You are viewing a plain text version of this content. The canonical link for it is here.
Posted to general@incubator.apache.org by ant elder <an...@gmail.com> on 2013/07/09 23:20:50 UTC
Podling new committer votes
Looking at the "Voting in a new committer" section at
http://incubator.apache.org/guides/ppmc.html i see that its been
updated since i last looked and now says the Incubator PMC should be
notified twice when a podling is voting in a new committer, once at
the start of the vote and again with the result. Is this really
necessary? The last big discussion about this that i recall was
http://apache.markmail.org/message/v5gwhbpbifqcafqu which resulted in
a single notification, before that in the "experiment" that Joe
started with podlings doing their own committer votes there wasn't any
requirement to notify the Incubator PMC at all -
http://apache.markmail.org/message/mihzgp7jfclj2dcv. So I'd like to
remove this dual notification suggestion from that ppmc guide page
again. Comments?
...ant
---------------------------------------------------------------------
To unsubscribe, e-mail: general-unsubscribe@incubator.apache.org
For additional commands, e-mail: general-help@incubator.apache.org
Re: Podling new committer votes
Posted by Ross Gardler <rg...@opendirective.com>.
+1 the IPMC should not be interfering and thus there is no need to be
involved. Mentors should be monitoring podling activity. IPMC members with
an interest should join the community lists.
On 9 Jul 2013 14:44, "Luciano Resende" <lu...@gmail.com> wrote:
> On Tue, Jul 9, 2013 at 2:20 PM, ant elder <an...@gmail.com> wrote:
>
> > Looking at the "Voting in a new committer" section at
> > http://incubator.apache.org/guides/ppmc.html i see that its been
> > updated since i last looked and now says the Incubator PMC should be
> > notified twice when a podling is voting in a new committer, once at
> > the start of the vote and again with the result. Is this really
> > necessary? The last big discussion about this that i recall was
> > http://apache.markmail.org/message/v5gwhbpbifqcafqu which resulted in
> > a single notification, before that in the "experiment" that Joe
> > started with podlings doing their own committer votes there wasn't any
> > requirement to notify the Incubator PMC at all -
> > http://apache.markmail.org/message/mihzgp7jfclj2dcv. So I'd like to
> > remove this dual notification suggestion from that ppmc guide page
> > again. Comments?
> >
> > ...ant
> >
> >
> Agree that initial notification is unnecessary, particularly considering
> that we will have the mentors oversight and still get notified at the end.
>
>
> --
> Luciano Resende
> http://people.apache.org/~lresende
> http://twitter.com/lresende1975
> http://lresende.blogspot.com/
>
Re: Podling new committer votes
Posted by Luciano Resende <lu...@gmail.com>.
On Tue, Jul 9, 2013 at 2:20 PM, ant elder <an...@gmail.com> wrote:
> Looking at the "Voting in a new committer" section at
> http://incubator.apache.org/guides/ppmc.html i see that its been
> updated since i last looked and now says the Incubator PMC should be
> notified twice when a podling is voting in a new committer, once at
> the start of the vote and again with the result. Is this really
> necessary? The last big discussion about this that i recall was
> http://apache.markmail.org/message/v5gwhbpbifqcafqu which resulted in
> a single notification, before that in the "experiment" that Joe
> started with podlings doing their own committer votes there wasn't any
> requirement to notify the Incubator PMC at all -
> http://apache.markmail.org/message/mihzgp7jfclj2dcv. So I'd like to
> remove this dual notification suggestion from that ppmc guide page
> again. Comments?
>
> ...ant
>
>
Agree that initial notification is unnecessary, particularly considering
that we will have the mentors oversight and still get notified at the end.
--
Luciano Resende
http://people.apache.org/~lresende
http://twitter.com/lresende1975
http://lresende.blogspot.com/
Re: Podling new committer votes
Posted by Bertrand Delacretaz <bd...@apache.org>.
