BIP30 and BIP34 interaction (was Re: [BIP Proposal] Buried Deployments)



Summary:

Eric Voskuil and another person are discussing the possibility of hash collisions in transactions. Eric believes that all future transactions will have different transaction IDs because rules guarantee it, but the other person argues that there is still a small chance of a collision happening. However, both parties agree that duplicate transaction IDs are impossible barring a hash collision. Eric suggests that claims requiring hash collisions should include a qualifier acknowledging the unlikelihood of such an event. He also notes that a re-org to before the enforcement height could make this possible, but checkpoints protect against it. The security of many parts of the system depends on hash collisions not being possible, so this is not something a node can dismiss.


Updated on: 2023-06-11T20:36:39.292123+00:00