As you have probably heard, Oracle has followed through with their earlier promise to “move OpenOffice.org to a purely community-based open source project.” OpenOffice is moving to Apache.
I’d like to offer you my own thoughts on this new opportunity and what it means. I recommend also the insights of my colleagues Ed Brill and Bob Sutor.
First, we should all be excited to see OpenOffice move to a foundation with the stature and track-record of Apache. If you are a web developer or server admin, then you of course know about the eponymous Apache http server and Tomcat. If you are a developer you know about Ant, Maven and Subversion. If you work with XML you know about Apache Xerces, Xalan, FOP and Batik. And if you don’t know about Apache Hadoop yet, then please do your résumé a favor and study up on it. All said Apache is custodian of nearly 170 open source projects, including 5 of the top 10 open source downloads. I’m hoping that soon, as OpenOffice transitions to Apache, they will be able to claim 6 of the top 10!
These diverse projects are run according to meritocratic development process, a tried and tested governance model, strong shared technical infrastructure, a pragmatic, commercially-friendly open source license and a set of social conventions known as the “Apache Way”.
I’d point out in particular that the Apache 2.0 open source license was recently blessed by the Free Software Foundation:
The Apache License 2.0 is the best non-copyleft license that does what a copyright license can to mitigate threats from software patents. It’s a well-established, mature license that users, developers, and distributors alike are all comfortable with. You can tell it’s important by the way that other free software licenses work to cooperate with it: the drafting processes for GPLv3 and the Mozilla Public License 2.0 named compatibility with the Apache License 2.0 as a goal from day one. The Apache Software Foundation deserves a lot of credit for pushing to do more to tackle software patents in a license, and implementing an effective strategy in the Apache License.
As you can tell, when it comes to Apache I’m a fan. I’ve experienced much of this first-hand. I was a committer in the Apache Xalan project many years ago (1999-2000). It was a great experience then, and when the opportunity came to add my name to the OpenOffice incubation proposal I did not hesitate. It was an honor. I look forward to coming back to Apache and participating in this continuation of OpenOffice. I am planning on getting directly involved with the engineering effort of this project.
So what are the next steps? As I understand it (and I’m not an Apache process expert), it is not accurate to say that “Apache OpenOffice” really exists yet. We’re not quite there. At Apache, you can’t just walk in off the street, drop some code and call yourself an Apache project. There is a multi-step process for initiating, reviewing and approving a new project. We’re at the first step, with the proposal submission, which Oracle made earlier today. This proposal will now be reviewed and voted on by the Apache Incubator Project Management Committee (PMC) over the next few days. If approved, the project then advances into incubation as a “Podling”. Incubation at Apache is a probationary stage, where the project recruits new members, reviews the code to establish IP provenance, adapts the project to the Apache infrastructure, and so on. We’ll undergo periodic reviews, and at some point, when we are ready, we can then be approved to “graduate” to be a new top-level Apache project, ideally something like http://openoffice.apache.org. There is no fixed time for how long this incubation state takes, but I’m told it generally takes several months. I don’t think we should rush it. Everything that occurs during incubation is for the long-term benefit of the project, so I think we want to soak up the help and special attention from our Apache mentors as much as we can. For more details on the process, I recommend the process diagram here and the associated process description.
The Apache process is based on a strong meritocracy. Developers who regularly provide high quality patches get elected as “Committers” and they then help review submitted patches as well as write their own code. And those Committers who remain active and have earned the respect of their peers typically then get elected to the Project Management Committee (PMC) and steer the direction of the project. And those who are most valued on the PMC may become the PMC Chair for their project, which also ranks them as an Apache Foundation Vice President. And some then have the opportunity to serve on the Apache Board of Directors. With this cursus honorum, it is recommended that those with leadership ambitions get involved early. When the Apache OpenOffice project begins, there will be project decisions to make and leadership roles to fill, and this will happen fast once we get started. Obviously, you can’t advance in the meritocracy if you are absent. Although, you can join anytime you want, there are clear advantages to “getting in on the ground floor”.
In particular, we need to attract a wide variety of project specialists. This includes C++ programmers (on Linux, Mac and Windows), QA (also on all platforms), help/documentation, UI/UCD, translation/globalization, accessibility, install, etc. Please keep your eyes open for an announcement from Apache in the next week or two, saying that the OpenOffice incubator project has been set up and is ready to accept members.
(I know that OpenOffice.org prided itself on a strong marketing committee as well. I think this is important, but it is not clear to me yet how that fits into an Apache project. Certainly this aspect is more critical to an end-user facing project like OpenOffice than it would be to a developer tool. Maybe someone out there in Apache-land will be able to offer some suggestions on how best to integrate this function into an Apache project?)
An extraordinarily cool thing to look forward to, once Apache OpenOffice graduates to be a top-level project (TLP), is that we will be able to sponsor other incubation proposals that would be a good fit for OpenOffice. So other components, plugins, toolkits, modules, even entire complementary productivity applications, could be brought into the project, via the incubation process, to help enhance the OpenOffice application suite. Imagine having a mind maps editor? Or a project planning tool? Or a standalone outliner? Or smaller, lighter, component-based editors? All these things — and more — are possible.
I have a vision of a free, high quality productivity suite, one based on open standards and open source, one that doesn’t treat the web and mobile and tablet form factors as a design afterthought, one that has a strong extensibility and programmability model that makes it the preferred platform for innovation, one that has a dedicated community of supporters. I’ll need your help to get there.
Thanks for the great post, Rob. I too fondly remember the Apache Xalan days, when XML was the Next Great Thing that was going to simplify everyone’s data forever! I suppose we’ll have to be happy enough that our Xalan code is now in the majority of Java distributions out there, even if XML isn’t quite as ubiquitous as we originally imagined.
As a non-profit, Apache’s purpose is to provide software for the public good. As we’ve grown, it’s become clear that it’s also helpful to be able to explain to the world what our software is, and why people might want to use it. So Apache does have both a VP of Marketing and Publicity (http://www.apache.org/press/), as well as a VP of Brand Management (http://www.apache.org/foundation/marks/), both of whom I imagine will be involved with any future Apache Office TLP once it graduates.
Indeed, as the first major end-user product, an Apache Office project would be quite a change. I welcome the chance to make popular Apache licensed software that normal end users can use!
I notice that you and your friends at Oracle do not mention the existing community at all – instead you talk about outsiders of all shades who have never been involved in OpenOffice.org forming a new activity with our source code. There are lots of people in the community and in its two projects – OpenOffice and LibreOffice – who don’t fit Apache at all, either becuase they are not involved in the “core code” or becuase they believe in copyleft and software freedom. Why have you and Oracle etc made no mention of them whatsoever? Is it your intention to isolate them and create division? Seems that way.
@Contrarian, Of course, anyone is welcome to join the project. Unlike LibreOffice, Apache does not have a membership committee to review and approve or reject developers. Anyone is welcome to join. But you need to agree with the project license. This is true of any open source project.
If you (or anyone) has a concrete proposal on how LibreOffice can or should related to Apache, I’d love to hear it. I think the time is now favorable for having that kind of discussion, more so now than it was when OpenOffice was run by Sun/Oracle with their CLA.
I didn’t know the Document Foundation rejects membership applications.
The main difference is of course that the Apache Foundation is based in the United States/Delaware and has no European branch. For a project mostly driven by European stakeholders that seems a disadvantage. Furthermore I don’t trust American governance methods. A board which can decide upon its own compensation is unacceptable to me.
Echoing The Contrarian’s sentiments: So, what are The Document Foundation and LibreOffice? Chopped liver?
It does seem that the post tiptoes around the elephant in the room (LibreOffice).
Saying “you need to agree with the project license” is begging the question, as that was only just decided. OpenOffice.org wasn’t under the Apache licence two days ago. Why Apache, do you think? Was IBM consulted on the licensing choice? What license did you suggest/propose? Did you encourage Oracle to make more efforts to reach out to the existing open source community surrounding this codebase? If you are right when you say that the stewardship of Sun/Oracle with their CLA wasn’t good, then surely you must recognise that the people who were working outside of that must be doing something right?
Gerv
Not sure LibreOffice requires any kind of membership, maybe you’re thinking of the Document Foundation – any hacker with a patch can submit it to the LibreOffice project.
In any event, I would have thought that the relevant question is how will this Apache project be related to LibreOffice? The codebase has moved on vastly, you just need to look at git – why would people want to move back to an older codebase? LO has already proved itself, Apache Office or whatever it will be called is a totally new project.
Great post rob, I guess the road ahead looks brighter to the OOo gloomy conversations. Now there is something to look forward to. With Apache running things hands off could also make a way to flatten the community which is more needed process. OpenOffice.org has certainly gained a lot of projects itself and hierarcy which has not as many technical expertise. Apache being more a technical culture refocus the project on what matter most, which is the code. Is my personal hope that we can guarantee a better quality project in the future.
Hi Rob. I must agree with Contrarian as well. Dismissing the LibreOffice and the Document Foundation is a big issue at least for me (I’m not so ever and in any way related to LibreOffice nor to Document Foundation).
My humble opinion, after the bumpy start, and the early embrace of LibreOffice by all major Linux distribution and getting the advantages from Go-oo, now it looks too late to to move OpenOffice.org to Apache foundation. It might make some sense to IBM, but for me it doesn’t. Oracle missed the train with OpenOffice almost a year ago.
Why not simply joint with Document Foundation and LibreOffice, make strong community and support? Why can’t Oracle, IBM and all other willing to help and participate, contribute (and on the end benefit from it) through the already established Document Foundation? Why fragment it instead of make it stronger?
