OpenJDK governing board, constitution

[Neal Gafter]

The OpenJDK governing board, having had its life extended by a year, is now scheduled to dissolve in four months, with two of its non-Sun positions remaining unfilled.  The last published meeting minutes were from April 2008, at which it was agreed that the GB would strive for a draft Constitution by the end of 2008.

Who are the seven members of the governing board?  Can we please see the minutes of meetings after April, and get a status report on the Constitution?

[Mark Wielaard]

Hi Neal,

On Wed, 2009-01-07 at 19:15 -0800, Neal Gafter wrote:
> The OpenJDK governing board, having had its life extended by a year,
> is now scheduled to dissolve in four months, with two of its non-Sun
> positions remaining unfilled.  The last published meeting minutes were
> from April 2008, at which it was agreed that the GB would strive for a
> draft Constitution by the end of 2008.
>
> Who are the seven members of the governing board?  Can we please see
> the minutes of meetings after April, and get a status report on the
> Constitution?

Good questions, and I don't know the answers.
May I ask a meta-question in return though?

What do we really expect from the governance board?

While it has been completely missing in action for the last half year, I
cannot say I have actually missed it. Things do seem to happen anyway.
What kind of issues do we as hackers really have that could be solved by
having an active governance board and a "constitution"?

Cheers,

Mark

[Andrew Haley]

Mark Wielaard wrote:

> On Wed, 2009-01-07 at 19:15 -0800, Neal Gafter wrote:
>> The OpenJDK governing board, having had its life extended by a year,
>> is now scheduled to dissolve in four months, with two of its non-Sun
>> positions remaining unfilled.  The last published meeting minutes were
>> from April 2008, at which it was agreed that the GB would strive for a
>> draft Constitution by the end of 2008.
>>
>> Who are the seven members of the governing board?  Can we please see
>> the minutes of meetings after April, and get a status report on the
>> Constitution?
>
> Good questions, and I don't know the answers.
> May I ask a meta-question in return though?
>
> What do we really expect from the governance board?
>
> While it has been completely missing in action for the last half year, I
> cannot say I have actually missed it. Things do seem to happen anyway.
> What kind of issues do we as hackers really have that could be solved by
> having an active governance board and a "constitution"?

gcc itself is a model.  The steering committee only gets involved in
political matters and appoints maintainers.  The maintainers control
all the aspects of gcc itself, including features and releases.  In
theory the steering committee could resolve deadlocks between
maintainers, but that has AFAIAA never happened.  Most gcc maintainers
don't ever have to deal with the steering committee.  The steering
committee makes sure that policies to do with freedom and licensing
are followed.  It negotiates with the FSF when problems involving
free sofwtare policy arise.

So, the less the steering committee does, the better.  An active OpenJDK
governance board and a "constitution", it hopefully would not have affected
our work at all.  Their job is to keep out of the way of the people
doing real work.  They've been doing this quite well.

Andrew.

[Neal Gafter]

The reason I ask is that I'm worried that openJDK may turn into the defacto mechanism for features getting into the platform.  The JCP used to play that role, but there has been little activity in forming a JSR for Java SE 7 in the past few years.  I've noticed that openjdk7 is more and more being called Java 7, JDK7, etc, even though it doesn't implement a platform specification approved by the JCP.  If openjdk is to become the mechanism by which features are added to the platform, it would be better for the governance model to acknowledge and support that.

On Thu, Jan 8, 2009 at 6:15 AM, Mark Wielaard <mark@...> wrote:

Hi Neal,

On Wed, 2009-01-07 at 19:15 -0800, Neal Gafter wrote:
> The OpenJDK governing board, having had its life extended by a year,
> is now scheduled to dissolve in four months, with two of its non-Sun
> positions remaining unfilled.  The last published meeting minutes were
> from April 2008, at which it was agreed that the GB would strive for a
> draft Constitution by the end of 2008.
>
> Who are the seven members of the governing board?  Can we please see
> the minutes of meetings after April, and get a status report on the
> Constitution?

