Re: [Mtgvenue] Last Call: <draft-ietf-mtgvenue-meeting-policy-04.txt> (High level guidance for the meeting policy of the IETF) to Best Current Practice

[Eric Rescorla <[e-mail]>]

On Sat, Apr 21, 2018 at 10:12 AM, Lou Berger <[e-mail]> wrote:

> While it is true that NA makes up a the biggest group overall, I think
> that it is worth noting that when we meet in Asia we have about 40%
> participation from Asia, and when we meet in Europe we have about 40% from
> Europe.
>
Hi Lou,

The data has evolved over time, so it's a bit hard to interpret, but I don't
think this is necessarily the best characterization of the situation. As you
said earlier, there have been 6 Asian meetings in the period IETF 72-IETF
102,
with steadily falling Asian attendance fractions:

3  Hiroshima 0.4995997
6    Beijing 0.4805389
9     Taipei 0.4005709
21  Yokohama 0.3866846
24     Seoul 0.3286418
27 Singapore 0.2966625

In the same period, we have had 10 European meetings, with more or less flat
European attendance:

         City   EU.Frac
2   Stockholm 0.3577236
5  Maastricht 0.3561013
7      Prague 0.3647416
10      Paris 0.3747547
14     Berlin 0.3842271
16     London 0.3885350
20     Prague 0.3627010
23     Berlin 0.3748686
26     Prague 0.3491124
28     London 0.3787654

In case people are curious, here's the data for all meetings and all
geographies:

   Meeting.Number          City Cont Asia.Frac   EU.Frac   NA.Frac
1              74 San Francisco   NA 0.1863261 0.2223892 0.5514651
2              75     Stockholm   EU 0.2357724 0.3577236 0.3617886
3              76     Hiroshima   AS 0.4995997 0.1905524 0.2818255
4              77       Anaheim   NA 0.2251852 0.1977778 0.5392593
5              78    Maastricht   EU 0.2709133 0.3561013 0.3376823
6              79       Beijing   AS 0.4805389 0.1826347 0.2979042
7              80        Prague   EU 0.2180851 0.3647416 0.3837386
8              81   Quebec City   NA 0.2419614 0.2114148 0.5072347
9              82        Taipei   AS 0.4005709 0.2007612 0.3720266
10             83         Paris   EU 0.2060170 0.3747547 0.3754088
11             84     Vancouver   NA 0.2330383 0.2028024 0.5044248
12             85       Atlanta   NA 0.2087343 0.1917098 0.5396003
13             86       Orlando   NA 0.1937858 0.2035977 0.5568275
14             87        Berlin   EU 0.1873817 0.3842271 0.3564669
15             88     Vancouver   NA 0.1782832 0.2017608 0.5436537
16             89        London   EU 0.1649682 0.3885350 0.3936306
17             90       Toronto   NA 0.1810402 0.2073733 0.5227123
18             91        Hawaii   NA 0.2098854 0.2098854 0.5150430
19             92        Dallas   NA 0.1787440 0.1994479 0.5383023
20             93        Prague   EU 0.1729904 0.3627010 0.4032154
21             94      Yokohama   AS 0.3866846 0.1997310 0.3443174
22             95  Buenos Aires   SA 0.2483553 0.1973684 0.3563596
23             96        Berlin   EU 0.1966351 0.3748686 0.3449001
24             97         Seoul   AS 0.3286418 0.2174525 0.3631246
25             98       Chicago   NA 0.1993846 0.2123077 0.5027692
26             99        Prague   EU 0.1791286 0.3491124 0.3717052
27            100     Singapore   AS 0.2966625 0.2626700 0.3751545
28            101        London   EU 0.1748148 0.3787654 0.3570370

