Open Stack

Hi stackers,

Seems likes after stackforge/rally -> openstack/rally Rally project started
being more attractive.
According recent stats we are on top 3 position (based on Patch sets stats)
http://stackalytics.com/?release=liberty&metric=patches&project_type=All
And if we compare half year ago we have 40 open reviews and now we have
about 140...
In other words we need to scale core reviewing process with keeping
quality.

<history_note>

I suggested in mailing thread:
[openstack-dev][all][infra][tc][ptl] Scaling up code review process (subdir
cores)
To create special rules & ACL groups to have fully automated system.

Instead of support I got raped by community.
Community was very polite & technical oriented in that thread and they
said:
1) I am bad PTL,
2) I don't know how to do open source
3) Rally project sux
4) Rally project community sux
5) Rally project has troubles
6) A lot of more constructive critics

So Instead of having NICE fully automated system for subcores we will use
ugly, not automated but very popular in community "trust" model based on
excel.....

</history_note>


Solution:
We will have single core team that can merge anything.
But there will be two types of core (based on trust ;()

I created page in docs, that explains who is who:
https://review.openstack.org/#/c/188843/1

Core reviewer
------------------
That are core for whole project

Plugin Core reviewer
----------------------------
That will just review/merge their component plugins and nothing else


I hope by end of this cycle each component will have own subteam which will
resolve
most of reviewing process scale issues..


Best regards,
Boris Pavlovic
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.openstack.org/pipermail/openstack-dev/attachments/20150605/49bf8fe5/attachment.html>