Well then let them follow the trajectory that SMS has - since in many ways twitter is just SMS over HTTP. Let twitter be coerced into interoperability like the telcos have ( http://bit.ly/jft6w ). Surely the necessary technical bits could be plucked from XMPP.
twitter is an SMS mailing list provider. Anyone can do this, but they quickly find that it costs lots and lots of money.
For whatever reason, twitter has been given lots and lots of money to provide this service. It's otherwise impossible to compete with "provide a very expensive service for free"
Well, google has jaiku as well as gmail, of course. And there's plurk, and identi.ca, and several others.
identi.ca is promising as it's federated and all that, but when you go there, you see the same thing over and over again: "How do I use this with SMS?" I don't think users understand the enormous cost twitter is just eating for no apparent reason.
And that's the main thing twitter has that is hard to reproduce: SMS in and out at scale. It's not that that's a hard technical problem, it's just incredibly expensive to do that.
It won't be hard to build competitors to Twitter—systems that do as much as it does but whose decentralized design ensures that they're not a single point of failure.
For a given, plucked-from-thin-air definition of "hard", no, it's not hard.
In the real world, yes, it's hard.
I don't see why the fact that people rely on Twitter should mean that they need to be broken up or opened up or any such thing.
We have one. It's extremely scalable and stable. It's well-tested, having been in use for 40 years, with a user base that has grown to around a billion worldwide. It's called email. See my comment above...
