Sun Urged to Give Up OpenOffice Control

inc_x writes "Developers from OpenOffice.org are urging Sun to set the project free and bring it under a foundation. Sun's dominance over the project makes other companies such as IBM, Redhat and Novell reluctant to contribute more. Both Mozilla and Eclipse managed to attract an increasing number of developers after the projects were moved over to an independent foundation."

should happen

[Anonymous Coward]

The next logical step - should have been done allready. I can't really se OO go very much further unless they go this way.

Re:should happen

[Daengbo]

How could Sun then relicense the program for sale as StarOffice? In my understanding, the Mozilla foundation can continue to operate on its own while Netscape Navigator is released because of the MPL license, but OO.o is under the LGPL, and Sun requires all submissions to be signed over to the company so that the program can be dual-licensed. How would this work if OO.o became its own Org, like Mozilla. I don't see it happening unless Sun gives up the StarOffice brand.

I for one..

[jimsteri]

I for one.. Just kidding. I don't actually see why Sun would not agree with this. But on the other hand I'm thinking from the user aspect and not corporation aspect. More developers sounds good for me, usually more and better features. But putting OO under its own foundation probably means less money for Sun?

Re:should happen

[Lussarn]

If Sun as you say has the copyright on the complete program today they can relicence it in any way they want. They don't have to use only LGPL, they can even use a BSD licence. I don't see the problem.

Don't

[Anonymous Coward]

While it's probably possible to fork it, it wouldn't be a good idea imho, especially seeing that the bulk of OO developers are from Sun anyway. So I don't think a fork would achieve much and I don't think a fork is really merited in any way at this point.

This isn't about forking however, but about better organizing OO developement, in order to attract more external developers, which would of course be a very good thing and probably also benefit Sun.

Maybe Sun should keep it?

[Anonymous Coward]

Mozilla, for all the support it has, still hasn't achieved any of their goals. 4 years later it's still essentially NS code, and it's plagued by code nobody likes, and bugs [slyerfox.com] both inherited and introduced.

With Sun at least you've got one company at the wheel so to speak.

Re:Being urged by developers is one thing

[TheRaven64]

Considering that over 80% of OpenOffice.org developers are employed by Sun (statistic provided by Novell), I wonder who, exactly, the developers asking for this are. I attended a talk by a Novell OpenOffice.org guy a while back, and his view was that the baroque build system was the biggest reason that new developers didn't get involved, and they had people working on simplifying that.

Re:It All Depends on Sun's Goals

[Decaff]

If Sun is interested in goodwill, then this seems a great way to go.

Open Office is possibly the single most important reason why Linux is useful as a workstation OS. Seems to me like they deserve all the goodwill anyway.

That's strange

[ThePhilips]

That's strange. We do hear that request from IBM.

But in fact I heard that most FLOSS developers are turned down by the size and overall (low) quality of OOo code.

As one developer said on blog (I failed to find that remark again) the thing is only paid Sun developers would work on it. And only because they are paid to do so. Compilation take ages and level of requirements for development is high - that all creates entry barrier to FLOSS developers, most of whome work in their own spare time.

To put in prospective: what would you want to spend you time on: hacking Linux kernel and then in 10 minutes seeing your changes or waiting N hours when OOo compilation finishes?

I never looked into OOo sources. But the pace of progress project makes - and the kind of progress it makes - tell quite much about how project is organized. I truly hope that KOffice would be able to run on Wind0ze - in office unfortunately I'm completely confined to the M$ Wind0ze. At the moment only OOo can read the SXW files OOo produces upon import from M$O... AbiWord fails completely to pick up styles in such documents. KOffice 1.4 is quite close to render the files the way as OOo does.

Pretentiousness

[NekoXP]

Mozilla has gotten new developers since Firefox, NOT because it's not controlled by AOL/Netscape anymore.

I wish developers would be less pretentious about their choice of projects. Surely successful projects
which have significant amounts of corporate backing, both financial and in terms of management, are some
of the better projects to work on. You have defined goals, a great infrastructure to work in, and nobody
ever complained about the way Mozilla was being run before The Foundation (in fact The Foundation works
exactly the same way for every developer in terms of bug tracking, IRC events, software testing and
releases, as it did during AOL's tenure)

OpenOffice could get more developers if it had some unsubstantial hype or managed to get a bunch of new
features it already had (get rid of Java and implement everything the same way, some other way :) but
not just because Sun would have dropped it. I actually think OpenOffice (like Seamonkey becoming a tiny
little sideproject in view of Firefox's popularity) would suffer for it.

