Are robots and content farms the future of the news?

There was some consternation in the media industry this week when the Chicago Tribune announced that it was letting more than 20 of its journalists go and handing over its local coverage to an outfit called Journatic, which looked to some like a “content farm” not unlike AOL’s hyper-local Patch unit. Meanwhile, Wired magazine wrote about another emerging competitor for the traditional news business — namely, the news-writing robots or algorithms employed by startup Narrative Science, which automatically generate sports and business stories. Is this what the future of the media industry looks like? Robots and content farms?

The announcement from the Tribune made it sound as though the newspaper was investing more resources into its hyper-local unit, which is known as TribLocal and has been providing coverage of smaller regions and communities both online and in print for about the past five years. Editor Gerould Kern said that the paper made the move because “we believe that it is a more effective way of providing hyperlocal news, and we think we can do more of it in this way.” And the Tribune isn’t just outsourcing its local coverage to Chicago-based Journatic — it has acquired a stake in the company as well, although the size of the investment hasn’t been disclosed.

Journatic pays writers $ 2 to $ 4 for each story

For journalists at the Tribune and elsewhere, however, the big news was probably the fact that about half of the 40 journalists who worked for TribLocal — reporters, editors, designers and web developers — were going to lose their jobs, with the rest being reassigned elsewhere inside the Tribune. And who is replacing them? Much like so-called “content farms” such as Demand Media, Journatic appears to use primarily freelance contributors who are paid on a piece-work basis: one ad for writers says they will be paid $ 2 to $ 4 per story, which the site says works out to about $ 12 an hour.

Position: Per Piece Writer
Treatment: 1099 Independent Contractor
Time: You choose when you work, but we are looking for day availability
Location: Remote. As a contractor, you choose where you work
Pay: Per-piece, roughly $ 12/hr. For example $ 4 stories take about 20 +/- minutes, and $ 2 stories take about 10 +/- minutes.

For some observers, including Streetfight columnist Tom Grubisch, this solution makes perfect sense: while some journalists lose their jobs, the Tribune gets local content that costs substantially less than the stories it was producing — and some of those stories don’t really deserve any more than 10 minutes or $ 2, says Grubisch, if they simply involve note-taking at a city council meeting. In fact, he says some stories may not require human beings at all: “Reporters, at $ 12 an hour or higher rates, aren’t needed for most data journalism. That can be produced by software and algorithms.”

Are robot writers freeing us up, or putting us out of work?

And that brings us inevitably to Narrative Science, a company that also happens to be based in Chicago. Founded by a couple of data scientists in 2010, it now produces sports and business coverage for a range of outlets, including Forbes magazine — where it provided this auto-generated article about the expected earnings from the New York Times Co. The Wired magazine piece makes much of the fact that Narrative Science pieces are almost indistinguishable from traditional sports or business coverage.

Rebecca Greenfield of The Atlantic, however, says that we shouldn’t be scared of “robot journalism” because much of it is barely even journalism at all — a point that she has made before. Instead, she says, most of what the company does is the meaningless data-driven drudgery that regular journalists usually hate doing to begin with, because it doesn’t add much value (and having written earnings summaries, I can attest that this is true). So why not let robots do it? A similar type of argument could be made for using Journatic.

Data specialist and journalism professor Matt Waite makes this argument in his defence of “algorithms for journalism.” In effect, he says, we can use robots or automated tools to do what those things have always done — namely, take away boring and repetitive tasks that human beings have always done, and make it easier for them to focus on the things they do that really add value, in ways that only humans can.

That’s a powerful argument, but it presumes that the journalists who are “freed up” because of Narrative Science or Journatic can actually find somewhere else that will pay them to do the really valuable work that machines can’t do. If they can’t, then they will simply be unemployed journalists. And newspapers like the Tribune that make use of a content-farm style approach need to be aware of what they are getting as well — as the Washington Post recently found with one of its bloggers, asking for dozens of stories a day can tend to reduce the quality to an unacceptable level.

Post and thumbnail images courtesy of Flickr users Connor Tarter and D. Miller

Related research and analysis from GigaOM Pro:
Subscriber content. Sign up for a free trial.

  • Content Farms: The Players, The Benefits, The Risks
  • NewNet Q1: Content Farms and Niche Networks on the Rise
  • Report: Monetizing Digital Content



GigaOM