You are viewing a plain text version of this content. The canonical link for it is here.
Posted to general@incubator.apache.org by Rodent of Unusual Size <Ke...@Golux.Com> on 2005/07/02 22:44:31 UTC
Re: svn commit: r178965 - /incubator/public/trunk/site-author/learn/newcommitters.html
-----BEGIN PGP SIGNED MESSAGE-----
noel@apache.org wrote:
>
> Discussion of a potential new committer should take place on the project's
> PPMC list (normally, it would take place on a project's TLP list). After
> vetting the new candidate, the vote can take place either on the PPMC list
> (with notice posted to the Incubator PMC list) or on the developer list (with
> a notice posted to the Incubator's general list). The latter practice of a
> private discussion followed by a public, pro-forma, vote is re-emerging as a
> Best Practice for ASF projects.
This last bit troubles me a bit. I can't say I've seen this 're-emergence'
but that's probably my shortcoming in not tracking discussions.
pro forma: made or carried out in a perfunctory manner or as a formality
In other words, lip-service.
Why?
I don't like the idea of discussions about people's merit (or the people
themselves, for that matter) happening in public right in their faces.
For one thing, people have a tendency to be less candid; for another
there's no anonymity to the vote; and finally (and most important) I
think it's unfair to the individual in question, who will get all anxious
and wound up as the voting proceeds. 'Will I make it?' 'What's *his*
problem?' and other such distractions from coding.
- --
#ken P-)}
Ken Coar, Sanagendamgagwedweinini http://Ken.Coar.Org/
Author, developer, opinionist http://Apache-Server.Com/
"Millennium hand and shrimp!"
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.2.4 (MingW32)
Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org
iQCVAwUBQsb8rprNPMCpn3XdAQFX5AQAvJF2+ms95JEPzIe8vvMyNIaOp3+yr8NE
FC5TaXYQA0wKJQvXNFenoix7cCQcBMKzuukguOoyKK8xbFEz4AW8A+9dTeSgMNeE
dEvZ7UFzw+gqRZuXRzg42/qWTGoBRMExC5xJnoKBefy/fQ+H8Gm/VOd/Qx+RuPjy
sh8y4iwGc/U=
=Tz/3
-----END PGP SIGNATURE-----
---------------------------------------------------------------------
To unsubscribe, e-mail: general-unsubscribe@incubator.apache.org
For additional commands, e-mail: general-help@incubator.apache.org
Re: svn commit: r178965 - /incubator/public/trunk/site-author/learn/newcommitters.html
Posted by David Crossley <cr...@apache.org>.
Rodent of Unusual Size wrote:
>
> I don't like the idea of discussions about people's merit (or the people
> themselves, for that matter) happening in public right in their faces.
> For one thing, people have a tendency to be less candid; for another
> there's no anonymity to the vote; and finally (and most important) I
> think it's unfair to the individual in question, who will get all anxious
> and wound up as the voting proceeds. 'Will I make it?' 'What's *his*
> problem?' and other such distractions from coding.
Thanks for following up on that Ken. It has been on my todo.
I agree entirely with you.
Forrest was a recent PMC creation. We reviewed how other projects
operated and decided that it was best to do such discussion and
vote on the private list, then announce in public. Our reasoning
was exactly as you describe.
-David
---------------------------------------------------------------------
To unsubscribe, e-mail: general-unsubscribe@incubator.apache.org
For additional commands, e-mail: general-help@incubator.apache.org
Re: svn commit: r178965 - /incubator/public/trunk/site-author/learn/newcommitters.html
Posted by Daniel John Debrunner <dj...@debrunners.com>.
William A. Rowe, Jr. wrote:
> At 05:42 AM 7/7/2005, Roy T. Fielding wrote:
>>Right now I see this process as a mixed bag that is made much
>>harder by the disjoint between the incubator PMC and the incubated
>>projects' private lists. I wonder what would happen if we made
>>a single "private@incubator" list for all the podlings instead
>>of the separate ppmc lists?
>
>
> or ppmc@incubator (until we rename these to private across the ASF).
>
> I'll give a huge +1 to your idea though. I agree that only people
> discussions and confidential negotiations belong on private@,
> and by watching other ppmc's grapple with these issues, all of our
> incubating projects will learn from one another.
A combined list has the possibility of a potential committer to
incubator project X seeing discussion about themselves because they are
a committer on incubator project Y.
Dan.
---------------------------------------------------------------------
To unsubscribe, e-mail: general-unsubscribe@incubator.apache.org
For additional commands, e-mail: general-help@incubator.apache.org
Re: svn commit: r178965 -
/incubator/public/trunk/site-author/learn/newcommitters.html
Posted by "William A. Rowe, Jr." <wr...@rowe-clan.net>.
At 05:42 AM 7/7/2005, Roy T. Fielding wrote:
>On Jul 6, 2005, at 3:42 AM, Ted Husted wrote:
>
>>IMHO, the idea of discussing the vote on the PMC list and then having
>>the vote on the dev@ list sounds better on paper than in practice. In
>>practice, the discussion becomes the vote, and creating a second
>>thread on the DEV list is just going through the motions.
>
>Some times it is better for the public to see us "go through the
>motions", and thereby get a handle on how we make decisions.
>In particular, I believe that the public vote provides benefits
>when people have to earn status as a committer. I don't see it
>as useful when commit access is handed out like candy.
As much as I agree that incubator is a good place for new people
to see the dirty laundry, observe the nuances of participation,
and learn the normal project flow...
>Perhaps we could change it to private votes and a slightly more
>formal process for the announcement (along with a suggestion that
>the existing committers congratulate the newbie in public)?
...and I agree that we should trumpet new committers...
>[...] I think that forbidding a project from making official
>decisions in private has benefits far beyond either form of
>committer votes.
...and I agree that no project decisions should be invisible
to the wider community...
...I disagree that people/personality decisions are productive
or easily resolved on a public list. I'm a believer that you
praise in public, chastise in private, and there's too much
opportunity 1) for such processes to become a flogging, or
2) for valid criticism to go unsaid due to politics or a desire
not to be critical on a public forum.
