Committers != PMC

[Carsten Ziegeler <cz...@apache.org>]

Hi,

as I found out the hard way, it seems that some of our committers are
not on the pmc. In general, the pmc should include all committers as the
pmc is the oversight of the project (of course there are exceptions to
this rule).

I think we should invite all committers who are currently not in the pmc
to join.

WDYT?

Carsten
-- 
Carsten Ziegeler
cziegeler@apache.org

Re: Committers != PMC

[Carsten Ziegeler <cz...@apache.org>]

Richard S. Hall wrote:
> We just voted on this in the Felix process document...
> 
Ups, did we? (its still august, isn't it?)

Ok, then we can close this thread... :(

Carsten



-- 
Carsten Ziegeler
cziegeler@apache.org

Re: Committers != PMC

["Richard S. Hall" <he...@ungoverned.org>]

We just voted on this in the Felix process document...

I don't think all committers should be PMC members by default. The PMC 
gets to set the direction for the project and, as such, has a big 
responsibility. The progression of contributor to committer to PMC 
member seems to be a reasonable way to earn karma with the other PMC 
members.

-> richard

Carsten Ziegeler wrote:
> Hi,
>
> as I found out the hard way, it seems that some of our committers are
> not on the pmc. In general, the pmc should include all committers as the
> pmc is the oversight of the project (of course there are exceptions to
> this rule).
>
> I think we should invite all committers who are currently not in the pmc
> to join.
>
> WDYT?
>
> Carsten
>   

Re: Committers != PMC

[Carsten Ziegeler <cz...@apache.org>]

Alex Karasulu wrote:
> +1 I concur with Marcel, Karl, and Richard.  I think Carsten is expressing
> what are
> various opinions; not necessarily concrete ASF rules.  Niclas gave a great
> synopsis
> as to how the various opinions arose.
> 
> Although I have a profound respect for Stefano and the Cocooners I think
> it's a potentially
> dangerous idea. Every community is different and has to decide for itself.
> The Cocoon
> folks went with it and it was OK but there can be problems.
> 
> The PMC might be young but it's got what it takes and does not need to clone
> the ways
> of the Cocoon PMC.  So if this is what they decide then lets lay it to rest
> without
> presuring them.
> 
I absolutely agree. But just for the record, it's not just Cocoon who is
doing it
this way. Nearly most projects I'm involved with here at Apache are
doing it this way (and that not because of me). I think both approaches
make sense in some way; we here have decided to explicitly distinguish
between committers and pmc members and that's fine. *If* we want, we can
change this in either direction at any time; but for now I guess we
should just leave it as it is.

Carsten
-- 
Carsten Ziegeler
cziegeler@apache.org

Re: Committers != PMC

[Alex Karasulu <ak...@apache.org>]

+1 I concur with Marcel, Karl, and Richard.  I think Carsten is expressing
what are
various opinions; not necessarily concrete ASF rules.  Niclas gave a great
synopsis
as to how the various opinions arose.

Although I have a profound respect for Stefano and the Cocooners I think
it's a potentially
dangerous idea. Every community is different and has to decide for itself.
The Cocoon
folks went with it and it was OK but there can be problems.

The PMC might be young but it's got what it takes and does not need to clone
the ways
of the Cocoon PMC.  So if this is what they decide then lets lay it to rest
without
presuring them.

Alex


On 8/28/07, Marcel Offermans <ma...@luminis.nl> wrote:
>
> On Aug 28, 2007, at 7:53 , Carsten Ziegeler wrote:
>
> > as I found out the hard way, it seems that some of our committers are
> > not on the pmc. In general, the pmc should include all committers
> > as the
> > pmc is the oversight of the project (of course there are exceptions to
> > this rule).
> >
> > I think we should invite all committers who are currently not in
> > the pmc
> > to join.
>
> I'm no Apache guru, if committers == PMC I don't see why there are
> two roles at all. By looking at http://apache.org/foundation/how-it-
> works.html#roles it seems to me that it's very well possible that the
> PMC is a subset of the "developers and committers". Looking at a
> couple of other projects (I could find lists for Cayenne and
> Geronimo, but did not look at all of them) they both have a
> difference between committers and PMC members.
>
> In short, as long as there are two roles I don't think we should
> invite all committers to become PMC members by default.
>
> Greetings, Marcel
>
>
>
>

Re: Committers != PMC

[Carsten Ziegeler <cz...@apache.org>]

Karl Pauls wrote:

> 
> If it is a rule it should be written done somewhere no? Furthermore,
> as Marcel points out, it definitely is not the case for a lot of other
> Apache projects. Let alone that we are free to define our on processes
> and rules I don't think I'd like a "lets just make all commiters PMC
> members" rule. We do have a way to do this in Felix btw. (the last
> person we added was Niclas) as you probably know.
Sure.

