Should Twitter have a built-in correction button?

In case you haven’t noticed, Twitter is becoming more and more of a media entity, both by curating and filtering the news and other content that flows through the real-time information network and also by offering to highlight content from certain providers. One of the things that most traditional and even digital-native media entities have is some kind of correction function — either a way of adding to a published piece, or posting an update to correct a fact, or in the case of newspapers and magazines a special section (often all but hidden) with corrections and retractions in it. Should Twitter have such a thing, now that so many people have come to rely on it as a news source? Craig Silverman at the Poynter Institute says yes, but Twitter says implementing such a thing would be, well… complicated.
The idea of a correction system for Twitter has been around for some time now. One of the first major incidents to trigger the debate was the shooting of Rep. Gabrielle Gifford in Arizona last year: although the congresswoman was only injured by a gunman, early reports — including some that were repeated by mainstream media sources — said that she had been killed. As with other errors, like the alleged death of Canadian singer Gordon Lightfoot and other misreported celebrity deaths (including several reports involving Apple founder Steve Jobs before he actually died in October 2011), retweets of these mistakes continued to circulate widely long after they had been corrected.

Corrections are hard because Twitter is a stream

The problem with corrections or updates to earlier tweets — as anyone who has tried to correct a spelling mistake in their own tweet knows — is that all communication through Twitter is inherently asynchronous and asymmetric. We can’t really know who is seeing them and when, and unless we keep an eye on the “interactions” feed within Twitter at all times, we don’t know who is retweeting them or saving them to possibly retweet later. That’s why people refer to Twitter as a “stream” — you can’t step into it at the same place twice. So even following up a mistaken tweet with a correction doesn’t always solve the problem, and often turns into a game of Whack-A-Mole.
As Silverman — who is also the author of a recent book about media fallibility called Regret The Error (and a personal friend) — describes at Poynter, the latest suggestion for a correction feature came from digital designer Oliver Reichenstein of Information Architects, who mocked up what such a feature might look like. In his view, Twitter could implement a “strikethrough” feature via a separate button, so that instead of deleting an incorrect tweet and making it vanish (which doesn’t always prevent people from repeating it), it would be obvious to everyone that a mistake had been made.
One of the biggest strengths of this approach, as Silverman notes, is that it doesn’t try to pretend that a mistake never happened — something that both digital and print-based media outlets have been known to do. Instead, it uses the correction method that has become a standard for honorable blogs (including GigaOM) in which both the mistake and its subsequent correction are made as obvious as possible. As Reichenstein describes it in his proposal:

The only format that clearly states a mistake is a fat strike through. It is a strong answer to any interpretations and accusations that follow. It clearly says: “Don’t read this. This is all wrong. I take it back. I’m sorry.”

Does Twitter have a duty to allow corrections?

In a series of responses to this idea on Twitter, however, Twitter designer Doug Bowman said such a feature would raise a number of concerns — which is presumably why it hasn’t been implemented yet. For one thing, he said, it would add “additional complexity” as opposed to simply having a “delete” button, which might confuse users and also be complicated to implement, since it would involve going back through the timeline to find a specific tweet and update it. Bowman also said that he didn’t think it was a feature many people would actually use, and therefore it likely isn’t worth the investment of time and resources it would require.
It’s easy to see how finding and updating a single tweet out of the 250 million 400 million or so Twitter handles every day might be a challenge — something that has also likely held back the company’s search function to the point where it is often useless (although Om argues that Twitter is trying hard to change that and become more of a social-search provider). If you wanted to find and delete a tweet that is more than seven days old, for example, Twitter’s own search function would be of no use, since it only goes back about a week.
At one point, Twitter was working on something that might have provided a solution, as I mentioned when this issue came up early last year in the wake of a bogus shooting incident in Oxford Circus in London: the company was experimenting with what it called “annotations” — a feature that would allow metadata about a tweet, such as tags or other information, to be included as a “payload” without affecting the length of the message. Theoretically this might have allowed for easier updating or correcting of tweets, but as far as I know the feature has never been implemented.
Does it even matter whether incorrect tweets are corrected visibly, if they are eventually updated or corrected by further tweets? Or should we just embrace the idea of a stream, and count on the truth to eventually come out, whether an individual tweet ever gets corrected or not? I tend to agree with Silverman that a correction feature would be useful — and not just for journalists or others who want to be conscientious about their mistakes, but also because it would help Twitter live up to its emerging status as a media entity. Whether it wants to do that or not remains to be seen.
Post and thumbnail images courtesy of Flickr users Florian Boyd and Rosaura Ochoa