Can we release an M1 with a dependency on a MINA-Snapshot version???

[John O'Hara <jo...@gmail.com>]

That's what we need -- some mentor time!

We can use any code (a license is a license is a license).
The question is whether the Incubator PMC will allow it to become M1?

Cliff, James, Paul - any opinion?

In a nutshell: we have a "fix" for MINA that Qpid depends on, but has not
yet been adopted by a released MINA version.
We want to "Release early and often" becuase the user base is building and
they'd like some more fixed targets....

Can Qpid still release an M1?

John

On 16/11/06, Steve Vinoski <vi...@iona.com> wrote:
>
>
> On Nov 16, 2006, at 11:26 AM, John O'Hara wrote:
>
> > On 15/11/06, Steve Vinoski <vi...@iona.com> wrote:
> >>
> >> Yes, it can be long. But those are the rules by which we must abide
> >> under Apache. As Rajith points out in a separate email, it's not even
> >> clear that M1 as it stands can get out the door given its dependency
> >> on an unreleased version of mina.
> >>
> >> I don't make the rules. I just want to live by them, as I'm sure we
> >> all do, and I know that maven will help greatly in that regard.
> >
> >
> > With Apache licensed software, we can use any checked in code that
> > bears the
> > license.
>
> Chapter and verse, please. We've had a number of people over the past
> few days assert what the "real" rules are, and now you've just added
> another. These assertions have greatly differed. If I were the M1
> release manager, I would want to see an official Apache document that
> clearly states what's possible in this regard.
>
> I am surprised that our mentors haven't chimed in yet.
>
> --steve
>

Re: Can we release an M1 with a dependency on a MINA-Snapshot version???

[Robert Greig <ro...@gmail.com>]

On 16/11/06, Daniel Kulp <da...@iona.com> wrote:
>
> I would say release with the non-snapshot version (all the tests do pass),

OK the tests may pass (our improvement addresses fairness and
throughput issues) just now but what happens in the case where there
is a relatively serious bug?

I should also say that we have seen cases on our hardware where the
heartbeats do not work properly due to the fairness issue our MINA
patch addresses (heavy writers can result in read starvation).

> but stick the "SNAPSHOT" up on the wiki/web page with a huge disclaimer
> with it that says something like "this fixes problems XYZ, but this is an
> unofficial snapshot that is not an official ASF release.

This is asking our users to make a choice they may not be comfortable
making or feel qualified to make. Surely it is our job to give clear
guidance to our users on what they should run? If we make an release
and simultaneously say "by the way we don't really like that version
and here's a patch" what would they think?

RG

Re: Can we release an M1 with a dependency on a MINA-Snapshot version???

[Paul Fremantle <pz...@gmail.com>]

On 11/16/06, Daniel Kulp <da...@iona.com> wrote:
>
> From my standpoint, I will vote -1 (non-binding) on the general@incubator
> list (as well as here) for a release with a SNAPSHOT of an Apache
> project.
>
> Allowed by the license or not, I don't think the ASF should do anything to
> promote having unofficial versions of ASF artifacts being distributed
> around under the guise of another "release".   I think that puts the ASF
> in a bad possition.   The ASF release processes are there to protect the
> ASF.   This basically circumvents it.

I don't agree. The PMC votes on a release. That includes all the
components of that release. All those components have to be in good
legal standing. If there are legal issues with the MINA snapshot then
that is a -1. But the fact that it is a specific snapshot is not it
itself an issue.

Paul

-- 
Paul Fremantle
VP/Technology, WSO2 and OASIS WS-RX TC Co-chair

http://bloglines.com/blog/paulfremantle
paul@wso2.com

"Oxygenating the Web Service Platform", www.wso2.com

Re: Can we release an M1 with a dependency on a MINA-Snapshot version???

[Daniel Kulp <da...@iona.com>]

From my standpoint, I will vote -1 (non-binding) on the general@incubator 
list (as well as here) for a release with a SNAPSHOT of an Apache 
project.

Allowed by the license or not, I don't think the ASF should do anything to 
promote having unofficial versions of ASF artifacts being distributed 
around under the guise of another "release".   I think that puts the ASF 
in a bad possition.   The ASF release processes are there to protect the 
ASF.   This basically circumvents it.


If JPMC or RedHat or IONA or someone wants to create a release with that 
snapshot, sure.   The license will definitely let them.   But from an ASF 
standpoint, I don't think WE want to do it.


I would say release with the non-snapshot version (all the tests do pass), 
but stick the "SNAPSHOT" up on the wiki/web page with a huge disclaimer 
with it that says something like "this fixes problems XYZ, but this is an 
unofficial snapshot that is not an official ASF release.   It may cause 
other issues. blah blah blah"   Even better would be to point to the MINA 
website if they have a nightly snapshot or something available.   Help 
drive some traffic there.

