After reading a post about Reporters’ Lab on the Nieman Journalism Lab blog, we had to know more about this innovative project and how the tools it’s creating to advance public affairs reporting. Founded and directed by Sarah Cohen, who won a Pulitzer Prize for her investigative reporting on DC’s child welfare system, Reporters’ Lab is part of Duke University’s DeWitt Wallace Center for Media and Democracy. The lab is working to adapt technology from other disciplines and develop tools that will reduce the cost of investigative journalism.
Cohen chatted with Ebyline about some of these tools as well as the challenges investigative journalists and freelance writers face. Excerpts follow:
What were some of the toughest problems you identified in investigative journalism?
I was an investigative reporter for a long time, particularly in computers into reporting. And then when I took this job, I went on about a year’s worth of interviews with reporters, editors, technologists, some producers; a whole range of people who are involved in and care about investigative reporting to see if my perception was right. And those interviews weren’t necessarily to find the biggest problems in investigative reporting, but the biggest problems that shouldn’t be there. They’re unnecessary problems in some ways. There are some simple little things that are really hard.
You’re recording this interview as an example. And there are very expensive programs, special pens and things like that, that will record and match it up with your notes so that you can look at it later. And some speech recognition like Dragon Speech. But there’s really no tool for reporters to go through a recording of some kind, audio or video, mark it up the way they need to, to write their stories and to find the quotes. That’s one simple thing that we can take care of. It’s a really big issue now that a lot of county and school boards don’t have recorded minutes anymore of their meetings, and they just put the video up. You have no way to find out what happened in there if you weren’t there or if you don’t watch the whole thing. So that’s one area, I’m working with audio and video in particular as reporting tools.
How might Reporters Lab help solve some of these problems?
We’re building something called the Video Notebook, which does two things. It should be ready for alpha release pretty soon, we’re setting up testing on it for the next couple months. Say you are a reporter who has to cover a school board meeting, which reporters still do believe it or not. A lot of reporters are out there every night at a school board meeting. They have to live tweet it, and they have to cover it, which means they can’t really take good notes. They have to use their Twitter stream as their notes.
It will take your Twitter stream and match it up to the video recording, since most places do post their video recordings pretty quickly; often not that night but the next morning. And it will save your Twitter stream and let you use it as a navigational aide to the video. And so if you tweeted something at 9:15, it will take you to about 30 seconds before that so you can listen to where you were at that point. But you can also take notes on it, and so at any point while the video or audio is playing, you just start typing and it integrates your notes into the Twitter stream, and the recording stops while you’re typing, and as soon as you hit enter, it enters the note, and now you can go search that video for exactly what your notes are.
That’s an example of something that we feel is an easy, quick win in terms of reducing the amount of time that reporters spend. And we’re looking at that along a number of different areas. We’re looking to see—I don’t know if you’re familiar with Web Scraping, but when you have to get data off the web—like say there’s a database you can search on the web, but you can’t search it the way you need to, so you have to scrape it.
There is no real good tool out there that’s under $20,000 a year to do it easily, and everybody’s still writing Python, Ruby or Perl Scripts to do it. So we’re looking to see can we find, create, or commission, or adapt an existing tool that will make it easier to do that.
Those are the kinds of things that we’re interested in doing. None of them is going to change the world, but if we can extract an extra hour and a half out of everybody’s day, that’s a huge win for investigative reporting.
You mentioned the Video Notebook earlier. Are there other notable projects you’d like to highlight?
One of the other things that we did was a tool called “Time Flow.” It’s a way to keep your notes on a long-running story in a way where you can see temporal patterns, and you can pick things out by timing and chronology because in investigative and most long-form reporting, the first thing you always do is start creating a chronology of events.
There are some very good tools for publishing timelines. But they’re not interactive, and they can’t handle a large data set. One of the example datasets we have in there is just a month in the life of the BP oil spill that was just pulled together from news reports.
But say one day you’re reporting on just the birds, and you want to see how the timeline of things affecting the birds, and then the next day you have to do some reporting on how congress is reacting to the spill. And so you can filter, you can zoom in, you can zoom out, you can color code, you can group things. And you can look at it several different ways. You can look at it on a timeline, you can look at it as a list, you can look at it on a calendar.
It’s a way to deal with temporal information. I always like to say that most reporting is about time and place, and we have lots of tools to deal with place, and very few to deal with time. So this was an attempt to take a stab at that.
Why do you think the tools are so focused on place rather than time?
About 15 years ago, President Clinton basically said that the federal government would make all GIS information open and available, and it encouraged all the same local people to do it. It’s very difficult to deal with place if you can’t put it on a map, whereas with time, you could make a 40-page word document as most reporters do, and kind of get a handle on time.
It’s much more of an exploratory type of thing than maps are. There’s been a long tradition in mapping to make things open, and so apps have developed around them. Making a Google map is really easy, but if you think about it in Google, there’s no similar thing about time. We use time to uncover patterns, but not necessarily to publish them.
