[steering-discuss] Hello! ... and lurking :-)

Hi all,

I imagine you've all heard about the proposal[1] to contribute OO.o to
the Apache Software Foundation. I've been involved with Apache for
well over a decade, on its Board of Directors since 2001, its current
Vice Chairman, the VP of Apache Subversion, and was the Chairman for
five years. In short: lots of Apache experience.

I've been following and participating in the discussion around the
OO.o proposal on the general@incubator.apache.org list[1]. One of the
threads of that discussion was to reach out to the people in the
Document Foundation and the LibreOffice communities. So... that's this
email. I'm now subscribed to discuss@df, steering-discuss@df, and
libreoffice@freedesktop.

I intend to lurk regarding all the regular work that you all are doing
here. I'll be paying particular attention to any conversations or
concerns that you may have about the OOo/Apache stuff, and will
attempt to answer questions that you may have. I'm catching up on the
archives now.

If you have any comments, questions, or concerns, then please feel
free to direct them my way (on whatever list). I'm here to listen and
understand, and to offer up answers where I can.

Cheers,
-g

[1] http://wiki.apache.org/incubator/OpenOfficeProposal
[2] send mail to general-subscribe@incubator.apache.org if you would
like to subscribe and directly talk about the proposal

I have a question:
Why would Apache contemplate helping IBM pull a Jenkins/Hudson on us,
fragmenting the license of a project that has been with a uniform
licensing so far ?
(Oracle could merge our changes... they elected _not_ to do so because they
wanted a Copyright assignment on top of the code, but that was not
a licensing incompatibility)

You (Apache) are lending your good name to a nasty endeavor, for the
benefit of a company
that has an history of screwing you over (Harmony ?)

Ironically what seems to be happening at Apache is very reminiscent to
me to the ISO/MSXML debacle...
Some corporation exploiting the letter of your governance to better
abuse the spirit of it.
(that is _if_ I understand what Apache stand for... but maybe I'm misguided)

Norbert

PS: I strongly encourage you to read:
http://www.itworld.com/software/170521/big-winner-apache-openofficeorg#comment-9942111
That shed a very illuminating light on IBM's involvement in OOo, and
why it is hard to take seriously their grandiose promises... that
would by far not been the first time, and there is no reason to
believe that the outcome will be any different this time around...
except that both the OpenOffice brand and the Apache reputation will
be tarnished in the process...

Hi Greg,

Hi all,

I imagine you've all heard about the proposal[1] to contribute OO.o to
the Apache Software Foundation. I've been involved with Apache for
well over a decade, on its Board of Directors since 2001, its current
Vice Chairman, the VP of Apache Subversion, and was the Chairman for
five years. In short: lots of Apache experience.

I've been following and participating in the discussion around the
OO.o proposal on the general@incubator.apache.org list[1]. One of the
threads of that discussion was to reach out to the people in the
Document Foundation and the LibreOffice communities. So... that's this
email. I'm now subscribed to discuss@df, steering-discuss@df, and
libreoffice@freedesktop.

I intend to lurk regarding all the regular work that you all are doing
here. I'll be paying particular attention to any conversations or
concerns that you may have about the OOo/Apache stuff, and will
attempt to answer questions that you may have. I'm catching up on the
archives now.

If you have any comments, questions, or concerns, then please feel
free to direct them my way (on whatever list). I'm here to listen and
understand, and to offer up answers where I can.

A very warm welcome from my side. I'm subscribed to the general@incubator list too, but just lurk (and think :wink: for the moment.

To present myself a bit, I'm a member of the Steering Committee, a member of the Membership Committee and the liaison for TDF at SPI. I'm doing the French localization, I'm also working on other various tasks and areas on the project.
I was in the OOo community since the creation of the project, member of the CC representing the NLC, lead of the FR project, I represented the OOo community before the AB.

As you are on the most active lists of our project, don't hesitate to ask too if you have any question on concerns. Either directly to me or on any lists.

Welcome again and thanks a lot for coming to visit us on our project, it is highly appreciated.

