--On Tuesday, November 28, 2023 08:44 +1300 Brian E Carpenter
<brian.e.carpenter@xxxxxxxxx> wrote:
> John,
>
> Thanks for the detailed analysis. Cutting to the chase:
>
>> Personally, I think we are better off with the flexibility of
>> "Do the Right Thing" over getting ourselves more and more
>> tangled in increasingly specific rules and procedures.
>
> Violent agreement there.
>
>> ...But, if
>> we need such procedures, let's not pretend they are there when
>> they aren't.
>
> Sure, but at the moment I'm not sure that we have a gap to
> fill.
In case it was not clear, I'm not either. Setting aside the
other streams for the moment and looking at the IETF one -- not
because I think there is any ambiguity but because there is far
more experience with it-- I think the rules are quite clear. As
a quick summary:
(1) IANA "can" do whatever the IETF tells them to do in an RFC
and, subject to (3) below, nothing else. If IANA does not
think what the IESG is telling them is reasonable, they have the
right (perhaps the obligation) to take the issue up with the IAB.
(2) RFC 2860 very explicitly permits ICANN to tell IANA to take
on work for "other organizations". Presumably, what IANA could
or could not do under those arrangements would be up to ICANN
and those other bodies: speculating about it would be a waste of
time. I don't know, but assume, that the PTI arrangements did
not change that very much.
(3) There is an open question about how much discretion IANA can
apply -- either with regard to specific registries or to general
procedures -- if the instructions from the IETF are not
specific enough. The 1995 version (and, for that matter, the
1986 version) of the answer to that question was "whatever IANA
likes and IANA gets to reject instructions from the IETF that
seem too specific or otherwise wrong". The 2000 (vintage RFC
2860) version was, in practice, that IANA should do what they
thought best, asking questions only if they thought it necessary
and with preference given to assuming that anything the IETF did
not specify was left up to IANA's discretion... and that, if the
IESG didn't like it, they and IANA could take the disagreement
to the IAB. AFAICT, the 2023 version is that, if IANA is not
absolutely certain of what is wanted, they ask the IETF to
clarify.
Given that, you don't see a gap to fill and I don't understand
what question is being asked and why. I suspect that,
ultimately, we are saying the same thing from slightly different
perspectives and consequent choices of vocabulary.
john
