incubator-general mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From Roman Shaposhnik <>
Subject Re: Reform of Incubator {was; [DISCUSSION] Graduate Ignite from the Apache Incubator)
Date Mon, 03 Aug 2015 02:05:02 GMT
I've been waiting for a bout a week for other to chime in, but
it seems that nobody has so I'll repeat my question as of
a week ago: what would be the effective way to change the
status quo around IPMC an make it more board like?

Perhaps we can start from making the release policy actually
make sense along the lines that Ross has outlined. I guess
I can propose a change to the current policies (or to Ross'
point just get it back from the wayback machine :-)).

But seriously, who else thinks the movement towards empowering
PPMCs and making IPMC very much like the board makes sense?


On Sun, Jul 26, 2015 at 8:56 PM, Ross Gardler
<> wrote:
> Well this explains how it got this way, it was poorly worded from the start...
> The first part is about incoming code (the SGA) and nothing has changed there.
> The second part says " SHALL formally request the Incubator PMC approve such a release"
It does not say [VOTE] and it was never, IMHO, intended to be a separate vote (unless there
were insufficient IPMC votes).
> Speaking personally I have always (and I know many other mentors have also, certainly
all those that have co-mentored with me) treated a vote on the podling list as binding and
a "request" in the form of a notification (giving time to object if appropriate) when three
positive IPMC votes have been cast.
> In 2006 it said "Therefore, should a Podling decide it wishes to perform a release, the
Podling SHALL hold a vote on the Podling's public -dev list. At least three +1 votes are required
(see the Apache Voting Process page), and only the PPMC member votes are binding. If the majority
of all votes is positive, then the Podling SHALL send a summary of that vote to the Incubator's
general list and formally request the Incubator PMC approve such a release."
> That's still the wording today.
> I've never been challenged on this approach (having mentored many podlings). It was my
approach with the most recent release I was a part of, just last week (Ripple). The additional
reviews received from the IPMC were useful, spotting a couple of small items, and turned up
the one required +1 (only two binding mentor votes).
> Whether this is an accurate recollection of what was discussed way back, or whether this
is just a practice I (and others) have fallen into and not been challenged on I urge the IPMC
to consider this approach (of course, it does rely on proper oversight from mentors and the
IPMC, I'm comfortable with this approach because I never vote +1 without having done due diligence
on the release - I trust others do the same).
> Ross
> -----Original Message-----
> From: David Nalley []
> Sent: Sunday, July 26, 2015 6:05 PM
> To:
> Subject: Re: Reform of Incubator {was; [DISCUSSION] Graduate Ignite from the Apache Incubator)
> On Sun, Jul 26, 2015 at 8:41 PM, Ross Gardler <> wrote:
>> The proposed need to announce release votes on the IPMC list is how things were when
the incubator was created. The need for IPMC to control the process is another case of the
IPMC over-reaching itself and in so doing causing problems by creating a bottleneck in the
>> It used to be that it was only required to *notify* the IPMC of a release vote underway.
Thereby giving interested IPMC members the opportunity to review artifacts and processes during
the normal release cycle. IPMC members were expected to cast their votes on the PPMC list
where such things belong.
> I'd love to see links to that - my digging didn't find it. (see below)
>> I'm not sure where this idea that the vote actually occurs on the IPMC list came
from but it's flat wrong in my opinion (someone may dig through the archives and find a good
reason it was changed, but my feeling is that it changed gradually through edits-on-edits-on-edits
of the incubation policy).
>> The fact is that the PPMC (with IPMC representation from the mentors) should be in
charge of their releases, and pretty much everything else. The IPMC role is one of teaching
the PPMC how manage itself. Mentors should do this through mentoring and the IPMC should ensure
it is done through an appropriate level of oversight (not an inappropriate amount of control).
>> Consider this... The board does not bring TLP release votes to board@, why on earth
must the IPMC do so?
>> I've half a mind to got back the wayback machine and pull the original
>> incubator polices and propose them as the "new" policies (yes, some
>> changes have been good, but it seems to me that many have not)
> So I couldn't find anything in 2003, but 2004 has this page[1] which included the text:
> "Podlings in Incubation SHALL NOT perform any releases of software without the explicit
approval of the Incubator PMC. Such approval SHALL be given only after the Incubator PMC has
followed the process detailed in (Reference to Charter), and SHALL NOT occur until all source
has been legally transferred to the ASF."
> and
> "Therefore, should a Podling decide it wishes to perform a release, the Podling SHALL
formally request the Incubator PMC approve such a release. The request SHALL have the endorsement
of the Mentor."
> So it seems that this has been with us for a long while.
> [1]
> ---------------------------------------------------------------------
> To unsubscribe, e-mail:
> For additional commands, e-mail:

To unsubscribe, e-mail:
For additional commands, e-mail:

View raw message