<p dir="ltr"><br>
Le 4 août 2015 13:18, "Davanum Srinivas" <<a href="mailto:davanum@gmail.com">davanum@gmail.com</a>> a écrit :<br>
><br>
> +1 to "pushing for is creating Nova subteams with their own core<br>
> reviewers, which would be experts and trusted to +2 on a defined<br>
> subset of code."<br>
><br>
> Oslo has cores who can +2 any of the oslo libraries and each Oslo<br>
> library has its own cores as well. We tend to elevate karma quickly<br>
> for cores between libraries and from library core to the main core. So<br>
> this is working well and could be something Nova can adopt. (oslo<br>
> library == nova sub team).<br>
><br>
> I used to think that subteam +2 karma with informal boundaries may be<br>
> better, but nova cores may be more comfortable with explicit<br>
> boundaries and karma elevation.<br>
><br>
> ...But anything is better than status quo!<br>
><br>
> Thanks,<br>
> dims<br>
><br></p>
<p dir="ltr">Well, we already tried to create subteams in Liberty, so please make sure it's not something on status quo.</p>
<p dir="ltr">That said, it's understandable that delegating trust to subteams doesn't necessarily resolve the problem and actually proxies that to the trustability within the team.</p>
<p dir="ltr">Are all the subteam members trusting them together ? Can we assume that we can scale ? What could be the acceptance criterias for being a subteam member ? How can we identify that level of acceptance ? See, there are still unresolved questions, hence my call for waiting Tokyo on that.</p>
<p dir="ltr">-Sylvain</p>
<p dir="ltr">><br>
> On Tue, Aug 4, 2015 at 6:54 AM, Thierry Carrez <<a href="mailto:thierry@openstack.org">thierry@openstack.org</a>> wrote:<br>
> > John Garbutt wrote:<br>
> >> [...]<br>
> >> Personally I find a mix of coding and reviewing good to keep a decent<br>
> >> level of empathy and sanity. I don't have time for any coding this<br>
> >> release (only a bit of documenting), and its not something I can<br>
> >> honestly recommend as a best practice. If people don't maintain a good<br>
> >> level of reviews, we do tend to drop those folks from nova-core.<br>
> >><br>
> >> I know ttx has been pushing for dedicated reviewers. It would be nice<br>
> >> to find folks that can do that, but we just haven't found any of those<br>
> >> people to date.<br>
> ><br>
> > Hell no! I'd hate dedicated reviewers.<br>
> ><br>
> > I want everyone to be a reviewer and a commit author. It's the only way<br>
> > to keep current on the code and relevant enough to be trusted for the<br>
> > final +2s.<br>
> ><br>
> > The trick is that given how wide and diverse Nova's code is, it's<br>
> > impossible to be a reviewer / commit author / expert in all the Nova<br>
> > things. If you slowly grew with Nova over the past years, you may have a<br>
> > good grasp of most parts. But for newcomers, it's an impossible mountain<br>
> > to climb, and they prefer to go to smaller projects. And as we burn out<br>
> > old people with a giant set of reviews, we fail to replace them with new<br>
> > blood.<br>
> ><br>
> > This is why I advocate dividing code / reviewers / expertise along<br>
> > smaller areas within Nova, so that new people can focus and become a<br>
> > master again. What I'm pushing for is creating Nova subteams with their<br>
> > own core reviewers, which would be experts and trusted to +2 on a<br>
> > defined subset of code.<br>
> ><br>
> > --<br>
> > Thierry Carrez (ttx)<br>
> ><br>
> > __________________________________________________________________________<br>
> > OpenStack Development Mailing List (not for usage questions)<br>
> > Unsubscribe: <a href="http://OpenStack-dev-request@lists.openstack.org?subject:unsubscribe">OpenStack-dev-request@lists.openstack.org?subject:unsubscribe</a><br>
> > <a href="http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack-dev">http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack-dev</a><br>
><br>
><br>
><br>
> --<br>
> Davanum Srinivas :: <a href="https://twitter.com/dims">https://twitter.com/dims</a><br>
><br>
> __________________________________________________________________________<br>
> OpenStack Development Mailing List (not for usage questions)<br>
> Unsubscribe: <a href="http://OpenStack-dev-request@lists.openstack.org?subject:unsubscribe">OpenStack-dev-request@lists.openstack.org?subject:unsubscribe</a><br>
> <a href="http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack-dev">http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack-dev</a><br>
</p>