On Wed, Aug 13, 2014 at 3:29 PM, Doug Hellmann <d...@doughellmann.com> wrote:
> > On Aug 13, 2014, at 4:08 PM, Matthew Treinish <mtrein...@kortar.org> > wrote: > > > On Wed, Aug 13, 2014 at 03:43:21PM -0400, Eoghan Glynn wrote: > >> > >> > >>>>> Divert all cross project efforts from the following projects so we > can > >>>>> focus > >>>>> our cross project resources. Once we are in a bitter place we can > expand > >>>>> our > >>>>> cross project resources to cover these again. This doesn't mean > removing > >>>>> anything. > >>>>> * Sahara > >>>>> * Trove > >>>>> * Tripleo > >>>> > >>>> You write as if cross-project efforts are both of fixed size and > >>>> amenable to centralized command & control. > >>>> > >>>> Neither of which is actually the case, IMO. > >>>> > >>>> Additional cross-project resources can be ponied up by the large > >>>> contributor companies, and existing cross-project resources are not > >>>> necessarily divertable on command. > >>> > >>> What “cross-project efforts” are we talking about? The liaison program > in > >>> Oslo has been a qualified success so far. Would it make sense to > extend that > >>> to other programs and say that each project needs at least one > designated > >>> QA, Infra, Doc, etc. contact? > >> > >> Well my working assumption was that we were talking about people with > >> the appropriate domain knowledge who are focused primarily on standing > >> up the QA infrastructure. > >> > >> (as opposed to designated points-of-contact within the individual > >> project teams who would be the first port of call for the QA/infra/doc > >> folks if they needed a project-specific perspective on some live issue) > >> > >> That said however, I agree that it would be useful for the QA/infra/doc > >> teams to know who in each project is most domain-knowledgeable when they > >> need to reach out about a project-specific issue. > >> > > > > I actually hadn't considered doing a formal liaison program, like Oslo, > in QA > > before. Mostly, because at least myself and most of the QA cores have a > decent > > grasp on who to ping about certain topics or reviews. That being said, I > realize > > that probably is only disseminating information in a single direction. > So maybe > > having a formal liaison makes sense. > > > > I'll talk to Doug and others about this and see whether adopting > something > > similar for QA makes sense. > > > > > > -Matt Treinish > > The Oslo liaison program started out as a pure communication channel, but > many of the liaisons have stepped up to take on the task of merging changes > into their “home” projects. That has allowed adoption of libraries this > cycle at a rate far higher than we could have achieved if the Oslo team had > been responsible for submitting those changes ourselves. They’ve helped us > identify API issues in the process, which benefits the projects that have > been slower to adopt. So I really think the liaisons are key to library > graduation being successful at our current scale. > Yes, I was going to say that we use doc liaisons with varying success per project, but it has definitely helped me keep sane (mostly). We originally thought of it as a communication channel (you attend my meetings I'll attend yours) but it's also great as a point person that I can reach out to as PTL or to point others to when they have questions. Anne > > Doug > > > _______________________________________________ > > OpenStack-dev mailing list > > OpenStack-dev@lists.openstack.org > > http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack-dev > > > _______________________________________________ > OpenStack-dev mailing list > OpenStack-dev@lists.openstack.org > http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack-dev >
_______________________________________________ OpenStack-dev mailing list OpenStack-dev@lists.openstack.org http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack-dev
