On 8/8/20 9:30 PM, Nico Williams wrote:
> Supposing a proper analysis demonstrates no current or recent usage,
> then what? Would proponent(s) drop the matter, or press on? If press
> on, is the reason fear that uninformed authors might make use of these
> terms and that uninformed reviewers, shepherds, ADs, and RPC staff might
> allow their use to go unchallenged? Would that be sufficient? Would
> such a fear be well-founded? Are there other motivations I might be
> missing?
I don't fully understand your argument, but I'll point out
that the list of usages in existing documents that Fred sent
out earlier can be used as the basis for answering questions
about trends, etc. The document covers published RFCs and, I
think, current internet drafts, and contains 2060 instances of
"master," "slave," "whitelist," and "blacklist" once you
strip out citations of rfc-index.xml and rfc-ref*, representing
358 documents (yes, I stripped out duplicate .txt/.xml files).
Many are current ("master secret" ftw) but not all are, with RFCs
ranging from RFC 6 to RFC 8806. Gratitude to Fred for
doing the work and sharing it.
I don't think the IETF is capable of this kind of change (and
question if it's capable of any kind of change, for that matter),
and I expect the most effective approach is to catch problematic
usages on an informal basis during the review process. What to do
about editors who refuse to change problematic language is left
as an exercise, etc., if there's no WG consensus or IETF
consensus. It would be awesome if the RSE kept an eye out for
things that slip through but I'm not sure that we would want to
formalize that.
Melinda
--
Melinda Shore
melinda.shore@xxxxxxxxx
Software longa, hardware brevis
