I understand your point. Startups can be be obscure in the same ways as research: Interesting idea, but turned out to be wrong. Interesting, and right, but no s

I understand your point. Startups can be be obscure in the same ways as research: Interesting idea, but turned out to be wrong. Interesting, and right, but no significant application in the broad world. Or not even a very interesting idea. There's value to a system where plausible things get funded, even if only 1 turn out to pay for the 9 that suck.