On Thu, Feb 23, 2017 at 6:58 PM, Peter Todd wrote: > > So to be clear, do you agree or disagree with me that you *can* extract a > compact proof from a MMR that a given output is unspent? > After wading through your logic on how updates are done, I agree that that can be done, but apples to apples compact proofs can also be done in a utxo commitment, and proofs of the validity of updates can be done in a utxo commitment, so there isn't any performance advantage to all that extra complexity.