Kind regards
Sophie

Hi all,

If you have any comments, questions, or concerns, then please feel
free to direct them my way (on whatever list). I'm here to listen and
understand, and to offer up answers where I can.

Let me start with a request. If we are going to have a productive discussion, it would be best if it were done using respectful terms. If, however, as described below it is the intent of TDF to systematically exclude one or more participants then I think the answer as to why there might need to be a separate effort answers itself.

I have a question:
Why would Apache contemplate helping IBM pull a Jenkins/Hudson on us,
fragmenting the license of a project that has been with a uniform
licensing so far ?
(Oracle could merge our changes... they elected _not_ to do so because they
wanted a Copyright assignment on top of the code, but that was not
a licensing incompatibility)

OpenOffice.org is not uniformly licensed. Never has been. You yourself point to a comment on an IT World article that makes this very point.

If is licensed to many under one license, and LibreOffice has continued with that license. It is licensed to others under a different license. The copyright assignment was an essential part in making that dual licensing strategy happen.

Whatever the history, if we truly seek to build one community, we will need to work together to find a license that enables the most widespread possible use of the code. While I am generally an advocate for the Apache License, Version 2.0, I do believe that this license is particularly appropriate in this situation given both the history and the list of people interested in participating going forward.

You (Apache) are lending your good name to a nasty endeavor, for the
benefit of a company
that has an history of screwing you over (Harmony ?)

Ironically what seems to be happening at Apache is very reminiscent to
me to the ISO/MSXML debacle...
Some corporation exploiting the letter of your governance to better
abuse the spirit of it.
(that is _if_ I understand what Apache stand for... but maybe I'm misguided)

In the interest of full disclosure, I am an IBM employee. Next month it will be 30 years.

I've also been an active participant in the Apache Software Foundation for over 10 years. I've been on the Board of Directors for the foundation for most of that time, and I am the VP of Legal Affairs. These positions are selected based solely on merit, and in the case of the board of directors are a result of an open election amongst the membership. It is not possible to buy a seat on the board in the ASF, the only way to get considered is by individual and sustained involvement.

IBM is a big company. I do not work in the part of the company that previously was involved in Harmony. I do not work in the part of the company that is interested in contributing to whatever project may be formed based the generous OpenOffice.org contribution. Others in my group participate in various ways in Apache. Others in many other groups within IBM participate in a number of Apache Projects. Some have even participated in Apache longer than I have.

What I will say is that one of the primary goals of Incubation at the ASF is to establish that there is a diverse community maintaining the code base, one diverse enough to survive the exiting of any particular contributor.

Without attempting to defend or excuse IBM's actions with respect to Harmony, I will state that in my opinion the primary obstacle that project has had in obtaining and retaining contributors is the failure to obtain access to the compatibility test kit which is essential part of obtaining certification.

Norbert

PS: I strongly encourage you to read:
http://www.itworld.com/software/170521/big-winner-apache-openofficeorg#comment-9942111
That shed a very illuminating light on IBM's involvement in OOo, and
why it is hard to take seriously their grandiose promises... that
would by far not been the first time, and there is no reason to
believe that the outcome will be any different this time around...
except that both the OpenOffice brand and the Apache reputation will
be tarnished in the process...

I believe that the key phrase in that comment is "objections to contributing code to be used in proprietary apps". A fundamental goal of the Apache license is to satisfy the needs of those that wish to include the code in Free and proprietary software alike.

I believe that if we want to attract everyone alike to contributing to a common code base -- wherever it resides -- then we need to establish this as a common goal.

If we do this, clearly there is much work that would need to be done. It will involve getting the consent of those that participate in this foundation and LibreOffice to relicense their work. It won't be easy, but I will be a part of making it happen.

- Sam Ruby

Let me start with a request.  If we are going to have a productive
discussion, it would be best if it were done using respectful terms. If,
however, as described below it is the intent of TDF [...]

we can start by working under the predicat that, unless specified otherwise,
any comments are the author's own and not representing anybody else?