Good questions, and I don't know the answers.
May I ask a meta-question in return though?

What do we really expect from the governance board?

While it has been completely missing in action for the last half year, I
cannot say I have actually missed it. Things do seem to happen anyway.
What kind of issues do we as hackers really have that could be solved by
having an active governance board and a "constitution"?

Cheers,

Mark

[Mark Wielaard]

Hi Neal,

On Thu, 2009-01-15 at 10:58 -0800, Neal Gafter wrote:
> The reason I ask is that I'm worried that openJDK may turn into the
> defacto mechanism for features getting into the platform.  The JCP
> used to play that role, but there has been little activity in forming
> a JSR for Java SE 7 in the past few years.  I've noticed that openjdk7
> is more and more being called Java 7, JDK7, etc, even though it
> doesn't implement a platform specification approved by the JCP.  If
> openjdk is to become the mechanism by which features are added to the
> platform, it would be better for the governance model to acknowledge
> and support that.

That is a very good point. Thanks for bringing that up. Currently we act
as if the JCP has some kind of status that restricts certain kinds of
modifications to public APIs. But this has been kind of a problem since
access to JSRs and JCKs is not guaranteed to be free of restrictions
that are incompatible with our way of working in a public and open free
software project. (See the multiple months long [still unresolved!]
thread on pkg-distro-dev about the inability to use a lot of official
JSR documents for work on OpenJDK.)

If we can get a smoother way of working on standards through OpenJDK
that would indeed be very welcome. But does indeed need to come with
some kind of guide lines of how we would like to handle such
responsibilities.

Cheers,

Mark

[Andrew Haley]

Neal Gafter wrote:

>> On Wed, 2009-01-07 at 19:15 -0800, Neal Gafter wrote:

>>> The OpenJDK governing board, having had its life extended by a
>>> year, is now scheduled to dissolve in four months, with two of its
>>> non-Sun positions remaining unfilled.  The last published meeting
>>> minutes were from April 2008, at which it was agreed that the GB
>>> would strive for a draft Constitution by the end of 2008.

>>> Who are the seven members of the governing board?  Can we please
>>> see the minutes of meetings after April, and get a status report
>>> on the Constitution?

> The reason I ask is that I'm worried that openJDK may turn into the
> defacto mechanism for features getting into the platform.  The JCP
> used to play that role, but there has been little activity in
> forming a JSR for Java SE 7 in the past few years.  I've noticed
> that openjdk7 is more and more being called Java 7, JDK7, etc, even
> though it doesn't implement a platform specification approved by the
> JCP.  If openjdk is to become the mechanism by which features are
> added to the platform,

I don't see how that can happen.  For Java SE 7 to be released there
must be a platform specification, and there must be a TCK.  openjdk7
is a bunch of packages slated for Java SE 7 that may or may not get to
be in the platform.

> it would be better for the governance model to acknowledge and support that.

It would, yes, but it would be a huge change.

In the past there have undoubtedly been developments very much like
the openjdk7 tree, where platform integration has proceeded prior to
the formal platform specification.  This is essential: you need to
make sure that a design works in a reasonable way before its
specification is finalized.  The only difference now is that the
openjdk7 tree is open.

Andrew.

[Volker Simonis]

What I find most astonishing on this thread is that apparently neither
the Governance Board members nor a Sun representative have considered
it important enough to comment or respond. Strange...

On 1/16/09, Andrew Haley <aph@...> wrote:

