I worked on two that failed, one because of Napster-Induced Hysteria ("nobody will ever make money with online music!") and the other because we were myopic eno

I worked on two that failed, one because of Napster-Induced Hysteria ("nobody will ever make money with online music!") and the other because we were myopic enough to think Orkut might bury us. Those were days, right? From the first one I learned two things that are probably still applicable today, and from the second I learned one big thing. In order: 1. If people from outside the tech world end up controlling your software startup, that indicates you are doomed. Pretending otherwise will just make it worse. 2. Great teams are incredibly rare, and if you have the good fortune to be part of one you should really try to keep it together for the next challenge. (We dissipated so hard it was ridiculous... one went into motorcycles, one got a PhD, two floated back to Europe, one became a bartender...) 3. If the founders can not convincingly dogfood their own product, you are doubly doomed. I remember sitting at a table with founders while they tried to think of things they could maybe do with The Product we'd just spent months building, and feeling the rats of doom nibbling at my toes. It had not occurred to me that this was a possibility. Engineers: vet your founders!