• In his latest article, Christopher Allen takes on a question that I struggled with at one point: how do we handle our social networks when they grow too large? Too large, in this case, is defined with respect to Dunbar’s Number. Some interesting thoughts, especially on how we handled the problem in a pre-technology age. Part of the issue which I don’t think he addresses is that the current social networking services don’t have any graceful way for us to rate our strength of relationship. In real life, I spend more time with my close circle of friends than with certain other acquaintances. I may like those acquaintances, but I won’t go out of my way to see them, or I will only interact with them in certain defined contexts such as chorus or ultimate. With the publicly articulated crude tools currently available online, though, I can’t distinguish between those two “friend” relationships without offending the acquaintances. It’s a tough question. Nuance is lost online, especially in the realm of autistic social software.
  • Speaking of autistic social software, danah boyd just gave a talk at Terry Winograd’s HCI seminar at Stanford. I really liked that seminar series when I was at Stanford, and attended a bunch of the talks there. Alas, I could not get down there yesterday to see danah speak, but fortunately the video of the talk is available online, so I just watched it. Nothing too surprising, given that I read her blog and a lot of her informally published work, but it was kind of neat to see her talk. Now that I know Winograd’s seminar is online, as well as Rheingold’s, I think I’ll be wasting time watching those instead of TV. Or at least I should.
  • I still haven’t read Malcolm Gladwell’s Blink (it’s in the Amazon order that I’m going to submit soon), but he keeps on popping up in weird places. Like this interview on, where he’s asked to apply his theories of thin-slicing to analyzing the Super Bowl.
  • And, as usual, you can see other links that I’ve found that were worth bookmarking, but not interesting enough to comment on, over on