I've been toying with the idea for a while, but there are still many obstacles to overcome:
a) Efficiency (communication, administration)
There's a reason most startups have 2-4 founders, and why certain open source projects eventually collapse under their own weight
b) Dividing the profits.
How do you measure User1233s? contribution vs. User252s?
c) More minds =/= better product
An enlightened leader (say, Jobs) is sometimes better than a democratic process. Also, massive collaboration tends to result in lowest-common-denominator results (ex. Digg, Reddit)
..but then again, a lot of these problems could be solved with some sort of (online web app?) system:
a) Basecamp tweaked for massive collaboration.
b) StackOverflow-esque reputation system + karma + uservoice demand identification?
c) No suggestions here yet, but HN seems to be doing a pretty good job at engineering away the Digg-problem.
