news.
  65. We kind of got our little scoop, our investigative piece, out of the tip, the FOIA, and the data
  66. entry.
  67. With that, I kind of was hooked, one, I saw the data was kind of a shortcut to getting
  68. to work on really cool stories, more than just kind of the run of the mill stuff, things
  69. I found exciting.
  70. Hey, as the guy who could sit down and figure out the tech, it might be a niche for me or
  71. a place for me to be the guy who fills that gap and helps make those stories happen.
  72. Those two things kind of got me going on it.
  73. I wanted to go further down that road, so I went to the University of Missouri and was
  74. a graduate assistant at IRE NYCAR, where I really got to focus on not just faking it,
  75. but actually learning how to do these skills, learning how to code, learning what real analysis
  76. was, getting a lot more practice.
  77. That set me off onto the path I still am today, now more than 20 years later.
  78. On that respect, I think that you'd grow so much as a data journalist by obviously knowing
  79. the practical experience, but also talking to people and collaboration and learning other
  80. people's tips and tricks.
  81. So I have to talk about your website, PowerWire, because I was like, this is amazing.
  82. I wish I had seen this last year.
  83. But yeah, like what made you want to compile all that information?
  84. Because it's not something that you're required to do, but it's amazingly helpful.
  85. Well, you know, it's nothing that happened overnight.
  86. It's the accumulation of kind of years and years of chipping stuff together.
  87. But you know, for me, the annual IRE NYCAR conference is really just kind of an opportunity
  88. to challenge myself and for all of us in our community to come together, to say, what do
  89. we want to teach each other?
  90. What's important?
  91. What have we learned in the last year that's worth passing on to other people or that we're
  92. excited about or that we want to share?
  93. You know, and I look at every year's conference as kind of that type of moment or opportunity
  94. and try to drum up new stuff.
  95. And I've, you know, been doing it long enough that I kind of got a little playbook for it,
  96. a little routine, and I've just been adding, you know, incrementally more and more of these
  97. kind of lessons that I put together with other people who helped me, by the way, of course.
  98. And it's become, you know, over time, this collection.
  99. And do you have any favorites or any talks that you have that, you know, you look forward
  100. to saying?
  101. Because I know some of them you repeat and some of them you tweak and things like that.
  102. But is there ever one where you kind of are like, I wish that something like this existed
  103. when I first got started?
  104. Well, I don't know if I have a favorite child, so to speak.
  105. To me, the best thing about it for me, there's a lot of great things about it.
  106. I love coming together with peers and people I respect to help create the classes and that
  107. sort of formative experience of us collaborating to really take an idea or a technique that
  108. hasn't really been fully fleshed out and trying to put it on paper and get it across.
  109. Like, that creative process is really stimulating and always rewarding to me, and that keeps
  110. me doing it.
  111. And then, you know, the conference itself is a great experience, but, you know, maybe most
  112. of all is the thing that comes out of putting it on the web and just kind of having a little
  113. bit of faith in the world or giving it out there is I am just often stunned to hear from
  114. people on the other side of the planet who I've never met, never will, who will reach
  115. out and say they learned this or that just based on a resource that we put on the web,
  116. not really knowing what would happen with it.
  117. And there's something that is kind of magical to that, that is a hit I keep chasing or that
  118. I just love, and it really reminds me of my own formative experiences on the internet
  119. as a teenager in the 90s, you know, I'm old enough that, you know, I remember life before
  120. the web and as a kid in rural Iowa, its arrival was a big deal.
  121. And the internet was really my kind of gateway to the world and such an exciting, stimulating
  122. place to learn and connect and to still have experiences like that 30 years later with people
  123. in Ukraine or Brazil or wherever else is just, it's the magic of technology that I find so
  124. rewarding.
  125. You know, I'm learning about your plight to where you are today and you went from not
  126. knowing that journalism was something that you even were interested in to creating an
  127. entire news applications desk.
  128. How did you know that that is something that that newsroom needed and how do you operate
  129. in that position?
  130. Not sounding too grandiose or self-assured, I think we are entering kind of a new era
  131. of data journalism in kind of the commercial space, commercial media, which is I think
  132. one of industrialization and specialization.
  133. You know, when I started out at the LA Times in 2007, the whole idea of putting data on
  134. the internet on a newspaper website was still a pretty new one.
