You are viewing a plain text version of this content. The canonical link for it is here.
Posted to general@incubator.apache.org by Martin Cooper <ma...@apache.org> on 2005/12/28 21:48:01 UTC
The line between full incubation and IP clearance
Is there a well-defined line between these two mechanisms for bringing new
code to the ASF? Not having seen the code yet, I can't say for sure, but I
would expect the Oracle ADF Faces donation to be substantial, and include
a framework that ties everything together. I would have thought that would
have warranted full incubation, but it appears to be going through only
the UP Clearance procedure. See:
http://mail-archives.apache.org/mod_mbox/myfaces-dev/200512.mbox/%3c8b3ce3790512281044k7aa44090o8a35863f71717399@mail.gmail.com%3e
I'm new in Incubator-land, so I'm not altogether savvy about processes and
procedures, but wanted to ask about this one.
--
Martin Cooper
---------------------------------------------------------------------
To unsubscribe, e-mail: general-unsubscribe@incubator.apache.org
For additional commands, e-mail: general-help@incubator.apache.org
Re: The line between full incubation and IP clearance
Posted by Ted Husted <te...@gmail.com>.
If we are moving from the general case to the specific, then, yes, it
may be that the MyFaces PMC will decide the existing community cannot
work with the code on their own.
The MyFaces PMC might then choose to bring a proposal to the Incubator
PMC to create a podling, and see if the Incubator PMC is willing to
incubate the ADF community along with the codebase, and then later
decide if ADF should be a MyFaces subproject or a TLP (or retired).
Once the code is exposed, it's even possible that the MyFaces
community will pass on the donation, and nothing will come of this at
all.
As I've said on MyFaces dev@, these types of ADF dicussions are
premature, because no one has put the code on the table yet.
-Ted.
On 1/1/06, Martin Cooper <ma...@apache.org> wrote:
> In some cases, I would agree. In others, I would not. In the particular
> case of MyFaces accepting the ADF Faces donation, for example, I believe
> that MyFaces has a general JSF community, but I disagree that there will
> therefore be a de facto community around ADF Faces just because the MyFaces
> PMC decides to accept the code. This has to do with the scale of the
> donation. While none of us have seen the code yet, my expectation is that
> the donation will be at least as large as the existing MyFaces code base, if
> not larger. That is not something that would be naturally absorbed by an
> existing community.
>
> In fact, in this particular case, it seems clear from the discussions on
> dev@myfaces that most people - from both MyFaces and ADF Faces - would
> prefer that the ADF Faces donation go through the full incubation process. I
> fully understand and respect your desire to close the "free committership"
> loophole, but I also believe that such a substantial donation warrants full
> incubation. Short of asking Oracle to keep the proposed committer list to a
> minimum (e.g. Adam & John), I'm not sure how we can resolve this dilemma.
>
> --
> Martin Cooper
>
>
> > What is before the Incubator PMC is the issue of whether the
> > code *can* be licensed to the Foundation. Whether the code *should* be
> > licensed to the Foundation is a decision best left to the receiving
> > PMC, who, by direction of the Board, already decide whether such code
> > should be licensed to the ASF every time there is a commit to that
> > project.
> >
> > -Ted.
---------------------------------------------------------------------
To unsubscribe, e-mail: general-unsubscribe@incubator.apache.org
For additional commands, e-mail: general-help@incubator.apache.org
Re: The line between full incubation and IP clearance
Posted by Martin Cooper <ma...@apache.org>.
On 1/1/06, Ted Husted <te...@gmail.com> wrote:
>
> On 12/28/05, Noel J. Bergman <no...@devtech.com> wrote:
> > Although that might be technically true, we do things collectively in
> the
> > ASF. Mind you, we've not had a process for voting on IP Clearance type
> > submissions, so that's been a potential loophole.
>
> I'm not sure what a vote would accomplish. Are we saying that we don't
> trust PMCs to do due diligence as to the action items on the
> checklist? That would be a strange perspective since the key role of
> every TLP PMC is to oversee the IP of each and every commit made to
> the project's repository.
>
> An attractive aspect of the IP Clearance protocol is that it *closes*
> the podling loophole that might bring committers into the Foundation
> outside of the meritocratic process. One thing that the votes on the
> proposal and podling graduation do is sanction the list of podling
> committers, who usually go on to become PMC members. In that case, we
> do need the Incubator PMC to vote on the new committers, and since the
> podling is an unproven community, we do need someone to decide if the
> community meets our standard of meritocracy.
