Bitcoin developers propose abandoning Replace-by-Fee: a signal that has become an anachronism
A discussion has begun in the Bitcoin developer community regarding a proposal to remove the Replace-by-Fee (RBF) feature from wallets. This mechanism, implemented through BIP-125, has long allowed users to increase the fee for a transaction that has already been sent but not yet confirmed, which is still in the mempool. However, in my opinion, its usefulness in modern conditions raises serious doubts.
The key argument for abandoning RBF is the network's transition to full-RBF. Previously, the feature was optional: not all nodes supported transaction replacement. Now that full replacement is becoming the standard, the RBF signal itself in transactions loses its meaning. It is no longer needed to indicate that a transaction can be replaced, as this is implied by default.
Moreover, the outdated signal creates a privacy issue. The presence of an RBF flag in a transaction allows third-party observers to determine with high accuracy which specific wallet or client the sender used. This is an undesirable information leak that violates the anonymity principles embedded in Bitcoin.
Why is this important right now?
With the growing popularity of second-layer protocols such as the Lightning Network and the increased efficiency of the mempool, the role of RBF as a tool for "accelerating" transactions is diminishing. In the current paradigm, when the network is congested less frequently and users increasingly resort to alternative methods of fee management, support for BIP-125 becomes an unnecessary burden both for the code and for privacy.
My analytical assessment: Removing RBF is a logical and timely step. It will simplify the protocol and eliminate a vector for deanonymization. However, the community must ensure that the transition does not disrupt the operation of older services that still rely on this mechanism. In any case, the evolution of Bitcoin does not tolerate compromises with outdated features.