Using the Interweb to connect communities

I was reading an article by Simon Willis of Cisco on connecting cities and municipal involvement in broadband from the recent New Statesman supplement on broadband ... (can be got from Anthony Townsend's excellent site. At the end of the article he bemoans the lack of good community applications. This is something I've been thinking about in the context of social software. There are interesting possibilities for the 'virtual' world to restore the idea of neighbours and community in a locality, or certainly to augment it. In a study done by (will update with names) of a new neighbourhood in Canada, they found that having broadband and community fora strengthened weak social connections between neighbours-- so people would know each other well enough to say hello. These kind of connections are very important in helping people to feel secure in a place and strengthening their sense of well-being.

The challenge is to determine what kind of applications would be useful in a context and then work out how they should be built. Taking the second point first, I think a centralised model of building applications, where a developer or council builds applications for the customer does have a place, but might ultimately be less successful than applications built by people living in the place. Centralised development would work for centralised services, e.g. transport, and lots of good applications exist to help people work out how to get places (e.g. Transport for London's Journeyplanner) and locate timetables.

But the exciting possibilities are for more informal even temporary applications built by people in a neighbourhood to be used by people in that neighbourhood. How those apps might be built, cheaply, is pointed to by Clay Shirkey's article on situated software. He talks about making applications without having to build in lots of security, authentication, scaling (which all cost) because the applications are designed to be used by a small user community who know each other (which gets over the problem of trust), or at least can be shamed in front of their peers if they misbehave...

There are lots of interesting questions arising from this... What would these applications look like? how would wireless change things-- allowing people to interact and get information about a space in that space engender new types of application and communication? what would be the role of the planner in this?
Applications like LinkedIn and Orkut promote social networks, but without much explicit reference to place-- how might you leverage them?

Many questions...