Open Stack

On 21/04/14 19:20, Akihiro Motoki wrote:
> Hi,
> 
> Previously Neutron team had ReviewDays [1] and core members made themselves
> focus on reviews and be available on IRC channel one or more day(s) of each week
> (as possible as they can).
> 
> The number of neutron reviews has increased compared to that time and
> I am afraid one or two days reviews cannot deal with neutron reviews, but
> the similar mechanism might work to make things better.
> 
> [1] https://wiki.openstack.org/wiki/Neutron/ReviewDays

Akihiro thanks very much for sharing this link. I searched through
openstack-dev for a discussion but missed the wiki :/

This is great - it makes more sense since as far as I can see the
community is really distributed and so it is difficult for all cores to
have consensus on a suitable time. As long as there is a minimum of 2
cores and any number of interested non-core (so that patches can
actually get pushed where possible).

The downside is losing the '2 meeting slots' advantage though I guess
that really is a separate issue (weekly meeting time slot convenience?)
that folks may i) agree/disagree is valid and ii) can be discussed as such.

thanks! marios

> 
> 
> On Tue, Apr 22, 2014 at 12:54 AM, marios at redhat.com <mandreou at redhat.com> wrote:
>> On 21/04/14 18:29, Kyle Mestery wrote:
>>> On Mon, Apr 21, 2014 at 9:38 AM, marios at redhat.com <mandreou at redhat.com> wrote:
>>>> Hi,
>>>>
>>>> I think both PTL candidates mentioned process improvements wrt
>>>> contributions and reviews in their candidacy announcements. As a new
>>>> Neutron dev I have seen that it is easy for reviews to go unnoticed,
>>>> especially when they are stand-alone bug fixes that aren't part of a
>>>> particular blueprint group (and so aren't discussed/highlighted at the
>>>> various sub-team meetings). Of course this is also compounded by a
>>>> seemingly heavy backlog of reviews. I realise that all core/non-core
>>>> devs are doing as much as they can and though more cores will help, it
>>>> takes a long time to develop people into this role.
>>>>
>>>> I was wondering if a 'review hour' would help. This is something Lucas
>>>> Gomez told me about; the Ironic core team has a weekly hour slot in
>>>> which they discuss x number of reviews and approve or -1 them. Besides
>>>> getting reviews cleared quicker, it also opens the process up. It will
>>>> allow new contributors to (more quickly) learn about the kinds of things
>>>> core reviewers look for in a patch and also give real-time feedback
>>>> (e.g. could just use #openstack-neutron for discussion during the hour).
>>>> I feel that this could have an impact on the review backlog even if this
>>>> only tackling the oldest 5 reviews for example.
>>>>
>>>> Do any of the core devs think this would be a good thing, and do you
>>>> think you have the time for it?
>>>>
>>> This is an interesting idea Marios, thanks for proposing it! Are you
>>> saying we should have a "Review Hour" on IRC, where we walk through XX
>>> number of reviews as a team? That's an interesting idea actually, and
>>> I'd be in favor of something like this. We could rotate timeslots so
>>> as to get everyone on the team (both core and non-core) a chance to
>>> participate.
>>>
>>> Can you attend our weekly meeting today [1] where we can discuss this as a team?
>>
>> Thanks very much for responding Kyle, I was worried about my message
>> sounding 'complainy' - I will try my best to attend the meeting today,
>> though it is at midnight here (CET +1hrs) so I typically only get to
>> catch up on the logs.
>>
>> Depending on whether others think setting up the "irc review hour" is a
>> good idea, one side effect would be that we then have a second 'neutron
>> meeting' slot during the week (even if this is only for reviews). If we
>> pick this time carefully we could even alternate between 'weekly
>> meeting' and 'review meeting' to make it easier for folks in Europe to
>> join the weekly meeting (and make it less harsh for people in Asia
>> Pacific who have to get up very early for the current slot [1]). Though
>> this is of course just speculation and I'm really getting ahead of
>> myself (I was going to include this last thought in my original email
>> but it was already long enough)
>>
>> thanks, marios
>>
>>
>> [1] [1]
>> http://www.timeanddate.com/worldclock/meetingtime.html?iso=20140421&p1=137&p2=179&p3=136&p4=37&p5=166&p6=248&p7=196&p8=47
>>
>>>
>>> Thanks!
>>> Kyle
>>>
>>> [1] https://wiki.openstack.org/wiki/Network/Meetings
>>>
>>>> thanks, marios
>>>>
>>>> _______________________________________________
>>>> OpenStack-dev mailing list
>>>> OpenStack-dev at lists.openstack.org
>>>> http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack-dev
>>>
>>> _______________________________________________
>>> OpenStack-dev mailing list
>>> OpenStack-dev at lists.openstack.org
>>> http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack-dev
>>>
>>
>>
>> _______________________________________________
>> OpenStack-dev mailing list
>> OpenStack-dev at lists.openstack.org
>> http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack-dev
> 
> _______________________________________________
> OpenStack-dev mailing list
> OpenStack-dev at lists.openstack.org
> http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack-dev
>