It comes down to a saying I’ve heard a lot lately, which I’ll bend to fit here. “More data beats better algorithms.” Getting your restaurant reviews from trusted friends might be a better algorithm than getting them from random web surfers. But I have a few hundred friends, and there are a few hundred million random web surfers, so there are many orders of magnitude more data on Yelp than will ever be on my Facebook friend list. And that’s why I don’t believe in the social graph. It will rarely be of any use when compared to the good old internet. I’ll stick with the anti-social graph, the hundreds of millions of randomly scattered data points on which search engines, eBay, Wikipedia, and just about everything else we’ve come to know and love in the Web 2.0 era are based.
