From before the time my first project entered the incubator, it was emphasized to me that each project (or podling) had to have at least three active members, *not* just for diversity, but so that each action taken by a project (or podling) would have three binding votes. And the most important votes are for new committers and for software releases.

To my understanding, this "three's a minimum" informed the requirement to have three mentors on a project entering incubation, so there would be at least three binding votes to add committers to a podling and three binding votes to release software from the incubator.

In reviewing the documented roles of participants in incubation, I now notice that nowhere in the Mentor responsibilities does it mention voting.

I think that changing this will have a profound impact on how the incubator runs, and in particular, address a few of the main pain points we have seen in podlings.

Pain point 1: Podlings have to beg for binding votes for new committers. If mentors are required to vote on new committers for their podlings, there are three binding votes and the subsequent incubator pmc vote is just a formality.

Pain point 2: Podlings have to beg for binding votes for releases. If mentors are required to vote on podling releases, there are three binding votes and the subsequent incubator pmc vote is just a formality.

So before we blow up the incubator, I'd propose modifying the Mentor contract to require all mentors to cast votes on new committers and releases. And see how much of the current turmoil is addressed.

Craig

just me.

p.s. I think the issue of pmc chair not handling new members timely is about to be resolved with the proposal to allow a delegate.


---------------------------------------------------------------------
To unsubscribe, e-mail: general-unsubscr...@incubator.apache.org
For additional commands, e-mail: general-h...@incubator.apache.org

Reply via email to