5 ms·
These faux rebuttals seem to intentionally miss the point Mark was making: Doing a startup, the kind that Mark means, the kind that he sees -- specifically tho
by wheels 16y ago
These faux rebuttals seem to intentionally miss the point Mark was making:
Doing a startup, the kind that Mark means, the kind that he sees -- specifically those with a story about how they're going to become billion dollar company -- works best with some specific personality traits. Is that really that hard to swallow?
Does it mean people without those can't start businesses? Heavens no.
These arguments are effectively a straw-man -- the point of contention is on the definition of a startup, not on who's fit to do one. They seem to redefine "startup" as "business doing something with the internet" and then proceed to debunk Mark's argument about who's well suited for those. But again, that's completely (intentionally?) missing the point.
Also the "lottery" hooky from DHH is dumb. Nobody compares getting to the top of other fields as pure chance -- why would it be different with startups? Did Michael Jordan and Zuck have a string of lucky breaks along the way? No doubt. But do you really think hard work had no more to do with it than it did for the person scratching the foil off of a lotto scratch-card? Do you really think there aren't recognizable personality patterns in top athletes? Really?
- brudgers 16y ago>"These faux rebuttals seem to intentionally miss the point Mark was making" I think much of it is a reaction to the way in which Suster has appropriated the term "entrepreneur." All entrepreneurs are not in startups and all entrepreneurs do not swing for the fences every time (just as Tiger Woods doesn't fire every shot at the pin). Most serial entrepreneurs do their deals with a pro forma not a prototype. Ted Turner built his media empire on sure things not with lottery tickets. The startup model creates a somewhat false impression of a radical division between entrepreneurs and investors. The damping effect for VC's in Suster's graphs is more a function of wealth than role. The short stack will always have more highs and lows at the final table than the chip leader.