You are viewing a plain text version of this content. The canonical link for it is here.
Posted to general@incubator.apache.org by Bertrand Delacretaz <bd...@apache.org> on 2007/10/13 09:43:42 UTC
Graduation: how do we check "three or more independent committers" ?
Do we have a solid process for checking this?
I'm not sure we do - each project's status page (like [1] for example)
contains the "Are there three or more independent committers?"
question, but unless I'm missing something we don't have a defined way
of checking this.
The simplest way to fix this might be to include, in the graduation
vote requests, the list of active committers and affiliations, so that
people can check it if they want.
-Bertrand
[1] http://incubator.apache.org/projects/tika.html
---------------------------------------------------------------------
To unsubscribe, e-mail: general-unsubscribe@incubator.apache.org
For additional commands, e-mail: general-help@incubator.apache.org
Re: Graduation: how do we check "three or more independent committers" ?
Posted by Guillaume Nodet <gn...@gmail.com>.
Well, 4 committers with 2 from the same organization: you still have 3
organizations involved. I think 3 is not that big a number:
committers do not have to be full time employee, and level of activity
may vary greatly between committers if some are full time employee and
others work on their spare time. But having 3 or more independant
committers is also a result of the openness of the project.
I know that when you have a bunch of full time employee devoted to a
project, it's much more difficult to attract new committers, but
still...
On 10/16/07, Matthieu Riou <ma...@offthelip.org> wrote:
> On 10/16/07, Robert Burrell Donkin <ro...@gmail.com> wrote:
> >
> > On 10/13/07, Bertrand Delacretaz <bd...@apache.org> wrote:
> > > Do we have a solid process for checking this?
> > >
> > > I'm not sure we do - each project's status page (like [1] for example)
> > > contains the "Are there three or more independent committers?"
> > > question, but unless I'm missing something we don't have a defined way
> > > of checking this.
> > >
> > > The simplest way to fix this might be to include, in the graduation
> > > vote requests, the list of active committers and affiliations, so that
> > > people can check it if they want.
> >
> > sounds like a reasonable approach
>
>
> As I mentioned in another thread I'm not so big on this rule. Having a fixed
> number is IMHO a bit too rigid. Do we forbid graduation for a project with 4
> active committers among which 2 are in the same company? How the project
> deals with its governance (are the independent committers opinion taken into
> account? are they ignored?) is, I think, more important. Am I the only one
> to think that way?
>
> That being said, having the list of active committers with affiliations in
> the graduation vote request would be helpful.
>
> Cheers,
> Matthieu
>
> - robert
> >
> > ---------------------------------------------------------------------
> > To unsubscribe, e-mail: general-unsubscribe@incubator.apache.org
> > For additional commands, e-mail: general-help@incubator.apache.org
> >
> >
>
--
Cheers,
Guillaume Nodet
------------------------
Blog: http://gnodet.blogspot.com/
---------------------------------------------------------------------
To unsubscribe, e-mail: general-unsubscribe@incubator.apache.org
For additional commands, e-mail: general-help@incubator.apache.org
Re: Graduation: how do we check "three or more independent committers" ?
Posted by Matthieu Riou <ma...@offthelip.org>.
On 10/16/07, Robert Burrell Donkin <ro...@gmail.com> wrote:
>
> On 10/13/07, Bertrand Delacretaz <bd...@apache.org> wrote:
> > Do we have a solid process for checking this?
> >
> > I'm not sure we do - each project's status page (like [1] for example)
> > contains the "Are there three or more independent committers?"
> > question, but unless I'm missing something we don't have a defined way
> > of checking this.
> >
> > The simplest way to fix this might be to include, in the graduation
> > vote requests, the list of active committers and affiliations, so that
> > people can check it if they want.
>
> sounds like a reasonable approach
As I mentioned in another thread I'm not so big on this rule. Having a fixed
number is IMHO a bit too rigid. Do we forbid graduation for a project with 4
active committers among which 2 are in the same company? How the project
deals with its governance (are the independent committers opinion taken into
account? are they ignored?) is, I think, more important. Am I the only one
to think that way?
That being said, having the list of active committers with affiliations in
the graduation vote request would be helpful.
Cheers,
Matthieu
- robert
>
> ---------------------------------------------------------------------
> To unsubscribe, e-mail: general-unsubscribe@incubator.apache.org
> For additional commands, e-mail: general-help@incubator.apache.org
>
>
Re: Graduation: how do we check "three or more independent committers" ?
Posted by Robert Burrell Donkin <ro...@gmail.com>.
On 10/13/07, Bertrand Delacretaz <bd...@apache.org> wrote:
> Do we have a solid process for checking this?
>
> I'm not sure we do - each project's status page (like [1] for example)
> contains the "Are there three or more independent committers?"