Re:Causation or correlation?

[Anonymous Coward]

Eclipse has become the dominant IDE because it was set free, otherwise there would have been no other reason to use Eclipse 1.0 over NetBeans or what have you. Because it targets Java programmers you can bet that it definately benefited by this (I see a Java bug in the tool I do my Java coding, I fix it)...

I'm not sure Mozilla is not so open and shut, devs would have had different reasons for working on it which may have been related to past loyalty to the browser, hatered of IE, or just plain curiousity... But I'm not sure what the OSS developer counts were for the Netscape/AOL Mozilla vs. Mozilla.org Mozilla releases.

OO.org biggest problems

[Libor Vanek]

In my view OO.org biggest problem is not that SUN pays most engeneers but huge complexity of OO.org. I've heard even some rumors that OO.org contains ASSEMBLER language in some parts!!!

To make much more flexible whole project needs to become much more modular (which equals trash all existing codebase and start from scratch):
- file modules (input/output) - in ideal world OO.org would share this part with AbiWord, KOffice etc...
- "processing" module (document "managment", scripting etc.) - imagine running OO.org without GUI (some server document processing etc.)
- "GUI" module - native Qt, GTK, Windows, MacOS, etc..

But I'm 99% sure that this will not ever happen. More probable is that KOffice will become much more usable and supported on Windows/MacOS.

Re:should happen

[Daengbo]

It's kind of my point that they really can't keep the current license and still sell StarOffice, because they wouldn't be able to take code which isn't theirs and relicense it. They would have to move OO.o to a BSD-style license to still sell StarOffice, right? And that would alienate a large number of developers who prefer the (L)GPL. Sun would also be seeing numerous, virtually identical competing offerings from other companies (e.g. IBM). I just don't see the motivation for Sun to do this. When Mozilla was cut loose, it looked to me to be a way to cut developer salaries, and since the Netscape brand was pretty much defunct (and free!) anyway, there was nofinancial disincentive to move Mozilla into its own org. StarOffice is, as far as I can tell, making "some" money for Sun, still, and is an up-and-comer, not a has-been. My two won (SKW).

Re:should happen

[Daengbo]

For Sun, the company, it probably is about whether it's a good financial move for them or not, wouldn't you agree. Since they're the ones making the decision, and not the developers, about whether to move OO.o or not, I expect that marketing will weigh in heavily.

Re:good step

[FireFury03]

MS could of course turn to just rely on MSN and Xbox

Didn't the XBox related activities make a loss?

Re:Causation or correlation?

[Savage-Rabbit]

Genuine question - did Mozilla and Eclipse gain developers because they were "set free", or is that just coincidence? (Remember - just because B followed A, doesn't mean that A caused B)

Genuine answer - Alot of developers have clauses in their employment contracts about what they can and cannot do in their spare time in terms of software develoment. In my own case (I had a lawyer check my contract) I can am free to work on OSS projects if they :

1) Do not undermine the business of my employer. That is the OSS project represents a competing product.
2) The project is not conntrolled by a competing company or corporation.

So I am guessing that it was at least partly a case of Mozilla and Eclipse gaining developers because they were 'set free'.

Some developers have truly draconian clauses in their contracts about the extent to which they can participate in OSS projects. I have even heard of people being forbidden by contract to develop software for anybody but their employer no matter what the circumstance or the nature of the development work (ie. even if it is an OSS project that is solely for their own enjoyment, unrelated to the employers line of business and not for profit). Such clauses would probably not hold up in court, at least not in most EU countries, but corporations include them in employment contracts anyway. The same goes for anti competition clauses, ie. "If you quit and start working for a rival corporation you must remain unemployed for N months before starting your new job". Supreme courts in a nubmer of European countries have have declared such anti competition clauses to be invalid but they keep being included in employment contracts regardless. I suppose employers are counting on their terror value since employees may be reluctant to take the matter to court even if they will win because of the legal cost and the time-demands and hassle of a court case.