>
> If code is developed outside of an ASF repository and mailing list,
> then it is appropriate that we provide a pedigree for the code. It is
> also appropriate that we have a central record of all such code that a
> PMC is bringing int our repository. It is also appropriate that we
> maintain that record at the Incubator, so that all external code
> arrives in one place. But, I would suggest that the people best suited
> to vote on the code itself are the people who are already making the
> technical decisions about such code: The receiving PMC.
>
> In the case of an IP Clearance, there is not a new community to
> consider.
In some cases, I would agree. In others, I would not. In the particular case
of MyFaces accepting the ADF Faces donation, for example, I believe that
MyFaces has a general JSF community, but I disagree that there will
therefore be a de facto community around ADF Faces just because the MyFaces
PMC decides to accept the code. This has to do with the scale of the
donation. While none of us have seen the code yet, my expectation is that
the donation will be at least as large as the existing MyFaces code base, if
not larger. That is not something that would be naturally absorbed by an
existing community.
In fact, in this particular case, it seems clear from the discussions on
dev@myfaces that most people - from both MyFaces and ADF Faces - would
prefer that the ADF Faces donation go through the full incubation process. I
fully understand and respect your desire to close the "free committership"
loophole, but I also believe that such a substantial donation warrants full
incubation. Short of asking Oracle to keep the proposed committer list to a
minimum (e.g. Adam & John), I'm not sure how we can resolve this dilemma.
--
Martin Cooper
What is before the Incubator PMC is the issue of whether the
> code *can* be licensed to the Foundation. Whether the code *should* be
> licensed to the Foundation is a decision best left to the receiving
> PMC, who, by direction of the Board, already decide whether such code
> should be licensed to the ASF every time there is a commit to that
> project.
>
> -Ted.
>
> ---------------------------------------------------------------------
> To unsubscribe, e-mail: general-unsubscribe@incubator.apache.org
> For additional commands, e-mail: general-help@incubator.apache.org
>
>
Re: The line between full incubation and IP clearance
Posted by Ted Husted <te...@gmail.com>.
On 12/28/05, Noel J. Bergman <no...@devtech.com> wrote:
> Although that might be technically true, we do things collectively in the
> ASF. Mind you, we've not had a process for voting on IP Clearance type
> submissions, so that's been a potential loophole.
I'm not sure what a vote would accomplish. Are we saying that we don't
trust PMCs to do due diligence as to the action items on the
checklist? That would be a strange perspective since the key role of
every TLP PMC is to oversee the IP of each and every commit made to
the project's repository.
An attractive aspect of the IP Clearance protocol is that it *closes*
the podling loophole that might bring committers into the Foundation
outside of the meritocratic process. One thing that the votes on the
proposal and podling graduation do is sanction the list of podling
committers, who usually go on to become PMC members. In that case, we
do need the Incubator PMC to vote on the new committers, and since the
podling is an unproven community, we do need someone to decide if the
community meets our standard of meritocracy.
If code is developed outside of an ASF repository and mailing list,
then it is appropriate that we provide a pedigree for the code. It is
also appropriate that we have a central record of all such code that a
PMC is bringing int our repository. It is also appropriate that we
maintain that record at the Incubator, so that all external code
arrives in one place. But, I would suggest that the people best suited
to vote on the code itself are the people who are already making the
technical decisions about such code: The receiving PMC.
In the case of an IP Clearance, there is not a new community to
consider. What is before the Incubator PMC is the issue of whether the
code *can* be licensed to the Foundation. Whether the code *should* be
licensed to the Foundation is a decision best left to the receiving
PMC, who, by direction of the Board, already decide whether such code
should be licensed to the ASF every time there is a commit to that
project.
-Ted.
---------------------------------------------------------------------
To unsubscribe, e-mail: general-unsubscribe@incubator.apache.org
For additional commands, e-mail: general-help@incubator.apache.org
Re: The line between full incubation and IP clearance
Posted by Davanum Srinivas <da...@gmail.com>.
Noel,
I agree, as we need a better polic/process for IP Clearance, it can't
be left to one person. I already insisted on full incubation for the
specific issue. Though mind you, i'd love to have some guidelines i
can check off of to back up why i think a full incubation is prudent.