OpenOffice.org is not uniformly licensed. [..]

If is licensed to many under one license, and LibreOffice has continued with
that license.  It is licensed to others under a different license.

I thought we were talking about FLOSS here... not about what
proprietary forks may or may not have done.

I believe that the key phrase in that comment is "objections to contributing
code to be used in proprietary apps".  A fundamental goal of the Apache
license is to satisfy the needs of those that wish to include the code in
Free and proprietary software alike.

Yes, I am aware of that 'feature'. I just happen to consider it a bug.

I believe that if we want to attract everyone alike to contributing to a
common code base -- wherever it resides -- then we need to establish this as
a common goal.

How exactly encouraging proprietary fork will attract contribution ?
you are relying on the ethic and moral sens of corporation ?
If only we had real life examples to give us clues on how realistic
that is... humm...

If we do this, clearly there is much work that would need to be done. It
will involve getting the consent of those that participate in this
foundation and LibreOffice to relicense their work.  It won't be easy, but I
will be a part of making it happen.

The very reason I decided to show-up was because 1/ the license was
copy-left and 2/ TDF dropped the copyright assignment.
(of course I found many other reasons to stay... but that's another topic :slight_smile: )
So I'd say... yes it won't be easy indeed, but I guess it won't be
harder than to convince AF not to cater to proprietary sink-hole...

But there is another solution, one that has happened in the past:
convince the company with a proprietary fork that if they are truly
interested in community
support and contributions, they should join that community under terms
that protect both our interest, not just its interests.

Norbert

Hello Sam,

(emptying interesting bits of discussion for clarity).

I think it goes without saying that everybody should be respectful to each
other.
Let me try to rephrase the general terms of this discussion from the broad
perspective of TDF (I'm not trying to suggest that this is an official
message from the Steering Committee of TDF, but... you get the point).

TDF started the LibreOffice eight months ago, and has so far released
software, deployed infrastructure and other resources, defined and
implemented processes while elaborating governance methods and structure
(that's still a work in progress).

TDF has had the ambition of being the future of the OpenOffice.org community
and after 8 months it is very safe to assume that it has largely succeeded
in being so. Granted, not everyone from OpenOffice.org has jumped to TDF;
existing Oracle employees have not joined the LibreOffice project
(obviously), and a few -but very few- people and teams have chosen either
not to choose, or to reject TDF mostly on personal grounds. Oracle has ended
up dropping the OpenOffice project as well as its commercial offerings
around it as this line of business and project were not complying to their
own internal criteria for profitability, and thus Oracle ended up dumping
the IPR assets to the Apache Foundation. I'm delighted to read from Andrew
Rist that Oracle will still support the existing OpenOffice infrastructure
throughout the transition. The announcement from Oracle came in 8 months
after the birth of the LibreOffice project. You will understand that any
argument framing the discussion along the lines that now that OpenOffice is
being transferred to the Apache Foundation we should all turn towards this
project sounds weird, chronologically anachronistic, and probably
counterproductive in our view.

But this is not a reason for TDF and the Apache Foundation to stop
discussing the matter at hand.

Going back to the OpenOffice project, I think it's safe to assume four key
elements in our discussion. Some of these are issues, but some others are
just factors to be considered.
- Oracle will not provide its existing OpenOffice engineers to continue the
OpenOffice project.
- Unless IBM pours engineers on it, it is likely that there will be a very
small community of developers working on OpenOffice
- IBM does have business and operational requirements that make the transfer
of OpenOffice to an Apache environment desirable and appealing.
- TDF and ASF have very different views on licensing.

Simon Phipps made some very interesting proposals yesterday on the Apache
list, and I'm glad to see that one of the many threads around this debate on
the Apache mailing list is furthering this discussion; namely, that the
OpenOffice project at Apache be not so much considered as an end to itself,
but rather as an engineering project catering to specific needs of IBM while
also helping the LibreOffice project in specific areas. The point on the
OpenOffice project not providing binaries is going in that direction.