On Wed, Jul 31, 2013 at 6:43 PM, Marvin Humphrey <ma...@rectangular.com> wrote:
> ...going
> forward, I'd like to suggest that we try to harmonize Incubator podling
> reporting with the Board's guidelines for TLPs....
+1 assuming the docs are updated to reflect this.
-Bertrand
---------------------------------------------------------------------
To unsubscribe, e-mail: general-unsubscribe@incubator.apache.org
For additional commands, e-mail: general-help@incubator.apache.org
Re: Podling new committer votes
Posted by ant elder <an...@gmail.com>.
On Wed, Aug 7, 2013 at 1:57 AM, Marvin Humphrey <ma...@rectangular.com> wrote:
>
> On Tue, Aug 6, 2013 at 1:04 AM, ant elder <an...@gmail.com> wrote:
> > On Wed, Jul 31, 2013 at 5:43 PM, Marvin Humphrey <ma...@rectangular.com> wrote:
> >>
> >> On Tue, Jul 30, 2013 at 6:24 AM, ant elder <an...@gmail.com> wrote:
> >> > I've been away so a little slow in finishing this but i have just now made
> >> > an update to the guide, see
> >> > http://svn.apache.org/viewvc?view=revision&revision=1508433
> >>
> >> I guess preserving the 2010 status quo was too much too much to ask for. :)
> >
> > Not sure what to make of that comment or the smiley at the end, i've
> > tried to do this change as gently as possible, can you live with the
> > current text or do you want an update?
>
> Personally, I believe that removing the IPMC notification requirement when
> adding podling committers is a positive change. The new language is expansive
> for my tastes -- the quarterly reporting requirement in particular seems out
> of place -- but I can live with it.
>
> It wasn't clear to me that there was a pressing need to change the policy,
> hence my pining for the previous hard-won consensus from 2010. But the die is
> cast...
>
I guess one of the things i'm looking at with this is finding ways to
try to minimize the points at which the Incubator PMC needs to
participate in day to day podling activities because its those that
can lead to the ISSUE03 type flair ups. The pTLP approach is an
interesting idea but an alternative might be to try to have most
podling contact be just via mentors, minimize the time incubating, and
have the Incubator PMC operate with the more hands off approach that
the board uses.
Agree about the sentence about reporting being out of place, i'd only
included that to try to placate anyone who may have been unhappy about
the notification removal. Lets remove the sentence here and move it up
the page to the section on Podling Status Reports, and while doing
that also look at your earlier suggestion of harmonizing the Incubator
podling reporting with the Board's guidelines for TLPs.
...ant
---------------------------------------------------------------------
To unsubscribe, e-mail: general-unsubscribe@incubator.apache.org
For additional commands, e-mail: general-help@incubator.apache.org
Re: Podling new committer votes
Posted by Marvin Humphrey <ma...@rectangular.com>.
On Tue, Aug 6, 2013 at 1:04 AM, ant elder <an...@gmail.com> wrote:
> On Wed, Jul 31, 2013 at 5:43 PM, Marvin Humphrey <ma...@rectangular.com> wrote:
>>
>> On Tue, Jul 30, 2013 at 6:24 AM, ant elder <an...@gmail.com> wrote:
>> > I've been away so a little slow in finishing this but i have just now made
>> > an update to the guide, see
>> > http://svn.apache.org/viewvc?view=revision&revision=1508433
>>
>> I guess preserving the 2010 status quo was too much too much to ask for. :)
>
> Not sure what to make of that comment or the smiley at the end, i've
> tried to do this change as gently as possible, can you live with the
> current text or do you want an update?
Personally, I believe that removing the IPMC notification requirement when
adding podling committers is a positive change. The new language is expansive
for my tastes -- the quarterly reporting requirement in particular seems out
of place -- but I can live with it.
It wasn't clear to me that there was a pressing need to change the policy,
hence my pining for the previous hard-won consensus from 2010. But the die is
cast...
I am especially fond of the fact that the change harmonizes Incubator policy
with top-level ASF policy. As we have talked about before, the more policy
that the Incubator can inherit rather than duplicate, the less material for us
to maintain and the better we will prepare our podlings for life after
graduation.