>Right now I see this process as a mixed bag that is made much
>harder by the disjoint between the incubator PMC and the incubated
>projects' private lists. I wonder what would happen if we made
>a single "private@incubator" list for all the podlings instead
>of the separate ppmc lists?
or ppmc@incubator (until we rename these to private across the ASF).
I'll give a huge +1 to your idea though. I agree that only people
discussions and confidential negotiations belong on private@,
and by watching other ppmc's grapple with these issues, all of our
incubating projects will learn from one another.
Before anyone objects due to the S/N ratio of such a combined
list, remember it will let few mentors monitor more activities,
and posts to private@ should be (and usually are) short, and
to the point. If not, it's probably a dev@ discussion.
Bill
---------------------------------------------------------------------
To unsubscribe, e-mail: general-unsubscribe@incubator.apache.org
For additional commands, e-mail: general-help@incubator.apache.org
Re: svn commit: r178965 - /incubator/public/trunk/site-author/learn/newcommitters.html
Posted by Rodent of Unusual Size <Ke...@Golux.Com>.
-----BEGIN PGP SIGNED MESSAGE-----
Roy T. Fielding wrote:
>
> Perhaps we could change it to private votes and a slightly more
> formal process for the announcement (along with a suggestion that
> the existing committers congratulate the newbie in public)?
In the scenario I personally consider the best and healthiest,
committers == voters == PMC. Clearly stated. I find disparity
a little upsetting, and somewhat analogous to the common U.S. law
that you can be forced to serve your nation at 18, but aren't
trusted to govern yourself drinking until you're 21. If you're
trusted with the code, you should be trusted to the extent of
having a formal voice in project direction.
(Note that I didn't say I thought such disparity is *wrong*,
just that it upsets me personally. It may be the right thing
for a group.)
Unfortunately (again IMHO) that's not univeral across the ASF.
I think we have multiple cases of 'committers != voters == PMC',
some 'committer == voters != PMC', and maybe some cases of all
three being unequivalent.
In a (IMHO) better world, the discussion and vote would happen on the
open list among the voters. Again unfortunately, that has three
terminal drawbacks: it happens in the face of the candidate, it
makes the candidate appear graceless and ungrateful if it declines,
and it falls afoul of the Janus effect: people have a tendency to
feel entitled to confront their accusers, but also long for the
ability to make denunciations anonymously.
These are all social concerns rather than technical, and therefore
have a direct bearing and effect, I think, on the community aspect.
Since there is no ASF rule that the three sets be identical, being
elected a committer does not equate to being elected to the PMC.
And it's only changes to the PMC that might have a documentation
need for legal reasons. (And I'm not even sure about that.)
So to defuse the social concerns, I think that votes on people
either need to occur on private lists, or else anonymously on
public ones.
- --
#ken P-)}
Ken Coar, Sanagendamgagwedweinini http://Ken.Coar.Org/
Author, developer, opinionist http://Apache-Server.Com/
"Millennium hand and shrimp!"
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.2.4 (MingW32)
Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org
iQCVAwUBQs1cWprNPMCpn3XdAQE/LQP+I3dmvK/Ckxq1dqtmiZzBthC/qsIzwfFi
vBNTZ2hDxT/SB+WvqL2f7I3JQOY6nGQyXkcXu4vpO9/gXElvLkDKo1TGZSxQws97
NSJsJx9JaTeeqSLW3zAGGYQ0p1rfXdDzSYR8owbNdrhDBjh/xjEkUMFKyWMGDBN/
1B40p5U1AY0=
=ft7t
-----END PGP SIGNATURE-----
---------------------------------------------------------------------
To unsubscribe, e-mail: general-unsubscribe@incubator.apache.org
For additional commands, e-mail: general-help@incubator.apache.org
Re: svn commit: r178965 - /incubator/public/trunk/site-author/learn/newcommitters.html
Posted by "Roy T. Fielding" <fi...@gbiv.com>.
On Jul 6, 2005, at 3:42 AM, Ted Husted wrote:
> IMHO, the idea of discussing the vote on the PMC list and then having
> the vote on the dev@ list sounds better on paper than in practice. In
> practice, the discussion becomes the vote, and creating a second
> thread on the DEV list is just going through the motions.
Some times it is better for the public to see us "go through the
motions", and thereby get a handle on how we make decisions.
In particular, I believe that the public vote provides benefits
when people have to earn status as a committer. I don't see it
as useful when commit access is handed out like candy.
Perhaps we could change it to private votes and a slightly more
formal process for the announcement (along with a suggestion that
the existing committers congratulate the newbie in public)?
> While transparency is good, honesty is better. When we start another
> thread on dev@, we create the illusion that the decision is being made
> on dev@. In practice, the decision has already been made. The PMC has
> discussed the candidate, and the candidate should have been contacted.
> (Yes, some candidates do decline!)
>
> While the PMC list is not public, it is archived, and the record is
> available to anyone who really needs to know, like root or the board.
> A pmc@ vote is every bit as legal and binding as a dev@ vote.
Only if the PMC is allowed to make decisions in private. That was
certainly not the case when I wrote the httpd guidelines, though
I see that got lost in more recent edits. I think that forbidding
a project from making official decisions in private has benefits
far beyond either form of committer votes.
Right now I see this process as a mixed bag that is made much
harder by the disjoint between the incubator PMC and the incubated
projects' private lists. I wonder what would happen if we made
a single "private@incubator" list for all the podlings instead
of the separate ppmc lists?
....Roy
---------------------------------------------------------------------
To unsubscribe, e-mail: general-unsubscribe@incubator.apache.org
For additional commands, e-mail: general-help@incubator.apache.org
Re: svn commit: r178965 - /incubator/public/trunk/site-author/learn/newcommitters.html
Posted by Ted Husted <te...@gmail.com>.
On 7/2/05, Niclas Hedhman <ni...@hedhman.org> wrote:
> On Sunday 03 July 2005 04:44, Rodent of Unusual Size wrote:
> > This last bit troubles me a bit. I can't say I've seen this 're-emergence'
> > but that's probably my shortcoming in not tracking discussions.