I thus think that there is room for TDF and ASF to cooperate even if the two
will not change their stance on licensing easily. This being said, many of
us here at TDF still question the whole relevance of an OpenOffice project
*anywhere* but inactive now that LibreOffice exists, runs and has been
releases several versions of stable software. I hope this helps you
understand more about our perception.

best,
Charles-H. Schulz.

Let me start with a request. If we are going to have a productive
discussion, it would be best if it were done using respectful terms. If,
however, as described below it is the intent of TDF [...]

we can start by working under the predicat that, unless specified otherwise,
any comments are the author's own and not representing anybody else?

I accept that predicate, but will not that it is orthogonal to my request, which I will make again: if we are to have a productive conversation please try to do so respectfully.

OpenOffice.org is not uniformly licensed. [..]

If is licensed to many under one license, and LibreOffice has continued with
that license. It is licensed to others under a different license.

I thought we were talking about FLOSS here... not about what
proprietary forks may or may not have done.

The Apache License was explicitly designed to support Free and proprietary use alike. Note that while many use the term proprietary in a pejorative sense, I won't shy away from using that term as it is accurate.

I believe that the key phrase in that comment is "objections to contributing
code to be used in proprietary apps". A fundamental goal of the Apache
license is to satisfy the needs of those that wish to include the code in
Free and proprietary software alike.

Yes, I am aware of that 'feature'. I just happen to consider it a bug.

We clearly disagree on this point. That's entirely OK and honest.

I believe that if we want to attract everyone alike to contributing to a
common code base -- wherever it resides -- then we need to establish this as
a common goal.

How exactly encouraging proprietary fork will attract contribution ?
you are relying on the ethic and moral sens of corporation ?
If only we had real life examples to give us clues on how realistic
that is... humm...

The ASF is full of successful examples of this. IBM WebSphere and Apache httpd is one such example. There are countless others based on different combinations of companies and products.

If we do this, clearly there is much work that would need to be done. It
will involve getting the consent of those that participate in this
foundation and LibreOffice to relicense their work. It won't be easy, but I
will be a part of making it happen.

The very reason I decided to show-up was because 1/ the license was
copy-left and 2/ TDF dropped the copyright assignment.
(of course I found many other reasons to stay... but that's another topic :slight_smile: )
So I'd say... yes it won't be easy indeed, but I guess it won't be
harder than to convince AF not to cater to proprietary sink-hole...

No one is suggesting catering. We either find common ground (as you outline in your following paragraph) or we go our separate ways, hopefully parting as friends.

But there is another solution, one that has happened in the past:
convince the company with a proprietary fork that if they are truly
interested in community
support and contributions, they should join that community under terms
that protect both our interest, not just its interests.

And I believe that you have just precisely captured why the Apache License, Version 2.0 is the most appropriate choice. It is fully compatible not only with GPLv3 and LGPLv3, but also with use in 100% proprietary software.

Norbert

- Sam Ruby

Hello Sam,

(emptying interesting bits of discussion for clarity).

I think it goes without saying that everybody should be respectful to each
other.

Thanks! And I appreciate both the content and tone of this entire message.

Let me try to rephrase the general terms of this discussion from the broad
perspective of TDF (I'm not trying to suggest that this is an official
message from the Steering Committee of TDF, but... you get the point).

TDF started the LibreOffice eight months ago, and has so far released
software, deployed infrastructure and other resources, defined and
implemented processes while elaborating governance methods and structure
(that's still a work in progress).

TDF has had the ambition of being the future of the OpenOffice.org community
and after 8 months it is very safe to assume that it has largely succeeded
in being so. Granted, not everyone from OpenOffice.org has jumped to TDF;
existing Oracle employees have not joined the LibreOffice project
(obviously), and a few -but very few- people and teams have chosen either
not to choose, or to reject TDF mostly on personal grounds. Oracle has ended
up dropping the OpenOffice project as well as its commercial offerings
around it as this line of business and project were not complying to their
own internal criteria for profitability, and thus Oracle ended up dumping
the IPR assets to the Apache Foundation. I'm delighted to read from Andrew
Rist that Oracle will still support the existing OpenOffice infrastructure
throughout the transition. The announcement from Oracle came in 8 months
after the birth of the LibreOffice project. You will understand that any
argument framing the discussion along the lines that now that OpenOffice is
being transferred to the Apache Foundation we should all turn towards this
project sounds weird, chronologically anachronistic, and probably
counterproductive in our view.

