Author: Anthony Towns 2022-02-02 10:21:16
Published on: 2022-02-02T10:21:16+00:00
In a discussion on bitcoin-dev, participants considered whether or not descendants matter for miner incentives. One edge case was discussed in which replacing transaction A with transaction X would obsolete descendant transactions (B, C, D...) and replace them with unrelated transactions (L, M, N...). However, if the total feerate/fees of A+B+C+D... is still higher than X+L+M+N, this scenario may be unusual. The group also discussed policies to address miner incentives and DoS issues, including requiring the transaction's ancestor absolute fees to be a certain percentage higher than the previous transaction's ancestor fees, and the transaction's ancestor feerate to be a certain percentage higher than the previous transaction's ancestor feerate. It was noted that "Y% higher" rather than just "higher" is only useful for rate-limiting, not incentive compatibility.
Updated on: 2023-05-22T17:00:09.308403+00:00