> 
>>> In short, as long as there are two roles I don't think we should invite
>>> all committers to become PMC members by default.
>> Hmm, why not? What do we loose? We have some committers here who are not
>> in the pmc and really should be on the pmc.
> 
> Well, your are on the PMC so it should be easy to propose those people
> you think really should be on it to the other PMC members. Honestly, I
> do not completely understand what you are trying to do here. The
> process is clear - if you want certain people on the PMC you are free
> to propose them and call a vote on the private list.
> 
Okay, in other projects at Apache we reverted the process, so its up to
the committers to say "hey, i want to be part of the pmc". But as you
said, it is up to us to define these rules. I think it makes sense to
give the people who do the stuff the power to make decisions about the
project. If they don't want to, well, no problem.
Anyway, so I'll just propose all committers who are not on the PMC as
new members, right? :)

Carsten
-- 
Carsten Ziegeler
cziegeler@apache.org

Re: Committers != PMC

[Karl Pauls <ka...@gmail.com>]

On 8/28/07, Carsten Ziegeler <cz...@apache.org> wrote:
> Marcel Offermans wrote:
> > On Aug 28, 2007, at 7:53 , Carsten Ziegeler wrote:
> >
> >> as I found out the hard way, it seems that some of our committers are
> >> not on the pmc. In general, the pmc should include all committers as the
> >> pmc is the oversight of the project (of course there are exceptions to
> >> this rule).
> >>
> >> I think we should invite all committers who are currently not in the pmc
> >> to join.
> >
> > I'm no Apache guru, if committers == PMC I don't see why there are two
> > roles at all.
> The difference is a very thin line. Putting it very simple, committers
> can commit and the PMC can make the decisions like voting (on project
> related stuff like new committers etc.), veto a commit etc.
>
> > By looking at
> > http://apache.org/foundation/how-it-works.html#roles it seems to me that
> > it's very well possible that the PMC is a subset of the "developers and
> > committers". Looking at a couple of other projects (I could find lists
> > for Cayenne and Geronimo, but did not look at all of them) they both
> > have a difference between committers and PMC members.
> The general rule (though not a written one) is to try to get all
> committers on the pmc.

If it is a rule it should be written done somewhere no? Furthermore,
as Marcel points out, it definitely is not the case for a lot of other
Apache projects. Let alone that we are free to define our on processes
and rules I don't think I'd like a "lets just make all commiters PMC
members" rule. We do have a way to do this in Felix btw. (the last
person we added was Niclas) as you probably know.

> > In short, as long as there are two roles I don't think we should invite
> > all committers to become PMC members by default.
> Hmm, why not? What do we loose? We have some committers here who are not
> in the pmc and really should be on the pmc.

Well, your are on the PMC so it should be easy to propose those people
you think really should be on it to the other PMC members. Honestly, I
do not completely understand what you are trying to do here. The
process is clear - if you want certain people on the PMC you are free
to propose them and call a vote on the private list.

regards,

Karl

> And it's just an invitation; if someone feels that she does not want to be on the pmc,
> that's fine as well.
>
> Carsten
>
> --
> Carsten Ziegeler
> cziegeler@apache.org
>


-- 
Karl Pauls
karlpauls@gmail.com

Re: Committers != PMC

[Niclas Hedhman <ni...@hedhman.org>]

On Wednesday 29 August 2007 02:59, Karl Pauls wrote:
> Not according to http://www.apache.org/foundation/voting.html. There it
> says:

Not easy to be in ASF, since I am pretty sure it wasn't like that a couple of 
years back, where PMC was only about oversight and legalities, incl releases. 
And day-to-day work (such as code vetoes) was the committers' headache.

So, somewhere between 2002 and now, it has obviously been modified... Sorry 
for the FUD!


Cheers
Niclas

Re: Committers != PMC

[Karl Pauls <ka...@gmail.com>]

On 8/28/07, Niclas Hedhman <ni...@hedhman.org> wrote:
> On Tuesday 28 August 2007 16:01, Carsten Ziegeler wrote:
> > Putting it very simple, committers
> > can commit and the PMC can make the decisions like voting (on project
> > related stuff like new committers etc.), veto a commit etc.
>
> There are several camps across the ASF. Anything that Stefano has had a hand
> in to start, seem to be "default PMC member" and many other projects in Java
> land has followed, but as others here point out, not all has done this.
>
> *I* have no strong opinion either way.
>
> AFAIK, committers(!) have the veto rights to individual commits.

