Sun, StarOffice, Calc, and Forks

Thanks to Scott for tipping me off to this.

Here's  something interesting. The question: what happens when contributors and, I'll just say publishers, want different things?

The result seems to be that Openoffice.org, as Linux did, is going to have different versions. Perhaps not for the same reasons.

Kohei Yoshida wrote a long post on the history of Calc Solver, which is an optimization solver module for the Calc component of OpenOffice.org. After three years, they don't agree on some aspects of the licensing.  Now Michael Meeks has announced ooo-build (previously just for build fixes) is now a formal fork of OpenOffice  to be located at http://go-oo.org/.

The fork looks pretty interesting as it includes several things that have not make it into the official build.

Intrigue in the OpenOffice.org world.  In ten years will there be as many versions of OOo as there are of Linux?

---

---

Comments

[Frank Peters]

Solveig,

there is numerous noise now about this infamous case. Your representation appears to lack some context which I'd like to add:

> Should OpenOffice.org (specifically, Sun)
> refuse to include any code whose rights
> aren't totally surrendered?

Read the JCA (or SCA). You do not surrender your rights. You still remain owner and copyright holder of your code. You share it with Sun representing the community. The reasons for that were discussed exhaustively before (e.g. here: http://danesecooper.blogs.com/divablog/2005/04/index.html).

> Sun says it will duplicate the work
> because Kohei doesn't want to sign over
> ownership of the code.

The rules of engagement of the OOo project are clearly defined (and they have been for quite a while). One of them being the requirement to contribute under JCA. These are the project's rules. You cannot just come along, not follow the rules although you knew them beforehand and then start whining about the fact that your contribution is not accepted.

If you don't agree to the project rules, don't contribute or use the processes to change them.

> Because of Sun's refusal to accept
> LPGL extensions in the upstream code,
> Michael Meeks has announced...

Like many you are mixing up copyright and licensing issues. LGPL is perfectly fine but the license does not specify any copyright assignments which are legally required for a project to work efficiently, so you need a CA as well (like most other OS projects do). You seem to have read Novell propaganda on this only (both Meeks and Kohei work for Novell, yet this fact is not mentioned in your post). To get a different view: Simon Phipps replies on the issue here: http://blogs.sun.com/webmink/entry/sca_r_office.

This is an ugly example of some individual (or company) trying to instrumentalize a contribution case for their obvious political purpose.

Meeks et al are on an FOSS crusade, and that would be fine if they weren't to risk community split for the higher sake of their political belief. That's the pathetic thing about this, not the "what" but the "how".

Never mind
Frank

[Javier]

I can understand Sun point of view. For me it's more difficult to understand why Novell prefers LGPL instead of GPL for the fork of OOo. The only reason I could think it's that they want to hurt Sun. In the same way that Sun seems to not be happy with Novell, but they have OOo and Novell not. For me it's more valuable the opinion of the people behind NeoOffice when they say: "Why use the full GPL? Using the full GPL ensures that anyone who makes improvements to NeoOffice will share them with the community. Why is NeoOffice separate from OpenOffice.org? When we started the NeoOffice project, our primary reason for being separate was that we did not like OpenOffice.org's SISSL license. However, now that the NeoOffice project has been active for a few years, license issues are no longer the primary reason that keeps us separate. The primary reason that we stay separate is that we can develop, release, and support a native Mac OS X office suite with much less time and money than we could if we worked within the OpenOffice.org project. Why would running a separate project be so much more efficient than using OpenOffice.org's infrastructure? Basically, the reason is that OpenOffice.org's processes and infrastructure are designed to handle the tens (or maybe hundreds) of millions of users on three very different platforms. This is a huge job that requires a huge amount of infrastructure and processes. Furthermore, OpenOffice.org is paid for by Sun Microsystems so its paid staff are most concerned with completing whatever goals Sun Microsystems sets. In comparison, NeoOffice averages less than a million downloads per month and NeoOffice only runs on a platform that Sun Microsystems has rarely released software for. Because of these differences, any Mac OS X work must be coordinated with the OpenOffice.org paid staff to ensure that the Mac OS X work does not conflict with any work on the Windows, Linux, or Solaris platforms. Both Ed Peterlin and Patrick Luby have worked within the OpenOffice.org project in the past and, in our experience, this coordination requires a significant amount of time. Since we are volunteers, we have very limited time and such coordination can quickly use up most of it. By running a separate project, we have eliminated most of this coordination time and have used that time savings on things that are important to NeoOffice users such as frequent bug fixes and responsive support." http://www.neooffice.org/neojava/en/faq.php#10

[Solveig Haugland]

Hi Frank,

Thanks for your clarification, and I know my phrasing isn't all that accurate. I've edited the post to neutralize the language and focus more on my angle on this, I think, is that perhaps inevitably there are now a lot more versions than just OOo and StarOffice. There are bound to be various differences of opinion on how to license, include, package, etc. software in the open source world and it's starting to show. I guess it's like Linux; there's certainly no longer one Linux, and the same thing is likely to happen to OOo. Which is nice. I think one version we need to see is the easy-peasy "all but essential features stripped out" beginner version that would be perfect for a certain audience. And would blow MS Office out of the water in terms of the usability/feature tradeoff.

[Frank Peters]

Hi Solveig,

your certainly right in that OOo is growing up. This is a critical phase in the project where we will see how mature the community is to cope with these kind of issues. Like I said, it's not so much what is being discussed but the tone that bugs me.

Frank