Don’t want to derail, but would TXHASH be sufficient for that? Or would you need to sign over it as well?
TBH I haven’t gone deeply into figuring this out. I guess you’d need to be able to check that the tx spending the current output has an output of equal (or some function) of the current output being spent which I guess TXHASH can’t do? This is for committing to the first commit tx of the channel in the address itself.
What are the requirements for this probably belongs elsewhere. I started some discussion some time ago here: Non-interactive channels without commiting to value in script pubkey · Issue #19 · ariard/bitcoin-contracting-primitives-wg · GitHub. Might be time to re-start it on this lovely new platform with latest thinking.
2 Likes
You can assert equality but not do math. You can request the hash of input N’s prevout amount and the hash of output N’s amount and check equality.
2 Likes
I doubt the pinning stuff needs consensus changes, or that ln-symmetry itself needs consensus-level checks any more than ln-penalty. Policy changes as I’ve described in BOLT/BIP form should get us there. Consensus changes to BIP118 could make it easier though, if you f.e. commit to number of inputs or similar.
Meanwhile, back to making sure package relay becomes a thing…
1 Like
so it turns out size doesn’t really matter. everything is just increasingly more political. especially forks.
1 Like