Re: IETF 109 Preliminary Agenda

[Lou Berger <[e-mail]>]

----------
On October 18, 2020 10:30:17 AM Martin Duke <[e-mail]> wrote:

> All of the virtual meetings have begun at 12:00 or 13:00 "local" time. No one seriously questioned this choice until it became inconvenient for Eastern Time.
>

Sorry, I think we've seen grumbling at each meeting. For examples I heard for many in California at the last meeting. It is true that the is the first time I am commenting on it.

> To start messing with the start time now would erode the principle of 1-1-1.
>

I have no interest in eroding that. but I think we could do something a little more intelligent for virtual ietfs going forward. For example on the virtuals that I scheduled, we call among the contributors and find a time that work best for them. Clearly polling is not a great way to do it for a whole meeting but, for example, we could be scheduling based on a majority of attendees rather than on what might be a hypothetical location in the future.  For example 110 could start based on what is a normal start time for the  weighted average of eu attendees for the last year.  I suspect that this will only make a small difference given the distribution of time zones in each region is only about 3-4 hours, but still seems less arbitrary and more reflective of today's reality to me...

Lou

> On Sun, Oct 18, 2020, 06:48 Lou Berger <[e-mail]<mailto:[e-mail]>> wrote:
>
> Andy,
>
> I'm fine for sharing the pain -- but let's not pretend this is fully scheduled based on Bangkok time.  I don't object *at all* to holding the meeting based on the typical meeting times at the planned location.  If we were doing so this for this meeting, we'd be starting 2-3 hours earlier -- which would more fairly spread the pain IMO.  (Full disclosure, you and I are catching the worst of it this time and starting at the normal Bangkok time would have made a material difference to me. )  Of course, I do freely admit that I really have nothing to complain about given what our Asian colleagues are subjected to week in and week out with meeting times generally biased to work for NA and EU.  I just hope that future meetings really do start at the normal local meeting time -- or, even better, we figure out a less apparently random+some-biased way to "share the pain".
>
> There's time enough to consider options before IETF110...
>
> Lou
>
> On 10/18/2020 9:16 AM, Andrew G. Malis wrote:
> David,
>
> Speaking as another person in the upcoming midnight-6 AM zone ....
>
> Think about how the folks in Asia feel when we schedule a virtual meeting for our daylight hours. It's convenient for us, but not for them. We do 1+1+1 for a reason - to spread the pain. It's only fair. And to your other point, the survey results were very clear that people prefer a shorter day (6 hours) for the online IETF meetings.
>
> You talked about turning your life upside down to attend the meeting - if the meeting had actually been in Bangkok, then you would have had two 20+ hour bouts of travel going and coming, and a week away from home. At least this way, you can still sleep in your own bed. If you don't want to become nocturnal for the entire week, then pick and choose which sessions are most important to you and set your alarm.
>
> Cheers,
> Andy
>
>
> On Sun, Oct 18, 2020 at 6:34 AM David Noveck <[e-mail]<mailto:[e-mail]>> wrote:
>
>
> On Sat, Oct 17, 2020, 3:36 PM Benjamin Kaduk <[e-mail]<mailto:[e-mail]>> wrote:
> On Sat, Oct 17, 2020 at 07:33:09AM -0400, David Noveck wrote:
>> On Fri, Oct 16, 2020, 8:34 PM Carsten Bormann <[e-mail]<mailto:[e-mail]>> wrote:
>>
>> > On 2020-10-17, at 02:23, Michael Richardson <[e-mail]<mailto:[e-mail]>> wrote:
>> > >
>> > > It appears that session are scheduled starting at 05:00 UTC, which would
>> > be
>> > > 12:00(noon) Bangkok time.   It's really as if this meeting was being
>> > held in Madrid.
>> >
>> > I don’t think you are aware when Spanish people tend to get up :-)
>> >
>>
>> Fair enough but I don't know when people in Bangkok typically get up and
>> I'm not sure it is really relevant.  Typically on-site meetings start at
>> 10am local time and go on to 5:30, creating a 7.5-hour window for meeting
>> sessions.
>>
>> Given that this is a virtual meeting,  it would make sense to widen those
>> windows to allow some amelioration of the problems that this creates for
>> some time zones, in this case, for those in the Americas.  Instead, the
>> window has been shortened to six hours and I'm not sure why.
>
> Could you expound a bit on why you think widening the block of time in
> which sessions occur would be helpful?
>
> First of all, given that the current block (midnight to 6AM) was unworkable for me, I was looking to provide options that would allow us to schedule an nfsv4wg meeting during IETF.
>
> It never occurred to me that there were people who would attend the entire 12Am-6Am meeting and had no expectation of people attending an expanded version of that.  It now appears that there are such people but I expect there will be very few.
>
> IIRC we have survey results showing
> that in a virtual format even the 7.5-hour window is too long to be
> practical, and having a longer window is going to push more of the session
> timeslots into times that are quite painful for more participant timezones.
> My understanding was that the idea of having a consolidated virtual IETF
> was to enable the cross-polination that occurs at in-person IETF meetings
> where most participants are attending sessions in most of the timeslots.
>
> If that is the goal, it does not appear to be realizable for those in the Americas given the current ICT session times, except for a small group willing to turn their lives upside down to attend the meeting. Although these times were not chosen to make this difficult, if you were trying to do exclude us, it would be hard to pick a more forbidding time slot.
>
>
> Thanks,
>
> Ben