I think anyone who looks at bitcoin thinks "what we need is a proof-of-work scheme with positive externalities". But when you look at the actual constraints of

I think anyone who looks at bitcoin thinks "what we need is a proof-of-work scheme with positive externalities". But when you look at the actual constraints of the proof-of-work-- unpredictable, unsolved problems with automatic difficulty adjustment-- you realize implementing this idea in a decentralized manner will take almost as big a breakthrough as Satoshi himself made.