>
> I second that having "shadow plays" isn't serving any practical purposes.
>
> In the same spirit, committership for some people is not necessarily something
> one wants, so keep a door open for the individual to decline privately before
> a public announcement.
+1
In one ASF project or another, I've participated in every
configuration of committer vote.
IMHO, the idea of discussing the vote on the PMC list and then having
the vote on the dev@ list sounds better on paper than in practice. In
practice, the discussion becomes the vote, and creating a second
thread on the DEV list is just going through the motions.
While transparency is good, honesty is better. When we start another
thread on dev@, we create the illusion that the decision is being made
on dev@. In practice, the decision has already been made. The PMC has
discussed the candidate, and the candidate should have been contacted.
(Yes, some candidates do decline!)
While the PMC list is not public, it is archived, and the record is
available to anyone who really needs to know, like root or the board.
A pmc@ vote is every bit as legal and binding as a dev@ vote.
In my years at the ASF, I've seen way too many public committer votes
go wrong. It's not unusual for the votes to go off topic, and
suddenly, we're discussing voting rather than the candidate. These
types of votes are embarrassing for everyone. Bad things do happen to
these votes, and bad things will happen to these votes. It's the
nature of the beast.
In this case, I believe the simplest practice is best. My advice is to
have the vote on the PMC list and be done with it.
-Ted.
---------------------------------------------------------------------
To unsubscribe, e-mail: general-unsubscribe@incubator.apache.org
For additional commands, e-mail: general-help@incubator.apache.org
Re: svn commit: r178965 - /incubator/public/trunk/site-author/learn/newcommitters.html
Posted by Niclas Hedhman <ni...@hedhman.org>.
On Sunday 03 July 2005 04:44, Rodent of Unusual Size wrote:
> This last bit troubles me a bit. I can't say I've seen this 're-emergence'
> but that's probably my shortcoming in not tracking discussions.
I second that having "shadow plays" isn't serving any practical purposes.
In the same spirit, committership for some people is not necessarily something
one wants, so keep a door open for the individual to decline privately before
a public announcement.
Cheers
Niclas
---------------------------------------------------------------------
To unsubscribe, e-mail: general-unsubscribe@incubator.apache.org
For additional commands, e-mail: general-help@incubator.apache.org
Re: svn commit: r178965 - /incubator/public/trunk/site-author/learn/newcommitters.html
Posted by Niclas Hedhman <ni...@hedhman.org>.
On Wednesday 06 July 2005 15:49, David Crossley wrote:
> Roy T. Fielding wrote:
> > [snip]... What the vote does is three things:
> >
> > 1) it provides a legally binding, public decision of the project;
>
> I am confused by that statement. We conduct our committer votes on
> the project's private mailing list, following a procedure that
> our PMC has agreed upon. Our PMC is the group that makes our decisions,
> and everyone in the group is on that mailing list. Is that not legal?
I think Roy meant; The public list is transparent and is available in public
mail archives for anyone to scrutinize. The private PMC lists are not open to
outsiders, hence not a good way to promote transparency.
(Not sure what needs to be 'legal' in this case, since IMHO this is internal
affairs of ASF, that doesn't break any laws. :o) )
Cheers
Niclas
---------------------------------------------------------------------
To unsubscribe, e-mail: general-unsubscribe@incubator.apache.org
For additional commands, e-mail: general-help@incubator.apache.org
Re: svn commit: r178965 - /incubator/public/trunk/site-author/learn/newcommitters.html
Posted by Rodent of Unusual Size <Ke...@Golux.Com>.
-----BEGIN PGP SIGNED MESSAGE-----
Roy T. Fielding wrote:
> On Jul 2, 2005, at 1:44 PM, Rodent of Unusual Size wrote:
>> I don't like the idea of discussions about people's merit (or the
>> people
>> themselves, for that matter) happening in public right in their faces.
>
> That doesn't happen -- discussion takes place in private before
> the vote even starts, just like it did in the old Apache Group.
You're missing the point, I think -- I'm talking about the 'pro forma'
vote that happens on the *public* list. There, people who haven't
participated in the private vote can do *anything*. I don't think
there should be a public vote at all.
- --
#ken P-)}
Ken Coar, Sanagendamgagwedweinini http://Ken.Coar.Org/
Author, developer, opinionist http://Apache-Server.Com/
"Millennium hand and shrimp!"
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.2.4 (MingW32)
Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org
iQCVAwUBQsvKl5rNPMCpn3XdAQGY/AP/Q7BEsNHLalCpd+0kr3bfnoMc+fYms6kl
xxvZZEl0wOITgcpNG1ye72SHL9wJRdHJnFrB5PvpykD+9n07/hH/D/jETKNtZYGI
w06HWAFeFdfIpWWJBRuspd/WKKmGqyGuh6opODum9d46t+Z79SjmgWXzJNbEeSsp
S8+q1QSL6KY=
=KiuQ
-----END PGP SIGNATURE-----
---------------------------------------------------------------------
To unsubscribe, e-mail: general-unsubscribe@incubator.apache.org
For additional commands, e-mail: general-help@incubator.apache.org
RE: svn commit: r178965 -
/incubator/public/trunk/site-author/learn/newcommitters.html
Posted by Justin Erenkrantz <ju...@erenkrantz.com>.
--On July 7, 2005 3:31:15 PM -0500 "William A. Rowe, Jr."
<wr...@rowe-clan.net> wrote:
> So that Joe knows he's much more popular with 29 votes than
> poor Sam who only had 15 supporters?
>
> No thank you. Didn't we just decide this w.r.t. Membership
> votes and announcements?
+1. I don't like naming names - which is sort of the point of having the
discussions in private. -- justin
---------------------------------------------------------------------
To unsubscribe, e-mail: general-unsubscribe@incubator.apache.org
For additional commands, e-mail: general-help@incubator.apache.org
RE: svn commit: r178965 -
/incubator/public/trunk/site-author/learn/newcommitters.html
Posted by "William A. Rowe, Jr." <wr...@rowe-clan.net>.