In this specific case, i think we need a closer look at what's
happening given the circumstances and the spectre of a large code base
helps reinforce that thought.
thanks,
dims
On 12/28/05, Noel J. Bergman <no...@devtech.com> wrote:
> Dims,
>
> > it's been a line in the sand AFAIK, i guess it's Noel's call.
>
> -1
>
> Although that might be technically true, we do things collectively in the
> ASF. Mind you, we've not had a process for voting on IP Clearance type
> submissions, so that's been a potential loophole.
>
> > I think we should insist on a full incubation for
> > Cherokee/Tomahawk/ADF (whatever it's name is today)
>
> I responded on the generic issue, but as for this specific submission, I
> don't know enough about it, yet. I just got back on e-mail today for the
> first time since Friday, so I've got some reading to do. :-) Please feel
> free to reply to my e-mail to Martin with your observations, but change the
> subject to reflect the specific, rather than general. :-)
>
> --- Noel
>
>
--
Davanum Srinivas : http://wso2.com/blogs/
---------------------------------------------------------------------
To unsubscribe, e-mail: general-unsubscribe@incubator.apache.org
For additional commands, e-mail: general-help@incubator.apache.org
RE: The line between full incubation and IP clearance
Posted by "Noel J. Bergman" <no...@devtech.com>.
Dims,
> it's been a line in the sand AFAIK, i guess it's Noel's call.
-1
Although that might be technically true, we do things collectively in the
ASF. Mind you, we've not had a process for voting on IP Clearance type
submissions, so that's been a potential loophole.
> I think we should insist on a full incubation for
> Cherokee/Tomahawk/ADF (whatever it's name is today)
I responded on the generic issue, but as for this specific submission, I
don't know enough about it, yet. I just got back on e-mail today for the
first time since Friday, so I've got some reading to do. :-) Please feel
free to reply to my e-mail to Martin with your observations, but change the
subject to reflect the specific, rather than general. :-)
--- Noel
---------------------------------------------------------------------
To unsubscribe, e-mail: general-unsubscribe@incubator.apache.org
For additional commands, e-mail: general-help@incubator.apache.org
Re: The line between full incubation and IP clearance
Posted by Davanum Srinivas <da...@gmail.com>.
Martin,
it's been a line in the sand AFAIK, i guess it's Noel's call. I don't
remember voting on some of the things on the site
(http://svn.apache.org/repos/asf/incubator/public/trunk/site-author/ip-clearance/)
<incubator_pmc_member hat="on">
I think we should insist on a full incubation for
Cherokee/Tomahawk/ADF (whatever it's name is today)
</incubator_pmc_member>
thanks,
dims
On 12/28/05, Martin Cooper <ma...@apache.org> wrote:
> Is there a well-defined line between these two mechanisms for bringing new
> code to the ASF? Not having seen the code yet, I can't say for sure, but I
> would expect the Oracle ADF Faces donation to be substantial, and include
> a framework that ties everything together. I would have thought that would
> have warranted full incubation, but it appears to be going through only
> the UP Clearance procedure. See:
>
> http://mail-archives.apache.org/mod_mbox/myfaces-dev/200512.mbox/%3c8b3ce3790512281044k7aa44090o8a35863f71717399@mail.gmail.com%3e
>
> I'm new in Incubator-land, so I'm not altogether savvy about processes and
> procedures, but wanted to ask about this one.
>
> --
> Martin Cooper
>
> ---------------------------------------------------------------------
> To unsubscribe, e-mail: general-unsubscribe@incubator.apache.org
> For additional commands, e-mail: general-help@incubator.apache.org
>
>
--
Davanum Srinivas : http://wso2.com/blogs/
---------------------------------------------------------------------
To unsubscribe, e-mail: general-unsubscribe@incubator.apache.org
For additional commands, e-mail: general-help@incubator.apache.org
Re: The line between full incubation and IP clearance
Posted by Martin Cooper <ma...@apache.org>.
On 12/28/05, Noel J. Bergman <no...@devtech.com> wrote:
>
> Martin,
>
> > Is there a well-defined line between these two mechanisms for bringing
> new
> > code to the ASF?
>
> No, not a well-defined one. The IP Clearance is primarily intended for
> use
> with a relatively simple Software Grant situation into an existing
> project,
> such as libtool, which we can contrast with mod_cli or mod_ftp. And,
> honestly, I think that we have made a few mistakes in the past by using
> the
> simpler form when a fuller incubation would have yielded a better result
> in
> terms of community building around new code.