OOo needs a Firefox makeover

[idlake]

OOo is at the same stage as Mozilla was: a functional but bloated and messy codebase and system. Unfortunately, that's what big companies tend to produce (I think it's a consequence of having too many engineers, many of which are mediocre).

What needs to happen to it is what happened to Firefox: the thing needs to be split up, the GUI and cross platform toolkit need to be overhauled (or even replaced with Gtk+), and Java needs to be exorcised from it.

And, yes, severing the connection with Sun would be a good thing for OOo, and ultimately for Sun as well.

Not only OpenOffice, Linux kernel too

[Anonymous Coward]

It is not only OpenOffice, Linux kernel development also should bring under a foundation and set an example for the open source community. The "Linux" trademark also should bring it under that foundation.

Re:It All Depends on Sun's Goals

[Decaff]

I love OO.o, but I sometimes wonder if we would now have a significantly lighter, "cleaner" office suite had OO.o not dropped into the picture when it had.

You may be right, but, sadly, I don't think that users want a light clean suite - they want something that looks like MS Office.

Re:OOo needs a Firefox makeover

[NekoXP]

(I think it's a consequence of having too many engineers, many of which are mediocre).

How do big companies tend to produce that, but you forgot all those huge, bloated, never-controlled-by-a-corporation projects like GCC, XFree86, and suchlike?

Too many cooks spoiling the broth IS what causes it, but why make the dig at big companies?

Re:It would make sense.

[zenmojodaddy]

On the other hand, if IBM, Red Hat et al are encouraged to commit manpower to the project by a loosening if licence restrictions, wouldn't that help in cleaning up the codebase?

That's couple of pretty big ifs, of course. Many open source projects, for good or bad, tend to focus on the addition of new features rather than just making the old ones work properly and cleanly, every time; so chances are, more manpower might well be used deployed on tasks other than cleaning and streamlining, despite OOo sorely needing a bit of both. Still, a guy can dream.

Re:should happen

[MooUK]

There wouldn't be anything stopping them selling OO.o, or selling support for it (which is, as you said, WHY people buy it - the support) under the GPL or LGPL. There is no reason here for them not to sell it. The GPL does NOT prevent selling your product, contrary to most people's interpretations of it.

(In other words, the parent is right.)

Re:OOo needs a Firefox makeover

[jilles]

Agreed. Except the changes that need to take place are probably much more like netscape 4 vs mozilla 1.0 than mozilla 1.x vs Firefox 1.0. Essentially firefox was only about the UI whereas mozilla was a complete rewrite of netscape 4.x.

Open office consists of close to 10 million lines of code. Much of it is star office legacy code that very few people understand. Because of this legacy, feature development happens at a snailpace and the UI looks like shit.

A complete rewrite is not a realistic option for Sun. Doing so certainly killed Netscape and their product was a much smaller software product. A full rewrite would take the best part of this decade. A more realistic strategy would be to gradually replace the worst parts of the system. Identify the problematic components and fix or redesign them. Sure the UI sucks but it sort of works. The 2.0 development even made it look semi native on some platforms. Deoupling the components of the suit means that you can evolve them (or kill them off) seperately.

Guilty is not Java, or Sun.

[ahmetaa]

i am always buffled with the ignorance of the people who thinks Java is causing the problems with the OpenOffice project.
- Java is used in a small amount of the OpenOffice (database and some less known seperate components). i wish it would be used more. Especially in the GUI part, check NeoOffice if you want.
- Bloat is caused by the C++ side of the application. And not Sun's fault it was already rotten when Sun bought the company.
- Complexity is caused by the C++ side of the application. it would be hell if they tried to use C++ for all the database part.
- People thinks everybody will rush to participate code if it would e an organization are dreaming. if Sun does not pay, it will be in the hands of one or two stinky hacker. coding C++ code is no feast especially for this magnitude of the project. Firefox has a bunch of core developers and one million people who just rub the developers back, thay are suffering grately. Eclipse is lucky on that side because they use a better and easier programming language (Java).

Re:Being urged by developers is one thing

[Bill_Mische]

"...they automatically forfeit the copyright. "

C'mon you know better than that.

They merely allowed other people to copy it under the terms of the licence. If the licence is breached then normal copyright applies. Similarly in the event that you had a line of code included in a project then you would be able to enforce the terms of licence for *your* line of code.

