Author: Pieter Wuille 2020-10-16 21:09:04
Published on: 2020-10-16T21:09:04+00:00
Pieter is seeking information on the support for sending to witness v1 BIP173 addresses. Rusty has provided two options for deciding on future segwit versions, one of which is backwards compatible and the other restricts future versions and may require new encoding. Rusty believes that option 2 is preferable due to fewer places generating addresses than accepting them. Pieter suggests using the backwards compatible proposal from BIPs PR#945 and consensus restricting v1 witness program size. Both Pieter and Rusty agree that deferring a hard decision is not useful, as it will only get harder. They discuss the possibility of adoption of new technology based on certain infrastructure and codebases going out of business or being replaced with newer ones. Pieter explains that new witness version/length combinations are rarely needed and are required for changes to the commitment structure of script execution, upgrades to new signing cryptography, and changes to signatures outside of a commitment structure. They discuss the simplicity of the change and the presence of production codebases that just don't update at all may complicate this. Finally, they express their reluctance to commit to restrictions on future addresses without knowing enough and believing that there is no need for it.
Updated on: 2023-05-20T23:57:35.500925+00:00