I’m 100% behind your vision of a free, high quality productivity suite, one based on open standards and open source, and all the rest that you have mentioned, but I also can’t envision it by not taking in consideration the already available community.
On the end, not to be taken wrongly, I do live the moment to have a real contender and competition to MS Office in the corporate world, and I think we can achieve this, all together, not apart.
Hi Rob,
This is indeed good news, but I think long term rather than trying to run the project at the Apache Software Foundation the code would be much better merged into LibreOffice and let them take the lead on this.
Given that there is already an existing community around this code, merging with it rather than trying to create another from scratch makes more sense IMHO.
Jeremy.
You, Ed Brill and Bob Sutor avoided to mention the Document Foundation and its OpenOffice.org fork LibreOffice. You refused to mention that LibreOffice already has a developer community. That the Document Foundation already has legal recognition. That because of those two things it’s already more credible than this vaporware Apache project that’s at least a year away from becoming reality. That this announcement is doing nothing more throw confusing and uncertainty around a free/open source office suite that’s credible enough to face Microsoft Office.
That Oracle did the same thing with Hudson and a number of other projects. That Oracle required copyright assignment but Apache Foundation doesn’t. That OpenOffice.org uses a different license than Apache Foundation projects.
And a lot of other things that now fall into place: Oracle conflict with Apache Foundation about Apache Harmony; Oracle conflict with IBM about Apache Harmony but both are now “best friends for life”; Oracle conflict with Apache Foundation and Google about Android.
If you think your actions, IBM’s and Oracle’s contribute to create a credible alternative to Microsoft you’re leaving in a parallel universe. Denying reality to an extreme worthy of Microsoft’s brainwashing…
I am sorry to see that Oracle attempts to divide the open source community by poisoning the Apache Foundation. I am afraid this really needs to end up in the attic as ‘once a good product but currently obsolete’. Don’t waste money and effort on this – say your prayers and let it wither and die.
What’s up with all the LibreOffice comments saying essentially the same thing? Okay guys, we get it. I fail to see how Oracle or OpenOffice or Apache owe anything to the LibreOffice guys–it was their decision to fork. Now they have to live with the consequences. To me these comments seem like a bit of sour grapes that OpenOffice may have a future. That said, both projects can survive and that may be better for all.
It’s clear now that this whole process is nothing more than an attempt to sabotage LibreOffice’s future. Signs that this is only a code dump. No commitment by Oracle, and only a few very timid commitment by IBM to add “some developers” that were attach to IBM’s proprietary fork Lotus Symphony.
This is a very nice piece of revisionism by Rob Weir on the Apache mailing list about the Document Foundation and LibreOffice:
“My understanding is that the Linux distros never really included the core
OpenOffice.org. They included the Novell Edition of OpenOffice, since
Novell (and some volunteers) did the leg work to get the code into a form
suitable for the distros to consume (packaging, catalog metadata, etc.)
When LibreOffice was announced, Novell pulled their OpenOffice Novell
Edition and put the same engineers on LibreOffice. The distros could
simply continue working with the same engineers they had worked with
previously. This was not necessarily some ideological switch by the
distros. From their perspective LibreOffice was more a rebranding of
Novell Edition of OpenOffice. They include LibreOffice because it was
packaged, ready for their consumption.”
My question is: Why is the Apache Foundation helping 2 companies, IBM and Oracle, sabotage an healthy open source project, LibreOffice?
Oracle already did this with Hudson.
Some readers are offended that I did not mention LibreOffice in this post. I certainly did not intend any offense. Today’s announcement has nothing directly to do with LibreOffice. Of course, the announcement does have an impact on the larger market, and that will effect not only the future of LibreOffice, but also Symphony, EuropOffice, BrOffice, NeoOffice, RedOffice and others. I hope it will also impact, in some way, the market around Microsoft Office. To cover my bases I hereby apologize to all of them as well as anyone else who feels neglected by me not acknowledging their existence or their right to exist or whatever other hate crime I’ve supposedly committed.
But there is one big irony there that should be noted. IBM has been a long-time member of OpenOffice.org. We’ve contributed code. We’ve sponsored OpenOffice.org conferences. And of course, we’ve distribute OpenOffice in the form of Symphony. We never joined LibreOffice. So last September, when LibreOffice categorically announced that the OpenOffice community had moved to a new project, that wasn’t quite truthful was it? Certainly some moved. But IBM in particular did not. Now we could have raised a fuss, complained, said that were insulted for not being mentioned in the TDF inaugural press releases. But we didn’t. It would have been silly to have been insulted, since the TDF press release was obviously about those who did move, not those who did not.
Do you see the irony yet?
So flash forward 6 months, and now that same OpenOffice.org member (IBM) that was so insignificant that it was not mentioned in the TDF press release about the OpenOffice “community”, has now decided to go in another direction. And guess what? Now TDF members are irate because they were not mentioned in the IBM press release. Well, go figure!
@André, I refer to you to The Document Foundation Community Bylaws:
“Every membership application is reviewed by the Membership Committee, which approves or declines it”
I cannot say for a fact that they have rejected applications. I’m not aware of their records being open. But there is a committee, appointed by the self-appointed “Steering Committee” that has the job of reviewing and approving or rejecting membership applications. Apache does not have that. If a developer wants to join, they join. Of course, their code contributions may be rejected, after review. But we don’t reject the person.
@J David, I hope you don’t get the impression that they are “chopped liver” [I’ll translate that as “a lesser cut of meat” for non-English natives]. Neither are RedOffice, NeoOffice, BrOffice, etc. Ditto for AbiWord, Calligra Office, Gnumeric, etc. This announcement was specifically about OpenOffice.org and where that project is moving. LibreOffice members are welcome to ignore the move, join the move or just complain about the move, as they wish. It is a free country.
@Gerv, good question. There are a variety of possible destinations for this code. This included established, well-known open source foundations, including Apache, but also other familiar names. The open source license is one issue, but not the only one. Other criteria include governance, track record/organizational stability,technical infrastructure, diversity, strategic alignment, name recognition/brand value, etc. Of course, other individuals may have other priorities for their criteria and would make different choices. They have that right. In the end, Oracle made a choice of Apache, and I think that is a fine choice.
It isn’t clear to me that every LibreOffice member has exactly the same priorities or that their priorities are 100% aligned with the self-appointed, unelected Steering Committee. It would be quote remarkable if that were the case, wouldn’t it? I bet that for some, the copyleft license is their #1 priority. For others, running their own foundation was a priority. And for others, they simply went where the momentum seemed to be going at the time. It seems to be that now LibreOffice members have choices to pick from. I don’t think this is necessarily a bad thing.
@Alex, See above on the membership question. I think what I said was accurate.
To your second question, I agree that we need to figure out how OpenOffice and LibreOffice relate. But it is more than that. We also need to look at Symphony, RedOffice, BrOffice, EuroOffice, NeoOffice, etc., including other new customized distributions that will certainly now proliferate now that we are freed from the restrictions of the copyleft license. This conversation is already underway at Apache. I invite interested parties to join and help define the answer to that.
@JZA, I think the “focus” will come from the priorities of those who join and participate. That is the nature of a meritocracy. Those who “do” decide.
@Miroslav, as mentioned above, there are many open source foundations out there that could handle a project like this. The Document Foundation/LibreOffice is just one. Oracle picked Apache. I think that was the right choice.
@Jeremy, I’d be absolutely giddy with joy if LibreOffice developers would come over to Apache and run their project under the Apache 2.0 license under the Apache process. I’d even be open to calling it “LibreOffice”. But this is much more an issue of organizational capabilities than it is the rather narrow gulf between the current OpenOffice and LibreOffice source codes. I want an organization that will last, not something that could fall over in the next storm.
@Mario, The Document Foundation has no “Foundation” yet. I understand that paperwork for this has been worked on for several months now, but I’m not sure what “legal recognition” you ascribe to it. A post office box?
@Carey, the voice of reason. Who let you in? Thanks ;-)
@Mario, have you seen the recent interview with Mark Shuttleworth:
Maybe I’m not reading deeply enough, but this sure does not sound like someone who is ideologically aligned with the LibreOffice vision.
Rob wrote :
“@Jeremy, I’d be absolutely giddy with joy if LibreOffice developers would come over to Apache and run their project under the Apache 2.0 license under the Apache process”
That’s funny, I’d be giddy with joy if it happened the other way around :-). This is about copyleft vs. non-copyleft licensing IMHO. I personally believe the the LGPLv3 copyleft license is a better choice for this codebase, rather than the Apache one. The reasons for this are too complex to go into in this short post, but based around my own experiences on Samba, where copyleft is one of the only ways to break into a monopoly-dominated market.
The good news is that the Apache license is compatible with the LGPLv3 LibreOffice license, so they can take the OpenOffice code and merge any useful changes into their codebase. I don’t think this can happen the other way around. The reason I would like Apache OO developers (including those from IBM) to throw in their lot with LibreOffice is that otherwise you end up re-running the same experiment of Linux vs. FreeBSD. Unless you consider Apple a FreeBSD success (not sure I do, at least from the FreeBSD point of view :-) then that experiment didn’t go well for FreeBSD. But maybe you want to be Apple, in which case good luck (but remember in the best tradition of ‘Highlander’ – “There can be only one” :-).
Jeremy.
