There's a good introduction to Object Centered Sociality (OCS) and why it's bad for the much maligned LinkedIn. OCS is summarised:

Rather than viewing a social network as a mapping of relationships between people he argues that successful social networks connect people through shared objects. These social objects are the reason people connect to each other and they cause people to participate in and return to the network.
Interesting idea. If they are objects they might be tradeable like trophies in a MMO game. I haven't seen this happen so it is more likely artifacts in a reputation-based economy. The odd thing about successful networks is that they would seem to center around one social object (e.g. bookmarks). From the thesis you would imagine the more social objects to connect with the better for the service provider. The examples given of flickr, youtube, etc. are probably still standard online communities with 1% contributors, 5% add comments, and the 94% lurkers - not exactly a social panacea. Like LinkedIn they probably are not native social networks just online services with some social features.

On a theoretical level perhaps it's really Aspect Centered Sociality where each successful service represents one aspect of a persons overall social network. Different aspects are appropriate to different groups of people in your real network. The set of aspects that's important changes according to current life events.

LinkedIn comes from a time before the default access to objects is public - pre-Paris Hilton if you like. That may be why it's successful in recruitment. It may also be a generational thing. I sighed up for LinkedIn because it's moderately useful, and in my experience, without any spam-y side-effects.

After searching LinkedIn it looks like Paris Hilton doesn't have a (real or fake) LinkedIn account but it does give "Similar searches:, model, donald trump" :) I might try to "friend" Donald to find our more after I catch up with the real one of the two listed Tom Waits :)

Facebook is seen as real threat to LinkedIn. And sure enough Paris has several facebook accounts although some people have had a facebook nightmare just trying to register their own name :)

Recruitment is one of the 3 or so (Craigslist) ways of actually making money online so I think LinkedIn will survive many of the current Web 2.0 social network crop even if it is slow moving doesn't have data public by default. This is in contrast to cases where content in a MySpace-like account would seem to reduce your chances of getting a regular job :) General recruitment is a pretty conservative area and LinkedIn's lack of cool probably reflects that. LinkedIn handles the resume "aspect" pretty well and it's a great place to hide from the paparazzi :)