  135. We weren't the first to do it for sure, you know what I mean?
  136. But in our context, it was definitely new, and I think a lot of what was going on even
  137. at the leading news organizations then was really experimental.
  138. You had really small teams of people kind of goofing around trying to make things happen
  139. on the web, often totally separate from their mainline newsrooms.
  140. You know, it's good to remember that it was that 2008 election was probably really the
  141. first one where you had really widespread live election results on newspaper and news
  142. organization websites.
  143. When Obama was first elected, they did, there was examples of that before, but the really
  144. kind of first like widespread publication of it was that time.
  145. And it was really kind of an experiment that was then wildly successful.
  146. And I think there's been many other examples since then running up to coronavirus trackers.
  147. And you can go down the list of all these different ways that taking data directly to
  148. the audience in the form of dashboards and databases and visualizations that aren't traditional
  149. stories, you know what I mean?
  150. That has gone from an experiment to really a widget or a product that we want to manufacture,
  151. you know, inside the metaphorical assembly line or factory of your newsroom.
  152. And so it's my view as we enter, as we want to go from experiment to mass production,
  153. we need to, you know, institutionalize, specialize, and really ramp up how we are organized to
  154. create these widgets, you know?
  155. And in my view, that means making a news desk, which is its own little assembly line in the
  156. factory that focuses not on the kind of battleship investigative reporting projects, not necessarily
  157. on kind of one-off graphics to accompany traditional text stories, but instead on these data products,
  158. whatever you want to call them, that are incredibly popular with audiences and are also, it could
  159. be Capital J journalism as well, right?
  160. And so this news application's desk is our attempt to name something that does that.
  161. I don't know if I love the name, if you have a better name, I'd love to hear it, but that's
  162. what we mean.
  163. Basically, that's what we're trying to do here at Reuters, is to create a group of people
  164. who just focus on making that type of thing.
  165. I think that's amazing, especially because, like you said, we're moving towards a time
  166. in data journalism where I like to say that people almost expect data to be in the story.
  167. That's become way more normal, and so I think it's amazing that you guys are thinking about
  168. this.
  169. Yeah, I think we can't forget that.
  170. I think because data journalism grew out of the investigative reporting tradition, or at
  171. least a large part of it did, there's many streams and strands, there's not one way.
  172. But the IRE/NICAR tradition is really one that stems back to investigative reporting,
  173. which is really what I was trained in and what I love and respect, and I've done a lot of
  174. it in my career, but the justification for that type of work, for the investment, for
  175. the dollars to be spent to fund it, is typically one that's based on prestige and public service,
  176. which are two of the very high call-ins of our profession, but in a time of decreasing
  177. resources and brutal disruption to the economic model that supports journalism, it can be
  178. difficult to justify jobs and investment in data journalism if it's strictly seen as a
  179. prestige project, so that's bad news, baby.
  180. But the good news is, and I think we often undersell this good news, is that these data
  181. products I'm talking about, your live election results or your coronavirus tracker, these
  182. are, yes, they are public service, and yes, they ought to be seen as prestigious.
  183. They have won some of the highest awards, but they are also popular, right?
  184. I think last year, Reuters.com, the most popular thing all year, were the live election results
  185. done by our team in London, led by John McClure, and the most popular thing in the history
  186. of the Los Angeles Times, where I used to work for the most read piece of journalism
  187. in the 100-and-whatever-year history of the LA Times, was its coronavirus tracker, and
  188. I suspect the same is for the New York Times and many, many other news organizations, and
  189. I think we shouldn't sell ourselves short.
  190. We're making things people want.
  191. A lot of people say journalists and math don't go together, or a lot of journalists aren't,
  192. I guess, set up naturally to do data, stuff like that, but what is your advice for someone
  193. who wants to get into that realm of things?
  194. It is a hot commodity right now, and it's an amazing skill set to have.
  195. Well, then, that's your advantage.
  196. You know what I mean?
  197. If other people are avoiding it and you're willing to take it on, that's going to give
  198. you the edge, that's going to give you the front foot, and I think you should not listen
  199. to people who are discouraging you to not go down this road if it's where you want to
  200. go, because, as you just said, we know it's where things are headed, and we know that
  201. it's valuable, and so don't be afraid to have the confidence of that and surround yourself
  202. with people who agree with you and find mentors and role models that'll help you move down
  203. that path.
