Peter Kafka

Recent Posts by Peter Kafka

Twitter + Robots = Instant Stories, No Humans Required

Hey, how’s Newt Gingrich doing on Twitter?

You could try answering this question yourself by combing through Twitter and looking at what Newt says there, and what people are saying about Newt there. But that would take time and effort.

Much easier to have a computer track this down for you — and then ask the computer to write up a story about it, too.

That’s what Narrative Science, the start-up that specializes in robot-written stories, did earlier this week. The result isn’t elegant, but it gets the job done, in a brute force sort of way.


Newt Gingrich received the largest increase in Tweets about him today. Twitter activity associated with the candidate has shot up since yesterday, with most users tweeting about taxes and character issues. Newt Gingrich has been consistently popular on Twitter, as he has been the top riser on the site for the last four days. Conversely, the number of tweets about Ron Paul has dropped in the past 24 hours. Another traffic loser was Rick Santorum, who has also seen tweets about him fall off a bit.

While the overall tone of the Gingrich tweets is positive, public opinion regarding the candidate and character issues is trending negatively. In particular, @MommaVickers says, “Someone needs to put The Blood Arm’s ‘Suspicious Character’ to a photo montage of Newt Gingrich. #pimp”.

On the other hand, tweeters with a long reach are on the upside with regard to Newt Gingrich’s take on taxes. Tweeting about this issue, @elvisroy000 says, “Newt Gingrich Cut Taxes Balanced Budget, 1n 80s and 90s, Newt experienced Conservative with values”.

Maine recently held its primary, but it isn’t talking about Gingrich. Instead the focus is on Ron Paul and religious issues.

No Pulitzer for that one, but that’s not the point. This is a work in progress that’s meant to illustrate an interesting project Narrative Science is working on: The Chicago-based company got its start by creating stories out of “structured” data sets, like baseball box scores. Now it is looking at “unstructured” data — like the thousands of messages that Twitter puts out each second — as story fodder, too.

Stories about things people say on Twitter aren’t super compelling (though if you check Techmeme, you’ll see that my fellow typers and I sure make a lot of them). But if Narrative Science gets good at this, it could end up somewhere interesting.

“This is a really basic story, but it’s fully automated,” says CEO Stuart Frankel. “No human touched it at all. You can see where this thing is going.”

Meanwhile, Frankel says, Narrative Science’s core business is taking off. The company, which raised $6 million last year, now has 30 clients, many of whom use the company to produce stories and reports they use internally. But you can see public examples of it via publishers like Forbes, which is using the service to crank out automated earnings previews.

Again, those previews aren’t made up of sterling prose, but they are very serviceable. Which means Forbes can either save the money it would have spent paying someone to mash that stuff out — or it can free up a writer to do the interesting kind of stories that only humans can do.

And now I’ll save myself a few minutes of work by citing myself from last year, where I tried to assure myself that Narrative Science wasn’t going to put me out of business: “The trick for content makers like myself is to find work that only content makers like myself can do — work where human qualities like experience, judgment and creativity get rewarded. And if we can’t do that, we ought to be doing something else, anyway.”

Latest Video

View all videos »

Search »

Another gadget you don’t really need. Will not work once you get it home. New model out in 4 weeks. Battery life is too short to be of any use.

— From the fact sheet for a fake product entitled Useless Plasticbox 1.2 (an actual empty plastic box) placed in L.A.-area Best Buy stores by an artist called Plastic Jesus