Social Networking is all the hype and the core of Web 2.0. There are probably 50 "Facebooks for scientists", none of which are really catching on  why not?
This session drew a comparably large crowd (basically full room) and most people were on social networks themselves and basically everybody had something intelligent to say. It was hosted by Deepak Singh and Cameron Neylon. One of the successful sites where comparatively many scientists hang out and that is actually quite useful is Friendfeed (I use it). Nature Network is another such fairly successful site. Cameron characterized NN more as a place where scientists socialize rather than a social network for scientists (a notion which I would probably tend to support).
The discussion then moved into what makes social network sites successful. This discussion then veered into Nature Network and what it actually is and why people consider it successful. Is Nature Networks a social network for scientists or is it (developing into) a blogging network, more like ScienceBlogs? In the room there have been quite a few examples of people who sought expertise on social network sites. Deepak and Henry Gee made the very relevant remark that this particular usefulness can be traced to the usenet groups of the 1990s. A final point in this discussion was the idea that scientists are not inherently social. They're not interested in meeting people, they are looking for people. "It's not the scientists who are social, it's the data". This Friendfeed thread has some of the points Cameron threw out about how to create a successful online service for scientists. Some other fairly well-known sites which got bounced around in the discussion were CiteUlike, Connotea and Mendeley.