The thing is, Rob, according to all the public reports NO ONE talked to the existing community (and of that list of projects you list above, only BrOffice is actually community, the rest are downstream-only – and BrOffice joined LibreOffice). This move is a pure stitch-up between IBM and Oracle, masterfully leaving Apache to take the heat. No-one at either the OpenOffice.org project or the LibreOffice project seems to have known it was coming until a day or two ago and right now it seems there’s still not been any e-mail sent to their mailing lists.
You seem to be revelling in revenge for some slight you imagined when the community left Oracle instead of pausing and realizing that the free labor you hope to harness is that huge group of people you are ignoring and alienating – NeoOffice and RedOffice won’t refactor the core code for you. I know Oracle is tone-deaf on community but I thought you IBM people understood it a bit better.
Oh, by the way, Apache does have strict membership rules and there is no way anyone in the OpenOffice/LibreOffice community – or probably you – will get to be an Apache Member any time soon. That’s probably not what you meant when you smeared The Document Foundation hoping none of us understood, though. You would be very, very welcome to join in at TDF on about the same basis as you’re being welcomed by Apache. Please stop the spin.
@Contrarian, so let me see if I get this right. If there are “no public reports” that I have eaten pizza in the last four weeks, once should safely assume that I have not had a slice? That logic seems rather weak.
The fact that there are no public reports of our private meeting and private email conversations with TDF Steering Committee members means absolutely nothing. You might ask your favorite TDF SC member who they were talking to on April 28th at 9am EDT. Perhaps that will help refresh their memory. And if you think that the general membership of TDF was not made aware of this opportunity, then who is to blame: me, or provisional Steering Committee ? If you think that you should have been informed of this opportunity, and wish that your SC had been more open to cooperating in this move, then I think you should complain to them.
Good to know that Apache was glad with the recognition and support from the FSF.
Maybe Apache would enjoy more recognition and support should it welcome the existing LibreOffice community, under the licensing terms that OpenOffice.org *used* to have as well, returning the favor and recognizing that FSF’s licensing practices are also good.
Clearly Apache isn’t opposed to copyleft, since it has worked along with the FSF to make ensure compatibility, so perhaps the following is an option that can’t be entirely ruled out: considering that the alternative is to keep a divide, with the community and most commercial supporters on one side, and a couple of commercial supporters left behind on the other, perhaps Apache should consider the possiblity of keeping this project under copyleft and seeing how that goes, instead of moving right away in a direction that might lead to negative press for the foundation and negative consequences for the software project itself.
@Alexandre, Is that a threat?
Yeah, of course it’s a threat! ’cause I’m *that* powerful :-D
Doh! It’s just a fact that the community and most commercial vendors moved on, and a couple of commercial supporters were left behind. It’s to counter that that you guys are trying to move to a permissive license. I hope Apache will see through that move against the community and take the right stand. If you regard that a threat, well… what can I say? :-)
Rob,
You not mentioned what could happen with the ongoing user support forums, now sitting on Oracle sponsored servers. As a Hungarian forum admin I manage this forum more than a three year. You see any possibilities to continue this work, similar way, somebody give a server and bandwidth. or we needs fast solution to organize it.
The user forums is good place to get user views and pick bugs in OpenOffice.
I don’t know how Apache works on these type not development efforts.
What will happen with OOo wiki page etc. and other documentation efforts.
Zoltan
Of course IBM was consulted, contacted before the start of TDF. Their name could have been in the very first announcement. But, though a sort of positive, IBM never wanted to join. Their choice. But did IBM, Oracle consult TDF on a possible move to the ASF.. ?
I am very pragmatic: if the ASF is the best home for the open source office community, it’s OK for me. But before I see the wider non-corporate community join a possible openoffice.apache.org, I think they want to be very sure that both 8 months code improvement will not get lost and that the growing (developer-)community will find itself comfortable there too. And of course that’s not as simple as saying those people: just change your mind and do it our way.
Then for the code contribution from IBM to OpenOffice.org in the past. Year after year on OOo Conferences the contributions were announced. After years, only little came. Still a lot is not there. So – to say it politely – using that argument too me is a bit embarrassing. That attitude alas does not add trust to the possible factual contribution from IBM in a possible Apache project. I am afraid that too is not simple to change.
So, how much as long time active OpenOffice.org community member I look forward to a large reunion, I don’t think that this route with the ASF will be an easy go.
Rob, I think you’re being disingenuous at best. Look at http://www.documentfoundation.org/develop/ : there is no requirement to be a TDF member or sign any paperwork to submit a patch, it just needs the right license. It is therefore obviously inaccurate to portray the TDF mechanisms as “approving or rejecting *developers*”.
But back to the matter at hand. LibreOffice is here, right now. Apache OpenOffice.org is still off in the future somewhere. What are prospective contributors to do? Wait around and see what the results of the incubation process are going to be? Hope that in a year or whatever the nascent OpenOffice.org project is judged to be sufficiently hardy that it will stick around?
Whatever you think about the manner in which LibreOffice was set up, at least it’s here *now* and allowing people to progress the project at a rate faster than most would have hoped for. It’s jam today, not tomorrow, allows people to JFDI rather than waiting for these various committees and proposals to work through the machine. The bureaucracy is/was stifling and LibreOffice killed it. Painting LibreOffice as “just another fork” might be a convenient message, but it’s not accurate.
Hi Rob,
I would refrain from commenting many of your statements about TDF, which I understand come out of a very superficial knowledge of the organization (and I would leave to your wisdom to comment when you don’t know enough about a topic: it’s something that I usually refrain from doing, but I recognize that other people behave in a different way).
Anyway, I just wanted to comment on one of your statements: “Now TDF members are irate because they were not mentioned in the IBM press release”.
I must state here that we were happy about not being mentioned in IBM press release, and I don’t really see why we should have been mentioned here.
You know, donating the OOo code to ASF should have been an Oracle announcement, while here we have one short and dry Oracle statement against an IBM press release plus 3 IBM blog post (linking to each other) in less that 30 minutes after the Oracle announcement.
It looks like you were eagerly awaiting the announcement, and have been carefully preparing it for days. You managed your declaration of support as if this this was an IBM announcement.
Unfortunately, for ASF, and I’m really speaking out of my specific communication competence and my total development incompetence, the outside perception is that IBM is in the driver’s seat here.
It looks like we were having one community with two projects and we will be ending up with two communities (sorry, but I don’t think I would like to cooperate in a project where you – bashing TDF – are one of the committers: so, either you calm down for a while and improve your understanding of TDF, and start to behave in a mature way, or I’ll withdraw from the discussion as of now).
@Rob
Rob wrote :
« @Jeremy, I’d be absolutely giddy with joy if LibreOffice developers would come over to Apache and run their project under the Apache 2.0 license under the Apache process »
Well who wouldn’t be « giddy with joy » because of free labor that this new project is obviously hoping to attract. It seems that the hope is that through it’s support of this Apache project, IBM is banking on it’s mere weight and corporate status, that developers will be attracted over from the LibreOffice community. It also looks like the strategy to get this type of movement is by trying to make the LibreOffice project and SC look bad.
But maybe I am wrong, maybe there is plan to hire developers ??? No one has said much about that as far as I have seen.
So, as I said, it seems like the goal here in your (oddly defensive) comments is to make the LO SC look bad, by saying things such as the following :
Rob wrote :
« But there is a committee, appointed by the self-appointed “Steering Committee” that has the job of »…
… « the self-appointed, unelected Steering Committee. »
… « then who is to blame: me, or your own unelected, self-appointed Steering Committee ? »
Several times, Rob, it looks almost like fear mongering against the steering committee of The Document Foundation. It literally sounds like you are trying to make people think that they are illegitimate drivers of the community. This is sad to me really, because granted they may not be elected, but they are the ones who TOOK ALL THE RISK, and stuck their necks out to fork the OpenOffice project away from the death grip of Oracle, when everyone else was too afraid to move a little finger, everyone else trying to hold on to their status quo. So because they took the initiative that makes them what?…dangerous? That is what you make it sound like. Somebody had to do something. But I guess it’s hard to understand that coming from a corporate point of view where you are forced to sit on the fence and wait to see what the legions lawyers say every time an opportunity presents itself. No risk taking, means slow, VERY slow progress.
The steering committee of LibreOffice took huge risks because (excuse my language) they had the balls to do so. Don’t rag on them for that for pete’s sake!!!
Anyway, you contradict that point in your own comments :
Rob wrote :
« @JZA, I think the “focus” will come from the priorities of those who join and participate. That is the nature of a meritocracy. Those who “do” decide. »
The Document Foundation is a meritocracy. Those who « did » are the members of the The Document Foundation steering committee.
‘Nuff said.
Oh PS : TDF bylaws state : « The Steering Committee will end it’s existence 1 year the latest after public announcement, or earlier, if the Document Foundation has been established and a new steering body is operational. » So the current steering committee is acting for the establishment of the foundation, and elections will take place… So really the emphasis on « un-elected » in your comments is disingenuous.
Some key points, who will make or not the success of this story :
– the quality (first) and (second) the number of developpers
– the companies like IBM or other behavior : will they truly play the game and share the code (e.g. Symphony mess)?
– the Copyright Assignment
P.S. : please don’t forget to mention OOo4Kids -> http://wiki.ooo4kids.org and OOoLight ( http://wiki.ooolight.org) as forks too.
One talks to an open source community on public mailing lists, Rob. Show me the invitation to Apache on any public OOo list of your choice.
So from all this hostility towards TDF (which you assume I’m part of – you know there’s also a bunch of people still at OpenOffice.org who aren’t silenced Oracle employees, right? You’re aware this community you are scared to mention is enormous, right?) it seems safe to assume you do not support Oracle and ASF when they say in the proposal:
“Both Oracle and ASF agree that the OpenOffice.org development community, previously fragmented, would re-unite under ASF to ensure a stable and long term future for OpenOffice.org.”
since you have no respect whatever for members of that community?