>> Regarding Bertrand's request, I note that while the process for adding
>> committers has been relaxed, Mentor participation and IPMC notification are
>> still required for PPMC membership. In my opinion, this is a fine scheme, as
>> it is consistent with how the Board requires notification for PMC membership
>> but not committership.
>
> I was tempted to update that too while editing the new committer text
> but as we hadn't discussed it just left it and had planned a new
> thread to discuss changing it. I can see the logic in having it work
> the same way as the board notifications of new PMC members, but that
> also feels a little unnecessary as the PPMC doesn't have any real
> standing or binding votes on much at all so i can also appreciate the
> Incubator PMC just leaving them to it and staying out of their way.
Podlings left to their own devices are sometimes sloppy about who is on the
PPMC and who is not. That's bad, because uncertainty around the voting rolls
undermines the legitimacy of the voting mechanism.
Contentious votes are undesirable because they alienate the losing minority,
but in some cases they are the only way to resolve a disagreement. A vote
result which is open to dispute because it's not clear whose votes are
binding makes it more difficult for the community to move on.
The current instructions for adding a PPMC member are not well-drafted because
they needlessly require double notification: once by forwarding the kickoff
VOTE email, and once after the candidate has accepted. I would not support
removing IPMC notification altogether, but I think it would be great to
streamline away the duplicate requirement.
For TLPs, email notification to board@apache by the project VP provides a
canonical record establishing PMC membership (once the 72-hour window
expires). If we are going to make a change, it seems to me that our
procedures with private@incubator and PPMC membership should mimic the Board's
procedures as closely as possible.
Marvin Humphrey
---------------------------------------------------------------------
To unsubscribe, e-mail: general-unsubscribe@incubator.apache.org
For additional commands, e-mail: general-help@incubator.apache.org
Re: Podling new committer votes
Posted by ant elder <an...@gmail.com>.
On Wed, Jul 31, 2013 at 5:43 PM, Marvin Humphrey <ma...@rectangular.com> wrote:
>
> On Tue, Jul 30, 2013 at 6:24 AM, ant elder <an...@gmail.com> wrote:
> > I've been away so a little slow in finishing this but i have just now made
> > an update to the guide, see
> > http://svn.apache.org/viewvc?view=revision&revision=1508433
>
> I guess preserving the 2010 status quo was too much too much to ask for. :)
>
Not sure what to make of that comment or the smiley at the end, i've
tried to do this change as gently as possible, can you live with the
current text or do you want an update?
>
> > This does not include anything yet for the request from Bertrand to still
> > require at least one mentor vote but instead goes more towards the
> > suggestion that we don't even need to document a standard for
> > committerpromotions- projects and podlings alike are free to experiment with
> > whatever process they deem appropriate.
>
> Regarding Bertrand's request, I note that while the process for adding
> committers has been relaxed, Mentor participation and IPMC notification are
> still required for PPMC membership. In my opinion, this is a fine scheme, as
> it is consistent with how the Board requires notification for PMC membership
> but not committership.
>
I was tempted to update that too while editing the new committer text
but as we hadn't discussed it just left it and had planned a new
thread to discuss changing it. I can see the logic in having it work
the same way as the board notifications of new PMC members, but that
also feels a little unnecessary as the PPMC doesn't have any real
standing or binding votes on much at all so i can also appreciate the
Incubator PMC just leaving them to it and staying out of their way.
...ant
---------------------------------------------------------------------
To unsubscribe, e-mail: general-unsubscribe@incubator.apache.org
For additional commands, e-mail: general-help@incubator.apache.org
Re: Podling new committer votes
Posted by Marvin Humphrey <ma...@rectangular.com>.
