--On Friday, February 2, 2018 15:54 -0500 Andrew Sullivan
<ajs@xxxxxxxxxxxxxxxxxx> wrote:
> On Fri, Feb 02, 2018 at 02:07:26PM -0500, John C Klensin wrote:
>>
>> _That_ problem, which would be more than sufficient to justify
>> an appeal, is that, normally, when someone says "this draft is
>> ok because of a provision in some other draft"
> I think what I was arguing is that the topic you are asking
> about just isn't in
> draft-ietf-mtgvenue-iaoc-venue-selection-process. The policy
> is not the process, in the good tradition of the IETF, and
> therefore if you want the policy you need to look somewhere
> else. That somewhere else is
> draft-ietf-mtgvenue-meeting-policy, and I don't think it's
> been developed in secret so I don't think it's fair to suggest
> that's being snuck past anyone.
Andrew, I would be in complete agreement with your observations
above except that, in some cases, including this one, I don't
see the policy-process distinction/boundary quite as clearly as
you, and I presume others more deeply involved with the
documents do.
>> However, those "open network" criteria have been around for
>> years and years and, while the MTGVENUE drafts (IMO, quite
>> properly) repeat and document those criteria, I don't see that
>> they change anything in what we've been told is being
>> practice.
>
> Right. All I was saying is that the Bangkok selection is a
> great opportunity to see whether it matches the (currently in
> draft) output of the MTGVENUE work. If people think there is
> a problem in this case, but the selection matches the
> selection process and policies and the mandatory criteria,
> then that tells us the output of MTGVENUE is still not right
> and we need to do more work on it. _My_ no-hat and no special
> knowledge sense is that Bangkok indeed meets the criteria and
> will be a good place for a meeting.
Agreed and, based on personal experience, I think Bangkok would
be great place to hold an IETF meeting assuming the purely
quantitative network and bandwidth requirements are met (I
assume they are or we wouldn't be here). I can't be quite as
sure about the on-the-ground issues as some others just because
I haven't been there since the intermittent political unrest in
recent years, but have seen no evidence that those issues, even
should they arise during an meeting, would pos any but the most
minor inconvenience to IETF meeting participants.
It seems to me that I started just with a question (not
disagreement with the idea of going to Bangkok) and a suggestion
that the question might help inform review of
draft-ietf-mtgvenue-iaoc-venue-selection-process. It seems to
me that what you are saying is consistent with that but suggests
looking at draft-ietf-mtgvenue-meeting-policy as well...
something with which I agree.
_However_:
(1) It seems to me that taking a process document, one that is
presumably supposed to reflect a set of policies, putting it
through Last Call, and presumably approving it before the
relevant policies have been approved is, at best odd. It would
be reasonable only if one could argue that the document is
policy-independent and would work equally well with any
conceivably set of policies. I do not believe that current
process document meets that test. Indeed, its Section 2 and at
least part of Section 3 appear to me to be either policy
considerations or direct consequences of policy considerations
while Sections 4 and 5 are clearly about process.
(2) If someone believed that either some of the topics conversed
in the process document were actually policy matters or vice
versa, or someone believed either that the process being
suggested was inconsistent with the current draft of the policy
document or that language in the policy document (current or
future drafts) required spelling out of procedural steps that
are not in the process document, separating the two documents on
Last Call would effectively disenfranchise the people and
probably suppress toe point of view because the approved
process could be used to argue for constraints on the policy
document.
To me, either of the above would argue for putting the two
documents through Last Call together or at least tabling the
Last Call results on this document and then opening a quick
review of it for policy issues only when
draft-ietf-mtgvenue-meeting-policy is put through Last Call. If
one believes, as I now do, that the split between the two
documents is not actually policy versus process but "a
discussion of the 1-1-1* rule and its implementation" and "other
policies and associated process", that argument gets even
stronger.
Finally, returning to the question that started this thread, I
agree with you that "near IETF participants" is, at least
mostly, a policy matter. From my perspective, there have
always been two almost-separate criteria. One is about
continental (ore or less) unilt, expressed over time as 2-1-1,
1-1-1, and now 1-1-1*. I think the claim that Adelaide was an
"exploratory meeting" or "exceptional instance" is revisionist
history. At the time that meeting was held, I am fairly sure
that the assumption was this it was just an Asia-Pacific meeting
that we had decided, because of the concentration of IETF
participants in the area, to take well south in that region.
The second criterion is "near to active IETF participants". In
general, this has been a selection criterion within
continental-type regions. In North America, we don't generally
meet in Bozeman, Whitehorse, Tuscaloosa, Tulsa, or Calgary and I
don't think issues about finding adequate hotel and conference
space are the only, or even the more important, issues. We
have gone to Orlando --near which there have not bee a lot of
active IETF participants -- more than once and, at least IMO,
regretted it each time (although not equally).
That criterion appears to have been silently dropped from the
policy document. I didn't detect it until I looked in the
process document for information about participants near
Bangkok. From one point of view, that reinforces me comments
about how these documents interact. From another, if we are
dropping the locality rule/principle (which, IIR, predates any
discussion about 1-1-1), then probably
draft-ietf-mtgvenue-meeting-policy should devote a sentence to
saying that. Along the same lines, I see nothing in that
policy document that allows either the IAOC or one of its
committees to declare Vancouver part of Asia. If it is, could
such a group decide that post-Brexit London is part of North
America without violating the policy? And if the "nearby
participants" rule is st9ll a preference, is the "familiar with
both the locale and the IETF" statement in Section 3.3 of the
process document supposed to encode it, or would more specific
language about criteria for familiarity and/or who decides?
As an example, I've been in Bangkok several times starting in
the mid-1980s, always primarily for business/professional
reasons, but never to a meeting about the Internet or that put
heavy demands of host or h0tel communications facilities. I
also have not been there is the lat half-dozen years. I
certainly do not speak the language nor do I claim expertise on
the local culture or politics. Does that qualify me as
"familiar" or "not familiar"? The question is rhetorical but,
if this is a process document that needs to stand along because
the policy document doesn't say anyone beyond a continental
breakdown, I suggest it doesn't say enough.
best,
john