Not according to http://www.apache.org/foundation/voting.html. There it says:

"However, the basic rule is that only PMC members have binding votes,
and all others are either discouraged from voting (to keep the noise
down) or else have their votes considered of an indicative or advisory
nature only...Only votes by PMC members are considered binding on
code-modification issues, however."

regards,

Karl

> PMC decides "general direction" and "releases". And there are some arguments
> whether it must be the PMC who decides on new committers, since some
> documentation and practices show that it is the other committers.
>
> And rightfully so, each project defines its own rules and guidelines, as long
> as they don't contradict the ASF by-laws and other legal requirements. It is
> also expected that projects follow the "spirit of ASF", which are the
> unwritten parts that Carsten mentions.
>
>
> I hope this is a gentle "contemporary history" lesson... ;o)
>
>
> My own opinion is; Let's continue with past practice for now, and see where
> that brings us. This is still an "infant project", trying to stand on its own
> legs.
>
>
> Cheers
> Niclas
>


-- 
Karl Pauls
karlpauls@gmail.com

Re: Committers != PMC

[Niclas Hedhman <ni...@hedhman.org>]

On Tuesday 28 August 2007 16:01, Carsten Ziegeler wrote:
> Putting it very simple, committers
> can commit and the PMC can make the decisions like voting (on project
> related stuff like new committers etc.), veto a commit etc.

There are several camps across the ASF. Anything that Stefano has had a hand 
in to start, seem to be "default PMC member" and many other projects in Java 
land has followed, but as others here point out, not all has done this.

*I* have no strong opinion either way.

AFAIK, committers(!) have the veto rights to individual commits. PMC 
decides "general direction" and "releases". And there are some arguments 
whether it must be the PMC who decides on new committers, since some 
documentation and practices show that it is the other committers.

And rightfully so, each project defines its own rules and guidelines, as long 
as they don't contradict the ASF by-laws and other legal requirements. It is 
also expected that projects follow the "spirit of ASF", which are the 
unwritten parts that Carsten mentions. 


I hope this is a gentle "contemporary history" lesson... ;o)


My own opinion is; Let's continue with past practice for now, and see where 
that brings us. This is still an "infant project", trying to stand on its own 
legs.


Cheers
Niclas

Re: Committers != PMC

[Carsten Ziegeler <cz...@apache.org>]

Marcel Offermans wrote:
> On Aug 28, 2007, at 7:53 , Carsten Ziegeler wrote:
> 
>> as I found out the hard way, it seems that some of our committers are
>> not on the pmc. In general, the pmc should include all committers as the
>> pmc is the oversight of the project (of course there are exceptions to
>> this rule).
>>
>> I think we should invite all committers who are currently not in the pmc
>> to join.
> 
> I'm no Apache guru, if committers == PMC I don't see why there are two
> roles at all.
The difference is a very thin line. Putting it very simple, committers
can commit and the PMC can make the decisions like voting (on project
related stuff like new committers etc.), veto a commit etc.

> By looking at
> http://apache.org/foundation/how-it-works.html#roles it seems to me that
> it's very well possible that the PMC is a subset of the "developers and
> committers". Looking at a couple of other projects (I could find lists
> for Cayenne and Geronimo, but did not look at all of them) they both
> have a difference between committers and PMC members.
The general rule (though not a written one) is to try to get all
committers on the pmc.

> 
> In short, as long as there are two roles I don't think we should invite
> all committers to become PMC members by default.
Hmm, why not? What do we loose? We have some committers here who are not
in the pmc and really should be on the pmc. And it's just an invitation;
if someone feels that she does not want to be on the pmc, that's fine as
well.

Carsten

-- 
Carsten Ziegeler
cziegeler@apache.org

Re: Committers != PMC

[Marcel Offermans <ma...@luminis.nl>]

On Aug 28, 2007, at 7:53 , Carsten Ziegeler wrote:

> as I found out the hard way, it seems that some of our committers are
> not on the pmc. In general, the pmc should include all committers  
> as the
> pmc is the oversight of the project (of course there are exceptions to
> this rule).
>
> I think we should invite all committers who are currently not in  
> the pmc
> to join.

I'm no Apache guru, if committers == PMC I don't see why there are  
two roles at all. By looking at http://apache.org/foundation/how-it- 
works.html#roles it seems to me that it's very well possible that the  
PMC is a subset of the "developers and committers". Looking at a  
couple of other projects (I could find lists for Cayenne and  
Geronimo, but did not look at all of them) they both have a  
difference between committers and PMC members.

In short, as long as there are two roles I don't think we should  
invite all committers to become PMC members by default.

Greetings, Marcel