On Tue, Jul 30, 2013 at 6:24 AM, ant elder <an...@gmail.com> wrote:
> I've been away so a little slow in finishing this but i have just now made
> an update to the guide, see
> http://svn.apache.org/viewvc?view=revision&revision=1508433
I guess preserving the 2010 status quo was too much too much to ask for. :)
> This does not include anything yet for the request from Bertrand to still
> require at least one mentor vote but instead goes more towards the
> suggestion that we don't even need to document a standard for
> committerpromotions- projects and podlings alike are free to experiment with
> whatever process they deem appropriate.
Regarding Bertrand's request, I note that while the process for adding
committers has been relaxed, Mentor participation and IPMC notification are
still required for PPMC membership. In my opinion, this is a fine scheme, as
it is consistent with how the Board requires notification for PMC membership
but not committership.
(Podlings which unify committership and PPMC membership must simply follow the
more strict PPMC member processes for all additions, which seems fine as
well.)
We still need to clarify the notification guidelines for PPMC member votes,
though. Per the 2010 consensus, the email which kicks off the VOTE was to be
forwarded. However, this is not consistent with how the Board does things,
which explains Bertrand's "clarifying" commit later specifying that it is the
"RESULT" email which must be forwarded.
The Board recently passed a resolution changing the procedure for PMC member
additions.
http://www.apache.org/foundation/records/minutes/2013/board_minutes_2013_06_19.txt
1) A notice of the appointment shall be delivered to the Board's
mailing list and the PMC's private discussion list, as recorded
within the Foundation's mailing list archives, at least 72 hours
prior to the effective date of the appointment, where said notice
may be sent by the chairman, or by any existing member of the PMC
if it contains a link to a formal decision by the PMC approving
of the appointment;
2) If any Director objects to the appointment, in writing, prior to
its effective date, the appointment shall not be effective until
72 hours after all such objections are resolved, or until a
resolution making the appointment is adopted by the Board at some
regular or special meeting of the Board of Directors; and,
3) The appointment shall not be effective until the Foundation's
records of committee membership are updated accordingly.
To resolve the ambiguity, I suggest the Incubator adapt our own procedures to
mimic TLP procedures as closely as possible:
* Forward the RESULT email to private@incubator.
* 72 hour implicit approval.
> The change does update the podling report template with a section to note
> any committers added since the last report so that we are aware of what is
> happening.
The reporting guidelines for top-level projects request both the last
committer election and the last PMC member election.
http://www.apache.org/foundation/board/reporting
I need to generate the report template for next month imminently. I'm
reluctant to incorporate last-minute changes into the per-podling template,
so I plan to be conservative and use the old format. However, going
forward, I'd like to suggest that we try to harmonize Incubator podling
reporting with the Board's guidelines for TLPs.
Marvin Humphrey
---------------------------------------------------------------------
To unsubscribe, e-mail: general-unsubscribe@incubator.apache.org
For additional commands, e-mail: general-help@incubator.apache.org
Re: Podling new committer votes
Posted by ant elder <an...@gmail.com>.
I've been away so a little slow in finishing this but i have just now made
an update to the guide, see
http://svn.apache.org/viewvc?view=revision&revision=1508433
This does not include anything yet for the request from Bertrand to still
require at least one mentor vote but instead goes more towards the
suggestion that we don't even need to document a standard for
committerpromotions- projects and podlings alike are free to
experiment with
whatever process they deem appropriate. The change does update the podling
report template with a section to note any committers added since the last
report so that we are aware of what is happening.
...ant
Re: Podling new committer votes
Posted by ant elder <an...@gmail.com>.
Going once, going twice...
...ant
On Wed, Jul 10, 2013 at 8:10 PM, Joe Schaefer <jo...@yahoo.com>wrote:
> It doesn't. People were just being pedantic
> and untrusting of podling participants. Committers
> have accounts, not formal standing in the org. There
> is absolutely no reason for the IPMC to inject itself
> in a podling election of a new committer, so let's just
> leave oversight over the voting process to the mentors.
> As a matter of fact, we don't even need to document a
> standard for committer promotions- projects and podlings
> alike are free to experiment with whatever process they
> deem appropriate. Case in point is the Subversion process,
> which essentially promotes new committers through lazy
> consensus alone.
