incubator-general mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From Craig L Russell <Craig.Russ...@Sun.COM>
Subject Re: making project decisions with a small number of PMC members
Date Tue, 15 May 2007 14:31:06 GMT
Hi Brett,

+1 to most of what Leo says.

I'd point out that the main purpose of the PPMC is to get the podling  
to the point of making decisions "the Apache Way" and the IPMC then  
ratifies the decisions. This model should allow the decisions to be  
made by the people who know and care the most, while preserving the  
accountability that's legally required.

So if you properly vote in the PPMC to accept a new committer, or  
vote on a new release, following the process in the incubator policy  
and guides, the IPMC will likely approve pretty quickly any decisions  
that are made. If you're having difficulty getting the binding IPMC  
votes after the PPMC votes, then this is an issue that needs to be  

Good luck,


On May 15, 2007, at 1:16 AM, Leo Simons wrote:

> On May 15, 2007, at 12:26 AM, Brett Porter wrote:
>> Quick question (I hope). I was thinking about what will happen if the
>> NMaven podling would like to add a committer, make a release, etc.
>> Would votes by Maven PMC members (As the sponsoring project) be
>> considered binding in this case, or should we have IPMC members?
> No the sponsoring PMC votes are not binding. It should be the IPMC.
>> We currently only have 2 IPMC members on the project, so I was going
>> to ask for an additional mentor anyway, but I feel like the
>> appropriate people to be casting votes should be Maven PMC folk.
>> Thoughts?
> I share your feeling. I'm afraid it is not so easy to change.
>  * ASF is structured to have a PMC (or Officer) responsible and
>    accontable for most things
>  * ASF is not structured for multiple PMCs or Officers to share
>    responsibility
> Changing this is a lot of work (for example it screws up the  
> workflow for infrastructure since it becomes much harder to know if  
> a request came with the right authority), so I've always shunned  
> away from it.
> As to which PMC *should* be responsible for *what*, "it depends".  
> We've seen projects were it was really needed for the Incubator PMC  
> to take the responsibility for a TLP-sponsored podling, and then  
> some projects where the IPMC was hardly ever needed at all.
> (aside: some of the original designs for the incubator introduced  
> the "PPMC" that would consist of
>   * incubator PMC
>   * sponsoring PMC
>   * all committers on the new project
>   * all mentors of the project
> where the goal would sort-of be to have the committers make the  
> actual decisions and then the IPMC and sponsoring PMC would  
> reluctantly fill in the gaps. We had to move away from that because  
> it broke the "clear accountability" chain; Secondly, a lot of  
> sponsoring PMC members and IPMC members didn't want to be on that  
> many mailing lists.
> So now we have a PPMC which is really just the committers on the  
> project with their mentors, and it's supposedly no longer allowed  
> to make any real decisions. Not so great a situation either.)
> /LSD
> ---------------------------------------------------------------------
> To unsubscribe, e-mail:
> For additional commands, e-mail:

Craig Russell
Architect, Sun Java Enterprise System
408 276-5638
P.S. A good JDO? O, Gasp!

View raw message