I suppose all that legal stuff could go in the NOTICES file, but I'm still 
very uncomfortable with it from an ASF standpoint.  (caveat being I'm not 
an ASF member, but we still need look out for it as an ASF project)


Dan



On Thursday November 16 2006 12:06 pm, John O'Hara wrote:
> That's what we need -- some mentor time!
>
> We can use any code (a license is a license is a license).
> The question is whether the Incubator PMC will allow it to become M1?
>
> Cliff, James, Paul - any opinion?
>
> In a nutshell: we have a "fix" for MINA that Qpid depends on, but has
> not yet been adopted by a released MINA version.
> We want to "Release early and often" becuase the user base is building
> and they'd like some more fixed targets....
>
> Can Qpid still release an M1?
>
> John
>
> On 16/11/06, Steve Vinoski <vi...@iona.com> wrote:
> > On Nov 16, 2006, at 11:26 AM, John O'Hara wrote:
> > > On 15/11/06, Steve Vinoski <vi...@iona.com> wrote:
> > >> Yes, it can be long. But those are the rules by which we must
> > >> abide under Apache. As Rajith points out in a separate email, it's
> > >> not even clear that M1 as it stands can get out the door given its
> > >> dependency on an unreleased version of mina.
> > >>
> > >> I don't make the rules. I just want to live by them, as I'm sure
> > >> we all do, and I know that maven will help greatly in that regard.
> > >
> > > With Apache licensed software, we can use any checked in code that
> > > bears the
> > > license.
> >
> > Chapter and verse, please. We've had a number of people over the past
> > few days assert what the "real" rules are, and now you've just added
> > another. These assertions have greatly differed. If I were the M1
> > release manager, I would want to see an official Apache document that
> > clearly states what's possible in this regard.
> >
> > I am surprised that our mentors haven't chimed in yet.
> >
> > --steve

-- 
J. Daniel Kulp
Principal Engineer
IONA
P: 781-902-8727    C: 508-380-7194   F:781-902-8001
daniel.kulp@iona.com

Re: Can we release an M1 with a dependency on a MINA-Snapshot version???

[Daniel Kulp <da...@iona.com>]

On Thursday November 16 2006 12:54 pm, Paul Fremantle wrote:
> John
>
> There isn't any license issue about using the code. The only problem
> is that you release code will not be rebuildable at some point in the
> future. If you want a possible technical solution to that problem here
> is one.
>
> I suggest that you take a specific revision of the MINA code - say
> nnnnnn. Now build that, and rename the JARs to include the revision
> number.
>
> Store them somewhere on people.apache.org/~someone so that they can be
> retrieved by your build.

Until people.apache.org crashes and we lose 3 months of data....   :-(

Dan


>
> Now you have a replicable build.
>
> Paul
>
> On 11/16/06, John O'Hara <jo...@gmail.com> wrote:
> > That's what we need -- some mentor time!
> >
> > We can use any code (a license is a license is a license).
> > The question is whether the Incubator PMC will allow it to become M1?
> >
> > Cliff, James, Paul - any opinion?
> >
> > In a nutshell: we have a "fix" for MINA that Qpid depends on, but has
> > not yet been adopted by a released MINA version.
> > We want to "Release early and often" becuase the user base is
> > building and they'd like some more fixed targets....
> >
> > Can Qpid still release an M1?
> >
> > John
> >
> > On 16/11/06, Steve Vinoski <vi...@iona.com> wrote:
> > > On Nov 16, 2006, at 11:26 AM, John O'Hara wrote:
> > > > On 15/11/06, Steve Vinoski < vinoski@iona.com> wrote:
> > > >> Yes, it can be long. But those are the rules by which we must
> > > >> abide under Apache. As Rajith points out in a separate email,
> > > >> it's not even clear that M1 as it stands can get out the door
> > > >> given its dependency on an unreleased version of mina.
> > > >>
> > > >> I don't make the rules. I just want to live by them, as I'm sure
> > > >> we all do, and I know that maven will help greatly in that
> > > >> regard.
> > > >
> > > > With Apache licensed software, we can use any checked in code
> > > > that bears the
> > > > license.
> > >
> > > Chapter and verse, please. We've had a number of people over the
> > > past few days assert what the "real" rules are, and now you've just
> > > added another. These assertions have greatly differed. If I were
> > > the M1 release manager, I would want to see an official Apache
> > > document that clearly states what's possible in this regard.
> > >
> > > I am surprised that our mentors haven't chimed in yet.
> > >
> > > --steve

-- 
J. Daniel Kulp
Principal Engineer
IONA
P: 781-902-8727    C: 508-380-7194   F:781-902-8001
daniel.kulp@iona.com

Re: Can we release an M1 with a dependency on a MINA-Snapshot version???