> question, but unless I'm missing something we don't have a defined way
> of checking this.
>
> The simplest way to fix this might be to include, in the graduation
> vote requests, the list of active committers and affiliations, so that
> people can check it if they want.
sounds like a reasonable approach
- robert
---------------------------------------------------------------------
To unsubscribe, e-mail: general-unsubscribe@incubator.apache.org
For additional commands, e-mail: general-help@incubator.apache.org
Re: Graduation: how do we check "three or more independent committers"
?
Posted by Simon Nash <na...@hursley.ibm.com>.
Jim Jagielski wrote:
> IMO, the need for "Independent committers" is based on the
> following wants and needs:
>
> 1. Reduce the influence that any single entity has over
> the codebase and/or its direction.
>
> 2. Reduce the risk that, if a single entity suddenly drops
> its "support" for a codebase, that development does not
> immediately whither away and die.
>
> 3. Provide some sort of assurance that the committers are
> personally invested in the codebase, independent of their
> employment.
>
> 4. The health of a community is proportional to its diversity.
>
> 5. Reduce the risk associated when corporate drivers do not
> align with community drivers (corporate entity wants A
> but community wants B).
>
>
All of this makes sense. I think we should refine the meaning of
independent committer as others have suggested, going beyond just
looking at which company is the employer. The "day job" principle
that Matt suggested is one aspect. Others are the amount of
personal commitment (which I think is observable), and willingness
to put community drivers ahead of the corporate line when these
are not in sync (your point 5). This last one could be hard to
calibrate without knowing what goes on in private corporate
discussions, but I think people show by their actions how much
weight they give to community views and input when some topic is
up for discussion. The mentors should have a good sense of how
well the corporate committers follow these principles.
Simon
---------------------------------------------------------------------
To unsubscribe, e-mail: general-unsubscribe@incubator.apache.org
For additional commands, e-mail: general-help@incubator.apache.org
Re: Graduation: how do we check "three or more independent committers" ?
Posted by Jim Jagielski <ji...@jaguNET.com>.
IMO, the need for "Independent committers" is based on the
following wants and needs:
1. Reduce the influence that any single entity has over
the codebase and/or its direction.
2. Reduce the risk that, if a single entity suddenly drops
its "support" for a codebase, that development does not
immediately whither away and die.
3. Provide some sort of assurance that the committers are
personally invested in the codebase, independent of their
employment.
4. The health of a community is proportional to its diversity.
5. Reduce the risk associated when corporate drivers do not
align with community drivers (corporate entity wants A
but community wants B).
---------------------------------------------------------------------
To unsubscribe, e-mail: general-unsubscribe@incubator.apache.org
For additional commands, e-mail: general-help@incubator.apache.org
Re: Graduation: how do we check "three or more independent committers"
?
Posted by Martin Sebor <se...@roguewave.com>.
Roland Weber wrote:
> Bertrand Delacretaz wrote:
[...]
>> c) If yes, to what extent (along the lines of Matt's "more than 4
>> hours a workday" threshold)?
>
> 4 hours a day sounds like a very high threshold to me. I'm active
> on my private time, and I often can't spend 4 hours a week.
FWIW, it's not clear to me at which point of a committer's
involvement on a podling should this rule apply. When they
are voted in? Just prior to graduation? Or is it meant to
be an average over their entire participation (IMO, this
would be more fair than either of the other two options
but probably impractical to calculate with much accuracy).
People's schedules change over time. There might be committers
who initially get involved on their own only to be moved into
a part time or even full time role working on the project by
their employer, and the other way around.
Martin
> b and c sound like something that should better be merged into
> a single question, but I guess we're not at that stage of the
> discussion yet.
>
> cheers,
> Roland
>
>> d) Which other committers do you have a "backchannel" with, i.e. who
>> do you regularly talk to outside of the project's mailing lists?
>>
>> I think d) is also an important measure of independence, as it's hard
>> for people who meet for coffee every day to restate on the public
>> lists everything that happens on their backchannel.
>>
>> -Bertrand
>>
>> ---------------------------------------------------------------------
>> 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: Graduation: how do we check "three or more independent committers"
?
Posted by Roland Weber <os...@dubioso.net>.
Bertrand Delacretaz wrote:
> To expose committers affiliations before graduation, I suggest the
> following questions:
>
> a) What are your company affiliations?
Add "with respect to this project", to make sure that only those
affiliations are asked for.
> b) Does your activity in the project depend on being paid by those companies?
>
> c) If yes, to what extent (along the lines of Matt's "more than 4
> hours a workday" threshold)?
