Proof Market: Submit Coq proof, get paid with Bitcoin

Re:Proof Market: Submit Coq proof, get paid with Bitcoin by yetanotherphd on Wednesday January 01, 2014 @04:26PM (#6996771)

This is a bit ordinary, since it is fundamentally just an escrow service.

What would be really cool is to find a way to "lock" the bitcoins so that only a correct proof in Coq of the given theorem could unlock them. Kind of like holomorphic encryption.

[ Parent | on HN ]

Re:Proof Market: Submit Coq proof, get paid with Bitcoin by equivrel on Wednesday January 01, 2014 @04:43PM (#6996866)

You probably mean homomorphic encryption[1].

[1] - https://en.wikipedia.org/wiki/Homomorphic_encryption

[ Parent | on HN ]

Re:Proof Market: Submit Coq proof, get paid with Bitcoin by paulgb on Wednesday January 01, 2014 @05:20PM (#6997020)

Bitcoin does have a (non-turing-complete) scripting language built in for ensuring that transactions meet arbitrary requirements. I wonder what subset of proofs could be verified in that environment.

[ Parent | on HN ]

Re:Proof Market: Submit Coq proof, get paid with Bitcoin by velis_vel on Thursday January 02, 2014 @02:00AM (#6998548)

The language doesn't support loops; in terms of flow control, you get if statements and that's it. So probably not much of interest.

[ Parent | on HN ]

Re:Proof Market: Submit Coq proof, get paid with Bitcoin by pirapira on Wednesday January 01, 2014 @09:34PM (#6997871)

I wonder which is easier to encode, natural deduction or Hilbert style.

[ Parent | on HN ]

Re:Proof Market: Submit Coq proof, get paid with Bitcoin by AlexCoventry on Wednesday January 01, 2014 @09:43PM (#6997895)

I don't know how to do that, but I did think about a cryptocurrency in which people got paid for optimizing objective functions.

https://docs.google.com/file/d/0B3qaT-ZL6aeKOHNEQWdpZEtRYWc/...

[ Parent | on HN ]