At 02:16 PM 7/7/2005, Noel J. Bergman wrote:
>Roy T. Fielding wrote:
>
>> I understand the concerns, but I still feel that project decisions
>> should always be made in public. How about if the vote takes place
>> in private and the announcement includes the final vote naming
>> those who voted?
>
>Since the vote requires a consensus, and therefore there are no -1 votes
>cast if it is successful, that should work, since it is documenting the
>consensus.
So that Joe knows he's much more popular with 29 votes than
poor Sam who only had 15 supporters?
No thank you. Didn't we just decide this w.r.t. Membership
votes and announcements?
---------------------------------------------------------------------
To unsubscribe, e-mail: general-unsubscribe@incubator.apache.org
For additional commands, e-mail: general-help@incubator.apache.org
Re: svn commit: r178965 - /incubator/public/trunk/site-author/learn/newcommitters.html
Posted by "Jean T. Anderson" <jt...@bristowhill.com>.
William A. Rowe, Jr. wrote:
> At 01:12 PM 7/7/2005, Jean T. Anderson wrote:
>
>
>>For example, from various list threads (including some on this list) I'm gathering that a private list is necessary for discussions about:
>>
>> - personnel decisions
>> - legal points
>> - negotiations (could use a little expansion)
>
>
> Personnel decisions is, of course, the issue we are kicking
> back and forth here...
>
> Legal points; we would want to publicly discuss anything about
> licensing, regulatory compliance, etc. The community needs to
> understand how we came to whatever conclusion is reached.
> We would not want to discuss reacting to a cease & desist notice,
> etc, on the public forum; these would be forwarded to our internal
> legal team for reaction.
>
> Negotiations; If company "X" was contemplating a contribution,
> yet wasn't quite sure if the ASF was the best avenue, then of
> course that company might ask us to keep it private. Similarly,
> negotiating licensing terms etc with patent holders would be a
> private matter if they asked us.
>
> Security issues; If a reporter comes to us with a vulnerability
> prior to it's disclosure, we would want to address that privately
> for the benefit of all of our users, until a fix was found. In
> some cases, the reporter wishes for us to keep it quiet a while
> longer, while other vendors/projects with similar faults correct
> their bugs, and the final announcement/release is coordinated.
>
> The bottom line is this; must it be private? If not, then the
> entire community deserves to observe the process and contribute
> their input, and see the history of how we came to where we are.
>
> Bill
this is excellent clarification.
thanks,
-jean
---------------------------------------------------------------------
To unsubscribe, e-mail: general-unsubscribe@incubator.apache.org
For additional commands, e-mail: general-help@incubator.apache.org
Re: svn commit: r178965 -
/incubator/public/trunk/site-author/learn/newcommitters.html
Posted by "William A. Rowe, Jr." <wr...@rowe-clan.net>.
At 01:12 PM 7/7/2005, Jean T. Anderson wrote:
>For example, from various list threads (including some on this list) I'm gathering that a private list is necessary for discussions about:
>
> - personnel decisions
> - legal points
> - negotiations (could use a little expansion)
Personnel decisions is, of course, the issue we are kicking
back and forth here...
Legal points; we would want to publicly discuss anything about
licensing, regulatory compliance, etc. The community needs to
understand how we came to whatever conclusion is reached.
We would not want to discuss reacting to a cease & desist notice,
etc, on the public forum; these would be forwarded to our internal
legal team for reaction.
Negotiations; If company "X" was contemplating a contribution,
yet wasn't quite sure if the ASF was the best avenue, then of
course that company might ask us to keep it private. Similarly,
negotiating licensing terms etc with patent holders would be a
private matter if they asked us.
Security issues; If a reporter comes to us with a vulnerability
prior to it's disclosure, we would want to address that privately
for the benefit of all of our users, until a fix was found. In
some cases, the reporter wishes for us to keep it quiet a while
longer, while other vendors/projects with similar faults correct
their bugs, and the final announcement/release is coordinated.
The bottom line is this; must it be private? If not, then the
entire community deserves to observe the process and contribute
their input, and see the history of how we came to where we are.
Bill
---------------------------------------------------------------------
To unsubscribe, e-mail: general-unsubscribe@incubator.apache.org
For additional commands, e-mail: general-help@incubator.apache.org
Re: svn commit: r178965 - /incubator/public/trunk/site-author/learn/newcommitters.html
Posted by "Jean T. Anderson" <jt...@bristowhill.com>.
William A. Rowe, Jr. wrote:
> At 12:16 PM 7/7/2005, Ted Husted wrote:
>
>>On 7/7/05, Roy T. Fielding <fi...@gbiv.com> wrote:
>>
>>>I don't believe we should have PMC
>>>lists (only dev and private), which is why your scenario doesn't
>>>apply in my model of how a project works.
>>
>>Could someone clarify on the difference between "private" and "pmc".
>
>
> I can take this question, I was paying attention :)
>
> Roy has been concerned for a -long- while that too much traffic
> occurs on pmc, and believes it's frequently abused.
>
> He's proposed that we change to private@ to make it perfectly
> clear that only private communications occurs there, and all the
> other day-to-day business of the pmc would be more likely to occur
> on the dev list if the list's name wasn't so abstract.
>
> And, I concur.
>
It isn't always clear to me what should be "private communication".
Perhaps some clarifying wording could be added to the "Balancing
confidentiality and public discussion" section of
http://www.apache.org/foundation/how-it-works.html#management .
For example, from various list threads (including some on this list) I'm
gathering that a private list is necessary for discussions about:
- personnel decisions
- legal points
- negotiations (could use a little expansion)
What are typical examples of "abuse" (so we can avoid them)?
thanks,
-jean
---------------------------------------------------------------------
To unsubscribe, e-mail: general-unsubscribe@incubator.apache.org
For additional commands, e-mail: general-help@incubator.apache.org
Re: svn commit: r178965 -
/incubator/public/trunk/site-author/learn/newcommitters.html
Posted by "William A. Rowe, Jr." <wr...@rowe-clan.net>.
At 12:16 PM 7/7/2005, Ted Husted wrote:
>On 7/7/05, Roy T. Fielding <fi...@gbiv.com> wrote:
>> I don't believe we should have PMC
>> lists (only dev and private), which is why your scenario doesn't
>> apply in my model of how a project works.