4 hours a day sounds like a very high threshold to me. I'm active
on my private time, and I often can't spend 4 hours a week.
b and c sound like something that should better be merged into
a single question, but I guess we're not at that stage of the
discussion yet.
cheers,
Roland
> d) Which other committers do you have a "backchannel" with, i.e. who
> do you regularly talk to outside of the project's mailing lists?
>
> I think d) is also an important measure of independence, as it's hard
> for people who meet for coffee every day to restate on the public
> lists everything that happens on their backchannel.
>
> -Bertrand
>
> ---------------------------------------------------------------------
> 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: Graduation: how do we check "three or more independent committers" ?
Posted by Justin Erenkrantz <ju...@erenkrantz.com>.
On 10/16/07, Bertrand Delacretaz <bd...@apache.org> wrote:
> d) Which other committers do you have a "backchannel" with, i.e. who
> do you regularly talk to outside of the project's mailing lists?
>
> I think d) is also an important measure of independence, as it's hard
> for people who meet for coffee every day to restate on the public
> lists everything that happens on their backchannel.
I talk to lots of other httpd committers outside of the project
mailing list - so I'm not sure that's a useful metric. I get what you
mean, but I doubt its relevance. -- justin
---------------------------------------------------------------------
To unsubscribe, e-mail: general-unsubscribe@incubator.apache.org
For additional commands, e-mail: general-help@incubator.apache.org
Re: Graduation: how do we check "three or more independent committers" ?
Posted by Bertrand Delacretaz <bd...@apache.org>.
On 10/17/07, Craig L Russell <Cr...@sun.com> wrote:
> ...Is there some
> consensus that affiliation is project-specific, such that a committer
> can be independent on project A while working on another project B
> as their "day job"?...
I agree with that.
To expose committers affiliations before graduation, I suggest the
following questions:
a) What are your company affiliations?
b) Does your activity in the project depend on being paid by those companies?
c) If yes, to what extent (along the lines of Matt's "more than 4
hours a workday" threshold)?
d) Which other committers do you have a "backchannel" with, i.e. who
do you regularly talk to outside of the project's mailing lists?
I think d) is also an important measure of independence, as it's hard
for people who meet for coffee every day to restate on the public
lists everything that happens on their backchannel.
-Bertrand
---------------------------------------------------------------------
To unsubscribe, e-mail: general-unsubscribe@incubator.apache.org
For additional commands, e-mail: general-help@incubator.apache.org
Re: Graduation: how do we check "three or more independent committers"
?
Posted by Craig L Russell <Cr...@Sun.COM>.
On Oct 16, 2007, at 7:12 PM, Matt Hogstrom wrote:
> Here is a quick initial definition. I believe that it is the
> committer's responsibility to properly state their affiliation.
> For instance, if a committer contributed to a project and it was
> approved by their employer (company A) but not actually part of
> their immediate duties I wouldn't necessarily expect them to have
> to say I am affiliated with Company A and they might identify
> themselves as independent. An underpinning of this definition is
> trust.
>
> Affiliation: For purposes of identifying a community's diversity
> and independence it is useful to identify a committer's
> afiliation. It is useful to disclose to the community if a
> committer is working on a project as part of their primary job
> responsibility. This can be loosely defined to mean paid to work
> on a project more than 4-hours a day. This affiliation can be used
> to identify a project that would be in jeopardy should an
> organization that is supporting developers should those developers
> time be redirected to unrelated efforts.
I like this definition. It includes the concept that affiliation is
project-specific.
Craig
>
>
> On Oct 16, 2007, at 8:21 PM, Craig L Russell wrote:
>
>> I do think it would be good to clarify the meaning of
>> affiliations, and make it more flexible.
>>
>> This topic has been kicked around for a while. Is there some
>> consensus that affiliation is project-specific, such that a
>> committer can be independent on project A while working on another
>> project B as their "day job"?
>
Craig Russell
Architect, Sun Java Enterprise System http://java.sun.com/products/jdo
408 276-5638 mailto:Craig.Russell@sun.com
P.S. A good JDO? O, Gasp!
Re: Graduation: how do we check "three or more independent committers" ?
Posted by Matt Hogstrom <ma...@hogstrom.org>.
Here is a quick initial definition. I believe that it is the
committer's responsibility to properly state their affiliation. For
instance, if a committer contributed to a project and it was approved
by their employer (company A) but not actually part of their
immediate duties I wouldn't necessarily expect them to have to say I
am affiliated with Company A and they might identify themselves as
independent. An underpinning of this definition is trust.
Affiliation: For purposes of identifying a community's diversity and
independence it is useful to identify a committer's afiliation. It
is useful to disclose to the community if a committer is working on a
project as part of their primary job responsibility. This can be
loosely defined to mean paid to work on a project more than 4-hours a
day. This affiliation can be used to identify a project that would
be in jeopardy should an organization that is supporting developers
should those developers time be redirected to unrelated efforts.