>
> IOW +1 to roll back to pre-May 1, 2007.
>
>
> ----- Original Message -----
> > From: ant elder <an...@gmail.com>
> > To: general@incubator.apache.org
> > Cc:
> > Sent: Wednesday, July 10, 2013 3:04 PM
> > Subject: Re: Podling new committer votes
> >
> > Ok seems everyone so far is ok with changing this, so how far can this
> go...
> >
> > In the "experiment" Joe commented "...basically rolling back the
> > clock
> > to May 1, 2007 on guides/ppmc.html" which is this change
> >
> http://svn.apache.org/viewvc/incubator/public/trunk/content/guides/ppmc.xml?r1=517024&r2=542806
> > which added "Only votes cast by Incubator PMC members are binding".
> > Before then podlings did their own committer votes and the Incubator
> > PMC weren't involved.
> >
> > Good we go back to that? Why does the Incubator PMC need to be notified?
> >
> > ...ant
> >
> > ---------------------------------------------------------------------
> > To unsubscribe, e-mail: general-unsubscribe@incubator.apache.org
> > For additional commands, e-mail: general-help@incubator.apache.org
> >
>
> ---------------------------------------------------------------------
> To unsubscribe, e-mail: general-unsubscribe@incubator.apache.org
> For additional commands, e-mail: general-help@incubator.apache.org
>
>
Re: Podling new committer votes
Posted by Joe Schaefer <jo...@yahoo.com>.
It doesn't. People were just being pedantic
and untrusting of podling participants. Committers
have accounts, not formal standing in the org. There
is absolutely no reason for the IPMC to inject itself
in a podling election of a new committer, so let's just
leave oversight over the voting process to the mentors.
As a matter of fact, we don't even need to document a
standard for committer promotions- projects and podlings
alike are free to experiment with whatever process they
deem appropriate. Case in point is the Subversion process,
which essentially promotes new committers through lazy
consensus alone.
IOW +1 to roll back to pre-May 1, 2007.
----- Original Message -----
> From: ant elder <an...@gmail.com>
> To: general@incubator.apache.org
> Cc:
> Sent: Wednesday, July 10, 2013 3:04 PM
> Subject: Re: Podling new committer votes
>
> Ok seems everyone so far is ok with changing this, so how far can this go...
>
> In the "experiment" Joe commented "...basically rolling back the
> clock
> to May 1, 2007 on guides/ppmc.html" which is this change
> http://svn.apache.org/viewvc/incubator/public/trunk/content/guides/ppmc.xml?r1=517024&r2=542806
> which added "Only votes cast by Incubator PMC members are binding".
> Before then podlings did their own committer votes and the Incubator
> PMC weren't involved.
>
> Good we go back to that? Why does the Incubator PMC need to be notified?
>
> ...ant
>
> ---------------------------------------------------------------------
> To unsubscribe, e-mail: general-unsubscribe@incubator.apache.org
> For additional commands, e-mail: general-help@incubator.apache.org
>
---------------------------------------------------------------------
To unsubscribe, e-mail: general-unsubscribe@incubator.apache.org
For additional commands, e-mail: general-help@incubator.apache.org
Re: Podling new committer votes
Posted by Bertrand Delacretaz <bd...@apache.org>.
On Wed, Jul 10, 2013 at 9:04 PM, ant elder <an...@gmail.com> wrote:
> Ok seems everyone so far is ok with changing this, so how far can this go...
Whatever happens, IMO we must still require at least one mentor vote
on podling committers elections - if only to make sure mentors are
listening.
-Bertrand
---------------------------------------------------------------------
To unsubscribe, e-mail: general-unsubscribe@incubator.apache.org
For additional commands, e-mail: general-help@incubator.apache.org
Re: Podling new committer votes
Posted by ant elder <an...@gmail.com>.
Ok seems everyone so far is ok with changing this, so how far can this go...