> Neal Gafter wrote:
>
>  >> On Wed, 2009-01-07 at 19:15 -0800, Neal Gafter wrote:
>
>  >>> The OpenJDK governing board, having had its life extended by a
>  >>> year, is now scheduled to dissolve in four months, with two of its
>  >>> non-Sun positions remaining unfilled.  The last published meeting
>  >>> minutes were from April 2008, at which it was agreed that the GB
>  >>> would strive for a draft Constitution by the end of 2008.
>
>  >>> Who are the seven members of the governing board?  Can we please
>  >>> see the minutes of meetings after April, and get a status report
>  >>> on the Constitution?
>
>
> > The reason I ask is that I'm worried that openJDK may turn into the
>  > defacto mechanism for features getting into the platform.  The JCP
>  > used to play that role, but there has been little activity in
>  > forming a JSR for Java SE 7 in the past few years.  I've noticed
>  > that openjdk7 is more and more being called Java 7, JDK7, etc, even
>  > though it doesn't implement a platform specification approved by the
>  > JCP.  If openjdk is to become the mechanism by which features are
>  > added to the platform,
>
>
> I don't see how that can happen.  For Java SE 7 to be released there
>  must be a platform specification, and there must be a TCK.  openjdk7
>  is a bunch of packages slated for Java SE 7 that may or may not get to
>  be in the platform.
>
>
>  > it would be better for the governance model to acknowledge and support that.
>
>
> It would, yes, but it would be a huge change.
>
>  In the past there have undoubtedly been developments very much like
>  the openjdk7 tree, where platform integration has proceeded prior to
>  the formal platform specification.  This is essential: you need to
>  make sure that a design works in a reasonable way before its
>  specification is finalized.  The only difference now is that the
>  openjdk7 tree is open.
>
>
>  Andrew.
>

[Doug Lea]

Neal Gafter wrote:
> The reason I ask is that I'm worried that openJDK may turn into the
> defacto mechanism for features getting into the platform.  The JCP used
> to play that role, but there has been little activity in forming a JSR
> for Java SE 7 in the past few years.

There has been some (sadly unavoidable) lack of transparency
in JCP that has given many people this mis-impression. The
JCP and its Executive Committee (that I am on) very much wants
the JCP to play the central role in driving major releases.
However, it has been stuck for a long while in policy impasses
(such as disputes over terms of TCK tests involving undisclosable
legal matters) that have made it impossible to approve a Java7
Release JSR. Everyone involved hopes that these are resolved soon.
And many of us have invested a fair amount of time trying
to help resolve them.

In the absence of resolution, Java7 plans have been left in a
long gestation mode. Spec leads and contributors (not just
at Sun) involved with changes and additions likely to make it in
to a next major release seem to be proceeding with plans, but
without an identifiable central coordination point. Placing
likely Java7 release contributions in openJDK is a convenient
way to maintain progress in the mean time, and presence
of code in openJDK repositories is one good indication of
some of functionality that some of them intend to include.

When a Java7 Release JSR can be proposed, it will
probably be able proceed quickly. Until then, the whole
process has an unfortunate shadowy-cabal appearance,
which does indeed suck.

-Doug

[Simon Phipps-2]

On Jan 8, 2009, at 16:47, Andrew Haley wrote:

> So, the less the steering committee does, the better.  An active  
> OpenJDK
> governance board and a "constitution", it hopefully would not have  
> affected
> our work at all.  Their job is to keep out of the way of the people
> doing real work.  They've been doing this quite well.

I agree. I'm not 100% happy that there have been no meetings, but I do  
feel that the overall goal we agreed - to make sure that any kind of  
committee would stay out of the way of the actual work, and that we'd  
wait until it was clear what the need was until acting - still seems  
the right one and seems to have been achieved by default. It may well  
be smart to keep going like this rather than create some document for  
the sake of having it.


On Jan 15, 2009, at 19:58, Neal Gafter wrote:

> The reason I ask is that I'm worried that openJDK may turn into the  
> defacto mechanism for features getting into the platform.

There's already code in the wider OpenJDK community that's  
experimental or pragmatic and the world hasn't ended (in fact the Java  
platform is now freely available on Linux) so I'm not sure I  
understand why this is relevant to an open source community. I'm not  
aware (correct me if I'm wrong) that Mozilla has governance statements  
about adherence to W3C process, for example.

S.

[Simon Phipps-3]

On Jan 8, 2009, at 16:47, Andrew Haley wrote:

> So, the less the steering committee does, the better.  An active  
> OpenJDK
> governance board and a "constitution", it hopefully would not have  
> affected
> our work at all.  Their job is to keep out of the way of the people
> doing real work.  They've been doing this quite well.