>
>Could someone clarify on the difference between "private" and "pmc".
I can take this question, I was paying attention :)
Roy has been concerned for a -long- while that too much traffic
occurs on pmc, and believes it's frequently abused.
He's proposed that we change to private@ to make it perfectly
clear that only private communications occurs there, and all the
other day-to-day business of the pmc would be more likely to occur
on the dev list if the list's name wasn't so abstract.
And, I concur.
Bill
---------------------------------------------------------------------
To unsubscribe, e-mail: general-unsubscribe@incubator.apache.org
For additional commands, e-mail: general-help@incubator.apache.org
Re: svn commit: r178965 - /incubator/public/trunk/site-author/learn/newcommitters.html
Posted by Ted Husted <te...@gmail.com>.
On 7/7/05, Roy T. Fielding <fi...@gbiv.com> wrote:
> I don't believe we should have PMC
> lists (only dev and private), which is why your scenario doesn't
> apply in my model of how a project works.
Could someone clarify on the difference between "private" and "pmc".
Is it just nomenclature, or would a private list include non-pmc
members and/or non-committers?
On some of my projects, we've started to subscribe the new committers
to pmc@. This keeps everyone in the loop and helps transition someone
from committer to pmc member. Of course, in the normal course, we
expect that all committers will be voted into the PMC.
The committers==pmc is the best goal, but in practice it often makes
better sense to step someone up from developer to committer to PMC
member, rather than try to do it in one fell swoop.
-Ted.
---------------------------------------------------------------------
To unsubscribe, e-mail: general-unsubscribe@incubator.apache.org
For additional commands, e-mail: general-help@incubator.apache.org
Re: svn commit: r178965 - /incubator/public/trunk/site-author/learn/newcommitters.html
Posted by Rodent of Unusual Size <Ke...@Golux.Com>.
-----BEGIN PGP SIGNED MESSAGE-----
Roy T. Fielding wrote:
>>
>> I think this is a particularly likely scenario when the
>> voters-who-count != the project committers -- i.e., where the
>> PMC != the committers. Those non-on-the-PMC committers are probably
>> going to feel that their votes should count -- but they won't.
>
> Well, now, whose fault is that?
I'm going to get into a finger-pointing process. Those situations
exist; I don't think assigning blame is productive. If they're
considered officially pathological, then taking steps to correct
those in existence and keep more from spawning would be appropriate.
> I understand the concerns, but I still feel that project decisions
> should always be made in public. How about if the vote takes place
> in private and the announcement includes the final vote naming
> those who voted?
Or have the voting occur on the public list -- by anonymous vote.
- --
#ken P-)}
Ken Coar, Sanagendamgagwedweinini http://Ken.Coar.Org/
Author, developer, opinionist http://Apache-Server.Com/
"Millennium hand and shrimp!"
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.2.4 (MingW32)
Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org
iQCVAwUBQs1dt5rNPMCpn3XdAQE9DwP/ZyQWXPWsOpBIomBYGB7+9T+/DMZeL/+M
ZM6ce8TV5EqSp6s+Pcnld+j3MzWHqAYRFuHonBJxOnvDKVPahGtNscOQrAxTPiGP
R1UtHeItPaVMoEKn3ORjCn5JOQ5n8N9p51Z7gfkNO6mPQc8kE+82/C5jSiyYubqZ
RC6vO8uQVkw=
=76f1
-----END PGP SIGNATURE-----
---------------------------------------------------------------------
To unsubscribe, e-mail: general-unsubscribe@incubator.apache.org
For additional commands, e-mail: general-help@incubator.apache.org
RE: svn commit: r178965 - /incubator/public/trunk/site-author/learn/newcommitters.html
Posted by "Noel J. Bergman" <no...@devtech.com>.
Roy T. Fielding wrote:
> I understand the concerns, but I still feel that project decisions
> should always be made in public. How about if the vote takes place
> in private and the announcement includes the final vote naming
> those who voted?
Since the vote requires a consensus, and therefore there are no -1 votes
cast if it is successful, that should work, since it is documenting the
consensus.
--- Noel
---------------------------------------------------------------------
To unsubscribe, e-mail: general-unsubscribe@incubator.apache.org
For additional commands, e-mail: general-help@incubator.apache.org
Re: svn commit: r178965 - /incubator/public/trunk/site-author/learn/newcommitters.html
Posted by "Roy T. Fielding" <fi...@gbiv.com>.
On Jul 6, 2005, at 5:21 AM, Rodent of Unusual Size wrote:
> I think I prefer the private vote, acceptance by the candidate,
> followed by a public announcement. Or else a pro forma vote
> which explicitly states 'only PMC members can vote.' (Which I
> don't think is a particularly community-building approach.)
>
> I think this is a particularly likely scenario when the
> voters-who-count != the project committers -- i.e., where the
> PMC != the committers. Those non-on-the-PMC committers are probably
> going to feel that their votes should count -- but they won't.
Well, now, whose fault is that? People who have the right to vote
on a project had better be on the private list, since otherwise
they won't be informed. I don't believe we should have PMC
lists (only dev and private), which is why your scenario doesn't
apply in my model of how a project works. In my model, the only
time that PMC != committers is when a given person is provided
commit access for a specific purpose and they should be informed
of those limitations at that time.
I understand the concerns, but I still feel that project decisions
should always be made in public. How about if the vote takes place
in private and the announcement includes the final vote naming
those who voted?
....Roy
---------------------------------------------------------------------
To unsubscribe, e-mail: general-unsubscribe@incubator.apache.org
For additional commands, e-mail: general-help@incubator.apache.org
Re: svn commit: r178965 - /incubator/public/trunk/site-author/learn/newcommitters.html
Posted by Rodent of Unusual Size <Ke...@Golux.Com>.
-----BEGIN PGP SIGNED MESSAGE-----
Noel J. Bergman wrote:
> Ken expressed concern:
>
>> The candidate is now in the position of being privately
>> blessed but publicly vilified.
