You are viewing a plain text version of this content. The canonical link for it is here.
Posted to general@incubator.apache.org by "Roy T. Fielding" <fi...@gbiv.com> on 2011/11/11 08:58:50 UTC
incubator is a single group
Hi folks,
It has come to my attention that we are wasting resources and
time trying to manage separate committer lists within infra for
every podling. That is something that we can effectively manage
once a TLP has become self-governing and self-sufficient in its
interaction with infrastructure. It is not something that we can
effectively manage when we have dozens of podlings trying to make
changes through the incubator chair.
I *suggest* that incubator change the procedure such that all
committers (or at least all committers within a single LDAP
group) have access to all incubator areas and that new people
simply be requested to only commit within areas for which they
have been given permission by the podling developers. That
will significantly ease everyone's job and allow more direct
control by podlings for inviting/uninviting their devs.
....Roy
---------------------------------------------------------------------
To unsubscribe, e-mail: general-unsubscribe@incubator.apache.org
For additional commands, e-mail: general-help@incubator.apache.org
RE: incubator is a single group
Posted by "Dennis E. Hamilton" <de...@acm.org>.
Small items:
1. There are no podling PPMC members that have the necessary authorizations, even though they are committers. Mentors (and a champion) yes, chair no. This is partly why there are these regular calls for a mentor or someone on the incubator PMC to connect some dots.
2. The authz is currently the only structure that carries who the official committers of a podling are and this can be cross-checked with in-podling records for managing intake of new committers, PPMC members, and reflecting changes in status that arise. This can be maintained within the PPMC but it is great to have "ground truth" in a concrete artifact that does not depend on how well a podling does or does not manage its administrivia. Making more work for secretary@ also doesn't seem like a win.
I sympathize with Roy's concern for where this moves some burdensome work. My only concern that it not make it even more awkward/unreliable to operate within a podling, where not all the dots are visible, let alone a means to connect them.
- Dennis
-----Original Message-----
From: sebb [mailto:sebbaz@gmail.com]
Sent: Friday, November 11, 2011 05:24
To: general@incubator.apache.org
Subject: Re: incubator is a single group
On 11 November 2011 07:58, Roy T. Fielding <fi...@gbiv.com> wrote:
> Hi folks,
>
> It has come to my attention that we are wasting resources and
> time trying to manage separate committer lists within infra for
> every podling. That is something that we can effectively manage
> once a TLP has become self-governing and self-sufficient in its
> interaction with infrastructure. It is not something that we can
> effectively manage when we have dozens of podlings trying to make
> changes through the incubator chair.
SVN podling access is currently managed through a file which can be
updated by members of the pmc-chairs group.
[And a few others such as infra/execs]
So it is no only just the incubator chair who can update the file.
[AFAIK, there aren't any separate LDAP groups for podlings; all
podling SVN access groups are managed by that file.]
> I *suggest* that incubator change the procedure such that all
> committers (or at least all committers within a single LDAP
> group) have access to all incubator areas and that new people
> simply be requested to only commit within areas for which they
> have been given permission by the podling developers. That
> will significantly ease everyone's job and allow more direct
> control by podlings for inviting/uninviting their devs.
There is already an "incubator" LDAP group, to which all podling
committers should belong, as it gives access to website updates.
This could also be given access to podling SVN directories.
If the individual podling member lists were abandoned, I think there
would still need to be a way to document the current members of a
podling.
The lists in the SVN auth file are currently used for that purpose
(e.g. they are used in creating
http://people.apache.org/committers-by-project.html)
> ....Roy
>
> ---------------------------------------------------------------------
> 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
---------------------------------------------------------------------
To unsubscribe, e-mail: general-unsubscribe@incubator.apache.org
For additional commands, e-mail: general-help@incubator.apache.org
Re: incubator is a single group
Posted by sebb <se...@gmail.com>.
On 11 November 2011 07:58, Roy T. Fielding <fi...@gbiv.com> wrote:
> Hi folks,
>
> It has come to my attention that we are wasting resources and
> time trying to manage separate committer lists within infra for
> every podling. That is something that we can effectively manage
> once a TLP has become self-governing and self-sufficient in its
> interaction with infrastructure. It is not something that we can
> effectively manage when we have dozens of podlings trying to make
> changes through the incubator chair.
SVN podling access is currently managed through a file which can be
updated by members of the pmc-chairs group.
[And a few others such as infra/execs]
So it is no only just the incubator chair who can update the file.
[AFAIK, there aren't any separate LDAP groups for podlings; all
podling SVN access groups are managed by that file.]
