incubator-general mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From Stian Soiland-Reyes <>
Subject Re: Radical proposal: no initial list of committers
Date Tue, 27 Sep 2016 15:56:20 GMT
It's a radical proposal - but I think it would only work well where
the champion/mentors are also from same community - e.g. the kind of
"Already lots of Apache folks" projects that we previously suggested
for the straight-to-PLP mode. There's also the danger of the project
to seem hostile to non-Apache folks - and if the project has not
already have a well-working contributor model; then activity will die
out as it would have almost no committers.

But choreographed well by the mentors - who would then have to be
almost like a "project coach" - it could turn to a very open model as
everyone on the list would feel they have a large say of project
decisions. Some open source projects work well like this on GitHub
today - with a very small set of people using commit rights and all
activity done through pull requests.

Perhaps the radical mini-PMC model could be good for those - working
out who "is" and "isn't"  a committer in ASF sense can be tricky for
such projects - as you would have to go through all the discussions on
all pull requests - code review and "helping" in design decisions is
also part of being a committer!

Just to note - one advantage of the "emeritus committers allowed" for
a long-standing community (e.g. Netbeans - although I think we're
happy with Netbeans proposal now!) is that those people will carry
with them history and opinions - and could still work well with the
PMC hat on even if they don't commit a single line of code - acting
almost like an Advisory Board for the project.

Another is that moving to Apache is not required to be a hard-reset -
so just as Apache projects don't tend to vote out dormant committers
(but they might voluntarily retire) - so if a long-running project
moves to ASF it should not have to 'shed' committers.

On 27 September 2016 at 16:43, Julian Hyde <> wrote:
> +0.5 I like this idea a lot (but there are governance problems to be solved).
> I have long been frustrated at folks who jump on the bandwagon and become initial committers
then play no further part in the project.
> Do inactive committers harm the project? No, of course they don’t. But it isn’t fair
to those who get in by pulling their weight.
> Now, those inactive committers should be removed when the project graduates, right? No,
in my experience they get to be PMC members. The project is too polite to throw people out
(especially what they perceive as “Apache VIPs”).
> And while I’m on that topic: I believe that mentors should not automatically become
PMC members when the project graduates.
> Julian
>> On Sep 27, 2016, at 4:25 AM, Greg Stein <> wrote:
>> The NetBeans proposal (among many others in the past) has demonstrated a
>> significant "problem" with trying to establish an appropriate list of
>> initial committers. There are many people that want to be on, for various
>> reasons. Because they are committers, recent or historic. Or they want the
>> "prestige" to be there. Some people believe they "deserve" to be on the
>> list. etc etc
>> Establishing the list is particularly difficult for large and old
>> communities.
>> But. What if we just said "no such list" ?
>> This will shift the initial voting of committers upon the Champion/Mentors
>> who will construct the entirety of the PPMC. But hey: aren't they supposed
>> to be involved? Aren't they supposed to demonstrate how to earn merit, and
>> the committership that results?
>> This would also solve the problem of initial committers that have not
>> established any merit whatsoever. We've had many situations where people
>> simply add themselves to the list. Why? Cuz they chose to do so. It is sort
>> of silently allowed for IPMC members to add themselves. "I wanna join!"
>> BAM. It happens.
>> So yeah. Radical thought: NO initial list. The PPMC is just the Champion +
>> Mentors. They will build the committers and PPMC according to merit. (note:
>> this could be *very* fast for a particular few highly-engaged with bringing
>> the project to the ASF)
>> ???
>> Cheers,
>> -g
> ---------------------------------------------------------------------
> To unsubscribe, e-mail:
> For additional commands, e-mail:

Stian Soiland-Reyes

To unsubscribe, e-mail:
For additional commands, e-mail:

View raw message