  204. It's a lane that's wide open.
  205. What are some things that you live by when you're doing data analysis and news application
  206. creation?
  207. I don't know if I can rewind.
  208. What would I tell younger me?
  209. I mean, I've studied harder in high school, you know what I mean?
  210. I don't know.
  211. I don't have any regrets or any big thing like that.
  212. There's reassurance, I think, we all need when we're young and that I needed like everyone
  213. else, which is if you work hard, it will pay off.
  214. I've seen that in my own life, and I've seen it in many others, and I would just encourage
  215. people to ... I almost said lean in.
  216. I won't say lean in.
  217. I'll say work hard and be a good person and it'll work out.
  218. That's what I would tell myself because I was a very nervous, young, country kid in
  219. the city.
  220. As far as the work we're doing, I mean, to me, you got to have high standards for yourself
  221. and you can't cut corners and you got to do the work.
  222. I think that that means coming in every day like you got something to prove, not being
  223. afraid of challenges and realizing that by pushing through them is how you're going to
  224. do something you're proud of.
  225. Don't be lazy.
  226. It's the same message maybe, but it's really what I believe.
  227. I think it's all about progress and push and not giving up.
  228. I have to ask some things.
  229. Do you prefer Python or R?
  230. Why?
  231. I'm definitely a Python person, but I respect R and really about the results.
  232. I think there's a lot of great and talented people who work in R, but Python is obviously
  233. superior.
  234. I agree.
  235. I wholeheartedly agree.
  236. And then I wanted to ask, what is your favorite data journalism memory?
  237. Something that if it's a story or maybe an interview or just something that happened
  238. where you're just like, yeah, this is the career for me.
  239. Well, there's that first story I told you earlier where I kind of had the spark of the
  240. light bulb went on.
  241. This could be something for me.
  242. To me, the most powerful moments are often, and maybe it does go back to those investigative
  243. pieces, but when statistics and data lead me to someone's doorstep or to somewhere in
  244. the world and help both me, the journalist who's trying to figure out the story, but
  245. more importantly in the moment and maybe in the cosmic sense, the people who lived the
  246. story helped them better understand what the heck happened to them in life, particularly
  247. if it was something bad.
  248. I did a whole series of stories about the local 911 system in Los Angeles where we uncovered
  249. a lot of systemic flaws and how emergency care is delivered, how the fire department
  250. is run.
  251. And some of the most powerful memories they had is where the statistics would tell us,
  252. well, there's this kink in the system.
  253. There's this inefficiency, there's this problem, and it could be having really bad effects
  254. out there in the world.
  255. But then we followed the data, individual calls out to doors and knocked on them to
  256. see what really happened in this case or that case where the data says something maybe went
  257. wrong.
  258. And what we found often is, yeah, it did go wrong.
  259. And so there's this way in which you kind of discovered a story based on a spreadsheet.
  260. But then also there's this powerful moment of connection with the person on the other
  261. end of the journalistic exchange where you help them understand why something bad happened
  262. in their life and help them be part of an effort to describe and share that with their
  263. community so that maybe it can be fixed and it doesn't happen to someone else.
  264. And to me, those moments of where statistics and shoe leather and the kind of connection
  265. with a subject come together are kind of these magical moments that data does make possible
  266. because when you don't just go off your own bias or instincts, when you let the math and
  267. science lead the way, you can have these powerful moments of discovery.
  268. And they're not the same as Albert Einstein and E=MC squared or whatever.
  269. But they are powerful.
  270. And so I struggle to describe it and contextually it would be almost impossible to give you
  271. every detail of what that was really like without writing a novel or something.
  272. But like those moments of discovery and connection where statistics and shoe leather meat are
  273. intense.
  274. To access Ben Welsh's site and his resources, be sure to check out p-a-l-e-w-i-dot-r-e.
  275. You can also find the link in the description of this episode.
  276. Reporting for IRE and Nightcar, I'm Nikkyla Carter, and this has been an episode of the
  277. IRE Radio Podcast.
  278. Doug Meggs edited the script of this episode, and we are recorded in the studios of KBIA
  279. at the University of Missouri School of Journalism.

Downloads

Recording audio · Timestamped transcript