I'm following the internet kerfuffle about Malcolm Gladwell's recent article in the New Yorker about what he considers the false promise of net-activism. It follows his standard model (one that I really enjoy, btw) of a)introducing an exceptional anecdote, followed by b) presenting a seemingly-plausible theory and then c) taking it apart by using some sort of social or behavioral science. In this case A is played by the civil rights movement, B is the optimistic outlook on social media as sold by Clay Shirky et al, and C is the sociology of movements. Gladwell is perfectly justified in pointing out that twitter does not make a revolution, but he overreaches at times.
I think there are a couple of useful takeaways from the article+discussion, most of which I assumed, perhaps wrongly, that people already knew:
1-Techno-utopianism, like any other utopianism, needs to be taken in very measured doses to be practically useful. Just because there's new potential for change doesn't mean it's guaranteed to turn out that way, especially if people naively assume the latter.
2-'lazy activism,' whether by donations, magazines, or new media, is no substitute for personal investment in a cause and face-to-face direct action, but can be useful when taken on its own terms. I've taken this as a given this ever since reading "Bowling Alone."
These things seem obvious to me, and maybe not enough to hang an article on, but they are valid points.
The main thing I take issue in the article with are the all the binary oppositions in the article (strong/weak ties, networks/hierarchies) that oversimplify a messy system with lots of overlaps. It's rarely all one way or another. Strictly top-down military command structures of WWI, for instance, gave way to more flexible arrangements that allow subordinates to exploit opportunities as they come. (Gladwell himself has covered this change to "in command and out of control"--he wrote about Paul van Riper and Millenium Challenge '02 in "Blink").
Not to mention that centralization of communication is a definite vulnerability: the classic revolutionary move in the 1930s was to seize the telephone exchange and radio station, the top targets in the Gulf War were the enemy's command and communications structures, and the standard counter-revolutionary move today is to shut down cell networks and social media websites. That's the whole reason the internet was built on decentralized protocols: to survive the expected atomic attack on communications centers.
That does raise one problem with facebook/twitter: people are decentralized, but the service is not. Shut it down or compromise the operators, and there's not a lot that users can do. In contrast, the old web services like email, usenet, or IRC are nearly impossible to eliminate, even if very few people actually use them. That's one reason why I'm intrigued by the prospect of Thimbl, an attempt to do twitter-style microblogging with decentralized protocols.