Author: Rusty Russell 2017-07-30 05:36:32
Published on: 2017-07-30T05:36:32+00:00
Rusty Russell has provided some feedback on a productive discussion that he missed. Pierre suggested choosing HTLC (Hash Time-Locked Contracts) min high enough to avoid issues, and Laolu pointed out that it would limit the ability to send and claim small amounts on the chain as it would create a dust output. Laolu suggested that the trim output spec portion should state not to create a dust output on second-level TX, although Rusty believes it is already mentioned in the document. Laolu also recommended against using sighash-all on second-level HTLC transactions, which would allow coalescing many HTLC transactions into a single one to save on-chain footprint and add more fees, similar to Lighthouse by Hearn. Rusty thinks this approach could work but is more suitable for fee bumping, and there are two other cases where such an approach would be possible: commitment TX with no HTLC outputs and the closing transaction.
Updated on: 2023-05-24T02:03:45.635127+00:00