The proposal is based on a flawed approach. I demonstrated this general issue per an example. The author responded by addressing the specific mechanism used in the example instead of addressing the general flaw. I concluded that it is not a constructive use of my time to engage further.
The absence of an attack does not prove security, especially in presence of obvious simpler explanations.
Murch, you linked to your infinite cycle attack that would use infinite resources. Are you saying there is a general flaw with replace-by-fee-rate that has infinite cycle attacks? Peter’s proof that Libre Relay’s implementation does not seemed solid.
No-one can prove SHA256 isn’t broken either. Bitcoin continuing to exist is still evidence it’s ok.
