Should we call the system of highways and roads a social network?

[koganbot]

Over on Tumblr, in a reblog and response to Tom, I said that places such as Facebook and Tumblr and ilX shouldn't be called "social networks," this for the same reason that we shouldn't call highway systems and phone systems and bars and coffee houses "social networks." But thinking about it more, I'm not so sure that we can't call such things "social networks," if we want to. People will do so whether we want them to or not.



What we need to keep in mind, though, is that "Facebook" isn't analogous to "Tom and the people whom he has conversations with about music," most of the latter constituting what I consider a genuine social network, if not a strictly identified or clearly bounded one — that is, "Tom Ewing" may be a node that connects a lot of people in my rock-critic/musicwrite world, and it makes sense to call these people a network, say "Tom's music-convo network"; but that doesn't mean that we're always clear as to whether someone is in the network or not in the network, the network being a set of fairly loose associations. The term "network" is useful nonetheless. My point is that the terms "social network" and "social media" don't explain themselves, so we have to be alert to what in particular we're talking about when we use them in particular situations.

Comments

[petronia.livejournal.com]

This is really interesting. My feeling is really that the brands can't be trusted to come up with an intellectually rigorous definition of these words, because they come up with self-serving definitions of course, but the media adopts them because there's not much neutral expertise in the field. There's a consensus issue in the communication, as you note.

It's in the interest of brands to operate under the assumption that they can define, measure, and predict things like: networks size and shape, influence, virality... But also, the category "People Frank Talks To On The Phone" is of no use to anyone unless you create it (eg. by making an LJ filter or Facebook group, say), at which point the brand will assume you went to the effort for a reason. They might not know what your metric for linking these people together was, but they take it as given that there is one, and then that gets plugged into the black box that predicts relative influence (I don't know if anyone is doing specifically this, but I can see a workable model).

But on the other hand, the whole idea that these groups are meaningful is a red herring that only exists because there are these behavioural models to sell.