incubator-general mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From Christian Grobmeier <>
Subject Re: Does Apache Have a 'Rule' Problem, or does the Incubator sometimes just make it look that way?
Date Sun, 27 Nov 2011 19:27:09 GMT
On Sun, Nov 27, 2011 at 11:47 AM, Robert Burrell Donkin
<> wrote:
> <snip>
> On Sun, Nov 27, 2011 at 9:38 AM, ant elder <> wrote:
>> I do acknowledge that it can be really
>> hard to get some things changed and you may need a thick skin and lots
>> of perseverance. But lets try to demonstrate its possible - Christian
>> tell us three things you'd like changed and we'll pick one and try to
>> fix it right here right now just to show it can be done.
> +1
> Robert (wondering about an Incubator F2F meetup for rules stomping)

Thanks to both to you for your kind words.
Actually I feel a bit exhausted. I had several heavy discussions
recently on the ASF and now I feel tired. Actually I can't say I have
a thick skin. Speaking of the incubator changes, I have commented on
the various threads about the new "Champion". Anyway, I actually think
the Incubator needs a complete refactoring. So I can't speak of three
items I would like to change, it is one big thing. I have not proposed
it, because: no energy.


The Incubator has more than 100 IPMCs members. Constantly we are
discussing about inactive mentors and overall lack of energy. Now the
proposal is to extend the Champion term and to change the reporting to
the board. I have asked: what do we expect to be on that reports? I
even doubt, having an extended Champion role will cause people to be
more active. We will still have missing reports and inactive mentors.
Is it a surprise? No. We are 100.

And we are not really a community. In fact, we are not even having a
Meritocracy, as the other projects have. On other projects you need to
earn merit and then you are elected as a committer and after earning
more merit, you can become a PMC. At the incubator you earn merit
somewhere, and once you are a Member, you can request membership on
the IPMC. This is a huge exception to my knowledge. The incubator has
many exceptions to standard procedures. That being said, we are a few
active people here and many inactive. We are bloated and we cannot
grow together as a group, because you can walk in and step out as you

Question is, are we already unable to do incubators work? When there
are not the mentioned, few active people on the projects, sometimes
there is no mentoring at all. no actually we don't do a good job in my
eyes (as said, a few exceptions of course). Or we don't have a clue if
we do.

I think we need to change our thinking and finally make up a
community. Not only a list of loose coupled names.

What if we would completely reboot the Incubator? Lets assume:

1) Every ASF committer on a non-podling project can serve as a mentor.
The phrase "Search for a mentor" should be rephrased to: "search for
an ASF committer who is willing to mentor your project".

2) Every ASF committer, who is mentoring a project and shows an
interest in the Incubator, can be elected into the IPMC

3) A podling must have a Champion who helps to get the podling into
the incubator. The champion can be every ASF committer. After the
project has been accepted, the Champion is the one who needs to take
care on reporting. The champion can be elected among ASF committers
(which includes the podling committers).

4) Every podling must have an IPMC member reading the lists

5) Every mentor must sign the report. Mentors who do not sign the
reports three times in a row are not to be considered active

6) Reporting schedule is at it is now. Podlings which fail to report
three times in a row will be terminated, except they have good reasons

7) The IPMC makes up an report. As a community it is easier to report:
who joined, who left, which podling was accepted, which one was not.

8) 3 IPMC votes are necessary to get a release out of the door. 3
Mentor votes are necessary.

9) New Podlings need to bring in their code within 3 months and
CLAs/Grants, or incubation ends, except they good reasons.

10) Podlings in incubation for > 1 year need to be discussed quarterly
by the IPMC. If there is no community, the IPMC might decide to
terminate this podling, because: Community over Code. No Community, no

Now with just 10 rules we have a small IPMC taking care on the
oversight. We do not need to let every mentor join the IPMC. In fact,
we can get rid of the terms "Mentor" as this is the same as an ASF
committer. The only exception is the "Champion", which in fact is
serving as some kind of podling chair. We could even remove this term
and call it "Podling Chair" which is more similar to what we have in
other projects.

How can we achieve this?

Drop everybody from the IPMC list except our chair. Let him decide 5
people he knows who are active and they elect everybody else in who is
active or has merit to join the IPMC.

Probably there are some flaws in my thinking. But I think it is more
efficient to make tabula rasa and try to be as near to a "standard
apache project" as possible.


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

View raw message