I agree. I'm not 100% happy that there have been no meetings, but I do  
feel that the overall goal we agreed - to make sure that any kind of  
committee would stay out of the way of the actual work, and that we'd  
wait until it was clear what the need was until acting - still seems  
the right one and seems to have been achieved by default. It may well  
be smart to keep going like this rather than create some document for  
the sake of having it.


On Jan 15, 2009, at 19:58, Neal Gafter wrote:

> The reason I ask is that I'm worried that openJDK may turn into the  
> defacto mechanism for features getting into the platform.

There's already code in the wider OpenJDK community that's  
experimental or pragmatic and the world hasn't ended (in fact the Java  
platform is now freely available on Linux) so I'm not sure I  
understand why this is relevant to an open source community. I'm not  
aware (correct me if I'm wrong) that Mozilla has governance statements  
about adherence to W3C process, for example.

S.

[Simon Phipps-3]

Sorry for the multiple posting, folks, looks like the discuss list  
rejects mail from me no matter where it comes from :-)

[Andy Tripp-2]

Simon Phipps wrote:

>
> On Jan 8, 2009, at 16:47, Andrew Haley wrote:
>
>> So, the less the steering committee does, the better.  An active OpenJDK
>> governance board and a "constitution", it hopefully would not have
>> affected
>> our work at all.  Their job is to keep out of the way of the people
>> doing real work.  They've been doing this quite well.
>
> I agree. I'm not 100% happy that there have been no meetings, but I do
> feel that the overall goal we agreed - to make sure that any kind of
> committee would stay out of the way of the actual work, and that we'd
> wait until it was clear what the need was until acting - still seems the
> right one and seems to have been achieved by default.

I thought the overall goal of the GB was to create a constitution, not
"stay out of the way of actual work". In fact, looking again at the OpenJDK
charter, it's quite clear that that's the purpose of the GB. You say
"we agreed" above, and assuming the "we" is the GB, I don't see any
big discussion in the meeting minutes about "staying out of the way".

The meeting minutes seem to indicate serious dedication to creating a
constitution. There's no indication there that having no further meetings
and making no progress on a constitution might be a reasonable outcome.

Is there any documentation that the GB ever really agreed that
"we'd wait until it was clear what the need was"? Documented or not,
I'm having a lot of trouble believing that the GB really agreed on this.

> It may well be
> smart to keep going like this rather than create some document for the
> sake of having it.

So you create a GB who's main purposes are to create a constitution and
resolve disputes, and now apparently the thinking is that there's no need for
a constitution, so it's OK that the GB never produced anything? What has changed
so that a constitution is no longer needed? Who exactly is the "we"
that doesn't think a constitution is needed and that no GB meetings are needed?

And most importantly, doesn't it seem like there's now an issue that requires
the GB and/or a constitution...namely, the issue of who decides what goes
into "the platform" - openJDK or the JCP? Is the JCP dead, and Neal should
just just try to get closures into the openJDK code? Or is the JCP alive,
and will ensure that openJDK code won't leak into JDK without a JSR?

I think it's time the GB either do what it's supposed to do, or officially
disband and let Java drift where it may. It's just silly to have a group
that does nothing and considers that OK because "we" agreed that it should
"stay out of the way".

Andy

[Simon Phipps-2]

On Jan 16, 2009, at 19:09, Andy Tripp wrote:

>  What has changed so that a constitution is no longer needed?

Who said it wasn't needed? It's just that the only dispute that has  
arisen so far appears to be this one, and experience elsewhere shows  
that creating governance in a vacuum leads to bad decisions. What  
actually is the need precipitating your passion, beyond an arbitrary  
date passing (through, I agree, apparent neglect)?

S.

[Neal Gafter]

On Fri, Jan 16, 2009 at 10:16 AM, Simon Phipps <webmink@...> wrote:

On Jan 16, 2009, at 19:09, Andy Tripp wrote:

 What has changed so that a constitution is no longer needed?

