By the time we were ready to start development, we had enough ideas for improvement to occupy 1700 programmers for a few decades. Unfortunately, all we have is three programmers, and we wanted to be shipping next fall, so there had to be some prioritization.
The scary part of that for me is the "shipping next fall" thing. YIKES. I cannot imagine working on something for a year before sending it out the door. Working on my current web-based project and seeing the cool stuff people are doing with web apps and constant development makes the idea of working on something for a full year, then pressing it onto a CD and shipping it seem both scary as hell and completely insane. I know Joel has his reasons for not wanting to do web-based products but criminey, it seems to weird to make your people work on something for a year before customers get to see it and pick at it.
I'm planning on launching a semi-private beta of my site around the first of the new year with a fairly complete system. But it won't be perfectly complete, by design. I'll be adding and tweaking things constantly. Customers (assuming I get any, which I think I will) will hopefully give me ideas, I'll have new ideas, things will need to get fixed or improved or discarded, etc. I can make changes to the code on the fly. Shipping a piece of software on a CD is like scribing it into stone plates to me now, I can't get my mind around it. Joel seems to be doing really well with his FogBugz product but man, the "lock yourself away from customers, work on something for a year, scribe it into stone plates, send it to people, repeat" school of software just seems so dead to me, like Dell announcing they going to ship the latest and greatest in vacuum tube computers.