- that it's all going to be worth it in the end (it would have been comforting during those 100-hour-weeks) - Some of that quick & dirty temporary code would b

- that it's all going to be worth it in the end (it would have been comforting during those 100-hour-weeks) - Some of that quick & dirty temporary code would be used for the next 18-19 years. - I might as well have used PHP instead of Perl. Same (bad, messy) code quality, but even faster development and much easier hiring. - Costly hardware early on was a waste of money (we outgrew it so fast that a beefy desktop PC would have been a saner investment at that point). - Managing people is the one thing that you can't "fix" permanently. It's always an uphill battle unless (presumably) you're naturally talented/charismatic/psychopathic. - don't bother with marketing people, advisors, business consultants early on and don't create product dependencies (e.g. by building a specific version of your product for others). It's not worth it until your product is polished and proven. - Don't hire people carelessly because you don't want to invest precious time. Don't hire friends/acquaintances unless you've seen them working. Firing people is one of the most difficult tasks. - Bad things will happen. Don't spend too much time trying to prevent them, or worrying to much. Just make sure you know what your options are when you need to put out fires and manage basic info (don't go searching for your hosting provider's phone number when you need it).