Who said it wasn't needed? It's just that the only dispute that has arisen so far appears to be this one, and experience elsewhere shows that creating governance in a vacuum leads to bad decisions. What actually is the need precipitating your passion, beyond an arbitrary date passing (through, I agree, apparent neglect)?

I'm not sure I would characterize this as a dispute or a problem.  I'm not upset or anything, I'm just trying to find out what the processes are for things moving forward.  If people are happy with the current process for changes going into JDK7 before the platform JSR exists (whatever that process is), then that's OK with me.

[Andy Tripp-2]

Simon Phipps wrote:
>
> On Jan 16, 2009, at 19:09, Andy Tripp wrote:
>
>>  What has changed so that a constitution is no longer needed?
>
> Who said it wasn't needed?

I'm deducing that you believe that a constitution is not needed, because
the GB is the one who was supposed to create it, it hasn't, and that seems to be OK
with you. Do you think it's needed or not?

> It's just that the only dispute that has
> arisen so far appears to be this one,

Naturally, no issues are going to "arise" if there's nowhere for them to go.
Closures are a good example. As you know, there are several proposals and
at least one implementation out there. At this point, it looks like Neal
could probably get BGGA closures into openjdk just by committing the code.
Could that code then flow into the JDK (without a JSR) or not?

> and experience elsewhere shows
> that creating governance in a vacuum leads to bad decisions. What
> actually is the need precipitating your passion, beyond an arbitrary
> date passing (through, I agree, apparent neglect)?

I want to know if the JDK is now in the process of forking into
a) the JCP-controlled JDK, which at this point looks like it may never have a "7" release
b) OpenJDK/IcedTea, self-controlled, with ongoing changes.

For me, it's entirely academic, but I'm sure someone like Neal wants to know
where he should put his efforts. And I would think Doug Leah would not be too happy
to hear that all his GB efforts are for nil.

So again, I ask "what has changed?" Who decided that a GB was needed, wrote up an
OpenJDK charter, established a GB to write a constitution, and has now changed his mind?
Has the entire GB decided not to bother with a constitution?
Where there discussions? Was there a vote?

I guess I'm just curious.
Andy

>
> S.
>
>

Andy Tripp wrote:

> Simon Phipps wrote:
>>
>> On Jan 16, 2009, at 19:09, Andy Tripp wrote:
>>
>>>  What has changed so that a constitution is no longer needed?
>>
>> Who said it wasn't needed?
>
> I'm deducing that you believe that a constitution is not needed, because
> the GB is the one who was supposed to create it, it hasn't, and that
> seems to be OK
> with you. Do you think it's needed or not?
>
>> It's just that the only dispute that has arisen so far appears to be
>> this one,
>
> Naturally, no issues are going to "arise" if there's nowhere for them
> to go.
> Closures are a good example. As you know, there are several proposals and
> at least one implementation out there. At this point, it looks like Neal
> could probably get BGGA closures into openjdk just by committing the
> code. Could that code then flow into the JDK (without a JSR) or not?
>

You said 'JDK' in the last phrase so I suppose you mean the thing we
label 'JavaSE' and ship through java.sun.com and java.com.

The JCK is derived from JCP specs and tests for conformance.  Our
product named JavaSE (a.k.a. JDK) has to meet the JSR specs and hence is
run through the JCK just like all other Java implementations.  Assuming
the JCK can successfully reject a compiler that supports an extension
like closures if closures are not in the spec, then a Java compiler
could not support that extension and call itself a Java compiler.

Thinking off the top of my head I suppose that such support could be
turned on by an option and so long as the extension is "off" by default
it would meet JCP requirements.


>> and experience elsewhere shows that creating governance in a vacuum
>> leads to bad decisions. What actually is the need precipitating your
>> passion, beyond an arbitrary date passing (through, I agree, apparent
>> neglect)?
>
> I want to know if the JDK is now in the process of forking into
> a) the JCP-controlled JDK, which at this point looks like it may never
> have a "7" release
> b) OpenJDK/IcedTea, self-controlled, with ongoing changes.