As to the rest, I might agree with your summary but I find the "how dare they compete with us" attitude of the middle bit a little silly.

OpenOffice dosen't matter - ODF does

[antonallan]

A lot of the people here have high hopes for OpenOffice to be the killer app that could bring enterprise users to Linux. Hence the many suggestions and ideas about what Sun should do with it.

But to me, the key is not OpenOffice but OpenDocument. With the widespread adaptation of an open document format, in a few years people will stop caring what Office suite you are running - just like I don't care about your brand of email client, I just assume you can read the emails I send you.

For that reason, Suns ownership of OpenOffice has been all for the good - they could probably not have justified the expense to their shareholders if they had just given it all up to a foundation - and we should all be very grateful. Now we have ODF, we have a working implementation (OO.o), and all FOSS developers can choose to work on their Office software of choice, as long as they are ODF-compliant, which will in the end lead to real choice for all of us that are mostly users.

As for Sun keeping OpenOffice or giving it to a foundation - who cares?

Who cares?

[xander26]

It seems to me that there is a double agenda here. This is what I hate the most about the open source community. They think all software should be open, regardless of what the developer of the code actually wants. Mr. Torvalds (I applauded btw, when he stuck with GPL v2) can release the Linux kernel under what ever license he wants to, since he's the one who developed it. Likewise, OOo is Sun's baby, and they should be able to do with it what they want. Now, before you all think I'm a troll I like open source. I like the fact that I can go into the code and figure out how it works. I even run Linux as my main desktop OS at home, and I use it to run my home-built media center and firewall. However, it should be up to who ever owns the code...and in this case, it's Sun. This is a free society right?

Give them a break...

[benmhall]

Wow. They release the best office suite for Linux/Unix, not to mention the only truly cross-platform office suite, under the LGPL and it still isn't enough. No, they have to set up a foundation.

Come on.

The only reason for the Mozilla Foundation was because AOL/Netscape wanted to stop bleeding money into a project that was giving them nothing. Sure, it's been a good move for Mozilla projects, but Sun's ongoing commitment to OpenOffice/StarOffice just shows how strongly they believe in the project. If OpenOffice was languishing, then it might be time to ask them to step aside and establish a Foundation. This is clearly not the case. Moreover, OpenOffice's license is very cut and dried. You can easily fork the project (as has been done with Workplace and, to a lesser extent, with NeoOffice.) Heck, you can even fork it and setup a foundation. Good luck replacing those Sun engineers, though.

OpenOffice is hugely important to Linux, Unix and Open Source in general. For the most part, I think that Sun has been doing an outstanding job with the code. Why ask them to further distance themselves from the project now? It's not like they're doing it a disservice. (As was arguably the case with Netscape/Mozilla.)

And another thing, people taking potshots at OOo 2's use of Java in Base should realize that this was yet another significant contribution to the project from Sun. Base, even with it's faults, works very well. In fact, it has already allowed me to use MySQL/OOo in place of Access at work. Sure it uses Java, but this was done for legitimate reasons with an eye on compatibility. The proof of this being that Red Hat et al. were so quickly able to port Base to the gjc. I highly doubt that the relative ease of this task was a coincidence.

Really, that this has come up at all is a true shame. I fear that it shows that even if Sun were to open source Java, people would still find some way to complain. Sun certainly does some strange things, but their contributions to OpenOffice have been nothing short of fantastic. They should be commended for Giving Microsoft their first real competition in years.

Re:should happen

[kimvette]

Uh, the code in Photoshop CS2 isn't there just to take up space, and contrary to what you may think there are no loops like:

while (count 15 seconds)
{
      waste time crunching randomized numbers here, pinning the CPU and thrashing the hard disk for no good purpose other than to annoy you
}

Every bit of code is there to provide features for graphic designers. I gave my art director three options, all licensed and installed on his PC:

  - Adobe CS2
  - The Gimp
  - Paint Shop Pro (since uninstalled and given to my brother for the occasional work he does for us)

He has run The Gimp exactly once just to check it out. He found the GUI counterintuitive (especially when multitasking - when bringing the document forward, the palettes stay below everything else. That pisses ME off every single time I use The Gimp as well). He hates The Gimp. The only reason I use The Gimp instead of Photoshop is that Photoshop for Linux is not (yet) available.