> I *suggest* that incubator change the procedure such that all
> committers (or at least all committers within a single LDAP
> group) have access to all incubator areas and that new people
> simply be requested to only commit within areas for which they
> have been given permission by the podling developers. That
> will significantly ease everyone's job and allow more direct
> control by podlings for inviting/uninviting their devs.
There is already an "incubator" LDAP group, to which all podling
committers should belong, as it gives access to website updates.
This could also be given access to podling SVN directories.
If the individual podling member lists were abandoned, I think there
would still need to be a way to document the current members of a
podling.
The lists in the SVN auth file are currently used for that purpose
(e.g. they are used in creating
http://people.apache.org/committers-by-project.html)
> ....Roy
>
> ---------------------------------------------------------------------
> 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: incubator is a single group
Posted by David Crossley <cr...@apache.org>.
sebb wrote:
> David Crossley wrote:
> > Dennis E. Hamilton wrote:
> >> It is the case that Initial Committers are eligible to serve on the PPMC, though not all choose to do so.
> >>
> >> Also, it is apparently a recommendation that invited committers be invited to the PPMC, but it doesn't seem to be a requirement. Again, not all that are so invited choose to serve on the PPMC.
> >
> > Yes Seb, i intended that to be separate list,
> > as it is when they are a TLP.
> >
> > I know we do encourage new committers to simultaneously
> > be on PPMC. As is for TLPs, there are cases.
> >
> > Trying to align with current TLP processes,
> > if possible while keeping it simple.
>
> In that case, I agree that keeping PPMC members in a separate file
> that corresponds with committee-info.txt is a good idea, so long as
> that file has a standard name and location so it can easily be found.
>
> However, I don't think the committer list (currently in
> asf-auth-template) belongs in podlings.xml.
>
> I think that might be better as another file, or perhaps as a marker -
> or another section - in the PPMC file.
Okay. Actually i wondered that to start with, but suggested
the podlings.xml instead.
So podling-committee-info.txt with markers.
Good. Similar to TLP.
This could also align with the TLP way of managing the
report schedule in committee-info.txt
Rather than list the mentors/committers in each
podling status page, then could just link directly, e.g.
http://incubator.apache.org/projects/#giraph (fixme)
http://people.apache.org/committers-by-project.html#giraph
because those are correlated and generated from
the podlings.xml and podling-committee-info.txt etcetera.
Sorry for the rushed comments: spread too thin.
-David
---------------------------------------------------------------------
To unsubscribe, e-mail: general-unsubscribe@incubator.apache.org
For additional commands, e-mail: general-help@incubator.apache.org
Re: incubator is a single group
Posted by sebb <se...@gmail.com>.
On 12 November 2011 23:14, David Crossley <cr...@apache.org> wrote:
> Dennis E. Hamilton wrote:
>> It is the case that Initial Committers are eligible to serve on the PPMC, though not all choose to do so.
>>
>> Also, it is apparently a recommendation that invited committers be invited to the PPMC, but it doesn't seem to be a requirement. Again, not all that are so invited choose to serve on the PPMC.
>
> Yes Seb, i intended that to be separate list,
> as it is when they are a TLP.
>
> I know we do encourage new committers to simultaneously
> be on PPMC. As is for TLPs, there are cases.
>
> Trying to align with current TLP processes,
> if possible while keeping it simple.
In that case, I agree that keeping PPMC members in a separate file
that corresponds with committee-info.txt is a good idea, so long as
that file has a standard name and location so it can easily be found.
However, I don't think the committer list (currently in
asf-auth-template) belongs in podlings.xml.
I think that might be better as another file, or perhaps as a marker -
or another section - in the PPMC file.
> -David
>
>> -----Original Message-----
>> From: sebb [mailto:sebbaz@gmail.com]
>> Sent: Saturday, November 12, 2011 03:52
>> To: general@incubator.apache.org
>> Subject: Re: incubator is a single group
>>
>> [ ... ]
>>
>> Is this an alternative?
>> Aren't podling committers the same as PPMC members?
>>
>> [ ... ]
>>
>>
>> ---------------------------------------------------------------------
>> 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
>
>
---------------------------------------------------------------------
To unsubscribe, e-mail: general-unsubscribe@incubator.apache.org
For additional commands, e-mail: general-help@incubator.apache.org
Re: incubator is a single group
Posted by David Crossley <cr...@apache.org>.
Dennis E. Hamilton wrote:
> It is the case that Initial Committers are eligible to serve on the PPMC, though not all choose to do so.
