geronimo-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From Greg Stein <>
Subject Re: committer process
Date Tue, 16 Sep 2003 22:15:21 GMT
On Tue, Sep 16, 2003 at 07:41:46PM -0000, wrote:
> From: "Ryan Ackley" <>
> Date: Tue, 16 Sep 2003 08:07:51 -0500
> > The ASF requires that all active committers be part of the PMC. It is just
> > that simple. Say anything you want, but it won't change the facts.
> Greg, if this is the case, you need to add this to the documentation on the
> incubator pages.

If I had a lot of time on my hands, then I would. Simple fact is that I
don't. Many people don't, and so we end up with copies of stuff that don't
necessarily match up well with the ASF's needs.

The root of the problem here is that most of the ASF (including myself)
was not really aware of a lot of the rationale for why the httpd PMC was
constructed and managed in the fashion it is. About this time last year,
we had a *lot* of conversations around the structure and organization of
the ASF, and a lot of this was discovered and discussed. There is now a
lot of backfill that needs to occur, but (still) has not.

> Thats what I was going by. You seem to know official ASF
> policy since you are the chairman. However, it would be nice if you get
> disseminate to the little people (put up documentation) when you get a
> chance so we don't have to wait for you to make condescending comments on a
> mailing list.

Dood. Your tone is rather abrasive, and I'm really not enjoying the need
to have to respond here. Your comments and approach are pretty much
mandating some kind of response, which I resent being required to do.

If I was condescending, then I apologize. I am simply trying to provide
some insight into the ASF's requirements here. This *is* the incubator,
and I *do* see a lot of questions of the nature, "what is expected here?"
I am also seeing people carrying over Jakarta-style behavior which should
be avoided for new projects.

> If its ASF policy, why didn't you just say that last week when this whole
> conversation got started. sheesh!

Look back at my original posts. I suggested to the list to send a list of
committer candidates to I *did* provide a
recommendation which met the ASF's needs. But then you guys went all off
on "deciding a process", closing that vote up real fast, and then starting
to use it to "vote in" new committers.

> > In any case, what Jim Jagielski is saying that the ideal is that all
> > committers are on the PMC. Just because it is broken elsewhere
> Uh...everywhere. According to earlier posts, "almost all" committers are PMC
> members for httpd. We know that this isn't true for jakarta. A very large
> chunk of the other top-level project are jakarta spinoffs.

And what is the point? Majority(*) rules? I think not. The requirements of
the ASF are what rules. As explained by others, the ASF has certain needs
to meet its obligations. The primary factor here is the requirement for
oversight of the project. To make that oversight legally binding requires
a structure where we can demonstrate that all activity on a project is, in
the end, performed at the direction of the officers of the ASF, and
ultimately at the direction of the Board.

This oversight means the committers are part of the PMC, which is overseen
by the PMC Chair, which is overseen by the Board. Only with that chain,
can the ASF state that the software is developed and maintained at the
direction of the ASF rather than the individual contributors. *Without*
that chain, then the individual contributors are *liable* for the code
that they write. The ASF needs to be able to tell a judge, "Joe wrote that
code at our direction. He is not responsible for it. Drop him from the
suit -- we are the responsible party." The judge will go for that *iff* we
can demonstrate that chain.

> > Corollary: the Jakarta PMC is broken. They know it, and they have spent
> > some time increasing the size of the PMC to include all active committers.
> Wow this is news to me. I am a Jakarta committer. I haven't been contacted
> or heard any rumors. Now that I have your attention Mr. Chairman, when am I
> going to be made a Jakarta PMC member?

No need to be nasty. Go ask the PMC. It's their problem, not mine. I've
done a fair amount of poking and prodding, and they are aware of the
problems. I see some action, but as with all volunteer organizations, it
will take time based on how much free time people have. They *are* moving
forward, which is good. Completely ignoring the problem would not be, um,
a Good Choice(tm).


(*) of the PMCs, httpd, apr, incubator, and commons (at least) follow the
inclusive model with private voting when people are involved. I believe
that Ant, WS, and Cocoon use the inclusive model, but public voting. not
sure about tcl, perl, maven.

-- ... ASF Chairman ...

View raw message