>
> And this will differ how from an announcement that someone has been accepted
> as a Committer, and people who did not participate in the private vote
> expressing their views anyway?
In the latter case, they will be arguing with a decision, rather than
mistakenly thinking they are making it.
> Please note that the text did offer either approach as acceptable, and up to
> each PMC, so I am not entirely sure of the real point of contention. If the
> PMC prefers one over the other, use it.
I'm contending that on the one hand the allegation of it being accepted
as 'best practice' may be less than well founded, and on the other hand
that while it's a possibility it's not one we should recommend.
- --
#ken P-)}
Ken Coar, Sanagendamgagwedweinini http://Ken.Coar.Org/
Author, developer, opinionist http://Apache-Server.Com/
"Millennium hand and shrimp!"
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.2.4 (MingW32)
Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org
iQCVAwUBQs1WnZrNPMCpn3XdAQHMqQP+PuRduqxLp18gIetC4O/0F1yrZ0NnPqG4
paUe0w3sfJsnTdCNaRlf7GmmQbHI8AS6zmiNJVEhgRF44FVY8qF+9d2GTdkQNuCQ
Gs8UzO3urlz2H51CnRxgB4dhYU/DqgLyd631vvbFNWE6drI/MzlygInSDAj9Fngx
UPBAB4d58EQ=
=Pk7O
-----END PGP SIGNATURE-----
---------------------------------------------------------------------
To unsubscribe, e-mail: general-unsubscribe@incubator.apache.org
For additional commands, e-mail: general-help@incubator.apache.org
Re: Transparency in voting and reporting (was Re: svn commit: r178965 )
Posted by "Geir Magnusson Jr." <ge...@apache.org>.
I always copy Geronimo to pmc@, with the theory that if there's ASF-
only information then it's contained via board and pmc being member-
private-at-worst, and anything that is public will come out in the
minutes.
geir
On Jul 8, 2005, at 6:33 AM, Ted Husted wrote:
> On 7/7/05, Noel J. Bergman <no...@devtech.com> wrote:
>
>> And by documenting the existing practices, I "catalyzed" this
>> discussion,
>> which may result in further change. Roy seems to be proposing a
>> compromise
>> of sorts, where the discussion and vote are on the private list,
>> but the
>> announcement on the public list includes the itemized result of the
>> positive, private, vote.
>>
>
> On a related topic:
>
> * Should the status report to the board be copied to pmc/private@,
> dev@, user@, or not copied elsewhere?
>
> -Ted.
>
> ---------------------------------------------------------------------
> To unsubscribe, e-mail: general-unsubscribe@incubator.apache.org
> For additional commands, e-mail: general-help@incubator.apache.org
>
>
--
Geir Magnusson Jr +1-203-665-6437
geirm@apache.org
---------------------------------------------------------------------
To unsubscribe, e-mail: general-unsubscribe@incubator.apache.org
For additional commands, e-mail: general-help@incubator.apache.org
Transparency in voting and reporting (was Re: svn commit: r178965 )
Posted by Ted Husted <te...@gmail.com>.
On 7/7/05, Noel J. Bergman <no...@devtech.com> wrote:
> And by documenting the existing practices, I "catalyzed" this discussion,
> which may result in further change. Roy seems to be proposing a compromise
> of sorts, where the discussion and vote are on the private list, but the
> announcement on the public list includes the itemized result of the
> positive, private, vote.
On a related topic:
* Should the status report to the board be copied to pmc/private@,
dev@, user@, or not copied elsewhere?
-Ted.
---------------------------------------------------------------------
To unsubscribe, e-mail: general-unsubscribe@incubator.apache.org
For additional commands, e-mail: general-help@incubator.apache.org
RE: svn commit: r178965 - /incubator/public/trunk/site-author/learn/newcommitters.html
Posted by "Noel J. Bergman" <no...@devtech.com>.
Daniel John Debrunner wrote:
> The history of this discussion was that folks were somewhat unhappy that
> the Derby PPMC voted a committer in without following either of the ways
> described
I do not fault the Derby PPMC in any way. It has made every effort.
Sometimes changes occur as a reaction. Voting for Committers on a public
list has always been viewed as an anti-pattern, and came out of Jakarta,
which had no choice. So there has been a re-learning effort across the ASF.
> So to make "every one" happy we voted to changed to the "ASF
> best practice" as described on the incubator site, and then
> still there are complaints.
And by documenting the existing practices, I "catalyzed" this discussion,
which may result in further change. Roy seems to be proposing a compromise
of sorts, where the discussion and vote are on the private list, but the
announcement on the public list includes the itemized result of the
positive, private, vote.
> As an incubator project it sometimes feels you can't win
> because there is no single idea of how an ASF project
> should be run.
Any placing of blame on an incubator project for attempting to follow
current practices as best it understands is entirely misplaced.
The Incubator exists to help projects follow he Apache Way (including
clearing IP). The Incubator is all about Community. Code cannot be of any
consequence; we have far too broad a disjoint spectrum of projects. All
that can matter is that we help each to evolve into a properly governed ASF
project. In so doing, some practices that exist to help us implement and
maintain our core philosophy evolve, too.
Consider that the establishment of the PPMC structure came out of issues
dealing with Geronimo and other scaling of the Incubator. Derby had an
effect on how copyright is handled throughout the ASF. And now we are
examining how Committers are voted into place.
I would worry if we our practices were unreasonably static, rather than
seeing in operation a process that can help keep the ASF vibrant for the
long term. Here we have Roy --- an ASF founder and a past Chairman of the
Board --- and other long-established participants evolving via discussion,
presentation of concerns, and a sincere effort to build a consensus how to
address one of the oldest needs in a forum intended to inform the newest
participants.
In my view, these are all good things. And I hold Derby in high regard for
its contributions and best efforts.
--- Noel
---------------------------------------------------------------------
To unsubscribe, e-mail: general-unsubscribe@incubator.apache.org
For additional commands, e-mail: general-help@incubator.apache.org
Re: svn commit: r178965 - /incubator/public/trunk/site-author/learn/newcommitters.html
Posted by Daniel John Debrunner <dj...@debrunners.com>.