-Ekr






Lou
> ------------------------------
>
> On April 21, 2018 11:57:50 AM Cullen Jennings <[e-mail]> wrote:
>
>>
>> When we went to a 1-1-1 policy, we agreed the locations should roughly
>> match participants location to be fair. The argument was made that if we
>> had more meetings outside north america, the participation would come to
>> match approximately 1/3 , 1/3 , 1/3. That has clearly not happened so I
>> think we need set the rotation of where we do meeting to something we agree
>> is fair to our participants.
>>
>> I do not support the 1-1-1 policy as it is based on a false premise that
>> this mirrors our participation.
>>
>>
>> On Apr 19, 2018, at 5:36 PM, Eric Rescorla <[e-mail]> wrote:
>>
>> Andrew,
>>
>> Thanks for raising this.
>>
>> While the stated rationale in S 2. is to spread travel pain around,
>> I don't really think this gets us to 1-1-1-*.
>>
>> First, if you look at historical demographics, over the past 12 IETFs,
>> we have 23% Asia, 26% Europe, and 42% NA). Put another way, the last
>> time we had > 1/3 Asian attendance was IETF 94 in Yokohama, and the
>> last time we had less than 1/3 NA attendance was IETF 79 in Beijing.
>> So, a policy that was designed to match per-continent attendance would
>> be more like 2-1-1-*.
>>
>> Second, continent is not a very good proxy for travel pain, both
>> because Asia is so large (for instance, the shortest Tokyo to
>> Singapore route is 7:25 out and 7:10 back (on JAL) and the shortest
>> Tokyo - Honolulu route (ANA) is 7:20/8:10, so not really much
>> different at all) and because flight connections are such a big
>> contributor ( for instance, SFO-BKK is almost 20 hours, whereas
>> SFO-NRT is 11).
>>
>> Bottom line, if this is supposed to be real requirements rather
>> than just aspirations, I think it needs a rethink.
>>
>> -Ekr
>>
>>
>> On Wed, Apr 18, 2018 at 6:34 PM, Andrew Sullivan <[e-mail]>
>> wrote:
>>
>>> Dear IESG,
>>>
>>> On Thu, Apr 05, 2018 at 01:12:40PM -0700, The IESG wrote:
>>> >
>>> > The IESG has received a request from the Meeting Venue WG (mtgvenue) to
>>> > consider the following document: - 'High level guidance for the meeting
>>> > policy of the IETF'
>>> >   <draft-ietf-mtgvenue-meeting-policy-04.txt> as Best Current Practice
>>>
>>> In a recent discussion, the IAOC came to realise that the documents
>>> draft-ietf-mtgvenue-iaoc-venue-selection-process and
>>> draft-ietf-mtgvenue-meeting-policy may be in some tension.  One of
>>> them requires the IASA to balance meeting venues over time, and the
>>> other has requirements that a meeting must meet.
>>>
>>> One possible difficulty that arises from the combination is if one
>>> region turns out to be vastly more expensive than others.  In that
>>> case, some criteria for each venue may not be met in one region.  The
>>> result might also be financially ruinous for the IETF in general.
>>>
>>> The current IAOC interprets the drafts such that any of the criteria
>>> except those in section 3.1 of
>>> draft-ietf-mtgvenue-iaoc-venue-selection-process may be traded against
>>> any other, over several years if need be, in order to meet the
>>> geographic distribution policy described in
>>> draft-ietf-mtgvenue-meeting-policy.  Assuming the documents are
>>> published as they are currently written, we will use that
>>> interpretation as governing IASA implementation decisions.  It is
>>> worth noting that, among the criteria that could be traded are those
>>> of affordability.  If that is not the interpretation of the IETF
>>> community, then some clarification is needed to the text.
>>>
>>> Best regards,
>>>
>>> Andrew Sullivan
>>> for the IAOC
>>>
>>> --
>>> Andrew Sullivan
>>> [e-mail]
>>>
>>>
>> _______________________________________________
>> Mtgvenue mailing list
>> [e-mail]
>> https://www.ietf.org/mailman/listinfo/mtgvenue
>>
>>
>>