[sakai2-tcc] [cle-release-team] retiring the kernel team

[Aaron Zeckoski]

I think the idea here is that the kernel team would not be a thing
anymore. There would still be kernel committers and a kernel list but
the team itself with the official policies around it would go away.

To remake the point I make in ATL, the vast majority of kernel tickets
are only reviewed by 2 people (and some things are still going in
without any review - not as much now but it was an issue earlier in
the year). If people are not going to be active then I see no need to
have an official team in place with 2 people on it.
We can leave the team in place as it is, but it's not effective in
it's current state IMO.

-AZ


On Mon, Jun 18, 2012 at 10:16 AM, Matthew Jones <matthew at longsight.com> wrote:
> Hi Aaron,
>
> I believe that the 'kernel team' is a useful entity. A team to review code
> changes, discuss architectural issues and keep some sense of actually
> committing code to the the core is a good idea.
>
> My problems were:
>
> 1) In JIRA, you can only assign KNL issues to the kernel team. Even though
> many kernel issues may not be actually worked on by the kernel team, but by
> someone else and then submitted to the kernel team for review (either branch
> or patch). If I can't assign an issue to myself and have it appear on my
> dashboard I'll generally forget about it.
>   1a) Of the people listed in that drop-down for assignees, about half of
> them are very unlikely to resolve issues in the kernel in the short term
> anyway; Ray Davis, Ian Boston, Jim Eng, Seth Theriault, Zach Thomas, Steve
> Githens,
>
> So I'd either split the jira assign list from the actual "kernel team" (like
> have the assign list be the same as the SAK list) OR change the kernel team
> completely to just be the most active people in the CLE community (based on
> weekly commits and/or Thursday calls).
>
> As far as Crucible is concerned the plugin is in Jira, but it required a
> separate process to be installed on source.sakaiproject.org server. I can
> work with Chris Maurer and get that connected. It looks like it has a 30 day
> trial license you can generate, and afterward we'd need Anthony to use that
> to supply an open source license. I'll set aside some time and look at
> fisheye/crucible this week. I think the only holdup before was that I'd
> forgotten my password and/or got busy with something else.
>
> After I get it setup with the evaluation license I'll send this to Anthony.
>
> On Fri, Jun 15, 2012 at 10:01 AM, Aaron Zeckoski <azeckoski at unicon.net>
> wrote:
>>
>> It was discussed/proposed that the kernel team as an entity has
>> outlived it's usefulness and should be disbanded. Instead, there will
>> simply be a small (initially) set of kernel committers and all commits
>> going into the kernel will be required to go through a review
>> application (like atlassian crucible).
>>
>> The consensus was pretty strong at the conference but I wanted to give
>> folks a chance to comment on list. We would need to talk to our
>> atlassian hosting company to get crucible setup but I think Anthony
>> Whyte could probably arrange that.
>>
>> This isn't really a vote (yet).
>> -AZ
>>
>>
>> --
>> Aaron Zeckoski - Software Architect - http://tinyurl.com/azprofile
>> _______________________________________________
>> cle-release-team mailing list
>> cle-release-team at collab.sakaiproject.org
>> http://collab.sakaiproject.org/mailman/listinfo/cle-release-team
>
>
>
> _______________________________________________
> cle-release-team mailing list
> cle-release-team at collab.sakaiproject.org
> http://collab.sakaiproject.org/mailman/listinfo/cle-release-team
>



-- 
Aaron Zeckoski - Software Architect - http://tinyurl.com/azprofile