Hi!
Just wondering when I read your comment on how you evaluated the different possible places to go – did you consider community momentum? It seems to me as an outsider that LibreOffice won that one, and for an unpaid community member that’s probably more important than anything else. Maybe you see that differently, or maybe that criterium is less important for you?
Isn’t it ironic that Microsoft ist almost opensource-friendly, while Oracle and IBM try their best to fragment, divide and damage as many opensource projects as possible?
I guess no one would have imagined that 10 years ago.
Hi rob
I echo J David Eisenberg and The Contrarian, this is only about IBM and shows little respect for our communities.
@Jeremy, very well said. Thank you for saying it. I do agree that this is fundamentally about a dispute of copyleft vs. non-copyleft, and we shouldn’t permit it be couched in any other terms. I updated my my blog post on this subject with a link to your comment here.
Hi Rob
I asked Mr. Sutor this question also. Many of us are trying to unite the LO community and what is left of the OOo community. Please let me have you thoughts on how a dipository of code at the Apache foundation will help accomplish this.
Best
Russ
@André: the Board of the Apache Software Foundation receives no compensation for its volunteer efforts. In fact, the ASF has zero employees. We contract with some people for services (sysadmin and marketing, primarily), but that is all. I would be just as offended as you, if the Directors received compensation (and I’m one of them!)
@ruurd: I can guarantee you that Oracle is not poisoning the ASF :-P … Apache has a proven track record of being a vendor-neutral group of projects. Sure, some companies perform the bulk of development on some projects, but the Foundation never lets them “take over”.
@Mario: the ASF is not trying to “sabotage” anybody. There is a proposal submitted to the Incubator that is being evaluated. If there is a community behind that proposal, then why should the ASF refuse them? The ASF’s goal is to *help* open source communities. Not send them packing. Yes, it may cause problems from LibreOffice, but I imagine our HTTP Server also causes problems for lighttpd, nginx, IIS, and many others. I fully believe that LO can take advantage of any project that gets built at Apache.
@Alexandre: sorry to say, but the ASF is quite firm with shipping code licensed only under the Apache License, v2.0.
@Cor: if IBM makes zero ongoing contributions to a project at Apache, then I’m not sure how the community is worse off? That seems to be the status quo right now. It seems that people feel an *entitlement* to IBM’s work that I believe is unjustified. I just don’t see how a company (any company, not just IBM!) can hurt a project by *not* contributing to it.
@Italo: Rob and IBM are not in the driver’s seat. IBM has no vote on whether the proposal is accepted. They have no vote on who gets to join the podling. They have no vote on graduation. Any other belief is unsupported conspiracy theories.
There’s one additional point that I’d like to make, which @Alexandre hints at but doesn’t point out explicitly. Rob is quoting the FSF completely out of context in the main post here. Specifically, he leaves out this part of FSF’s post on the Apache 2.0 license:
The existing license of OpenOffice.org and LibreOffice is LGPLv3. Oracle, in coordination with IBM, unilaterally changed the license out from under the community, rather than cooperating with the existing licensing. Oracle of course had the legal right to do so as copyright holder, but this was an act in conflict with the existing community in a moral sense, even if, again, it was a permissible act under the OO.o “community” guidelines.
@Cor Nouws: you said “Then for the code contribution from IBM to OpenOffice.org in the past. Year after year on OOo Conferences the contributions were announced. After years, only little came. Still a lot is not there.” Perhaps not to OpenOffice.org itself, but there are quite a few IBM people working on the ODF toolkit, which I believe to be an important complement to OOo.
@Greg Stein
I really do hope that IBM is not in the driver seat, because having someone like Rob Weir there would be frightening for everyone.
The problem is that the way IBM handled the announcement, which should have been primarily an Oracle announcement and then an ASF announcement, raised the perception that IBM is in the driver’s seat. I’m not talking about myself here, but about the journalists – several of them – that have expressed this perception.
If Apache is happy, then I’m happy as well. It looks like, from what I’ve read so far, that I won’t contribute to the project (definitely not as a developer, but also as a community member).
I have founded TDF to be independent from corporations, and I don’t want to participate in a project where there is a clear influence of a single large corporation (although I understand that membership is managed on a individual basis, and I fully appreciate this, but I am also almost 60 and for the last 35 I’ve been in the IT market, and this has made me just a little bit suspicious about that corporation and his behavior in the market).
@Greg well, then, that still leaves open the possibility of ASF to accept the donation and, instead of releasing the code under ALv2, passing it on to some other organization that does maintain projects under copyleft licenses, such as TDF, FSF, FSC, or even Mozilla.
But, heck, I’m sure Oracle’s and IBM’s lawyers thought of that in advance, and the terms of the code donation prohibit this sort of community-oriented behavior. And even if they didn’t, I suppose Oracle could still preemptively release the code under ALv2 before the transfer. So…
TDF has an active community of developers, strong community support and a headstart of several months with copyleft contributions. We’ll see where that leads us all. Best wishes for ASF to seek success of the project rather than alienate its original community. Oracle and IBM took a bold anti-community bet, now it is upon us, the community, to avoid being manipulated and used, regardless of support or sabotage by the ASF.
Hi Rob,
I was wondering how this incubator project at ASF works?, Is there any time frame where we can see OOo development starting again after it graduates from the ASF incubator project?
Thanks.
As I commented on another IBM blog, I’m glad to see Openoffice.org on the way to becoming truly free for contribution.
Being a longtime Openoffice user, I moved to Libreoffice when it started, as it allows truly free contribution. And I have be very satisfied with Libreoffice’s progress. As for sponsor membership of the TDF which hosts Libreoffice, I don’t care. I’m interested in contributing to the development of Libreoffice itself, something I couldn’t easily do to Openoffice.
However, unlike many others in the Libreoffice community, I see a truly free Openoffice as positive. Undoubtedly, the interface will diverge somewhat, but we will have 2 well-known fully open office suites which are fully (and natively) compatible with the open ODF file formats. Which will, in the long run, be good for everyone. ODF becoming the major office formats will have benefits for all office suite users, even those who use neither office suite.
No more lock-in, formats accessible in perpetuety, a means of exchange available to all.
BTW, I still expect Libreoffice to be much more popular.
IBM might have contributed code, but more than five years ago they promised to include speech to text code in OOo. A promise that has yet to be realized. Breaking promises when it comes to those with accessibility requirements is IBM’s standard modus operandi.
Hello everyone. Obviously I can’t respond to all of you in depth, but trust me, I’ve read all of your comments.
@Zoltan, we’re discussing this topic at Apache. Oracle will give Apache the domain name for openoffice.org. The plan is to preserve all active projects and perhaps archive the inactive ones. Send me an email if you want more information. But we’re not planning on losing anything that is active.
@Cor, when Oracle announced several weeks ago that they were looking to move OpenOffice to a community lead project, we talked to TDF Steering Committee members. We also talked to Oracle. And I’m told that TDF talked to Oracle as well. We all had talks and exchanged emails. Did we continue talking with TDF after it became clear that they were not willing to consider moving to any non-copyleft foundation? No, of course not. But be honest with me. If we had asked you, to join Apache, and we asked you before the announcement, would that have changed your mind? Or would that just have given you more time to prepare blog posts, and contact the press and plan a media campaign to oppose the announcement. Be honest now. Aha!
@AlexH, If someone wants to contribute to Apache OpenOffice, they can join and do so, even during incubation. In fact, this is encouraged. Incubation is more about giving the “podling” (what they call the project under incubation) extra help and attention with regard to Apache infrastructure and procedures. But the development work occurs via the normal consensus process. In fact, showing that you are doing active development is necessary to “graduate” from incubation.
But certainly, there will be a stage in incubation where it is mainly server admin, web site deisgn, build repository work, etc. And not everyone is interested in that.
But all said, this is not something that takes a year. I hearing more like 3 months.
@Italo, It is probably a good thing to separate two things: 1) the contribution of the OpenOffice code, trademark, logo and domain name to Apache, and 2) the announcement of a new Apache OpenOffice project. The first is entirely Oracle’s work. Of course they issued a press release on it. For the project announcement, of course companies planning on joining the project are going to blog about. Does that make more sense?
@Bas, the self-appointed, unelected TDF Steering Committee remains self-appointed and unelected even if they have self-authored bylaws, unapproved by the general membership that say that they might have an election after a year. Maybe call then the “acting steering committee” or “provisional steering committee” or something like that?
When you say that they, “TOOK ALL THE RISK, and stuck their necks out”, what exactly did they risk? Their lives? Their jobs? Their reputations? It doesn’t seem like ranting about a large American software company is really risking anything. But I’m sure it is fun. The LO fork was more the culmination of a multi-year fight between Novell and Sun/Oracle than anything else.
@ericb, we’re putting together a list of things we want to contribute from Symphony. But we’ll need to have a discussion with the new project first to determine the priorities.
@Contrarian, I know OOo was contacted because we saw several OOo members join the Apache project today. I understand an email also went to TDF, But from what I’ve read, they are being told (by Meeks) “I would strongly prefer to see either all of us as initial committers [in Apache], or none at all, and that is a decision we need to make collectively”
@Ole, I see how one could look at momentum of an existing project, but I don’t see how you could do a “what if?” and estimate and compare it across other foundation options. Past performance is no indication of future results, etc.
