Posted by: structureofnews | March 17, 2014


TowIf machines could write stories as well as humans, what would you have them do? Write faster, more cheaply, and on more topics than humans can, almost certainly.  But what else?

That wasn’t the core part of Thursday’s panel discussion on “computational storytelling” organized by Nick Diakopoulos of the Tow Center for Digital Journalism at Columbia Journalism School, and moderated by yours truly, but it threw up a number of smart, interesting ideas that get us beyond doing what we’ve always done and into truly taking advantage of technology to help fulfil journalism’s mission.

Among the ideas: Different versions of the same story for different people; hypothetical stories to explore different theories, and “perturbation analysis” to understand how solid a story’s conclusions might be.

But first, some background on the panel, which featured three top-notch computer scientists:  Jichen Zhu of Drexel, Mark Riedl of Georgia Tech and Larry Birnbaum of Northwestern U and Narrative Science.  We spent nearly two hours exploring a wide range of questions about machine-generated narratives, from biases inherent in algorithms, data and interaction design to the importance of the provenance of data in a world where stories are created purely from data to our inherent (if unjustified) trust of narrative and anthropomorphic avatars.

You can watch it – telling insights, corny jokes and all – here.  It’s pretty interesting, even if I say so myself.

For example, we touched on the subject of how there’s some research to suggest that people respond better to avatars than to impersonal interfaces.  It may make sense, if that’s the case, to try and create more human-like user interactions for news sites.  But does it also mean that Amazon would sell more stuff if recommendations came to us in the form of conversations with a vaguely human system powered by machine-generated text?

And if humans are inherently biased towards creating and believing narratives, how can we help audiences better understand what assumptions are baked into the stories they’re reading?  If those stories are machine-generated, then there are likely fairly explicit ideas about how they were created; and if so, asked Larry Birnbaum, why not let people “tune” the assumptions to see how the story might change?  His analogy was that of a self-driving car, which can be – and presumably is – programmed to trade off speed for safety, or vice versa.  Why shouldn’t machine-generated stories offer the same kinds of controls to readers?

Similarly, if there are questions about how accurate the data is that’s used to generate stories, how about allowing readers to explore different scenarios by tweaking the data – what Mark Riedl described as “counterfactuals” and Larry framed in terms of “perturbation analysis.”  Regardless of what we call it, it’s a fascinating use case to help bring data to life for people and to help them explore the impact of it on their lives.  It’s also a way, as Larry noted, of understanding how “stable” a result is – something that any sports fan can understand when looking at a game result and trying to figure out it hinged on a few turning points or was a complete blowout.

Speaking of sports, Larry also described an early iteration of sports story that he had tried to build – one that emphasized suspense (“it was late in the eighth inning, and John Smith stepped up to the plate with the fate of the championship resting on his veteran shoulders…”) over results (“John Smith’s three-run homer in the eighth inning sealed the championship…”).  It turns out, of course, that most sports fan prefer a story that tells them what happens up front; they don’t need to be seduced into reading a story about their team.

On the other hand, non-sports fans may need such suspense to keep them interested in a topic they don’t care about.  With human sportswriters, of course, we can only write one version of that story, and generally we choose the one with the biggest audience.  But if machines can write such stories – and they can – why not write a different type for each type of reader?  Why not do that on stories about education, or health policy, or economic data?

If the purpose of journalism is to better inform society, not solely through the provision of dry facts but with engaging narrative and tales, can machines help us democratize such storytelling by providing the kind of story that each person needs?  Does letting readers explore hypotheticals help them better understand why things happened the way they did?

Maybe technology isn’t able to give us these capabilities yet.  But shouldn’t we be thinking about how best we can use them now, so that we’re really using them in the most useful way possible?


  1. […] If machines could write stories as well as humans, what would you have them do? Write faster, more cheaply, and on more topics than humans can, almost certainly. But what else?  […]

  2. […] But technology has advanced significantly so that the prospect of personalized narratives and exploratory “counterfactuals” and news-on-demand are fast becoming a […]

  3. […] give you useful responses when you ask whether you need an umbrella tomorrow; or help you understand what the key factors are in how a result came to be, as Narrative Science‘s Larry Birnbaum explained at a panel I […]

  4. […] Machines are already better at speed and breadth than humans are, and in the not-too-distant future will probably take on tasks that people can’t possibly compete on, such as highly personalized stories or creating news-on-demand. […]

  5. […] to write stories? Why not let readers decide when they want the information? Or even give them different ways of approaching the same information, from sports stories for non-sports fans to systems that generate different stories depending on […]

  6. […] provide is deeply bound up with how their sites, CMSes, and information are created. How well would natural language generation work across a host of platforms?  The creation of personalized news-on-demand? Immersive data […]

  7. […] And they also come to the conclusion that it isn’t just about better storage of what we’ve already done – although there’s a good argument for that – but more the need to be able to automatically (or algorithmically) recombine information into new stories to serve new needs, audiences or platforms – or even interests. […]

  8. […] That’s where, as Andreas also notes, automated journalism can really grow. And that includes creating stories with different angles or tones to suit different readers, and also stories that allow for hypothetical sc…. […]

  9. […] much more about it in the next week or so.  It’s a capability that opens up personalization, counterfactuals and so much […]

  10. […] hadn’t sold IBM last week, you idiot, you’d be up up 3%.  (Showing the value of doing perturbation analysis as well.  Although we may leave out the “you idiot” […]

  11. […] Google’s Living Stories – and something I’ve written a fair amount about, both as where journalism needs to be going and what structured journalism can help power.  And it moves us off the notion that the story is […]

  12. […] than humans.  It’s a perfectly valid question, but isn’t the better question how machines could change the way we find and provide information to […]

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s


%d bloggers like this: