So, the aim we have been set as I see it by Eben Moglen's great talk "Freedom in the Cloud" [1], is to demonstrate that we can have a completely p2p social network, that would end up fitting in our pockets, where we can decide what we share with whom, and we don't have a third party making any extra decisions for us.
As it happens this is simple to do without inventing anything much new. We don't need to invent anything, rather we have to learn how to use what we have, which is much more powerful than is usually realised. To learn this we need to start simple, so as not to get lost in unnecessary details. So the first use case should be as simple as: We need to each have a web server on our machines, and allow our friends to authenticate on it and do something. Perhaps they can write something, perhaps they can read something. They should do this without having to make an account on our software - that should happen automatically. Yet we should recognise them, as if they were our buddies on Facebook. So our software should be be able to call them by name - whatever they want to reveal about themselves. What they reveal about themselves is of course just the p2p mirror of the above. Namely it is what they allow us to see, when we connect to them and identify ourselves. Or perhaps we don't have to identify ourselves because the information is public to the world. Use Case 2: Same as the above but with friends of our friends. Perhaps we have software to invite foaf to our party. Our friends know who their friends are, and they may tell us (but only because we are friends). The resource to protect here is the party, a wiki page, or whatever. It's more fun if it's a party. But we don't want to do anything too complicated: we just need to prove we can protect a resource. That's it. Once we have done that we can build the rest. Those are the two initial recursive steps. Of course the whole thing has to be extensible in a way that we can provably attack any problem that we may be confronted with. This is what the aim of http://esw.w3.org/Foaf+ssl is. We have solved the initial technical hurdles, and we really just need more people to join the fun. It's not difficult to do - it just takes learning a few patent free standards, that are not so difficult to learn either, and are extremely powerful, while being deceptively simple. The only difficult thing is implementation. And that is because for some $...@! reason bugs always pop into code. ;-) Henry [1] http://bit.ly/brQmJz Social Web Architect http://bblfish.net/