@Russell, go back a few years, when Sun was running OpenOffice. Core work was done in OpenOffice.org. Others took that code and customized it. Novell Edition, NeoOffice, Symphony, RedOffice, OOo4Kids, etc. We had one big community, with distinguished roles and we got together and partied at an annual conference. Of course, the world was not all roses. There was the CLA to Sun, that many did not like. Sun as the corporate gatekeeper was something that that many, especially Novell, did not like. I’d like to restore what was good in that picture, while removing what was bad. So Apache instead of Sun, no copyright assignment, no gatekeeper. But also no fork. Separate packaging and customization, yes, of course. But core work on editors all done in Apache.
@Bradley, thank you for your moral tutelage. I’ll do ten Hail Mary’s.
@Italo, again, you are confusing the contribution announcement with the project announcement blog posts. Two different things. As for project diversity, Apache has a requirement that for a project to graduate from incubation, it must have an “open and diverse community”. The guidelines state one aspect of this requirement as, “there is no single company or entity that is vital to the success of the project “. It is rules like this that make Apache a valuable place to host a project for the long term. They are thinking about what is needed for success and stability. Something to consider, is whether LO can match that same level of diversity.
@Alexandre, the OOo code has not been sent to Apache yet. It is in two stages. First the signed grant paperwork, then the proposal is reviewed and approved and if approved the code is transferred over and the headers updated.
@Mike, Apache incubation is result driven, not time driven. When we meet the criteria for graduation and are approved then we graduate. There are quarterly reviews and after each review we’re rated as continue, cancel or graduate. My guess is we’ll take 1 or 2 quarters. Much will depend on whether we want to make a release during incubation. Or wait until after for our first release. In other words, is our priority to get a release out quickly, or to graduate quickly.
@Andre, thanks for the comment.
@Amber, actually we did contribute the accessibility work,. But it sat in a CWS and Sun never pushed it. That is one advantage of moving to Apache. We will make feature decisions among in the project, by consensus, with no corporate gatekeeper. And we’ll get that accessibility work finally into the release!
@Rob, “IBM has been a long-time member of OpenOffice.org. We’ve contributed code”. Is there some more detail about what and how much code was contributed ? e.g. <a href="http://www-03.ibm.com/software/lotus/symphony/buzz.nsf/web_DisplayPlugin?openform&unid=62E0A7822C460B45862574A40074E651&category=announcements" in 2007 announces "IBM's contribution of technology and engineering resources" to OpenOffice.org. I'd be interested in how much tangible integrated code this amounted to ?
As an member of the still existing germanophone OpenOffice.org-community I want to clarify, that the OpenOffice.org-project ist more than a big bundle of code.
There are also people, who have played a strong part in the success and high profile of OpenOffice.org not only by developing code or writing documentation but also on many other ways. So developing code or writing documentation is not the only way to make valuable contributions to the project. You could meet some of these people for example last year in Budapest.
I think it would be the best way to continue the success of OpenOffice.org proceeding the good partnership between enterprises (and their employees) and volunteers.
@Rob, A few thoughts:
1) It cannot be denied that some Free Software supporters will find it difficult to contribute under a non-copyleft open source license such as Apache License 2.0. As a result, I anticipate that many contributors to LGPL-based forks such as LibreOffice will not be inclined to contribute their code to Apache OO.org podling.
2) You seem to have a serious issues with The Document Foundation, especially with regards to the “unelected” Engineering Steering Committee. There seems to be sort of a bootstrapping problem here, though. Who would hold these elections? Who would be allowed to vote? Who would be eligible to run? Who decide what the criteria are for eligibility?
Furthermore, do you feel mere governance was a deal breaker for LibreOffice taking over OO.org, or is it simply a personal gripe you have regarding TDF? I would prefer to see a better governance model for TDF with clearer language on how people are elected and appointed, but considering the number of “benevolent dictators for life” we already have in the open source world, I may not give it the same import as you do.
3) Note that setting up an Apache incubator for OO.org independently from Oracle was impossible because ASF only allows their incubators to use own license, and OO.org was under the LGPL. Was there an appropriate existing open source organization (with governance more you’re liking) that could have absorbed an LGPL fork of OO.org at the time? The Free Software Foundation, perhaps?
4) You said: “Did we continue talking with TDF after it became clear that they were not willing to consider moving to any non-copyleft foundation? No, of course not. But be honest with me. If we had asked you, to join Apache, and we asked you before the announcement, would that have changed your mind?”
I don’t understand what you’re trying to say. It sounds like you’re saying that you stopped talking with TDF after it became obvious that they weren’t willing to disband and have all their members and contributors join the Apache project. I don’t think community Seppuku was a realistic demand, so I’m guessing I misunderstand you. Please clarify.
Your astounding hostility to TDF is shocking, Rob.
This episode is actually inspiring me to contribute to LO and TDF.
I wonder if it could be good for recruiting at TDF.
I certainly trust the LGPLv3 and an “unelected” Engineering Steering Committee more than a couple of giant corporates fragmenting a community out of spite and trying to get some free labor/code into their proprietary code bases.
Thanks for replying Rob!, I was wondering the following once the OOo project continues at Apache we will see a change on what versions of OOo will be distributed? (i.e. OOo for Windows, OOo for Linux, OOo for Mac, etc.)
Otherwise, I am very happy on seeing OOo moving into Apache I think there it will go into new directions that we never expected.
@Jeremy
Just concerning the FreeBSD-Apple relationship I have to say, as a FreeBSD user-contributor for more than 10 years, that it has been awesome for us. Apple has hired many BSD guys, and they have added to our code in significant ways. We were never actually expecting them to develop FreeBSD themselves but they have generally been very easy going about relicensing stuff when we need it. They have also enhanced some of our stuff, like smbfs, and very recently they have been helping us in our attempts to avoid the GPLv3 in the toolchain.
And yes, we are *very* happy to see BSD code redistributed in MacOS X.
Concerning OpenOffice and the TDF.. libreoffice is already under the license TDF finds the best in their interest. I don’t see why people have to complain because the code is now *also* under a more liberal license for other people to use. What happened to the speech about freedom, anyways?
That just seems dismissive to me, and frankly slightly rude. I realize that it’s amusing to joke that those who believe in software freedom as a moral issue are akin to Roman Catholics (as most know, RMS is fond of making this joke himself). However, using that joke as a cop out in a serious discussion about an important Free Software political issue is just a rhetorical trick to dodge the issues at hand. I hoped for better discourse here. I still believe you’ve quoted FSF out of context and have made no effort to correct it.
In short, I’ll state again, the FSF’s belief that Apache-2.0 is a preferred permissive license doesn’t mean that the FSF encourages moving from LGPLv3 to Apache-2.0 as Oracle and IBM have done here.
@Rob
If I was a young unexperienced guy, I could believe your words about the separation between the announcement from Oracle, and the different statements of support from IBM. But it happens that I have several years of experience in the specific field of communications, and (surprise) I have even been contracted by IBM in the past for this specific experience (but I found the experience very unpleasant, and I gave up after a couple of months, because IBM is just impossible to work with). Please find a more credible answer, because the one that you have provided is just too naive (or too stupid, if you prefer). From what I’ve seen and read so far, it looks like IBM is the only interested party in Oracle decision.
@ Rob: Maybe, apparently I missed that there has been spoken specifically about non-copyleft in the weeks before the announcement. But the name Apache was indeed very new, when I heard it one/two days before announcement.
On the other hand, I now have read some of the discussion on the Apache list, also in the light that for the LibreOffice project both the non-copy left, and the assurance that contributions cannot be used without changes being given back, are vital. Your explanantion makes it clear to me, that these items are IBM’s reason not to join TDF, and apparently IBM has made Oracle join them in this. Alas, sh* happens.
Dear Greg,
I am very much in favour of a professional board that is compensated for their work. But in the US governance of non-profits is “strange”. When you look in the ASF bylaws you’ll find that compensation is decided by the board. You also find rules that a general quorum can kick members out for whatever reasons. etc.
I have no idea how TDF is currently institutionalised but Cor Nouws said it would be on a mediocratic base. What counts for me is European governance. When most relevant stakeholders are from Europe you would not like a delaware company to run it, and prefer to meet for phone conferences in UTC.
@Caolán, For example, check one of your own mailing lists for a discussion of how LO could benefit from our IAccesibility2 work. That was a multi-million effort for IBM, spread across OpenOffice work, API design, integration with assisted technology devices, etc.
@Michael, this is very true. I tried to make that point stronger in the blog post that followed this one.
@Matthew, do you have a sense for what portion of the 200 TDF developers think that copyleft is the most important issue, above all other? I think it is rather small, less than 25%, but that is just my hunch.
I understand the bootstrapping issue, but a year to hold elections? A year? That is more than bootstrapping. In any case, this is a factor for evaluating organizational stability. An organization that can demonstrate a healthy pattern of having elections, transferring power, grooming new leaders and seeing leadership successions has demonstrated stability. TDF has not shown yet this capability. This is not a personal gripe. Remember, we’re a business. It is very hard to justify to executives an investment in a 6-month old unincorporated “foundation” run by a self-appointed, unelected board, over an alternative that had more than decade long experience running many open source projects, operating as a legally incorporated foundation.
No one asked TDF to “disband”. There are many forms in which TDF and Apache could collaborate, although the most powerful ones are difficult to the extent TDF is unable effectively to make upstream contributions. But we’re discussing the possibilities now, with TDF Steering Committee members on the Apache list. I’m hopeful.
@Zizzle, I wish you luck with the recruitment efforts.