He has used Paint Shop Pro a couple of timezs just for a couple of filters it had that Photoshop didn't, but he has since implemented his own filters.

Sit a professional graphic designer in front of The Gimp and he'll ask if you're joking. I would have to agree, even though graphic design is not my primary focus. Effects which come free in CS2 in the form of layer effects take roughly 30 clicks and 8 to 10 passes of various filters in The Gimp to achieve similar results, and if you're not happy with the results in CS2, it's the tweaking of a setting or two to remove, modify, or reapply an effect, whereas it's having to repeat the same many steps in The Gimp, and you're not likely to ever get the same effect exactly the same multiple times because there are so many steps involved. When I have to design something for a client that isn't just image editing but requires design, I usually go to my designer's workstation and use Adobe CS2. I may be pro-Linux and Pro-OSS, but I definitely see and appreciate the value in commercial products. If Adobe CS2 is ever released for Linux, I'll definitely be one of the first to buy it. Well, I take that back - I'll wait for the first patch to be released.

There is definitely value in Adobe CS2, and the "bloat" as you put it does not go to waste. If you think it's waste, you should be choosing a lighter tool because you obviously don't need the functionality it provides, but I assure you it is not bloat. Photoshop is not and was never intended to be a lightweight tool. If lightweight is what you need, check out Paint Shop Pro, The Gimp, or maybe even Photoshop Elements.

Open Projects

[Doc Ruby]

If Sun releases control over the project, that will make it more likely that an independent project will split OO.o into a server and AJAX clients. And that another project will cluster the OO.o server. Then Web forms and GUIs can finally have the kinds of editing control we deserve in the 21st Century. Like automatic version control. Or anything that beats these slate-like TEXTAREAs.

Great way to discourage companies from OS

[Augusto]

I thought the goal was to get more companies to open source products?

Sun has spent a lot of money on this product, and now people are demanding that they give this up too? I don't get it. Wouldn't this discourage people from opening up their projects to OS?

You can already fork this and do your own thing right? There's no incentive for Sun to "give up" more control over this, unless they want to fire a bunch of developers Sun pays for to lower their costs.

Re:should happen

[richlv]

they do not relicense anything. actually, you could take oo.org code, add something and sell it as ohmygodoffice right now. there are companies/individuals that are/were doing this.

actually, having all copyright dually assigned allowed sun to drop sissl (that is, dual licensing) at launch of oo.org 2.0 an thus oo.org is licensed as lgpl only now.

http://www.desktoplinux.com/news/NS3294924491.html [desktoplinux.com]

there also is a faq linked from that article (and you probably could find a lot of info in oo.org mailing list archives / website and other places)
http://www.openoffice.org/FAQs/license-change.html [openoffice.org]

Sounds like the latest news ...

[hritcu]

... from the IBM P.R. department with the title: "IBM urges Sun to give up OpenOffice control"

The truth is that if a high enough percentage of the OpenOffice.org people wanted to break up with Sun nobody could stop them - they could just FORK. But wait ... more than half of the OpenOffice.org developers are Sun EMPLOYEES! So how could _OpenOffice.org_ want "Sun Microsystems to give up control over the OpenOffice productivity suite, and donate the intellectual property to an independent not-for-profit foundation" (quote from the article)? And what developer would consciously use a term like intellectual property [gnu.org] in the first place?

Re:Give them a break...

[hritcu]

Sun's contribution to open source is extremely valuable, and so it's IBM's. I really hate to see the two undermining each other's efforts. They should know better.

As for the slashdotters ... they will always find something to pick on, no matter what the subject is.

Re:It All Depends on Sun's Goals

[Anonymous Coward]

Much of the rapid development that was happening to bring KOffice along went quiet when OO.o was released, if I remember correctly.

Ahh... the famous KDE fallacy. Some nice screenshots and a bit of user interface code and a hype article on slashdot == finished app. It's been a common theme on slashdot for 5 years now.

Clue: KOffice slowed down because all the low-hanging fruit had been picked. It entered the phase of development in which all much of the fun easy pay-off stuff had already been done... and it moved in the phase that all software developers know about the "elbow-grease" phase, in which you must buckle down and get all the tough stuff done.