I thought you might be interested in a idea of mine that I'm in the process of getting off the ground:
Idea: create a net.project who's purpose was to research and experiment with collaborative hypermedia. Here's the pitch, off the cuff:
Experiment with various strategies to build stable collaborative hypermedia. Collaborative is defined as requiring or permitting more than one person to build an information structure. That is, no webart done by lone artists. If it's going to built up with individual parts, it's going to build up like Burning Man into an overarching structure.
Also, since other approaches exist too, study them.
It's a meta-project.
It's not going to be, "Let's build a Wiki with this, that and the other thing." It's going to be, we need to collaborate on theorem proving, on politics, on art, on story telling, on interactive realms. Each focus would likely result in unique structures. For instance, a theorem proving structure would likely not tolerate unreviewed submissions, whereas a political arena might--but it would also likely require authentication and non-repudiation.
Also, some systems don't need to be scalable or have identity, like a museum exhibit. Other systems do need to scale, like a pattern repository.
So, as a result, several interesting projects may be spawned.
I'm thinking of calling the system Meatball as a hypermedium generally looks like meatball spaghetti, with pages of content = meatball, links = spaghetti.
The focus will be on the content, rather than the technology.
And you can say this really bad pun: when a system really takes off and gets really hopping, we can say it meatballed ~ snowballed. ;)
Would you be interested in this?