>
> Also, it is apparently a recommendation that invited committers be invited to the PPMC, but it doesn't seem to be a requirement. Again, not all that are so invited choose to serve on the PPMC.
Yes Seb, i intended that to be separate list,
as it is when they are a TLP.
I know we do encourage new committers to simultaneously
be on PPMC. As is for TLPs, there are cases.
Trying to align with current TLP processes,
if possible while keeping it simple.
-David
> -----Original Message-----
> From: sebb [mailto:sebbaz@gmail.com]
> Sent: Saturday, November 12, 2011 03:52
> To: general@incubator.apache.org
> Subject: Re: incubator is a single group
>
> [ ... ]
>
> Is this an alternative?
> Aren't podling committers the same as PPMC members?
>
> [ ... ]
>
>
> ---------------------------------------------------------------------
> 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: incubator is a single group
Posted by "Dennis E. Hamilton" <de...@acm.org>.
It is the case that Initial Committers are eligible to serve on the PPMC, though not all choose to do so.
Also, it is apparently a recommendation that invited committers be invited to the PPMC, but it doesn't seem to be a requirement. Again, not all that are so invited choose to serve on the PPMC.
-----Original Message-----
From: sebb [mailto:sebbaz@gmail.com]
Sent: Saturday, November 12, 2011 03:52
To: general@incubator.apache.org
Subject: Re: incubator is a single group
[ ... ]
Is this an alternative?
Aren't podling committers the same as PPMC members?
[ ... ]
---------------------------------------------------------------------
To unsubscribe, e-mail: general-unsubscribe@incubator.apache.org
For additional commands, e-mail: general-help@incubator.apache.org
Re: incubator is a single group
Posted by sebb <se...@gmail.com>.
On 12 November 2011 06:54, David Crossley <cr...@apache.org> wrote:
> William A. Rowe Jr. wrote:
>> Roy T. Fielding wrote:
>> >
>> >I *suggest* that incubator change the procedure such that all
>> >committers (or at least all committers within a single LDAP
>> >group) have access to all incubator areas and that new people
>> >simply be requested to only commit within areas for which they
>> >have been given permission by the podling developers.
>>
>> +1 [provided that there be some machine-parsable record of the
>> current PPMC composition, both committers and members].
>
> A suggestion:
>
> Add committers to the new site-author/podlings.xml
> This already has mentors.
>
> Could that be utilised by the infra tools/validation/* scripts and
> also feed into the generation of p.a.o/committers-by-project.html ?
Should be possible.
> Record PPMC members in a text file podling-committee-info.txt
> similar to TLP committee-info.txt file, which gets them
> used to doing that TLP job.
Is this an alternative?
Aren't podling committers the same as PPMC members?
>> I would rather see each PPMC come to a consensus decision on
>> its lists of PMC membership and initial committers as a final
>> step for graduation, since the mapping of "I'd support this"
>> at proposal phase can vary widely from "I've supported this"
>> at the end of the incubation phase. But its probably a topic
>> for an entirely different thread ;-)
>
> I reckon that should happen also.
>
> -David
>
> ---------------------------------------------------------------------
> 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: incubator is a single group
Posted by David Crossley <cr...@apache.org>.
William A. Rowe Jr. wrote:
> Roy T. Fielding wrote:
> >
> >I *suggest* that incubator change the procedure such that all
> >committers (or at least all committers within a single LDAP
> >group) have access to all incubator areas and that new people
> >simply be requested to only commit within areas for which they
> >have been given permission by the podling developers.
>
> +1 [provided that there be some machine-parsable record of the
> current PPMC composition, both committers and members].
A suggestion:
Add committers to the new site-author/podlings.xml
This already has mentors.
Could that be utilised by the infra tools/validation/* scripts and
also feed into the generation of p.a.o/committers-by-project.html ?
Record PPMC members in a text file podling-committee-info.txt
similar to TLP committee-info.txt file, which gets them
used to doing that TLP job.
> I would rather see each PPMC come to a consensus decision on
> its lists of PMC membership and initial committers as a final
> step for graduation, since the mapping of "I'd support this"
> at proposal phase can vary widely from "I've supported this"
> at the end of the incubation phase. But its probably a topic
> for an entirely different thread ;-)
I reckon that should happen also.
-David
---------------------------------------------------------------------
To unsubscribe, e-mail: general-unsubscribe@incubator.apache.org
For additional commands, e-mail: general-help@incubator.apache.org
Re: incubator is a single group
Posted by "William A. Rowe Jr." <wr...@rowe-clan.net>.