But this is not a reason for TDF and the Apache Foundation to stop
discussing the matter at hand.

Going back to the OpenOffice project, I think it's safe to assume four key
elements in our discussion. Some of these are issues, but some others are
just factors to be considered.
- Oracle will not provide its existing OpenOffice engineers to continue the
OpenOffice project.
- Unless IBM pours engineers on it, it is likely that there will be a very
small community of developers working on OpenOffice

This point is not clear to me. What is clear to me is that if it turns out that this (not yet accepted, and therefore only potential) incubating project never establishes a diverse set of contributors that could survive the exit of any one of them, including IBM that the ASF would undoubtedly conclude that incubation was unsuccessful and terminate the incubation unsuccessfully.

- IBM does have business and operational requirements that make the transfer
of OpenOffice to an Apache environment desirable and appealing.
- TDF and ASF have very different views on licensing.

Simon Phipps made some very interesting proposals yesterday on the Apache
list, and I'm glad to see that one of the many threads around this debate on
the Apache mailing list is furthering this discussion; namely, that the
OpenOffice project at Apache be not so much considered as an end to itself,
but rather as an engineering project catering to specific needs of IBM while
also helping the LibreOffice project in specific areas. The point on the
OpenOffice project not providing binaries is going in that direction.

I thus think that there is room for TDF and ASF to cooperate even if the two
will not change their stance on licensing easily. This being said, many of
us here at TDF still question the whole relevance of an OpenOffice project
*anywhere* but inactive now that LibreOffice exists, runs and has been
releases several versions of stable software. I hope this helps you
understand more about our perception.

Acknowledged. That being said, different people will come to different conclusions as to what is relevant to their particular interests, and I will say that choice of license is a factor in that decision.

best,
Charles-H. Schulz.

- Sam Ruby

Hi Sam, *

Sam Ruby schrieb:

[...]What is clear to me is that if it turns
out that this (not yet accepted, and therefore only potential)
incubating project never establishes a diverse set of contributors that
could survive the exit of any one of them, including IBM that the ASF
would undoubtedly conclude that incubation was unsuccessful and
terminate the incubation unsuccessfully.

And this might be even worse for OpenOffice.org than ASF not accepting the incubating project:

While in case of unsuccessful incubation the trademark OpenOffice.org will be finally dead (because it has not been maintained by a vivid community as we did it for ten years) and buried.

We could discuss if this might help LibreOffice - but all of us who started supporting OpenOffice.org years ago don't want OOo to be lost in the dark. It has been a great name and stands for an (r)evolution in office software.

With the public recognition of being the ones to drop OpenOffice.org (because it didn't get the necessary momentum to survive) the Apache project might be damaged too in this case.

If the proposed incubation will be rejected on the other hand because
it's preparation had not been sufficiently elaborated (necessity to keep the project running during the incubation time), it would be Oracle and IBM to think about their choice:

Keeping the community out of the decision making process has never been a good plan for OpenOffice.org, and it would not Apache to be blamed for a possible failure if they keep on this way...

Most of this context might be totally irrelevant for Apache - damage to the existing community, damage to the trademark - because at your side the donation is a "fait accompli".

But possible damage to the Apache reputation by a failure of the new incubator project well known all over the world might be something different.

With their donation to Apache Oracle and IBM got the best out of their situation, as several other already said. Neither the remnant OOo project nor the larger part of the OOo community now working on LibreOffice have any chance to influence their decision.

Apache has - and it has got some kind of responsibility for OpenOffice.org (the existing one, with the ".org" ending) already now, and even more when they accept the incubation project.

Best regards

Bernhard