Twitter Is A Truth Machine »

barthel:

gofwd:

By John Herrman

(This full post is now available on BuzzFeed, which was down for a time during the storm.)

There was no shark in Brigantine, and certainly no beached seal in Manhattan. The NYSE trading floor did not flood, and the 10 or more Con Edison workers trapped at a damaged plant turned out not to exist. These rumors were briefly and embarrassingly juxtaposed in users’ Twitter timelines with real and often devastating stories about lives and property that had been destroyed, people in need of help, and a city’s infrastructure buckling under the weight of a historic storm.

But more important, perhaps, we already know they’re false. 

Here is a post in which John Herrman argues that it’s OK for erroneous information to spread widely on Twitter, because Twitter is so good at refuting erroneous information that it won’t persist. Since Twitter is (he says) far better and faster at getting out accurate information than the mainstream media, it’s worth the occasional outbreak of false rumors.

It’s a widely-shared view in the tech community, and it’s therefore important to note where this ideology originates: the free software mantra of “release early, release often.” The justification there is that the more often you put your software out to be tested, the more bugs can be fixed earlier in the process, and the more fixes other people can contribute, thus building a product that’s far more robust overall. Herrman is here applying that to Twitter, arguing that information should be released early and often so that the right stuff can be of use and the wrong stuff can get shot down.

The problem with this line of argument is that information is not software. Software is a tool that either works or doesn’t; there’s no reason to keep using an old version of a program if it doesn’t do what you want it to do. It’s binary, yes-or-no. But the truth value of information, which Herrman is here emphasizing, is just one of the many purposes people use information for. It’s been widely found, for instance, that people pay attention primarily to information that confirms their existing beliefs and ignore the rest. And this is the case whether or not that confirmatory information is true, as Maura has noted.

The vast majority of what we hear about in the media isn’t of use to us personally, and so the utilitarian value of information that Herrman is emphasizing is not really the primary purpose for the information distributed by the media. For the vast majority of people unable to physically confirm a given piece of information, something that sounds true is as good as something that has been proven true, as anyone who’s been on Facebook or ventured outside the media bubble on Twitter can attest. Everything we’ve found about how humans process information points to a process in which truth matters very little, and so the later disproving of a fact matters just as little compared to the initial existence of the falsehood. Once they’re exposed to the falsehood in what seems like a credible format, it has a tendency to stick.

This matters when we start to bring the media into the equation, as Herrman does. For all the very real problems with the media, the one thing they do well is distribute accurate information. This is because accurate information distributed in a timely fashion is the only product the press has to offer. If they can’t do that, there’s no use for them. So they have some pretty great systems in place for ensuring truth! And the big advantage they have over Twitter is that they can prevent untruths from getting out to the public in the first place. An untruth not spoken aloud cannot spread.

Of course, we find ourselves in a situation where anyone can say anything on Twitter, and so there’s always an opportunity for untruths to find a wide audience. What matters, then, is how highly we value Twitter as a source of accurate information - whether we’re all agreeing to see it as a credible source or not. And that’s why Herrman’s argument is an important one to attend to. Twitter’s going to be there and people are going to be saying silly things on it forever. But if we regard it as a “truth machine” then we’re far more likely to assume some of those silly things are true. And that’s the case even if they’re later shown to be false.