On 11/11/2011 1:58 AM, Roy T. Fielding wrote:
>
> I *suggest* that incubator change the procedure such that all
> committers (or at least all committers within a single LDAP
> group) have access to all incubator areas and that new people
> simply be requested to only commit within areas for which they
> have been given permission by the podling developers.
+1 [provided that there be some machine-parsable record of the
current PPMC composition, both committers and members].
I would rather see each PPMC come to a consensus decision on
its lists of PMC membership and initial committers as a final
step for graduation, since the mapping of "I'd support this"
at proposal phase can vary widely from "I've supported this"
at the end of the incubation phase. But its probably a topic
for an entirely different thread ;-)
---------------------------------------------------------------------
To unsubscribe, e-mail: general-unsubscribe@incubator.apache.org
For additional commands, e-mail: general-help@incubator.apache.org
Re: incubator is a single group
Posted by Tommaso Teofili <to...@gmail.com>.
same here, big +1.
Tommaso
2011/11/11 Jean-Baptiste Onofré <jb...@nanthrax.net>
> Hi guys,
>
> I second Chris on that point.
>
> Regards
> JB
>
>
> On 11/11/2011 04:37 PM, Mattmann, Chris A (388J) wrote:
>
>> Hi Roy,
>>
>> I would be super +1 for this. I think it would really help people out, and
>> let those mentoring the podlings and those doing the work manage
>> the addition/removal of people to their PPMCs. It would also hopefully
>> ease the burden on the PMC chairs and (hopefully) simplify infra.
>>
>> Thanks for suggesting this.
>>
>> Cheers,
>> Chris
>>
>> On Nov 10, 2011, at 11:58 PM, Roy T. Fielding wrote:
>>
>> Hi folks,
>>>
>>> It has come to my attention that we are wasting resources and
>>> time trying to manage separate committer lists within infra for
>>> every podling. That is something that we can effectively manage
>>> once a TLP has become self-governing and self-sufficient in its
>>> interaction with infrastructure. It is not something that we can
>>> effectively manage when we have dozens of podlings trying to make
>>> changes through the incubator chair.
>>>
>>> I *suggest* that incubator change the procedure such that all
>>> committers (or at least all committers within a single LDAP
>>> group) have access to all incubator areas and that new people
>>> simply be requested to only commit within areas for which they
>>> have been given permission by the podling developers. That
>>> will significantly ease everyone's job and allow more direct
>>> control by podlings for inviting/uninviting their devs.
>>>
>>> ....Roy
>>>
>>> ------------------------------**------------------------------**
>>> ---------
>>> To unsubscribe, e-mail: general-unsubscribe@incubator.**apache.org<ge...@incubator.apache.org>
>>> For additional commands, e-mail: general-help@incubator.apache.**org<ge...@incubator.apache.org>
>>>
>>>
>>
>> ++++++++++++++++++++++++++++++**++++++++++++++++++++++++++++++**++++++
>> Chris Mattmann, Ph.D.
>> Senior Computer Scientist
>> NASA Jet Propulsion Laboratory Pasadena, CA 91109 USA
>> Office: 171-266B, Mailstop: 171-246
>> Email: chris.a.mattmann@nasa.gov
>> WWW: http://sunset.usc.edu/~**mattmann/<http://sunset.usc.edu/~mattmann/>
>> ++++++++++++++++++++++++++++++**++++++++++++++++++++++++++++++**++++++
>> Adjunct Assistant Professor, Computer Science Department
>> University of Southern California, Los Angeles, CA 90089 USA
>> ++++++++++++++++++++++++++++++**++++++++++++++++++++++++++++++**++++++
>>
>>
>> ------------------------------**------------------------------**---------
>> To unsubscribe, e-mail: general-unsubscribe@incubator.**apache.org<ge...@incubator.apache.org>
>> For additional commands, e-mail: general-help@incubator.apache.**org<ge...@incubator.apache.org>
>>
>>
> --
> Jean-Baptiste Onofré
> jbonofre@apache.org
> http://blog.nanthrax.net
> Talend - http://www.talend.com
>
>
> ------------------------------**------------------------------**---------
> To unsubscribe, e-mail: general-unsubscribe@incubator.**apache.org<ge...@incubator.apache.org>
> For additional commands, e-mail: general-help@incubator.apache.**org<ge...@incubator.apache.org>
>
>
Re: incubator is a single group
Posted by Jean-Baptiste Onofré <jb...@nanthrax.net>.
