Author: Jeremy Rubin 2022-02-15 21:38:11
Published on: 2022-02-15T21:38:11+00:00
In a recent discussion on the bitcoin-dev mailing list, Suhas raised an issue with a variant of sponsors attempting to address a second criticism. The distinction between this issue and sponsors is subtle and can be explained by defining different properties of reorgability. These include strong reorgability, simple existential reorgability, epsilon-strong reorgability, epsilon-simple existential reorgability, and perfect reorgability. However, perfect reorgability does not exist in Bitcoin due to unconfirmed transactions that can be double-spent, which invalidates descendants. The sponsors proposal aims to change from epsilon-strong reorgability to epsilon-weak reorgability, which seems to be a sufficient property. There is no clear reason to rely on strongness when Bitcoin's reorgability is already imperfect, and it is susceptible to malicious reorg generators. In response to Suhas' concern about valid transactions becoming invalid later on, Jeremy Rubin argued that there is no practical downside to transactions becoming invalid as long as people wait for standard finalization. The downside is only apparent for people who do not wait for standard finalization. Russell O'Connor pointed out that in a 6 block reorg, any transaction that has passed its expiration date becomes invalid along with its descendants. The current consensus threshold for transactions to become invalid is a 100 block reorg, and O'Connor sees no reason to change this threshold.
Updated on: 2023-06-15T16:30:50.539701+00:00