We didn't open source the spec process, we open sourced an
implementation.  The JSR's are the spec and are controlled by the JCP.

I think that questions about opening the spec process need to be brought
to the JCP.  The OpenJDK project is not the best place to pursue opening
up the spec process.

For prior Java<n> cycles for values of n less than 7 .. "little changes"
were rolled into the "Platform JSR" which allows for changes on the size
of a bug fix or minor feature which aren't significant enough to require
a JSR.  The same would be true for the Java7 cycle assuming we're able
to settle the issues which currently prevent a Java7 JSR from existing.  
Hence such ongoing small changes in OpenJDK7 would be rolled into the
platform JSR I'm assuming will eventually exist.

- David Herron

[Mark Reinhold]

The OpenJDK Governance Board (GB) has not met since last April because
Sun has not yet appointed anyone to the two open GB seats.  Sun and the
GB considered various candidates early on but then the effort faded into
the background over the summer in the face of more pressing issues.

Without a full GB, there has been no further work on the Constitution.

In the meantime, as others have noted, work is getting done and the
OpenJDK Community is, slowly, expanding.

It is definitely the view of the current GB Members that the GB is, and
ought to be, more of a legislative and judiciary body than an executive.
The GB exists to define the high-level processes by which the Community
operates, and to adjudicate disputes when they arise.  It should never,
ever be in the position of making technical decisions.  I fully expect
this view to be codified in the Constitution.

With regard to the JCP, as Doug observed there is not yet a Java SE 7
Platform JSR due to a policy impasse within the Executive Committee.
Sun will submit such a JSR as soon as that issue is resolved.  In the
meantime the JDK 7 Project here in the OpenJDK Community will move
forward to prototype some features and enhancements that might -- or
might not -- wind up in the SE 7 Platform Specification.

Some of these improvements will come from outside Sun.  Several major
contributions from non-Sun developers are already lined up, and we'll
shortly announce a process by which anyone can propose a new feature.
Stay tuned.

- Mark

[Simon Phipps-2]

On Jan 16, 2009, at 19:51, Andy Tripp wrote:

> Simon Phipps wrote:
>> On Jan 16, 2009, at 19:09, Andy Tripp wrote:
>>> What has changed so that a constitution is no longer needed?
>> Who said it wasn't needed?
>
> I'm deducing that you believe that a constitution is not needed,  
> because
> the GB is the one who was supposed to create it, it hasn't, and that  
> seems to be OK
> with you. Do you think it's needed or not?

Yes, I think a document on procedures will be needed at some stage. We  
already have an interim statement that seems OK with the people who  
have commented and seems to be causing no problems, so pragmatically  
there doesn't seem to be a gap at present. By the way, I am not  
setting the agenda for the OGB, I'm a participant like all the others.

>> It's just that the only dispute that has arisen so far appears to  
>> be this one,
>
> Naturally, no issues are going to "arise" if there's nowhere for  
> them to go.

The OGB exists and has a mailing list, hence this discussion.

> Closures are a good example. As you know, there are several  
> proposals and
> at least one implementation out there. At this point, it looks like  
> Neal
> could probably get BGGA closures into openjdk just by committing the  
> code. Could that code then flow into the JDK (without a JSR) or not?

I hope that whatever governance the OpenJDK community finally creates,  
it will never give the OGB a role of technical judgement.  I agree  
with Andrew that open source steering bodies are there to establish  
order, not architecture. As David has pointed out, the content of the  
reference implementation is dictated by the spec, not by the work of  
any code contributor of any affiliation.

S.

