river-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Geir Magnusson Jr." <g...@pobox.com>
Subject Re: Responsibilities of the PPMC
Date Fri, 16 Feb 2007 09:23:35 GMT

On Feb 16, 2007, at 2:46 AM, Mark Brouwer wrote:

> Craig L Russell wrote:
>>>> Podlings are encouraged to make all committers members of the  
>>>> Podling's PPMC. Thus, the vote to accept a new committer should  
>>>> also be a vote to accept a new member of the PPMC. The initial  
>>>> membership of the PPMC should include the initial committers,  
>>>> the Champion, and the Mentor.
> Can someone with a thorough understanding of the ASF tell whether  
> there
> are projects that keep a distinction between committers and PMC  
> members,
> if so for what reasons and for what kind of projects might it be  
> better
> to maintain this separation.

All projects do in the sense that there *is* a difference between  
committers and PMC members.  A PMC member has a binding, legal vote  
on project issues (like releases) and the committer doesn't.  PMC  
members, by definition, are responsible for oversight over the whole  
project, whereas a committer technically is responsible for the  
commits they make.

There are different styles of how projects do this.  Some keep the  
bar somewhat low for committership and then over a period of time, a  
few months (?) watch the committers behavior and if "compatible" (as  
defined by that PMC) with the community, offer that person PMC  
membership.  Others keep a higher bar for committership, and then  
offer PMC quicker.  This is something this community has to decide.

My personal taste lately is a reasonably high bar for commit to  
ensure working social and technical compatibility, and then faster PMC.

I also believe that all committers should eventually be on the PMC.   
You will have cases where people choose to not be - they don't have  
time, for example.  But those tend to be exceptions, dictated by life  
experiences ("I'm going to have a baby next week, and won't have the  

> I have very mixed feelings with regard to this subject and no  
> experience
> at all, but I realize that as soon as all the committers are PMC  
> members
> and making committer == PMC member turns out to be a mistake it is
> likely too late to correct it.

A PMC can always get rid of someone if there is a problem, but it has  
to be for a good reason, and really, really is an unpleasant  
experience. :)

> Please enlighten me :-)

Hope that helped.  In practice, it works well - in a project with  
active social interaction, you can usually figure out problems before  
they happen, and people tend to be self-selecting.  IOW, if I'm a  
PITA to work with, we're going to probably have friction from the get- 
go, and it won't be fun for anyone, including me, and therefore I  
probably won't hang around anyway.  We're all volunteers :)


> -- 
> Mark

View raw message