In the "experiment" Joe commented "...basically rolling back the clock
to May 1, 2007 on guides/ppmc.html" which is this change
http://svn.apache.org/viewvc/incubator/public/trunk/content/guides/ppmc.xml?r1=517024&r2=542806
which added "Only votes cast by Incubator PMC members are binding".
Before then podlings did their own committer votes and the Incubator
PMC weren't involved.
Good we go back to that? Why does the Incubator PMC need to be notified?
...ant
---------------------------------------------------------------------
To unsubscribe, e-mail: general-unsubscribe@incubator.apache.org
For additional commands, e-mail: general-help@incubator.apache.org
Re: Podling new committer votes
Posted by Ross Gardler <rg...@opendirective.com>.
Great archeology
On 9 Jul 2013 18:02, "Marvin Humphrey" <ma...@rectangular.com> wrote:
> On Tue, Jul 9, 2013 at 2:20 PM, ant elder <an...@gmail.com> wrote:
> > Looking at the "Voting in a new committer" section at
> > http://incubator.apache.org/guides/ppmc.html i see that its been
> > updated since i last looked and now says the Incubator PMC should be
> > notified twice when a podling is voting in a new committer, once at
> > the start of the vote and again with the result. Is this really
> > necessary? The last big discussion about this that i recall was
> > http://apache.markmail.org/message/v5gwhbpbifqcafqu which resulted in
> > a single notification, before that in the "experiment" that Joe
> > started with podlings doing their own committer votes there wasn't any
> > requirement to notify the Incubator PMC at all -
> > http://apache.markmail.org/message/mihzgp7jfclj2dcv. So I'd like to
> > remove this dual notification suggestion from that ppmc guide page
> > again. Comments?
>
> Continuously relitigating rules like this is tiresome.
>
> In the longish thread from 2010, you made a specific proposal:
>
> http://markmail.org/message/eb4477634gkaomgk
>
> Nobody objected at the time. The language was perfectly clear, too.
>
> Six months later, there was a "clarification" which actually changed the
> rule.
>
>
> http://svn.apache.org/viewvc/incubator/public/trunk/content/guides/ppmc.xml?r1=1036361&r2=1140654&diff_format=h
>
> +1 to revert to the rule that you built consensus for.
>
> Marvin Humphrey
>
> ---------------------------------------------------------------------
> To unsubscribe, e-mail: general-unsubscribe@incubator.apache.org
> For additional commands, e-mail: general-help@incubator.apache.org
>
>
Re: Podling new committer votes
Posted by Marvin Humphrey <ma...@rectangular.com>.
On Tue, Jul 9, 2013 at 2:20 PM, ant elder <an...@gmail.com> wrote:
> Looking at the "Voting in a new committer" section at
> http://incubator.apache.org/guides/ppmc.html i see that its been
> updated since i last looked and now says the Incubator PMC should be
> notified twice when a podling is voting in a new committer, once at
> the start of the vote and again with the result. Is this really
> necessary? The last big discussion about this that i recall was
> http://apache.markmail.org/message/v5gwhbpbifqcafqu which resulted in
> a single notification, before that in the "experiment" that Joe
> started with podlings doing their own committer votes there wasn't any
> requirement to notify the Incubator PMC at all -
> http://apache.markmail.org/message/mihzgp7jfclj2dcv. So I'd like to
> remove this dual notification suggestion from that ppmc guide page
> again. Comments?
Continuously relitigating rules like this is tiresome.
In the longish thread from 2010, you made a specific proposal:
http://markmail.org/message/eb4477634gkaomgk
Nobody objected at the time. The language was perfectly clear, too.
Six months later, there was a "clarification" which actually changed the rule.
http://svn.apache.org/viewvc/incubator/public/trunk/content/guides/ppmc.xml?r1=1036361&r2=1140654&diff_format=h
+1 to revert to the rule that you built consensus for.
Marvin Humphrey
---------------------------------------------------------------------
To unsubscribe, e-mail: general-unsubscribe@incubator.apache.org
For additional commands, e-mail: general-help@incubator.apache.org