Hi guys,
I second Chris on that point.
Regards
JB
On 11/11/2011 04:37 PM, Mattmann, Chris A (388J) wrote:
> Hi Roy,
>
> I would be super +1 for this. I think it would really help people out, and
> let those mentoring the podlings and those doing the work manage
> the addition/removal of people to their PPMCs. It would also hopefully
> ease the burden on the PMC chairs and (hopefully) simplify infra.
>
> Thanks for suggesting this.
>
> Cheers,
> Chris
>
> On Nov 10, 2011, at 11:58 PM, Roy T. Fielding wrote:
>
>> Hi folks,
>>
>> It has come to my attention that we are wasting resources and
>> time trying to manage separate committer lists within infra for
>> every podling. That is something that we can effectively manage
>> once a TLP has become self-governing and self-sufficient in its
>> interaction with infrastructure. It is not something that we can
>> effectively manage when we have dozens of podlings trying to make
>> changes through the incubator chair.
>>
>> I *suggest* that incubator change the procedure such that all
>> committers (or at least all committers within a single LDAP
>> group) have access to all incubator areas and that new people
>> simply be requested to only commit within areas for which they
>> have been given permission by the podling developers. That
>> will significantly ease everyone's job and allow more direct
>> control by podlings for inviting/uninviting their devs.
>>
>> ....Roy
>>
>> ---------------------------------------------------------------------
>> To unsubscribe, e-mail: general-unsubscribe@incubator.apache.org
>> For additional commands, e-mail: general-help@incubator.apache.org
>>
>
>
> ++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++
> Chris Mattmann, Ph.D.
> Senior Computer Scientist
> NASA Jet Propulsion Laboratory Pasadena, CA 91109 USA
> Office: 171-266B, Mailstop: 171-246
> Email: chris.a.mattmann@nasa.gov
> WWW: http://sunset.usc.edu/~mattmann/
> ++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++
> Adjunct Assistant Professor, Computer Science Department
> University of Southern California, Los Angeles, CA 90089 USA
> ++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++
>
>
> ---------------------------------------------------------------------
> To unsubscribe, e-mail: general-unsubscribe@incubator.apache.org
> For additional commands, e-mail: general-help@incubator.apache.org
>
--
Jean-Baptiste Onofré
jbonofre@apache.org
http://blog.nanthrax.net
Talend - http://www.talend.com
---------------------------------------------------------------------
To unsubscribe, e-mail: general-unsubscribe@incubator.apache.org
For additional commands, e-mail: general-help@incubator.apache.org
Re: incubator is a single group
Posted by "Mattmann, Chris A (388J)" <ch...@jpl.nasa.gov>.
Hi Roy,
I would be super +1 for this. I think it would really help people out, and
let those mentoring the podlings and those doing the work manage
the addition/removal of people to their PPMCs. It would also hopefully
ease the burden on the PMC chairs and (hopefully) simplify infra.
Thanks for suggesting this.
Cheers,
Chris
On Nov 10, 2011, at 11:58 PM, Roy T. Fielding wrote:
> Hi folks,
>
> It has come to my attention that we are wasting resources and
> time trying to manage separate committer lists within infra for
> every podling. That is something that we can effectively manage
> once a TLP has become self-governing and self-sufficient in its
> interaction with infrastructure. It is not something that we can
> effectively manage when we have dozens of podlings trying to make
> changes through the incubator chair.
>
> I *suggest* that incubator change the procedure such that all
> committers (or at least all committers within a single LDAP
> group) have access to all incubator areas and that new people
> simply be requested to only commit within areas for which they
> have been given permission by the podling developers. That
> will significantly ease everyone's job and allow more direct
> control by podlings for inviting/uninviting their devs.
>
> ....Roy
>
> ---------------------------------------------------------------------
> To unsubscribe, e-mail: general-unsubscribe@incubator.apache.org
> For additional commands, e-mail: general-help@incubator.apache.org
>
++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++
Chris Mattmann, Ph.D.
Senior Computer Scientist
NASA Jet Propulsion Laboratory Pasadena, CA 91109 USA
Office: 171-266B, Mailstop: 171-246
Email: chris.a.mattmann@nasa.gov
WWW: http://sunset.usc.edu/~mattmann/
++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++
Adjunct Assistant Professor, Computer Science Department
University of Southern California, Los Angeles, CA 90089 USA
++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++
---------------------------------------------------------------------
To unsubscribe, e-mail: general-unsubscribe@incubator.apache.org
For additional commands, e-mail: general-help@incubator.apache.org