On Oct 16, 2007, at 8:21 PM, Craig L Russell wrote:
> I do think it would be good to clarify the meaning of affiliations,
> and make it more flexible.
>
> This topic has been kicked around for a while. Is there some
> consensus that affiliation is project-specific, such that a
> committer can be independent on project A while working on another
> project B as their "day job"?
Re: Graduation: how do we check "three or more independent committers"
?
Posted by Craig L Russell <Cr...@Sun.COM>.
On Oct 16, 2007, at 4:05 PM, Simon Nash wrote:
>
> Matt Hogstrom wrote:
>
>> Not to go down a rat hole but I think that clarifying the
>> affiliations needs some refinement. On Apache Geronimo we have a
>> few committers that work for IBM but their participation is not
>> part of their day job. Would the committer in question for this
>> situation state that they are independent? I'm looking for
>> clarification as I expect that this same situation may exist for
>> other incubator projects and like Matthieu I would be concerned
>> about creating a set of hard rules where what I think we're
>> looking for is project independence based on individual merit and
>> contribution.
> Good point, Matt. This applies to Tuscany as well.
I do think it would be good to clarify the meaning of affiliations,
and make it more flexible.
This topic has been kicked around for a while. Is there some
consensus that affiliation is project-specific, such that a committer
can be independent on project A while working on another project B
as their "day job"?
Craig
>
> Simon
>
>> On Oct 13, 2007, at 3:43 AM, Bertrand Delacretaz wrote:
>>> Do we have a solid process for checking this?
>>>
>>> I'm not sure we do - each project's status page (like [1] for
>>> example)
>>> contains the "Are there three or more independent committers?"
>>> question, but unless I'm missing something we don't have a
>>> defined way
>>> of checking this.
>>>
>>> The simplest way to fix this might be to include, in the graduation
>>> vote requests, the list of active committers and affiliations, so
>>> that
>>> people can check it if they want.
>>>
>>> -Bertrand
>>>
>>> [1] http://incubator.apache.org/projects/tika.html
>>>
>>> --------------------------------------------------------------------
>>> -
>>> 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
>
Craig Russell
Architect, Sun Java Enterprise System http://java.sun.com/products/jdo
408 276-5638 mailto:Craig.Russell@sun.com
P.S. A good JDO? O, Gasp!
Re: Graduation: how do we check "three or more independent committers"
?
Posted by Simon Nash <na...@hursley.ibm.com>.
Matt Hogstrom wrote:
> Not to go down a rat hole but I think that clarifying the affiliations
> needs some refinement. On Apache Geronimo we have a few committers
> that work for IBM but their participation is not part of their day
> job. Would the committer in question for this situation state that
> they are independent? I'm looking for clarification as I expect that
> this same situation may exist for other incubator projects and like
> Matthieu I would be concerned about creating a set of hard rules where
> what I think we're looking for is project independence based on
> individual merit and contribution.
>
Good point, Matt. This applies to Tuscany as well.
Simon
>
> On Oct 13, 2007, at 3:43 AM, Bertrand Delacretaz wrote:
>
>> Do we have a solid process for checking this?
>>
>> I'm not sure we do - each project's status page (like [1] for example)
>> contains the "Are there three or more independent committers?"
>> question, but unless I'm missing something we don't have a defined way
>> of checking this.
>>
>> The simplest way to fix this might be to include, in the graduation
>> vote requests, the list of active committers and affiliations, so that
>> people can check it if they want.
>>
>> -Bertrand
>>
>> [1] http://incubator.apache.org/projects/tika.html
>>
>> ---------------------------------------------------------------------
>> 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: Graduation: how do we check "three or more independent committers" ?
Posted by Matt Hogstrom <ma...@hogstrom.org>.
Not to go down a rat hole but I think that clarifying the
affiliations needs some refinement. On Apache Geronimo we have a few
committers that work for IBM but their participation is not part of
their day job. Would the committer in question for this situation
state that they are independent? I'm looking for clarification as I
expect that this same situation may exist for other incubator
projects and like Matthieu I would be concerned about creating a set
of hard rules where what I think we're looking for is project
independence based on individual merit and contribution.
On Oct 13, 2007, at 3:43 AM, Bertrand Delacretaz wrote:
> Do we have a solid process for checking this?
>
> I'm not sure we do - each project's status page (like [1] for example)
> contains the "Are there three or more independent committers?"
> question, but unless I'm missing something we don't have a defined way
> of checking this.
>
> The simplest way to fix this might be to include, in the graduation
> vote requests, the list of active committers and affiliations, so that
> people can check it if they want.
>
> -Bertrand
>
> [1] http://incubator.apache.org/projects/tika.html
>
> ---------------------------------------------------------------------
> 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