'"We studied these online gangs at the same time I was looking at the offline gangs and it turned out the model we were developing to explain the behaviour of the online guilds began to coincide with the offline gangs," says Johnson. "We could explain the data using the same mathematical ideas."' Which all makes sense, you know, but it's still interesting to see this stuff being done and taken seriously.

Absolutely excellent. A little depressed but also pleased at the overlap with my NLGD talk. Here's hoping I can munge together something good. And give credit where credit is due.

"...the same topic kept coming up, over and over... you can think of it as an amalgamation of crowd theory, human terrain mapping, and social simulation. It is the science of groups; it is a new kind of quantitative political science."

A perceptive and detailed analysis of the Dunbar Number - and why some of the hype around it is misguided.