@Mike, the current proposal is to serve all existing OpenOffice users on all existing supported platforms.
@Pedro, thanks for the comment.
@Bradley, you thought my response *seemed* dismissive? I apologize for the ambiguity.
@Italo, I’m sorry, what exactly are you accusing me of now? I’ve lost track.
@Cor, I’m still waiting for you to answer my question. Every TDF member refuses to answer this simple question. “If we had asked you, to join Apache, and we asked you before the announcement, would that have changed your mind? “
Rob, TDF *is* upstream now, and if ASF takes your bait, TDF wll be perfectly capable of legally accepting the contributions made to the ASF project. So let me see if I got this right: you guys wanted a more permissively-licensed project whose contributions could be used by the wider community, but that the wider community wasn’t willing to contribute back to under such a permissive licenses, so that the community project would get the resulted of the combined efforts, while you’d stick with the proprietarizable lagging-behind product? Wow, sounds like a really good deal to the community! Now, being a business (thanks for the reminder!), a corporation driven by the executives that wouldn’t entrust the code to TDF, why oh why would you be oh so nice, but still create this divide rather than contribute directly upstream?!? Puzzling, if one tries to assume it was a well-meant move, isn’t it?
As for looking for a stable foundation, have you ever heard of the FSF? A foundation that has been around for more than a quarter of a century, and that coordinates and supports the development of an operating system that IBM and Oracle have surely heard of, and that’s key to their businesses, in which they have invested billions of dollars, even though they prefer to refer to it by the name of the kernel that runs with it.
Has it been considered? Were there discussions with it about donating the code and keeping the project under its current weak copyleft license? Or was the removal of copyleft the key ingredient to rule it out and make Apache preferable?
I’m just asking because I know the FSF supports the TDF and LibreOffice, so donating to the FSF would be a good means to contribute upstream and merge the projects, rather than dividing the community.
Any change can be done in LibreOffice.
Consider the massive complexity of this technology, and the amount of work to be done.
IBM could help the community by having the core OO developers work in LibreOffice. Note also that LibreOffice could use this help. The proprietary stuff (Notes, Symphony, etc.) doesn’t matter to us, but why not build that on LO as well?
-Keith
@Rob: It is not about if I/we would have changed our mind if we would have been included earlier in talks about Apache. Of course that would have changed little in that respect. But what it would have given, is an opportunity for us all to look if there is a more common ground for joined development.
Now the move from Oracle/IBM gives extra food the feeling that there is something not to trust. And alas there is ground for that from the past.
I again invite you to come up a sense-full proposal from IBM to rebuild trust, about really contributing in a fair way, and not in the way we know from the past in OpenOffice.org.
I think that is a key step to get as much as possible people join in whatever place to work on a as large as possible common code base.
Apologies if the way I ask this is a bit blunt. I’m Dutch you see. An answer to my question is just the logic extension to what I see from people pointing to what went wrong in the past in this respect, and from the many writing about licensing/(non-) copyleft differences.
Rob, it should be pretty obvious that this move has created a PR disaster for IBM, Apache and Oracle (not that Oracle really appeared to care to begin with). Perhaps you should have thought about writing an article on why IBM chose not to work with TDF? If you don’t have the hearts and minds of developers with you on this, the project may survive, but be weakened and continue being tainted.
I’m curious on why IBM originally chose not to join TDF when the fork was imminent (IBM could have had the chance to guide a young organisation considering their experience with open source communities), and my suspicion tends to be the copyleft licence, leading me to believe that this had a much heavier weight in IBMs support for the Openoffice project being hosted with Apache.
@Alexandre
Even when the we owe a lot to the FSF, especially for the early days, I am afraid that free/opensource software has already evolved way beyond GNU is not Unix days to make the FSF irrelevant. The FSF is still important for license zealots that try to give some philosophical coherence to their actions but the FSF and particularly it’s founder, doesn’t code much nowadays. I would think that gcc is the most succesful FSF project and it’s so uglily broken in many ways, that some companies we all know have moved away from it and the GPLv3, as there’s not really much to gain for them there.
Giving Openoffice to the FSF would’ve probably end up making it as popular as the Hurd. For the type of things that remain to be done we are much better of with the Apache guys. I would expect, for example, some work being done to build OpenOffice with CLANG, and further integration with web technologies, perhaps even competing with Google docs, and that is not something that will happen in an organization that has little else than a political agenda.
@Rob: Seems obvious to me that there will be two projects, one based on a permissive license and another on a copyleft license. Should we not just accept this and move forward from there? There is nothing to stop those without too much concern to contribute to both and some to each individually. Who knows, with some efforts on cooperation it might work well :-)
@rob: “check one of your own mailing lists for a discussion of how LO could benefit from our IAccesibility2 work.” I’m sure it’s good work, but the question was more about gathering some sense of the scale and status of IBM’s previous code and engineering contributions back to OpenOffice.org. Taking the ia2 work as an example, in 2007 I believe that was announced initially, e.g. http://blogs.oracle.com/korn/entry/ibm_joins_openoffice_org_community
Was that ia2 work integrated into OpenOffice.org ? and if so when. i.e. http://wiki.services.openoffice.org/wiki/Accessibility/IAccessible2_support suggests that “ooo11ia2” is an unintegrated ia2 branch based on the OpenOffice 1.1 code-base (https://tools.services.openoffice.org/EIS2/cws.ShowCWS?logon=true&Id=6175&Path=SRX645%2Fooo11ia2). While digging in the OpenOffice.org “cws” tooling I also see “ooo31ia2” which seems to be an ia2 branch around OpenOffice.org 3.1, where work started in 2010-05-24, but is also an unintegrated branch (https://tools.services.openoffice.org/EIS2/cws.ShowCWS?logon=true&Id=9444&Path=OOO310%2Fooo31ia2).
Considering the recent spat with Java/JCP, I’m surprised you had the audacity to say what you did, Greg!
It was disappointing and could be predictive to read your article mentioning support for and welcome to users of Linux, Windows and Mac OS X, yet no mention of the only other Free/Open Source Software (FOSS) Operating System group – the *BSDs.
If you have any inside knowledge or understanding of *BSD being left out in the Aapache cold, please indicate so, so that those of us using FOSS OS primarily will know where the emphasis of OpenOffice lies.
Possibly the more “popular” OS, including the proprietary guys to the exclusion of other reputable and established FOSS OS?
W. Anderson
[email protected]
Pedro, your smearing comments about GNU and the FSF are either misinformed or malicious. While it is true that there are lots of Free applications and libraries nowadays, and a lot of the job of creating an operating system is done, a lot of work is still done in GNU, maintaining the hundreds of programs that people rely on (many without even realizing it, many without realizing they’re part of the GNU project), and developing new ones. Should we go about counting the number of actively-maintained GNU programs, to compare them with Apache, and the proceed to assessing how many of them are essential for a majority of users of GNU/Linux, Oracle Solaris and even IBM AIX?
That the president of the FSF codes little recently is totally irrelevant: he is indeed devoted more to political activities, which are a far more important contribution that writing lines of code, but the GNU project advances regardless of his code-writing or even day-to-day coordination of individual projects. Many people perceive changes from code monkey to management and then to executive as career advancement, even if you don’t get to code much any more. Do you not value the contributions of your manager, or of the CEO or chairman of a company?
Besides, conflating GNU with the FSF is a common mistake by those who disagree with the FSF political agenda and want to dismiss GNU’s technical achievements by association. Although the FSF is a political organization, GNU is a software development project. The suggestion that the FSF could be the legal entity to offer legal support for the OpenOffice project has nothing to do with its becoming part of the GNU project. Indeed, I explicitly mentioned that the FSF, as supporter of the TDF, could just make it part of LibreOffice, which AFAIK is not part of GNU.
As for GCC, I wish you could mention the PR numbers of at least one of these serious problems you allude to. From the way you phrased it, I get confirmation of my earlier impression that the ugliest issue that some monopoly-friendly companies find in recent versions of GCC is precisely GPLv3, which has caused some freedom detractors to begin reinventing GCC and other GNU tools, poorly. Your highlighting GPLv3 brings us back to one of the highlights in IBM’s announcement: the monopoly-friendly Apache license. Coincidence?
@W. Anderson, no offense intended by omitting BSD. Do you know of anyone who can help us maintain that port? If you know anyone, have them send me an email. I’d be happy to propose that support includes BSD, but we need the expertise.
@Alexandre
Do check google for “How much GNU is there in GNU/Linux?”. It is an interesting reading (by another Pedro, not me), and please note that I am not confusing the GNU license with the FSF (which used to point to the same web page some years ago).
I’ve been following the efforts in the BSD´s to get rid of GNU software in their codebases, and guess what … they are doing fine. They are not only replacing things because of the license, they are actually improving the system.
For comments on gcc, you will find Marc Espie’s comments in undeadly.org on why he would prefer pcc quite enlightening. I have some other private comments from a developer that I won’t name here because the mail is.. private “GCC on the other hand is admittedly really pushing that old junky codebase around.. I see it as an Aeroflot airplane with sticky tape on the wings.. It flies, but gosh it’s scary..”. gcc’s PR’s? quite honestly I stopped looking on after gcc 4.2.1.
Hmm… IBM, I recall we asked them to make JFS available under a non-copyleft license many years ago and they didn’t .. good thing they have learned something about free software with the years.
@Rob
FreeBSD has a specific mailinglist for openoffice which is also listed as the maintainer for the FreeBSD port.
I forwarded a link to your blog there.
.
Nice to see ASF involved with OOo.
