I feel like I don't have the requisite understanding of all the tradeoffs with various improvement proposals for RBF and CPFP, but given the importance of the mempool to L2s and the importance of L2s succeeding for Bitcoin to scale, turning a lot of attention to improving it seems well worth it.

There appears to be three main considerations in proposals: DoS avoidance vs UX vs miner incentives.

Fee accounts seem kind of ugly (stateful) relative to my understanding of RBF, but ugly problems often require ugly solutions.