On 20. Dec 2024, at 05:14, Brian E Carpenter <brian.e.carpen...@gmail.com> wrote: > >> Agreed. In fact, that's what HFDU should mean: this error requires >> us to update the RFC. >> This is not how I interpret HDFU. Here's the IESG statement on this topic >> "*Hold for Document Update* - The erratum is not a necessary update to the >> RFC. However, any future update of the document might consider it and >> determine whether it merits including in an update." >> In other words, HDFU errata could still ultimately be determined to be right >> or wrong; HDFU just defers them. > > Right. And my "should" above is what I would prefer; it should be a call for > action. If it's just a shrug of the shoulders, it seems like wasted joules, > and calling it "No action" would be more honest.
I think this categorization misses that we sometimes find defects that * are serious, i.e., require an update * require design work to fix them, so they need to be done in a WG * are, at the time of noting down the erratum, being handled by informal agreements outside the RFC errata reporting process These are currently lumped under HFDU and don’t fit the above IESG description. (Also, when the problem has been fixed by additional design work and publication of another RFC not completely obsoleting the old one, we don’t seem to have a process to note the old HFDU having been covered.) If you need an example: RFC 9682 fixes HFDU defects in RFC 8610. Example HFDU reports: https://www.rfc-editor.org/errata/eid6278 https://www.rfc-editor.org/errata/eid6527 These carry the "Note: This RFC has been updated by RFC 9682” … good!, but do not outright say that this errata report is now fully obsoleted by this update. Grüße, Carsten -- rswg mailing list -- rswg@rfc-editor.org To unsubscribe send an email to rswg-le...@rfc-editor.org
