As an attendee of the SF startup weekend, I can say that it's mostly just an exercise for large group management and not actually product development. It's probably a great event for anyone who wants to show off their leadership skills, and not very useful for a top coder that wants to show off his production capacity. The stuff produced in the SF startup weekend could have easily been done by a single person under 24hrs without interruptions from "product managers" who want to change the direction of the product every 10 seconds...
If you want proof, just compare the startup weekend final products with those produced by 1 day contests like railsday or mashupday.
I remember the SF startup weekend, they had programming team and a database team that didn't really consult each other... the programming team went off and did a few simple data modeling and developed some interfaces and the database team went and created this huge database schema that nobody could conceivably develop for in 24hrs. They had all these fields for special case scenarios that were all basically neglected in the final product. I'd say there were probably at least 100 people in the SF weekend... all to produce one simple facebook app.
Yes, it does. Skribit is still alive and well from last year's Atlanta Startup Weekend.
In defense of the other projects from this year, I will say that 6/8 projects are alive past the weekend. 1/8 died, and 1/8 is on ebay. I'm not negating anything Blake says - he hits the money on the head with his lessons learned - you just get to learn them a lot faster in a failed project. I just wanted to point out that Startup Weekends can generate something else - if the team composition is right.