[Mark Wielaard]

Hi Geir,

On Sat, 2009-01-17 at 06:40 -0500, Geir Magnusson Jr. wrote:

> On Jan 15, 2009, at 7:11 PM, Mark Wielaard wrote:
> > That is a very good point. Thanks for bringing that up. Currently we  
> > act
> > as if the JCP has some kind of status that restricts certain kinds of
> > modifications to public APIs.
>
> Yes, the JCP has *exactly* that status - public APIs must comply with  
> the Java SE specification as produced by the expert group, or else the  
> software is not Java (or, in deference to Sun's ownership of the Java  
> trademark, "Java compatible").  I think this reliable consistency is  
> one of the great things about Java The Ecosystem (as well as Java the  
> Platform).

Sure, consistency and having unambiguous, strong standards are great for
everybody. No argument there. But if the rules around (producing) those
standards are counter to the spirit of free software then it takes away
from the user instead of adding value. We have to strive for public,
open free specifications with testsuites that are free for everybody to
use to verify any implementation's claim of compatibility with such
standards. I am not convinced the current state of the JCP encourages
that though. If through OpenJDK we can improve the process of producing
specs, the reference implementation and free test suites, then I am all
for it.

> > But this has been kind of a problem since
> > access to JSRs and JCKs is not guaranteed to be free of restrictions
> > that are incompatible with our way of working in a public and open  
> > free software project.
>
> It turns out that's the least of your problems.

Yes, the biggest problem was getting a full free reference platform for
the Java platform. Although we worked very hard on that through the
various efforts around GNU Classpath and friends, gcj, kaffe, and
finally with harmony, it cannot be denied that Sun's liberation of
almost all of their core platform implementation code base helped
enormously. And doing it in a way that united their effort with almost
all of the existing libre-java community can only be given the highest
praise.

I might be highly critical about some of the processes, the non-open
specs, the TCK being non-free and only available under a NDA forcing
people to work in secret cut of from the rest of the community (but
again high praise for Sun coming up with something that at least lets
people produce Free Software and doesn't get in the way of releasing the
results under the GPL) and the non-transparent trademark rules. But I do
realize that the biggest and most important hurdle has been taken now.
That we are slowly but surely creating a community that produces a fully
free Java platform together, even if some of the steps forward might be
still tricky.

>   As you know, the ASF  
> is engaged in what is now a multi-year battle to get the Java SE 5 TCK  
> under terms compatible with being able to distribute the resulting  
> tested binary under an open source license.

Yes, we started that process 12 years ago, and even before we started
Harmony we tried to unite the free java groups and get access to old
TCKs. http://lwn.net/Articles/184967/ One of the reasons that I was one
of the Harmony founders, which I and lots of others hoped would be the
ultimate unification of all the java-libre efforts that would not only
bring us a solid, full, free java implementation shared by lots of
groups, but also would give us the political cloud with the JCP
community. We all know how that ended. As I said before I think your
actions were not helpful.
http://gnu.wildebeest.org/diary/2007/04/21/openjck/

I hope we can move past that sad history though and focus on the future.
Now that we have a full free Java compatible platform for Java SE 6 lets
focus on making the processes for getting the same for Java SE 7 and not
just having free code, but also open processes (including open and free
specifications and finally a free TCK!) instead of harping on the past
failures.

> Java will never really be free until we get past all of this.  Please  
> inform RMS.

You seem hung up on the term Java(TM). Yes, it would be great if we had
a more open, transparent and Free Software compatible way of handling
the trademark issue. But don't confuse naming with code. The code is all
out there, under free software licenses. And even some binaries produced
have been certified as passing the JCK - in a way that is less from
ideal seeing the TCK itself isn't Free Software, but the resulting code
is fully free software. I do talk with RMS from time to time and he
knows my position, goals and the work that still has to be done.

Cheers,

Mark

[Mark Wielaard]

Hi Geir,

On Sat, 2009-01-17 at 09:53 -0500, Geir Magnusson Jr. wrote:
> What do you think we've been fighting for at the ASF for the last 8  
> years, and specifically, the last 2 for the JCK?