[Paul Fremantle <pz...@gmail.com>]

John

There isn't any license issue about using the code. The only problem
is that you release code will not be rebuildable at some point in the
future. If you want a possible technical solution to that problem here
is one.

I suggest that you take a specific revision of the MINA code - say
nnnnnn. Now build that, and rename the JARs to include the revision
number.

Store them somewhere on people.apache.org/~someone so that they can be
retrieved by your build.

Now you have a replicable build.

Paul

On 11/16/06, John O'Hara <jo...@gmail.com> wrote:
> That's what we need -- some mentor time!
>
> We can use any code (a license is a license is a license).
> The question is whether the Incubator PMC will allow it to become M1?
>
> Cliff, James, Paul - any opinion?
>
> In a nutshell: we have a "fix" for MINA that Qpid depends on, but has not
> yet been adopted by a released MINA version.
> We want to "Release early and often" becuase the user base is building and
> they'd like some more fixed targets....
>
> Can Qpid still release an M1?
>
> John
>
> On 16/11/06, Steve Vinoski <vi...@iona.com> wrote:
> >
> > On Nov 16, 2006, at 11:26 AM, John O'Hara wrote:
> >
> > > On 15/11/06, Steve Vinoski < vinoski@iona.com> wrote:
> > >>
> > >> Yes, it can be long. But those are the rules by which we must abide
> > >> under Apache. As Rajith points out in a separate email, it's not even
> > >> clear that M1 as it stands can get out the door given its dependency
> > >> on an unreleased version of mina.
> > >>
> > >> I don't make the rules. I just want to live by them, as I'm sure we
> > >> all do, and I know that maven will help greatly in that regard.
> > >
> > >
> > > With Apache licensed software, we can use any checked in code that
> > > bears the
> > > license.
> >
> > Chapter and verse, please. We've had a number of people over the past
> > few days assert what the "real" rules are, and now you've just added
> > another. These assertions have greatly differed. If I were the M1
> > release manager, I would want to see an official Apache document that
> > clearly states what's possible in this regard.
> >
> > I am surprised that our mentors haven't chimed in yet.
> >
> > --steve
> >
>
>


-- 
Paul Fremantle
VP/Technology, WSO2 and OASIS WS-RX TC Co-chair

http://bloglines.com/blog/paulfremantle
paul@wso2.com

"Oxygenating the Web Service Platform", www.wso2.com

Re: Can we release an M1 with a dependency on a MINA-Snapshot version???

[Cliff Schmidt <cl...@gmail.com>]

On 11/16/06, John O'Hara <jo...@gmail.com> wrote:
> That's what we need -- some mentor time!
>
> We can use any code (a license is a license is a license).
> The question is whether the Incubator PMC will allow it to become M1?
>
> Cliff, James, Paul - any opinion?
>
> In a nutshell: we have a "fix" for MINA that Qpid depends on, but has not
> yet been adopted by a released MINA version.
> We want to "Release early and often" becuase the user base is building and
> they'd like some more fixed targets....
>
> Can Qpid still release an M1?
>
> John
>
> On 16/11/06, Steve Vinoski <vi...@iona.com> wrote:
> >
> > On Nov 16, 2006, at 11:26 AM, John O'Hara wrote:
> >
> > > On 15/11/06, Steve Vinoski < vinoski@iona.com> wrote:
> > >>
> > >> Yes, it can be long. But those are the rules by which we must abide
> > >> under Apache. As Rajith points out in a separate email, it's not even
> > >> clear that M1 as it stands can get out the door given its dependency
> > >> on an unreleased version of mina.
> > >>
> > >> I don't make the rules. I just want to live by them, as I'm sure we
> > >> all do, and I know that maven will help greatly in that regard.
> > >
> > >
> > > With Apache licensed software, we can use any checked in code that
> > > bears the
> > > license.
> >
> > Chapter and verse, please. We've had a number of people over the past
> > few days assert what the "real" rules are, and now you've just added
> > another. These assertions have greatly differed. If I were the M1
> > release manager, I would want to see an official Apache document that
> > clearly states what's possible in this regard.
> >
> > I am surprised that our mentors haven't chimed in yet.

There is definitely no official policy describing what you're talking about.

Have qpid committers discussed this with the mina folks yet?  I would
think that's the first step if not already done.  The ideal thing
would be that you all are able to help mina put out the release you
need.

There are definitely no legal problems with this, as long as the PMC
(Incubator PMC, but first with the qpid PPMC) takes responsibility for
what they are releasing.  By taking responsibility, I mean being as
sure about the legal issues of the mina code as you are about your own
code; I also mean taking responsibility for dealing with any incompat
issues caused by the snapshot and helping to get this stuff merged
back together.

I bet there are other ASF projects wondering about the same issue; so
I might raise a thread on the incubator general list to see if we can
get some consensus there.

Cliff