incubator-general mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From Upayavira ...@odoko.co.uk>
Subject Re: Adding new committers process
Date Tue, 29 May 2007 19:13:52 GMT
Martin Ritchie wrote:
> Noel,
> 
> It has been a while since I posted this and the conversation has gone
> cold. I'd like to get some consensus on what the PPMC's role is so
> that we can update the documentation.
> 
> See embedded comments below.
> 
> On 11/04/07, Martin Ritchie <ritchiem@apache.org> wrote:
>> On 11/04/07, Noel J. Bergman <noel@devtech.com> wrote:
>> > Cliff Schmidt wrote:
>> >
>> > > 1. Only IPMC members (e.g. mentors) should send root requests for new
>> > >    podling committers.
>> > > 2. A podling committer vote requires three IPMC +1s to be approved
>> > >   (ideally the mentors, assuming the project still has three 
>> mentors).
>> >
>> > > This [is] not how I read what we have documented at
>> > > http://incubator.apache.org/guides/ppmc.html
>> >
>> > Then we need to fix the documentation.
>> >
>> > > From Noel's comments, it sounds like those "(P)"s should be removed
>> > > from the above sentence.
>> >
>> > The PPMC has no standing within the ASF.  It is a useful structure 
>> for the
>> > Incubator, but the only binding votes on a PPMC are those of the 
>> Incubator
>> > PMC members casting them.  The PMC is the recognized entity within 
>> the ASF
>> > structure responsible for the management of a project, and we need 
>> to ensure
>> > that decisions go through the PMC in order to maintain that role.  
>> Why do
>> > you think that I keep pushing the minimum of three (active) Mentors
>> > recommendation?
> 
>  So should the PPMC's role be to organise, select and perform the
> vote and then forward the vote to general@ for ratification that we
> have performed the process correctly.

Well, it really depends upon whether you've got IPMC members (e.g. 
mentors) on your private list. If your private@ list vote had three +1s 
from IPMC members (e.g. your mentors), then IMO all you need to do is 
inform the IPMC of the vote when complete. Also, the request for an 
account should be sent by an IPMC member, likely one of your mentors.

>  At which point the IPMC takes control of the vote such that when three
>  IPMC memebers have voted (Which may have already occured IF the
>  podlings mentors have voted) they create the account requests and send
>  them to root@ copying the <podling>-private@ list.

I think the thing is to remember that your mentor(s) is going to be an 
IPMC member. The idea of the "IPMC taking control" seems a little 
strange - I cannot imagine the IPMC being sufficiently 
organised/motivated to do that work. Your mentor however, who is also an 
IPMC member, should be sufficiently motivated. And if they're not, 
you're at liberty to kick their butts :-)

> This would then give a bigger pool of recognized people that could
> pickup the completed votes and create the account requests.

>  Alternatively the IPMC could then notifiy the podling-private list
>  that their vote was successfull so that the PPMC could create the
>  account request (Learning that process) and send it to the IPMC for
> forwarding to root@.

I don't personally want to see the IPMC private list given over to 
voting in new committers on podlings. That would make, IMO that list 
pretty unusable. My take on Noel's comments is that the podling should 
ensure that it has got 3 +1s from IPMC members. If it hasn't, it could 
use a mail to private@incubator to solicit them, but those votes would 
go to the podling's private list.

>  That way the IPMC gets to over see the PPMCs ensuring they are moving
> towards the "Apache Way" and root@ only gets emails from people that
> they know have the right to request the action.
> 
> Thoughts?

For all intents and purposes, IPMC == podling mentors. Other IPMC 
members are not likely to be sufficiently motivated to make what you 
mention above actually work.

>> > > I honestly don't know if this is a case of things evolving rules, or
>> > That's fine.  My comment to Martin Ritchie was entirely procedural, 
>> and not
>> > intended to be any sort of slap.  I, too, am favorably disposed 
>> towards QPid
>> > (questions about the specification process aside).  If you don't have
>> > sufficient votes, let me know, and I will review the archives in 
>> order to
>> > determine my own vote.
>> >
>> >         --- Noel
>>
> I didn't take the comments as a slap of any sort. It is just good to
> clear up the procedural work. I would have thought that the incubation
>  documenation would have been more RC than beta as it appears in places
> but I understand the organic process that is going on here. :)

Well, I think we're still working some of this out. And you're helping 
with that :-)

Regards, Upayavira

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


Mime
View raw message