First, and most generally, Jacobs suggests a few cautions that people creating online friendstering tools willy-nilly might want to think about. At least, the notion of uninhibited friending and information-sharing and social networking probably needs to run up against her observations about privacy and togetherness. In particular, her observation that in the city (as online), where there are too many people to be friends with them all, forcing people to choose between togetherness and nothing often leads them to choose nothing. What she suggests is that these casual, loosely-defined sidewalk relationships are exactly what people want. And we might suppose that the friendstering emphasis on explicitizing one's social network would make people as queasy as inviting one's neighbors into one's home.

Second, Jacobs makes some interesting observations about networks and connections. Again, she's talking about cities, neighborhoods, and sometimes political organizing, but they may have analogs in the online world. First is her discussion of public people like the shopkeeper who is a central character in his neighborhood. He provides all kinds of casual services for people around him and is sometimes a conduit by which they connect, but, interestingly, he would never take the role of explicitly introducing people. Second is her discussion of district-wide networking, which starts in fairly narrow environments like streets and shared-interest clubs and then extends by way of "hop-skip people" across the district. Analogs in the online world might include the sorts of hub people who bring together people from different realms of interest. Furthermore, her discussion of cross-district ad-hoc organizing might parallel people getting together for larger political or other purposes; an example that comes to mind is the way the friendster/fakester controversy was, in some ways, an ad hoc point of connection for people across friendster.

Finally, another point Jacobs touches on is that these relationships need time and trust to develop, and often originate in shared goals. This is almost opposite to the sort of rapid friendster-collecting that we see online. Perhaps collecting happens mostly because people are playing with the novelty of the things. But possibly designing for a slower friendstering process, more grounded in activities, interests, goals might make for stronger and more permanent links (though of course, I don't want to make the mistake I mentioned earlier of assuming that people are looking for intimate friends).

This point touches on a particular bias of mine: my preference for online social networking in the context of some other activity, such as journalling, as opposed to just naked and on its own. I mostly don't find much point in friendster, orkut, etc., except in the sort of meta-game of collecting friendsters and seeing who's on it and occasionally checking messages. But in the livejournal context, the network is embedded in people's journals. On the one hand, a journal gives much more evidence as to whether a person is someone you want to friend, and on the other, the journals provide a primary reason for friending someone: to read their journal.

Of course, there's a lot more to the Jacobs book than the few things I've picked out here as relating to social networks. It's an interesting and provocative book, a bit of a lost classic, and well worth reading.