4 ms·
Proof of work isn't a big concern if the hard problems are also economically valuable to solve; there are myriad such problems.
by pishpash 9y ago
Proof of work isn't a big concern if the hard problems are also economically valuable to solve; there are myriad such problems.
- TylerE 9y agoNo they aren't. For a PoW to be valuable it needs to be cheap to verify. Hard problems usually aren't.
- Ar-Curunir 9y agoWhat you want are "moderately hard functions": functions that are neither easy nor excessively hard.
- nambit 9y agoMost solutions to NP problems are cheap to verify.
- setr 9y agoIsn't the entire NP class hard to solve, cheap to verify? Im sure some economically valuable problem can boil down to an NP problem...
- pishpash 9y agoWhat? Any NP-complete problem is "cheap" to verify vs. to solve. You're telling me there are no economically valuable NP-complete problems?
- TylerE 9y agoIf there are, why has no one built a coin around it unstead of using untold watts to calculate hashes?
- kirillseva 9y agoumm, generating a block whose hash starts with N consecutive zeroes (e.g. 000000009A8C3...) seems to be exactly that - an NP hard problem that's trivial to verify in polynomial time.
- TylerE 9y agoyes, but it's not useful in someway outside of the blockchain, which was the thrust of GP's argument.
- setr 9y agoThey did, like protein folding. Part of it is that if its going to be a currency, its not particularly appealing that some group gets arbitrary benefits from the act of mining, for free. The government has the power to force such a currency on us, but otherwise, unless the economically valuable activity is globally valuable, its a difficult proposition to justify.
- TD-Linux 9y agoIf a proof of work solution is economically valuable to solve, the hashrate will just increase to the point where it's just marginally profitable again.