Noel J. Bergman wrote:
> Ken expressed concern:
>
> Please note that the text did offer either approach as acceptable, and up to
> each PMC, so I am not entirely sure of the real point of contention. If the
> PMC prefers one over the other, use it.
But only one is described as '... Best Practice for ASF projects ...'
and since text is aimed at incubator projects, it seems such projects
will try to follow the ASF best practice, so it's not really a balanced
choice between the two.
The history of this discussion was that folks were somewhat unhappy that
the Derby PPMC voted a committer in without following either of the ways
described, even though that document was added after the vote was
complete! So to make "every one" happy we voted to changed to the "ASF
best practice" as described on the incubator site, and then still there
are complaints. As an incubator project it sometimes feels you can't win
because there is no single idea of how an ASF project should be run.
Dan.
---------------------------------------------------------------------
To unsubscribe, e-mail: general-unsubscribe@incubator.apache.org
For additional commands, e-mail: general-help@incubator.apache.org
RE: svn commit: r178965 - /incubator/public/trunk/site-author/learn/newcommitters.html
Posted by "Noel J. Bergman" <no...@devtech.com>.
Ken expressed concern:
> The candidate is now in the position of being privately
> blessed but publicly vilified.
And this will differ how from an announcement that someone has been accepted
as a Committer, and people who did not participate in the private vote
expressing their views anyway?
> Or else a pro forma vote
As the author who initiated this use of the term pro forma, I am fine to
remove the adjective. My thought was that if we've had the discussion in
private, the public vote really is pro-forma, since it would not have
occurred unless there were a consensus for the positive outcome.
Please note that the text did offer either approach as acceptable, and up to
each PMC, so I am not entirely sure of the real point of contention. If the
PMC prefers one over the other, use it.
> non-on-the-PMC committers are probably going to feel
> that their votes should count -- but they won't.
Nor do they with code, either. Committers should want to be on the PMC.
IMO, this is an education issue that needs to be better understood across
the ASF, especially as we continue to dismantle the umbrellas that created
the issue. We want all new projects to start off with the desired approach.
--- Noel
---------------------------------------------------------------------
To unsubscribe, e-mail: general-unsubscribe@incubator.apache.org
For additional commands, e-mail: general-help@incubator.apache.org
Re: svn commit: r178965 - /incubator/public/trunk/site-author/learn/newcommitters.html
Posted by Rodent of Unusual Size <Ke...@Golux.Com>.
-----BEGIN PGP SIGNED MESSAGE-----
Roy T. Fielding wrote:
>
> That doesn't happen -- discussion takes place in private before
> the vote even starts, just like it did in the old Apache Group.
Oh, perhaps you're assuming that the only people permitted to
participate in the 'pro forma' vote are the ones whoe participated
in the private one. If so, that is *far* from clear to me, particularly
in the (I fear too common) case of a project for which 'who can vote'
set isn't clearly or explicitly defined.
*I'm* concerned about the effect of people voting on the public
list who weren't on the private one. They may not *know* that it's
lip-service and that their vote doesn't count.
And it still sets up what I consider to be an unfortunate situation
for the candidate. Told that the private list has voted him for
committer, and he accepts, and then there's the pro forma public
vote, and *there* people express concern. The candidate is now in
the position of being privately blessed but publicly vilified.
I think I prefer the private vote, acceptance by the candidate,
followed by a public announcement. Or else a pro forma vote
which explicitly states 'only PMC members can vote.' (Which I
don't think is a particularly community-building approach.)
I think this is a particularly likely scenario when the
voters-who-count != the project committers -- i.e., where the
PMC != the committers. Those non-on-the-PMC committers are probably
going to feel that their votes should count -- but they won't.
- --
#ken P-)}
Ken Coar, Sanagendamgagwedweinini http://Ken.Coar.Org/
Author, developer, opinionist http://Apache-Server.Com/
"Millennium hand and shrimp!"
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.2.4 (MingW32)
Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org
iQCVAwUBQsvM3ZrNPMCpn3XdAQHzBgQAiI866GBYynOiTRnvvUJgPYROCv3gGh4l
10Ms6DyNJEjYt12YzZlbdMRcAFHpSx36blySV4cZxigzAShXuSELJUnDIdKMhz9/
Y6eGAS/nBpNmQxXUGGA5UY8HdOAbfOFMP9cs3GSyQ15XJWT2dt7p8la4KWeGa4+v
nihY+2BB5eY=
=KRNk
-----END PGP SIGNATURE-----
---------------------------------------------------------------------
To unsubscribe, e-mail: general-unsubscribe@incubator.apache.org
For additional commands, e-mail: general-help@incubator.apache.org
Re: svn commit: r178965 - /incubator/public/trunk/site-author/learn/newcommitters.html
Posted by David Crossley <cr...@apache.org>.
Noel J. Bergman wrote:
> David Crossley wrote:
>
> > Does the vote need to happen in view of the general public
> > for that to be considered a legal decision of the project PMC?
>
> Of course not, nor have I seen anything that would inspire such a question.
Great, glad that is resolved. I could see that Roy's comment could
be misinterpreted. So that is not good to leave dangling in the
incubator mail archives. Thanks for helping to clarify.
-David
---------------------------------------------------------------------
To unsubscribe, e-mail: general-unsubscribe@incubator.apache.org
For additional commands, e-mail: general-help@incubator.apache.org
RE: svn commit: r178965 - /incubator/public/trunk/site-author/learn/newcommitters.html
Posted by "Noel J. Bergman" <no...@devtech.com>.
David Crossley wrote:
> Does the vote need to happen in view of the general public
> for that to be considered a legal decision of the project PMC?
Of course not, nor have I seen anything that would inspire such a question.
To the contrary, the commit that people are discussing documented two
approaches, one of which was an entirely private vote.
--- Noel
---------------------------------------------------------------------
To unsubscribe, e-mail: general-unsubscribe@incubator.apache.org
For additional commands, e-mail: general-help@incubator.apache.org
Re: svn commit: r178965 - /incubator/public/trunk/site-author/learn/newcommitters.html
Posted by David Crossley <cr...@apache.org>.
