I think what you're proposing for different solutions for serendipity is a local maximums thing. What I mean by that is slightly different than how you're using the term. What I mean by it is that it's really useful to know what's the most interesting topic to a different group of people – not the diametrically opposite group of people, just a whole bunch of different groups of people. Facebook Pages has a directory. And for a while the directory was one of the best places to find out what people were really into in Turkey, for instance, or Indonesia. And you'd go to the directory, and you'd go to what's popular, starting with a given letter. Some of the letters are wholly dispiriting, like a letter that's commonly used like an "S" is just going to give you brand names and American rappers and athletes. But when you start to get slightly more obscure like "V," "V" turns out to be a really good letter. And as you dig into "V" – or maybe it was O, I'll send you the slide – you end up with, listed at the very top of all of it, an Italian motorcyclist, and underground viral video from Mexico, a Turkish folk singer, and like five other things that you would never find. And what's interesting about them is that they're almost impossible to find information on in English. They're basically what's interesting to a very small niche, but it's an interesting way to sort of think about how you would find what another community is psyched about, and navigate that space based on: popular-enough-that-all-the-conservatives-are-talking-about-it, but different-enough-from-you.
So, I went to make a slide of this for CHI, where I was speaking last week, and I tried to go to the directory. This was so cool when I found it three months ago, and Facebook has now redone their directory so that it's now highly personalized to you. So the first page that you get of the directory tells me that I'm likely to like Global Voices and Craig Newmark, which, in fact, is true, and then tries to personalize a whole bunch of other stuff. So I logged out and logged back in, but I'm in Vancouver, so the first thing it does is pull up Tim Hortons as something that I'm likely to like as their attempt to geo-target and geolocate. So even if you're looking for a way out of the bubble, Facebook is quite actively fighting it.
Yeah, I think Facebook is the most ingenious in a way. Personally, I think they probably do this really well for the thing they're maximizing for – I just think they're way ahead of the kinds of companies that I've talked to in terms of actually finding clever ways of making those kind of connections, and [they're] the most amoral. The most like, "Hey, who cares? People are using Facebook. That's great." And not really much beyond that.
Eli Pariser speaks at the Harvard Book Store on June 1.