Is amusing to see so much people does not like the idea and in the same time use the Free Software Flag :). Apparently ASF has not the right to develop some kind of software because there is a “Free Community” with some god send rights.
As a Free Software supporter and user, I welcome ASF in the Office arena. ASF develop very clever software and in OOo this is the most important to me and for almost every user. For those people complaining about the move, in the worst case scenario (no agree at all with LO/TDF), is handy to remember that competition is good.
Pedro, I read the “““study””” that selectively counted GNU components to getto an intended result, and still didn’t. You might find my comments on that LWN thread, as well as the *real* data points provided there, an interesting read as well.
As for GCC (even though its development has nothing to do with the FSF), it does have a legacy codebase, having been one of the first GNU programs started. But the parsers was rewritten recently, the Tree-SSA infrastructure that amounts to most of the middle-end is all recent, so the legacy is mostly in RTL. Even there, the register allocator is all new, and so are important parts of the instruction scheduler. Still, RTL is where most of the machine-dependent bits are encoded, and where some of the trickiest passes such as reload are implemented; these will likely be with us for a while. I couldn’t find Marc’s comments on GCC at the site you mentioned, other than a link to an interview given in 2004. Was that it?
Rob, I was giving you the benefit of the doubt. I didn’t want to assume you were purposely being rude, but apparently you were. That leads me to the conclusion that you probably didn’t err by quoting the FSF out of context; you most probably did it on purpose and are just trying to avoid the conversation now that I’ve pointed it out.
@Bradley, I’m not intending to be rude, but just dismissing your ideas as irrelevant. I assume you would do the same if I commented on your blog telling you whether or not you should eat meat, wear red on Tuesdays, or cross the path of a black cat. Your moral views are of no significance to me. The fact is, Oracle has clean title to the copyright of this code. Or at least no one disputes this. It is primarily the product of their employees. The remainder was from 3rd parties who signed over the right to re-license it. No one was coerced to do anything. Oracle is free to contribute it to Apache, contribute it to LibreOffice, contribute it to the Committee to Reelect Barack Obama or to print it out and burn it in the parking lot. We’re all entitled to our opinions on this. But when you start talking about your opinion as being the “moral” choice, then yes, I dismiss you.
Rob, if the notion of morals is so unimportant to you, to IBM and to Oracle, it means that we humans are failing, because our *only* way to keep corporations in check and bring benefit to humans rather than exploiting us, is to ensure companies behave in ways that are compatible with ethics and morals. We can do that through laws, in true democracies, and through citizen and consumer activism (voting with our acts and our wallets). I hope enough people realize we have that power, and we can choose not to be manipulated, exploited and used, so that companies that choose to dismiss issues that are important to humans because they don’t regard them as valuable to their business find out the hard way that we people are relevant after all.
“But when you start talking about your opinion as being the “moral” choice, then yes, I dismiss you.”
You seem to be of the opinion that there is no moral implications but regardless of the legal agreements, there is. If IBM fails to recognize this, then it is IBM’s big failure in this PR disaster
@Alexandre/@Jeffrey, I never said morals are not important. But if we’ve learned anything from the last 30 centuries is that nothing good comes from attempting to impose your morals on others.
Personally, my value system is based on freedom, including the freedom of association, the freedom to own property, the freedom to enter into uncoerced contracts regarding that property. A man should be free to do pretty much whatever he wants so long as it does not interfere with someone else’s freedom. Oracle donating the code it owns to Apache, aside from being generous of its own right, takes away no one else’s freedom.
That’s my personal view, but I have no interest in pushing this on any one else as moral obligation.
@alexander
Hmm.. you are counting some stuff that I wouldn’t consider part of an OS .. emacs, gcc, gdb? OTOH FreeBSD carries groff which you didn’t count so I will concede such reports can be biased easily.On FreeBSD we could get completely rid of GPL software in the base if we wanted to by using packaged binutils (ld and as only + perhaps sort) as for the rest we have equivalent or superior replacements in the works.
The story behind gcc is particularly interesting. If the FSF hadn’t given up control over gcc, egcs would be the dominant fork.
Marc’s comments are here:
http://undeadly.org/cgi?action=article&sid=20070915195203&pid=52
On FreeBSD we got tired of trying to have gcc include some (-f) format extensions that we use in the kernel, so the packaged gcc will not be able to produce kernels out of the box. CLANG, OTOH, did include those and they have been very receptive with our requirements, plus their nice warnings and C static checking facilities have helped a lot, so the way ahead seems clear.
– don’t forget the total nightmare of autoconf/libtool/automake. Heck, even the GCC people have taken years to update their infrastructure to a recent autoconf. And GCC is *the only program in the ports tree* that actually uses its own libtool. Its configuration and reconfiguration fails abysmally when you try to use a system-wide libtool.
I could actually go on for pages…
I’ve actually been de facto maintainer of GCC on OpenBSD for a few years by now, and I will happily switch to another compiler, so frustrating has been the road with GCC.
(oops .. I actually included an incomplete quote of Marc Espie’s comments in my last post.. sorry.
Please ignore anything after ” so the way ahead seems clear.”
Rob, that is true, what Oracle is doing with OpenOffice.org code doesn’t take away anyone’s freedom. It is what IBM wants to do with it, that makes a non-copyleft license so important, that will. That’s why I’m writing about this here. That’s why a non-copyleft license is a poor choice: although it doesn’t take away any freedom from anyone, it does offer power to take away others’ freedoms. That is the key difference between copyleft licenses and non-copyleft licenses. Please don’t conflate one’s freedom with power over others. Copyleft vs non-copyleft has to do with the latter, not the former, so your bringing up freedom is an attempt to confuse (assuming you’re smart enough to not be confused about these issues yourself; if you are, I apologize for not giving you the benefit of the doubt on this one, and I’m truly sorry for your naïvette :-).
@Rob, you continue to ignore the fact that I raised two issues with you. I raised the question of morality of the relicensing action, and I raised the fact that you’d quoted the FSF out of context in the primary post above. In your responses, you’ve consistently over-focused on the former, presumably to avoid the latter. That’s keen debating on your part, but it does indicate that you probably have no remorse for misrepresenting the FSF’s positions for your own political gain.
(Feel free to reject all my recent replies .. this is getting completely off-topic wrt OpenOffice so I won’t reply more on your blog … I’ll just bitch Alexander in private by email ;-). )
You talk about imposition and coercion. It is not about that. Contributors to Openoffice.org including competing vendors like Red Hat and Novell but also many volunteers did so in good faith that Sun would setup a foundation in the near future as it had promised and continue to license code under a copyleft license. Sun and its commercial agreement with IBM which allows IBM to sell Symphony as a proprietary fork of Openoffice.org changed the dynamics of how Openoffice.org as a open source ecosystem worked. Oracle is now unilaterally relicensing the code under a non-copyleft license. While Sun might have been granted the legal right to do so which has been inherited by Oracle now, doing so breaks the inherent trust of the contributors which is beneficial to IBM since IBM can continue to derive benefit from taking all the Apache code contributed by anyone else and building a proprietary product however it isn’t of benefit to a community build around a LGPL licensed codebase. This does present some moral questions and also the question of trust which is important when mixing commercial vendors with volunteer contributors in a critical project. I think Oracle and IBM are off to a bad start here and you don’t seem to appreciate the importance of it and without taking the issues into consideration, I don’t see a possibility of reconciliation with the LibreOffice community which you seem to detest very much. Whatever works for you.
Haha, Rob, you are describing the Calligra Suite perfectly. Was that the intention? :-)
@Alexandre, I fail to see what “power over another” that Apache 2.0 would give that they would not already have via their copyright as the author of a creative work. Remember, copyleft is just a license. It only works legally because of copyright law. You cannot have GPL without respect for copyright. So copyleft only works because of the power the author has over users of that work, e.g., the power to sue them if they do not respect the copyleft license, the power to deny them use of the work. So I don’t see any less use of power over others with GPL. It is just in a different form. I could just as well write a license that says “You can use my code so long as you volunteer at a hospital for one hour” That would also give me, as the original author, power over you, if you wanted to use my work. I could also have a license that says, “You can use my code in exchange for $1000”. That is another form of power over you. Ditto for any other form of license. What makes Apache 2.0 so nice, is that it gives almost no conditions on what the receiver does with the code. It is much closer to a true commons.
@Rahul, Apache OpenOffice would very much benefit the LGPL LibreOffice since they would be able to take any of our Apache 2.0 code and reuse it in LibreOffice. The licenses are compatible in that direction.
@Inge, good point! I think certainly in the mobile/web space, Calligra Suite is far ahead of OpenOffice.org.
@Bradley, the excerpt I gave was from Brett Smith’s blog post. The passage that you gave is from an entirely different web page. I didn’t quote anything out of context. But perhaps Brett did? But I did give the link to Brett’s post and via the magic of hypertext, the reader can read it all. Ultimately, “context” is subjective.
In any case, if you want to get into the “How to choose a license” text, then I think you are guilty of cherry picking what paragraph you choose to invoke. I find that this part is much more applicable to OpenOffice.org:
OpenOffice, as the leading (marketshare wise) open source implementation of the ODF document standard, competes against the proprietary Microsoft formats.
Rob, this FSF recommendation you cite would indeed support releasing the bits that read and write ODF under a permissive license, for broader adoption of the format in other programs. But arguing that this small portion should imply that the entire suite, with all filters for various different formats, user interface and applications should, by association, get the same treatment is fallacious at best.
I’ll address your misunderstanding of power vs freedom in a separate, longer post.
