I agree with Joel that we shouldn't launch a company, but an experiment. Very true! And the goal of that experiment should be to find a product/market fit. Agai

I agree with Joel that we shouldn't launch a company, but an experiment. Very true! And the goal of that experiment should be to find a product/market fit. Again, true! But I don't agree that the above points mean that we should delay incorporating. Creating an LLC is cheap and simple, and it doesn't have any tax complications. Moreover, it gives tremendous benefits in terms of reducing liabilities (what happens when a pedophile decides to use your website to look for victims, and you get sued?), and clearly defining ownership (can you be sure that your friend, who worked on the experiment for the first two months, and then quit for a job at Facebook, won't come back to claim ownership when you see any traction one year later?) His reason for not incorporating -- your ideas will change -- can be easily taken care of. Just don't tie the ideas to the LLC. Maybe to remind yourself of that, you can name the LLC, Joel's Experiments LLC. And when your experiment is ready to be a VC-backed business, converting the LLC to a C-Corp is easy.