[Cryptography] Proposal of a fair contract signing protocol
jon at callas.org
Wed Jun 22 15:06:08 EDT 2016
-----BEGIN PGP SIGNED MESSAGE-----
I apologize in advance if I'm being clueless on some piece of this, because I have *not* spent enough time to thoroughly understand Mok-Kong's protocol. (Furthermore, I have picky questions that I don't know if they're relevant about parts the protocol, like what properties should "signcryption" have?)
There are plenty of unsolvable problems that you can solve is useful ways. Yeah, they're "limited" but they're often limited to the real world and often because the full problem is unsolvable.
For example, it's often completely reasonable to solve the halting problem by setting a timeout. Split-brain problems are solved all the time with a quorum system. You can solve the Generals/Agreement problems either by allowing it to go a wait state or declaring an exit condition, etc. In short, these unsolvable problems are solvable when you put in an escape valve that says that when you get into the unsolvable state, you declare a solution.
Mok-Kong has this note at the end of the protocol:
Note that after step (2) Alice cannot innocently refuse to perform
step (3), since the pair (X,Y) stems from her. In other words, after
step (2) the contract is de facto completed.
To me, it appears that "de facto completed" means there's an escape valve. If the protocol says that Alice does X, Bob does Y, and Alice is supposed to do Z, but if she doesn't, tough -- then the protocol terminates.
Of course, this might be unsatisfactory on some level and there may need to be software duct tape and chicken wire to make it work. But it sounds to me like there's the escape clause here.
Am I right or not?
-----BEGIN PGP SIGNATURE-----
Version: PGP Universal 3.3.0 (Build 9060)
-----END PGP SIGNATURE-----
More information about the cryptography