Hi Noel,
On Jun 8, 2007, at 12:18 PM, Noel J. Bergman wrote:
> Craig Russell wrote:
>
>>> Does being a Mentor grant any special powers/rights?
>
>> Well, yes. There are a dozen references to the Mentor role in Policy
>> [1]. The responsibilities, requirements, karma grants are all defined
>> in Policy.
>
> Yes, but read it again. Consider: there are no new rights
> granted. There
> are new responsibilities, but the person accepting those
> responsibilities
> already has all of the rights necessary to do them, by virtue of
> being a
> PMC Member.
I just read it again, and I see a few new rights granted. Maybe these
are rights that all incubator PMC members have, so please enlighten me.
1. On acceptance of a candidate project, the assigned Mentors shall
be given access to the Podling's repository for the duration of the
incubation process. This is to allow the Mentors to perform their
incubation duties, and is for administrative purposes only.
2. The Mentors form the initial podling PPMC and are subscribed to
the podling private email list. Incubator PMC members who are Apache
members can review the archives but can't post to the private list
without being moderated.
>>> Now, let us say that you have a vote. The result is 6 to 4.
>>> Majority, even 60%. But I'd hardly consider that a consensus.
>>> On the other hand, if there is a clear consensus, do we always
>>> need to explicitly count it?
>
>> I think the term is "lazy consensus" which is not used to describe
>> anything beyond the acceptance of the podling by the incubator.
>
> We don't use Lazy Consensus to accept a podling, we use mandatory
> majority
> approval (in HTTP Server terms, q.v.,
> http://httpd.apache.org/dev/guidelines.html).
I just read the Policy again, and it sure appears that acceptance of
a podling by the incubator is a lazy approval process. No vote occurs
unless an incubator PMC member says hold.
<policy>
Acceptance By Incubator
Upon a successful result, the PMC Chair of the Sponsor SHOULD request
the Incubator PMC take on the Candidate as a new Podling. The
request, which should be sent to the Incubator PMC on the general
list, MUST contain the following information:
a reference to the results of the vote (so as to provide an audit
trail for the records);
a reference to the Candidate's proposal;
the Mentors, nominated by the Sponsor, who will guide the Candidate
through the Incubation Process. At least one nominated Mentor MUST be
a member of the Apache Software Foundation.
Any Incubator PMC member can send an acknowledgement that the request
was received, then a 72 hour waiting period starts. After this time
has elapsed and no Incubator PMC member objects, the status file may
be committed and the podling started. If any Incubator PMC member
says "hold" before the 72 hours are up, a formal discussion/vote will
be conducted.
The nominated Mentors MAY be immediately accepted by the Incubator
PMC. However the Incubator PMC MAY also suggest replacement Mentors.
The Incubator PMC has the final choice of Mentors.</policy>
>>> Consider: do we need to vote for a release manager? The answer is
>>> no, by the way.
>
>> The release manager is not a defined Incubator role. The Mentor is.
>
> It was an analogy. The question is whether or not it even requires a
> declared vote at all, unless someone raises the issue. Sometimes
> projects
> get a bit vote happy, and then complain about the process overhead,
> without realizing that the undesired overhead was entirely self-
> imposed.
As it is today, adding a new Mentor seems to be undefined, which
invites self-imposed anarchy.
I'm mostly responding to the situation here in which the PPMC was not
functioning and needed a jump-start [1]. No one seemed to know how to
get it to start. Noel's suggestion that the PPMC vote new members [2]
didn't work because Dims was the only active PPMC member, and it's
hard to claim consensus if there is only one voice.
So I suggested that the project get a new Mentor in order to get the
PPMC to work. Which brought us to this point where there is (still)
no defined process to get a new Mentor on a podling.
Craig
[1]
> On Jun 6, 2007, at 1:18 PM, Craig L Russell wrote:
>
>> On Jun 6, 2007, at 12:37 PM, Matthieu Riou wrote:
>
>>> As a community I believe we're very healthy, it's just that we
>>> don't have an official PPMC that represent this community and
>>> have no way to get one voted properly.
>>
>> That's why I'd push for you and Dims to get together a collection
>> of committers that do represent the community and get the Mentors'
>> attention at least long enough to vote them in. Once you have a
>> functioning PPMC you should be able to get it together.
>>
>> If you cannot get the current Mentors to vote in a slate of
>> community representatives, then you'll have to get some new Mentors.
>
[2]
From: "Noel J. Bergman" <noel@devtech.com>
Date: May 15, 2007 9:05:44 AM PDT
To: private@incubator.apache.org
Cc: ode-private@incubator.apache.org
Subject: RE: [VOTE][ODE] Populating the ODE PPMC
Dims,
Since a PPMC has no standing in ASF structure, the Incubator PMC really
couldn't care less who is on it. OK that's overstating, but the
point being
that you can elect whomever you want to the PPMC. That's purely a PPMC
vote. When you are done, send an notice to the Incubator PMC and
record it
in incubator-info.txt, and the various other locations of redundantly
redundant data.
--- Noel
Craig Russell
DB PMC, OpenJPA PMC
clr@apache.org http://db.apache.org/jdo
|