>
> > I would expect the Oracle ADF Faces donation to be substantial,
> > and include a framework that ties everything together. I would
> > have thought that would have warranted full incubation
>
> Is there community, or just code? If a community comes with it, too, I
> would certainly expect incubation, however, the e-mail you referenced says
> (in part): "[after clearance,] the MyFaces committers treat the code as if
> we wrote it ourselves", and I don't see any mention of a community coming,
> too. But here is a sample grey area: one or two committers absorbed into
> a
> healthy community might be deemed acceptable, whereas a substantial number
> of new ones might not. So if you buy that argument, where do we draw the
> line, and how do we address potential abuse?
I suppose I'm asking prematurely, since, of course, I haven't seen a
proposal either. ;-) However, up until now, ADF Faces has been part of (or
an add-on to - I'm not sure) a commercial Oracle product (JDeveloper). There
may be OTN forums that it's discussed on; not sure if that counts as a
community. Also, since it's no longer all of ADF Faces that's going to be
offered to the ASF, I don't know how that would affect any existing
community.
I expect that some Oracle developers will be listed in the forthcoming
proposal, and they will most likely be needed, too. How many, I have no
idea. How much cross-pollination between the ADF Faces code base and the
existing MyFaces code is also an open question in my mind. At least
initially, I would expect that the ADF Faces code base will be set up as a
separate build and download from the existing MyFaces code, until the
collective team decides how to integrate the two, or even if integrating
them makes sense, versus treating ADF Faces as a separate sub-project.
--
Martin Cooper
> I'm new in Incubator-land, so I'm not altogether savvy about processes and
> > procedures, but wanted to ask about this one.
>
> It is a fine question. Hypothetically, without judging the merits of this
> submission, you raise the spectre of a type of situation that I'd want to
> ward against as a general rule.
>
> --- Noel
>
>
> ---------------------------------------------------------------------
> To unsubscribe, e-mail: general-unsubscribe@incubator.apache.org
> For additional commands, e-mail: general-help@incubator.apache.org
>
>
Re: The line between full incubation and IP clearance
Posted by Ted Husted <te...@gmail.com>.
On 12/28/05, Noel J. Bergman <no...@devtech.com> wrote:
> Is there community, or just code? If a community comes with it, too, I
> would certainly expect incubation,
+1
I would say that the line is whether the existing community is ready,
willing, and able to work with the code without adding untested
volunteers.
My suggestion that we use IP Clearance for the ADF donation is
predicated on the existing community being up to the challenge of
working with the code, and adding any new committers in the usual way.
-Ted.
---------------------------------------------------------------------
To unsubscribe, e-mail: general-unsubscribe@incubator.apache.org
For additional commands, e-mail: general-help@incubator.apache.org
RE: The line between full incubation and IP clearance
Posted by "Noel J. Bergman" <no...@devtech.com>.
Martin,
> Is there a well-defined line between these two mechanisms for bringing new
> code to the ASF?
No, not a well-defined one. The IP Clearance is primarily intended for use
with a relatively simple Software Grant situation into an existing project,
such as libtool, which we can contrast with mod_cli or mod_ftp. And,
honestly, I think that we have made a few mistakes in the past by using the
simpler form when a fuller incubation would have yielded a better result in
terms of community building around new code.
> I would expect the Oracle ADF Faces donation to be substantial,
> and include a framework that ties everything together. I would
> have thought that would have warranted full incubation
Is there community, or just code? If a community comes with it, too, I
would certainly expect incubation, however, the e-mail you referenced says
(in part): "[after clearance,] the MyFaces committers treat the code as if
we wrote it ourselves", and I don't see any mention of a community coming,
too. But here is a sample grey area: one or two committers absorbed into a
healthy community might be deemed acceptable, whereas a substantial number
of new ones might not. So if you buy that argument, where do we draw the
line, and how do we address potential abuse?
> I'm new in Incubator-land, so I'm not altogether savvy about processes and
> procedures, but wanted to ask about this one.
It is a fine question. Hypothetically, without judging the merits of this
submission, you raise the spectre of a type of situation that I'd want to
ward against as a general rule.
--- Noel
---------------------------------------------------------------------
To unsubscribe, e-mail: general-unsubscribe@incubator.apache.org
For additional commands, e-mail: general-help@incubator.apache.org