[Executive summary: Betteridge’s Law (probably) applies]
The Twitter furore over the #GodIsNotGreat hash tag has pretty much died down now, but there’s one branch of the debate that is still getting comments and retweets. Here’s an example from johnwilander.
#GodIsNotGreat pulled from trends because christians protest. But#ReasonsToBeatYourGirlfriend was allowed. Stay classy,@Twitter.
As I mentioned a couple of days ago, the hashtag vanished from the list of global trending topics on Friday morning. And this conspiracy theory leapt up almost immediately. As far as I can see, none of the people repeating this claim have any evidence to back it up – which is more than somewhat ironic given Hitchens’ evidence-driven view of the world.
The argument seems to go like this: At one point the hashtag was trending. Then Christians got upset and starting making death threats aimed at the people who started the trend. Soon after that, the hashtag was no longer trending. Therefore Twitter must have given in to Christian bullying and censored the hashtag.
Whilst it all sounds frighteningly possible, I hope I don’t have to spell out the flaws in the logic. If you can’t work it out for yourself then I recommend the Wikipedia article on Correlation does not imply Causation.
I could be wrong here. There might be some irrefutable piece of evidence proving conclusively that Twitter deliberately censored the hashtag. If there is, then I haven’t seen it and I’d be grateful to anyone who could bring it to my attention.
There is, however, some evidence that Twitter didn’t censor the hashtag. On Friday morning, as the debate still raging, a Facebook friend in Canada pointed out that it was still trending there. In the middle of the afternoon someone pointed out that it was still trending in San Francisco. So if Twitter were censoring it, they weren’t doing a very good job. There’s even someone who apparently works for Twitter saying that they didn’t do it.
Of course, none of this is conclusive evidence that Twitter didn’t censor the hashtag. But balancing some evidence for non-censorship against absolutely no evidence at all for the censorship I know which side I come out on.
All of which leaves us searching for an explanation for the sudden disappearance. And, to be honest, I don’t think we really need to look too hard. Things stop being trending topics all the time. Things have to drop out of the list so that new things can come in. Otherwise the list would constantly be full of nonsense about Justin Bieber and Twilight. The Twitter trending topics algorithm can’t possibly just measure the popularity of topics. That would be incredibly dull. Instead, what it does is to look for changes in popularity. A steady buzz of the same few million people talking about a particular topic doesn’t get noticed, but a sudden increase in the number of people discussing the same topic does. The Buffer blog has a good explanation of this and the official Twitter blog says much the same thing.
I’m sure that this won’t convince the conspiracy theorists. “Ah,” they’ll say, “That’s all very convenient. But that just gives Twitter an easy way to cover up their censorship..” Which is true, I suppose, but hardly a basis for a rational discussion.
And that’s the most disappointing thing to come out of this affair. The people making this accusations are fans of Christopher Hitchens. You would hope they’d be from the more rational end of the spectrum. You’d hope that they would be above making accusations like this without evidence. I guess no-one is immune from irrationality.
But I’m going to go out on a limb here. And lay my cards on the table. And other clichés that Hitchens would despise.
Twitter (probably) didn’t censor the #GodIsNotGreat hashtag.
Update: The author of the tweet I quoted above seems to agree with me.