spec'ing a website and other nightmares

i'm spending the weeknd at a web-based social innovation boot camp - see www.sicamp.org. Perfect opportunity to meet loads of geeks and show them my quite sad 10 page powerpoint website spec and ask them how on earth to get started. Great opportunity to learn and meet but what i'm finding most of all is how not-easy and nonlinear the process of spec'ing and launching a web-based service is.

Do you go quick and dirty to save money and time, spend a few K to get smething up, get reactions from people, and see if there's traction? Pros are quick and cheap, get out to users fast, don't do too much without their input. Con is that if it takes off in 6 months you will have to scrap and rebuild.

And then what kind of geek do you get? How do you know if s/he is any good? His input will have a major impact on the site. It feels like getting married. Or at least agreeing to go on a long road trip with a complete stranger and no air conditioning in the car. What if the other person litens only to meatloaf (the singer, not the meal) and sucks at map reading?

And all of this is supposed to take place sprinting alongside me doing user interviews, focus groups, and strategizing. And of course, i haveno proof until it's up and running that it's a good idea and well executed.

Back to camp now...
JS.