Noel J. Bergman wrote:
> David Crossley wrote:
>
> > Roy T. Fielding wrote:
> > > [snip]... What the vote does is three things:
> > > 1) it provides a legally binding, public decision of the project;
>
> > I am confused by that statement. We conduct our committer votes on
> > the project's private mailing list
>
> > Is that not legal?
>
> Notice that Roy said legal AND public.
Gee, this is getting very cryptic.
How about i just ask a straight question...
Does the vote need to happen in view of the general public
for that to be considered a legal decision of the project PMC?
Answering myself, i say no: whatever way the PMC does it
is entirely up to them.
Does anyone have a different answer?
> Haven't you ever noticed the seeming
> disconnect between having the discussion (and vote) in private, and the
> instructions to the PMC that they should send a link to the vote thread as
> part of the account request?
The "link" that we provide is the date that the vote
occurred on our pmc list. Perhaps we should also provide
the MessageId. So no, i don't see a disconnect other than
the work link is a bit ambiguous. I take it to mean:
"provide a way to correlate the PMC decision with this action
of creating a new committer account".
-David
---------------------------------------------------------------------
To unsubscribe, e-mail: general-unsubscribe@incubator.apache.org
For additional commands, e-mail: general-help@incubator.apache.org
RE: svn commit: r178965 -
/incubator/public/trunk/site-author/learn/newcommitters.html
Posted by "William A. Rowe, Jr." <wr...@rowe-clan.net>.
At 10:24 PM 7/6/2005, Noel J. Bergman wrote:
>Notice that Roy said legal AND public. Haven't you ever noticed the seeming
>disconnect between having the discussion (and vote) in private, and the
>instructions to the PMC that they should send a link to the vote thread as
>part of the account request?
root has access to all mail archives, public and private.
---------------------------------------------------------------------
To unsubscribe, e-mail: general-unsubscribe@incubator.apache.org
For additional commands, e-mail: general-help@incubator.apache.org
Re: svn commit: r178965 -
/incubator/public/trunk/site-author/learn/newcommitters.html
Posted by "William A. Rowe, Jr." <wr...@rowe-clan.net>.
At 02:49 AM 7/6/2005, David Crossley wrote:
>Roy T. Fielding wrote:
>>
>> [snip]... What the vote does is three things:
>>
>> 1) it provides a legally binding, public decision of the project;
>
>I am confused by that statement. We conduct our committer votes on
>the project's private mailing list, following a procedure that
>our PMC has agreed upon. Our PMC is the group that makes our decisions,
>and everyone in the group is on that mailing list. Is that not legal?
As all PMC traffic is also archived, of course it's legal.
It isn't (as Roy points out) public, but private archives
are equally binding.
In fact, most actions by the PMC Chairman (a VP of the ASF)
are binding. Ideally the PMC Chairman executes the will of
the project. This is why people-requests to our infrastructure
or root should always come from the Chair him-or-herself.
Bill
---------------------------------------------------------------------
To unsubscribe, e-mail: general-unsubscribe@incubator.apache.org
For additional commands, e-mail: general-help@incubator.apache.org
RE: svn commit: r178965 - /incubator/public/trunk/site-author/learn/newcommitters.html
Posted by "Noel J. Bergman" <no...@devtech.com>.
David Crossley wrote:
> Roy T. Fielding wrote:
> > [snip]... What the vote does is three things:
> > 1) it provides a legally binding, public decision of the project;
> I am confused by that statement. We conduct our committer votes on
> the project's private mailing list
> Is that not legal?
Notice that Roy said legal AND public. Haven't you ever noticed the seeming
disconnect between having the discussion (and vote) in private, and the
instructions to the PMC that they should send a link to the vote thread as
part of the account request?
--- Noel
---------------------------------------------------------------------
To unsubscribe, e-mail: general-unsubscribe@incubator.apache.org
For additional commands, e-mail: general-help@incubator.apache.org
Re: svn commit: r178965 - /incubator/public/trunk/site-author/learn/newcommitters.html
Posted by David Crossley <cr...@apache.org>.
Roy T. Fielding wrote:
>
> [snip]... What the vote does is three things:
>
> 1) it provides a legally binding, public decision of the project;
I am confused by that statement. We conduct our committer votes on
the project's private mailing list, following a procedure that
our PMC has agreed upon. Our PMC is the group that makes our decisions,
and everyone in the group is on that mailing list. Is that not legal?
A note for the mail archives: there is a parallel thread at
http://marc.theaimsgroup.com/?t=111936409900007
-David
---------------------------------------------------------------------
To unsubscribe, e-mail: general-unsubscribe@incubator.apache.org
For additional commands, e-mail: general-help@incubator.apache.org
Re: svn commit: r178965 - /incubator/public/trunk/site-author/learn/newcommitters.html
Posted by "Roy T. Fielding" <fi...@gbiv.com>.
On Jul 2, 2005, at 1:44 PM, Rodent of Unusual Size wrote:
> I don't like the idea of discussions about people's merit (or the
> people
> themselves, for that matter) happening in public right in their faces.
That doesn't happen -- discussion takes place in private before
the vote even starts, just like it did in the old Apache Group.
> For one thing, people have a tendency to be less candid; for another
> there's no anonymity to the vote; and finally (and most important) I
> think it's unfair to the individual in question, who will get all
> anxious
> and wound up as the voting proceeds. 'Will I make it?' 'What's *his*
> problem?' and other such distractions from coding.
Again, that just doesn't happen. If there are objections, the vote
doesn't occur at all. What the vote does is three things:
1) it provides a legally binding, public decision of the project;
2) it provides group affirmation to the individual;
3) it makes the entire project feel good, including both long-time
committers and newbies lurking on the sidelines.
I remembered that from early httpd. After some of the other projects
described how their voting actually works, I applied it to Jackrabbit
and, sure enough, it has the desired effect without exposing the
prior private discussions. The only problem we've had is keeping the
separate Incubator PMC in the loop.
....Roy
---------------------------------------------------------------------
To unsubscribe, e-mail: general-unsubscribe@incubator.apache.org
For additional commands, e-mail: general-help@incubator.apache.org