I agree, but that's not easy to do. If you're going for VC, you get the same crap from investors. If you're going to bootstrap, that requires an orthogonal skil

I agree, but that's not easy to do. If you're going for VC, you get the same crap from investors. If you're going to bootstrap, that requires an orthogonal skill set in addition to a bit of luck. I think that the 2010s will be dominated by self-started firms that evolve from consulting into product spaces, but most of that activity will not be in the startup hubs (e.g. Silicon Valley, New York, Boston) because the burn rates are too high. The geography of that growth spell will be very different from what we're used to. I think the way out is to become an "X Who Programs". You become something else-- financial quant, startup founder, game designer-- who also happens to be a programmer. The reason this is so powerful is that no one (except a top programmer) can evaluate programmers, which creates nasty politics as people jockey for visibility, so you establish yourself in your more-visible X but use programming as a secret weapon. It keeps you from being treated like a commodity programmer and also gives you an edge on the best projects. XWP is what the data science phenomenon, in truth, is. There, the X is statistical knowledge.