Just got off a Skype conversation with my boy Tom Smyth over at George Tech.  He's doing an interesting project in Liberia: a mobile in-country YouTube where people within the country can get a feel for what people are talking about from around the country.  Catch a pre-publication pic.

What's interesting is that you have a crowd of people trying to "use" (in a non-Western sense of the word) a system simultaneously.  This is reminiscent of Udai's work at MSR India with single-display groupware as well as Mischief.  It would be interesting to brainstorm ways of allowing multiple people to "use" the system at the same time, e.g. using computer vision to allow input and to interact with the system.

One idea is to use the webcam and show 2 buttons side-by-side on the screen and ask participants to move their hands who are on that side of the system -- the side that moves more 'wins' and the system goes down that path.  Use-by-voting is not necessarily a good thing and may be culturally-dependent.  Smyth was saying that people weren't crowding the system and took turns nicely.  Maybe due to novelty effect and the fact that foreigners were watching the system be used? 

Anyway, interesting work from the great Tom Smyth!