Rob, WRT freedom vs power, I just wrote what follows down, and posted it also on my blog:
http://fsfla.org/blogs/lxo/pub/freedom-vs-power
Some definitions and premises:
* freedom is the ability to make decisions about yourself and your own
future, and to carry them out
* power is the ability to make decisions about others and their
future, and to carry them out
* one’s freedom ends where the neighbors’s freedoms begin
* software users and developers deserve freedom over each program they
use to carry out their computations, specifically:
** the freedom to run the program for any purpose
** the freedom to study the source code of the program, to learn from
it and to verify that it behaves as intended, and to adapt the
program so that it does what the user wishes
** the freedom to make copies of the program, and to distribute them
** the freedom to improve the software, and to distribute the
improvements
* rights are freedoms recognized and defended by social institutions
* privileges are power granted and enforced by social institutions
* copyright is the power to prohibit others from copying, modifying,
distributing and/or publicly performing a work of authorship
* copyleft is a technique, first developed to defend software users’
freedoms, by curtailing power while respecting freedoms
So you see, copyright is a misnomer, for it is does not amount to a
right to copy. It is rather a privilege, that empowers its grantee to
stop others from copying (and more). Therefore, copying is not a
right, for it is not defended (it is rather attacked) by social
institutions, and copyright is a privilege, but not a right, for it is
power (decision over others), not freedom (decision over oneself).
Copyleft was first implemented through copyright licenses, using the
restrictive power of copyright to permit the distribution of software
only in forms that respect recipients’ freedoms. The conditions set
forth in copyleft licenses do not prevent anyone from running,
studying, adapting, copying, improving or distributing the software or
improvements over it, so they don’t interfere with freedoms. The
conditions rather refrain from granting recipients a privilege: the
power to deprive other recipients’ of their freedoms.
The effects of copyleft do not depend on copyright; that’s confusing
specification with implementation. Although GPL, AGPL, LGPL, MPL,
EPL, etc implement copyleft through copyright license, similar effects
could be attained in the absence of copyright through distribution
only under contractual commitments, through consumer-protection laws
(empowering consumers to demand respect for their freedoms), or
through preservation of freedom of expressions.
http://fsfla.org/blogs/lxo/pub/manifesto-livre-express
Now, per copyright laws worldwide, authors of works of authorship are
granted a privilege, the legal power to exclude others from various
uses of the works. Recipients, without permission from the authors of
a work covered by copyright, cannot distribute the work to third
parties, so as to exert power through selective (e.g. sourceless)
distribution, nor can they, without permission, become co-authors, by
modifying the work, and then distributing modified copies to third
parties, so as to exert copyright powers.
Copyright law grants to each (co-)author of a work the power to veto
others’ freedoms as to the work, and the power to extend (or not) the
privilege to potential co-authors of derived works. Copyleft uses
neither power: it grants permissions so as to respect freedoms, and it
refrains from extending the privilege to co-authors, so that they
don’t get to veto others’ freedoms. Permissive licenses, such as the
X11 license, variants of the BSD licenses, and the Apache licenses,
don’t use the power to veto others’ freedoms, but they do extend to
others this power to veto.
Therefore, for recipients, the difference between copyleft licenses
and permissive licenses is not in the freedoms, but rather in the
power to not respect others’ freedoms. Now, given that one’s freedom
ends where the neighbors’s freedoms begin, this is an illegitimate
privilege. For uses grounded on legitimate rights and freedoms,
copyleft and permissive licenses make little or no difference. Thus,
permissive licenses don’t respect more freedom than copyleft licenses;
they rather enable illegitimate power over others.
Of course, ethical recipients won’t abuse this illegitimate power, but
why take the risk of granting this power to recipients in the first
place, if you can avoid the tragedy of the commons through a credible
commitment that keeps unethical recipients honest, such as copyleft?
http://fsfla.org/~lxoliva/papers/free-software/BMind.pdf
@Alexandre, that fact remains that copyright is the law and is considered a form of property. That fact that you think this form of property is unjust or merely a social convention is irrelevant. We can point to societies that considered real property to be unjust and eliminated private ownership of land. We’ve also seen societies where even the most basic ownership of “self” was denied. The general correlation has been that where property rights are denied, that this comes at the same time as denial of civil liberties as well. Economic freedoms and social freedoms are connected.
It scares me that the same arguments that you make against the social institution of copyright can also be used to deny me my house and my belongings. You could just as well say that the exclusive control I exert over my house and my bed is a privilege granted by society and that this represents power over others, by denying them entry, and that my lock on the door vetoes their right to enter.
In any case I fail to see the problem with Oracle taking that code that they own and releasing it as open source under the Apache 2.0 license. If any person or group of persons feels strongly about having a copyleft version then they are free to take that code and modify it and put it into a GPL fork. No one denies them this freedom. And those who use the code under that license are of course obligated to follow the terms of that license, since that license is what gives them rights to use the copyrighted work.
Maybe the confusion is that you think that Oracle has somehow licensed itself the code under LPGL and is somehow obligated under LGPL? That would be ridiculous. Oracle owns of the code. They control the copyright. They are not under terms of any copyleft license for OpenOffice. They have freedom to do what they want with this code.
Copyright was not regarded as a form of property for almost 90% of the time in which it existed. That’s a very recent attempt to reframe it. That copyright can be used to exclude others from doing whatever they like with copies of a work in their possession has nothing whatsoever to do with your ability to exclude others from your tangible property in your possession. This kind of faulty reasoning of yours is just one of the reasons that show how nonsensical it is to try to fit copyright in the property framework. Property doesn’t expire into the public domain, nor does it apply to copies of an object.
Again, my issue is not with the freedoms Oracle respects of others, it is rather with what others (such as IBM) will do with the powers that Oracle is going to give them. I’ve already written than, and you don’t show good faith trying to pretend that the issue is a different one.
I’m fully aware Oracle, as sole copyright holder, is the licensor, not a licensee, so the provisions of the LGPL for licensees don’t apply to them, no matter how much lured-in contributors may have expected Sun to retain those licensing terms. It’s a perfect lesson on why volunteer contributors should stick to respecting freedoms, rather than giving more power to profit-driven corporations. Fortunately LibreOffice doesn’t seem inclined to make this mistake.
@Alexandre, Let’s look again at the FSF statement on licensing recommendations. Their guidance is to use a permissive license for “projects that implement free standards that are competing against proprietary standards”. But then you suggest that this applies only to the “bits that read and write ODF”.
I think you confuse what the ODF standard actually is. The ODF standard specifies more than just the serialized format of the document. It also runtime requirement, such as how spreadsheet formulas are calculated. etc. The ODF standard “specifies the characteristics of an XML-based application-independent and platform-independent digital document file format, as
well as the characteristics of software applications which read, write and process such documents.”
So an ODF is more than just reading and writing. It is the “processing” as well. So in a spreadsheet. the display of charts, the calculation for formulas, the styling of text, the formatting of numbers and dates, as well as the reading and writing of the document, are all things specified by ODF.
Rob, even if your argument holds any water as to the extent of ODF, the FSF has made its position regarding OpenOffice under a permissive license pretty clear, so they evidently don’t agree with your too-broad interpretation of their own recommendations.
http://www.fsf.org/news/openoffice-apache-libreoffice
@Alexandre: Sorry, I didn’t mean to suggest that logic or consistency is necessary to the FSF’s position. They can say whatever they want. I’m just saying that their position, and your argument, would be more persuasive if it were based on facts, including facts concerning what the ODF standard actually defines.
But please, don’t let the facts get in the way of your propaganda. I’m told the fear of Apache OpenOffice has caused TDF’s donations to triple. Maybe your scare tactics will increase donations to FSF, to save the world from the big bad Apache 2.0 license? I wish you greater luck in your future endeavors. But this one does not seem to be going your way.
I just hope you meant it when you wrote you were for freedom, and that you didn’t mean for power over others, and that you practice what you claim to be for, even when the time comes to make decisions regarding OpenOffice and its release under a license that permits abuse, and when the company you work for proceeds to use that permission to deny freedom to its customers. If you’re for freedom as you wrote, it would seem that this path isn’t quite going your way either.
@Caolán McNamara
If the IAccessible2 implementation is in an OpenOffice.org 3.1 branch, that’s because that’s the version Sun/Oracle asked IBM to contribute it to, as far as I know. By the time the code was donated, Oracle had released a new version of OpenOffice.org. Oracle was still integrating and testing the IAccessible2 code in February 2011 (FOSDEM Conference, Brussels; I have not seen any public statements on the status of this work since FOSDEM). I hope it will be part of the donation to the Apache Software Foundation; it is an important contribution that should not be lost.
Being a plain user, I will look at the quality and features of the respective projects and then decide – I am not using politics, but software.
(Hint: Whosoever truly supports opentype-font-features, probably wins.)
Couldn’t Apache decide to dual-license it Apache 2.0/LGPLv3+ so that LibreOffice and Apache OpenOffice.org could freely contribute in both directions? Or does the Apache Foundation prohibit such a scheme?
I see it as a welcome move by Oracle in having given openoffice to Apache. I submit the following:
1. The development by the community to continue and the product may be renamed as ApacheOffice.
2. As stated earlier in some forum, let there be common modules (and code) each of which could be made as efficient as possible or as it could get by a given time and then different office suites may combine them ingeniously to deliver a product that have varying degree of features.
3. Is it possible to make the database component (base) to natively use postgresql?
4. LibreOffice, openoffice, etc all probably have a common base and doing good service in this segment.
5. The community is also becoming more active and are contributing useful addons.