I appreciate you trying to fight. And trying to change things from
within the JCP. Which is indeed brave. You share that optimism with
Dalibor who also always urges to not burn down that which produces bad
results and be positive that change can happen. I might have become to
cynical with the results not being free, open and community friendly
over that many years, that I focus my efforts there where I can actually
produce free code in the open. You risk legitimating an institution
which doesn't guarantee software freedom without achieving any of your
goals. And alienating the community that you say you are fighting for
because you hide any talks and results behind the shadowy-cabal that you
have become part of.

> > If through OpenJDK we can improve the process of producing
> > specs, the reference implementation and free test suites, then I am  
> > all for it.
>
> You might argue that it's better, because you can get the TCK for use  
> in OpenJDK and derivatives

It is better because it is under terms that allow publishing any code
that is tested and/or passes with it under a free software license that
guarantees that the source, and not unimportantly, all patent claims
must be shared under reciprocal terms, without any restrictions on use
for any purpose by any user.

That said the current terms are certainly not good enough. Having the
TCK as proprietary software is bad, having people cut of from the rest
of the community through NDAs is anti-social and not giving anybody the
change to test any implementation as you wish is just very unfair and
unproductive.

> Another view is that they masterfully split the free/libre/open java  
> community, exploiting long-standing license fault-lines, in order to  
> counteract the threat that Harmony represented - a quality, performant  
> open source implementation with an *open, free community* under a  
> permissive license.

We started Harmony to unite the various free java efforts that we were
working on in the hope we could also work closer with the Apache
community. That it then turned out to split the community with an
apache-only effort was never what I, and other founders, like Dalibor
and Tom, intended it to be. I am glad Sun talked to the libre-java
community before starting their own effort and kept us in the loop about
their plans and desires to work together. I am not saying the
cooperation is perfect, there is a lot to improve. But we keep talking
and trying to work together. Our renewed Fosdem talks cooperation is
very indicative of that effort, and I am happy that everybody will take
the time again to come and exchange views.

> But it didn't work out, mainly because you never could consider  
> yourself producing software under the AL because of your views towards  
> "software hoarding"

No, creating an alternative code base incompatible with almost all the
existing efforts and not considering working together on a shared common
interface to all the components that 30 existing runtimes, class
libraries, jits, compilers, etc. already were using and working on
together was what made the harmony effort fail. I might not like
"software hoarding", and I certainly prefer using copyleft licenses that
are fairly reciprocal, but being expressly incompatible was what I
objected to. Any license that would be compatible with what the exiting
communities were using would have been OK. Luckily then the FSF did
solve a lot of those issue though by finally upgrading the GPL and
making compatibility an explicit goal. Please do reread "Toward a Free
Java" http://lwn.net/Articles/184967/ if you don't get what the history
is here.

> And if you're not providing code, you're providing "air-
> cover" by letting them point to openjdk as a model open free software  
> community.

It is an open free software community, even though some derivatives are
not fully free software. Something I greatly regret. And you will always
see me being very critical of that and help out any alternative effort
to work around that. And that works. There has never been any attempt to
stifle anybody or any group creating any derivative of the code, whether
it be IcedTea as shipped most GNU/Linux distros now, nor any of the
other hybrid implementations
http://www.infoq.com/news/2007/06/openjdk-hybrids

> So whatever problems you see has historical actually exists, and is  
> still very real and harmful for another group of people with the same
> interests and aspirations as you, who managed to actually bring an  
> independent implementation together to the point of being ready for  
> compatibility testing.

Sure, I know. We had this back in 2005 already:
http://advogato.org/person/robilad/diary/64.html
And even though Dalibor and Onno pushed for it, we never succeeded back
then with 1.5. Keeping chasing after these old issues instead of
focusing on the future seems not very productive though. There are
indeed still serious issues with non-open specs and anti-social TCK
usage restrictions for 1.6 (I would say the processes around them are
not really workable right now). But we do have free implementations now.

Instead, lets work together on fixing these issues going forward for
1.7, and make sure that we will not just have free code as reference
implementation, but also with a fully free community process we all seem
to want. Either through the JCP if you feel that can still be saved, or
by going around it if it ends up not being able to produce results that
are free for all.

Cheers,

Mark