lightning-dev
Lightning, the death of BIP62, and Segregated Witness
Posted on: November 19, 2015 02:53 UTC
The author of a lightning variant that used only non-experimental, in-planning BIPs realized that anchor malleability would be an issue and that BIP62, which was assumed to protect against it, would never be deployed.
The options were either to ignore it as non-standard, add a timeout to the anchor, propose a reduced BIP62, or assume that Segregated Witness fixes all malleability problems. The author eventually settled on the fourth option, assuming that SW will fix all malleability issues. This is because Pieter Wuille is working on a prototype, and ignoring malleability will simplify other parts of the lightning code. It is also consistent with what is best for Bitcoin and smart contract systems. Although this result has no effect on lightning in the short term, if malleability support needs to be added later, handling it correctly will be difficult.