Open Stack

On Thu, Mar 12, 2020 at 15:45, Sylvain Bauza <sbauza at redhat.com> wrote:
> 
> 
> On Wed, Mar 11, 2020 at 7:11 PM Sean Mooney <smooney at redhat.com> 
> wrote:
>> On Wed, 2020-03-11 at 17:36 +0000, Sean Mooney wrote:
>>  > On Wed, 2020-03-11 at 18:17 +0100, Balázs Gibizer wrote:
>>  > > Hi,
>>  > >
>>  > > I've just got the news from my employer that due to COVID19 I 
>> cannot
>>  > > travel to Vancouver in June.
>> 

Since then the in-person PTG in Vancuver has been canceled [1]. So we 
have to come up with a virtual PTG plan that works for most of the 
people and allows us to discuss nova and cross project related 
questions regarding the Victoria cycle.

>> >
>>  > From a redhat perspective i dont think a decision has been made 
>> on if we should attend or not.
>>  ill clarify that slightly in that we do have guidence that "Red 
>> Hatters may not travel to attend external events or
>>  conferences with 1000+ attendees, even within their home country."
>> 
>>  in the past when the ptg and summit were combinined and we had the 
>> devsumit have ment that travel to the openstack even
>>  would not be allowed. At its current size its kind of in a gray 
>> zone where its is not banned as a public event but if it
>>  was an internal event the number of redhat employee that would be 
>> attending woudl be over the limit we have and the
>>  physical event would be canceled and converted to a virtual only 
>> event.
>> 
>>  so its tbd if i will be attending too although i have not heard a 
>> definitive No at this point but i also cant really
>>  book tickets and flight yet either however the guidance we have 
>> been given is to try and default to virtual attendance
>>  were possible.
>> 
>>  > last time i spoke to my manager we were still waiting to see how 
>> thing progress with the assumtion we would attend
>>  > but we might want to plan for a virtual PTG (via video 
>> conference, etherpad and email) in the event many cant travel
>>  > or
>>  > that things escalate to a point where the PTG event could be 
>> canceled.
>>  >
>>  > i dont think the foundation has indicated that that is likely to 
>> happen but im sure they are monitoring
>>  > things closely as our employers will be so having a plan b might 
>> now be a bad thing in either case.
>>  > if there isnt a diverse quoram physically at the ptg it would 
>> limit our ability to make desisions as happend to
>>  > some extent in china. it would be still good to get operator 
>> feedback but they may also be under similar travel
>>  > restrictions.
>>  > >
> 
> 
> I personnally have concerns with virtual events that are large like 
> our PTG :
>  - which timezone should we use for the virtual PTG ? Anyway, Asia, 
> Europe or North America folks could have problems with one.
>  - some folks (like me) have bandwidth issues in their (home) 
> offices. They could be having problems when trying to discuss
>  - how can we be sure that all of us can discuss by a virtual meeting 
> ? Not saying it won't work, but discussions could take more time.

I agree that his is a problem. I think we cannot make a 3 days, 8 hours 
long video meeting due to timezone, bandwidth and other restrictions. 
Here is what I propose:

Until two weeks before the PTG:
* Collect discussion topics on the etherpad [2] as usual
* you also free to start reflecting on and discussing such topics that 
are already on the etherpad.

Two weeks before the PTG (May 25 - May 29)
* I will scan the etherpad and for each open topics I will start a mail 
thread to have focused discussion.
* If the discussion is not finished on a given topic within this week 
then I will ask the proposer of the topic to organize a real-time 
meeting (chat, voice, video) either for the week before the PTG (Jun 1 
- June 5) or for the week of the PTG (Jun 8 - Jun 12). I think based on 
the active participants of the mail (or etherpad) discussion the 
proposer will know who to involve to the real-time discussion. So not 
every nova devs needs to sit through every video meeting in obscure 
local times.

I'm proposing this longish three weeks schedule to
i) allow enough time for an ML discussion to conclude if the topic is 
straight forward
ii) make it possible at the end of the first week to start organizing 
real time discussions and still have two weeks for those discussion to 
be scheduled. It might take time to find and agree on a common time 
slot and there will be popular time slots. So in both case more 
available days can help.

What do you think?

> 
> Anyway, I'm sad for you, gibi. I'm not -2 about a virtual PTG, I just 
> want to make sure we think about the above concerns before agreeing 
> with any virtual PTG.

I agree. We need to figure out what could work.

Cheers,
gibi

[1] https://www.openstack.org/events/opendev-ptg-2020/
[2] https://etherpad.openstack.org/p/nova-victoria-ptg