forrest-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From David Crossley <cross...@apache.org>
Subject Re: [DRAFT] Forrest Project Guidelines
Date Wed, 09 Jun 2004 13:26:28 GMT
Nicola Ken Barozzi wrote:
> David Crossley wrote:
> > Nicola Ken Barozzi wrote:
> > 
> >>David Crossley wrote:
> >>...
> >>
> >>>Developers
> >>>----------
> >>>These people are contributers who make more substantial
> >>>contributions. They get more involved in discussion,
> >>>especially discussion about design issues, help with
> >>>solving major issues, provide patches with new and
> >>>improved functionality. They would participate on both
> >>>the "dev" and "user" mailing lists.
> >>
> >>Some developers can be given partial or time limited commit access, to 
> >>be able to give a more substantial contribution to the project without 
> >>necessarily becoming stably part of it.
> > 
> > Then we would need to qualify our statement about how
> > a committer can become part of the PMC.
> 
> All committers are part of the PMC. Committers are the ones committed to 
> the project. What about PMC="Project Management Committers"?

Good idea. What a difference a name makes.

> Developers are not part of the PMC and *may* have partial commit access.

Yes that solves it. Some developers are provided task-based
access from time-to-time.

So the roles are User, Developer, Project Management Committer
(called "Committer" for short).

> > Gee i am still not happy with this levels of committership,
> > but i might be getting warmer.
> 
> It's something that we should use sparingly, focusing not on the person, 
> but on the task.
> 
> If we want someone on board, then this is not the thing to use.
> In instead we want a developer to give a hand on a feature for a limited 
> period of time, or for example be in charge of maintaining a skin, then 
> this is what to use.
> 
> It's very similar to what we have now.
> 
>   current definition of committer  -> developer with extra access
>   current definition of pmc member -> committer
> 
> In fact, did not we have the OO feature committer by a developer? He can 
> be given access to that part. And the lenya skin? The person can be in 
> charge just of that for example. We don't necessarily need to put blocks 
> to svn, just define these spaces in the STATUS file.
> 
> I am making committer===PMC member===binding voter, as it seems to be on 
> the ASF bylaws. If someone has blanket commit access, votes, and is in 
> the group, he *must* be on the PMC.

This is great. It simplifies everything.

> If he is not on the PMC, he has no legal oversight obbligation.
> All non PMC member that have commit access are thus in the developers' 
> group.

Beaut.

> >>All commit accesses to developers must be given with the direct 
> >>supervision of at least one PMC mmember, and must be voted favorably by 
> >>the PMC.
> > 
> > I am not clear what "supervison" means. I would have thought
> > that all committers are supervising anyway, because it is
> > one responsibility to keep oversight on the svn@forrest
> > commit logs.
> 
> True. The role is taken from other projects that use this method, and is 
> a "mentor" like figure. Let's leave this out for now, and keep the 
> global committer group responsible.

It could be mentioned in the definition of the developer
with extra ability.

> >>Being a developer, with or without commit access, is a necessary step 
> >>for becoming a project committer.
> > 
> > Good point.

--David



Mime
View raw message