« previous 1 2 3 4 next »

The fun has gone

Meeks takes the view that OpenOffice.org should be developer driven, with less hands-on involvement from Sun staff. Successful developer driven projects, such as the Linux kernel, Samba or GNOME, tend to be open, democratic, noisy, argumentative, divisive, and chaotic, but are often highly creative and successful because they promote developer initiative and attract a greater number of developers. Organisations participate in such projects for selfish reasons, because it works, and because it brings twice the resources at half the price. Individuals participate for a multiplicity of reasons, and some, like Linus Torvalds, became involved purely for fun. "The most important design issue...", he wrote in 1995, "is that Linux is supposed to be fun..."

A characteristic of such projects is that they are often fiercely independent. The idealistic view of an open source project is that the IBM employee who is paid to contribute to the Linux kernel or Samba or GNOME is first a hacker, and second an employee of IBM. (This, of course, is more true of some projects than others). Almost as important as the code to the integrity of the project is the amount of noise and discussion on the mailing lists. The theory is that anyone can contribute to the whole, and become an equal partner in a world where the measure of your worth is the quality of your contribution, and while many of the long term contributors to projects such as the Linux kernel are now employed by outside interests, most continue to work just as they did before.

Meeks feels that the fun has gone out of OpenOffice.org development, precisely because of Sun's grip on the process, and that this is inhibiting the creative hubbub common to most open source projects, something that Novell has tried to replicate in its "freer" branch of OpenOffice.org at go-oo.org. The Sun project lacks the free and easy exchange of ideas, and has moved away from the "release early, release often" philosophy common to most free software projects, where bugs get tracked down quickly, and are dealt with on a first come, first served basis.

"It's certainly possible to cruise along talking about all the marketing advantages of end-user communities," Meeks says of OpenOffice.org, "but in the end-game, without a focus on developers, and making OO.o truly fair and fun to contribute to - any amount of spin will not end up selling a dying horse."

A philospohical impasse

This judgement may be harsh, and is countered by the likes of Thorsten Ziehm, who uses similar metrics to Meeks to show that 2008 was a uniquely successful year for OO.o, in which nearly 900 child work spaces were integrated into the code, more than 4300 issues (features, enhancements, bug fixes etc.) were dealt with, and more than 12750 issues were reported in IssueTracker, demonstrating a healthy user community. There have also been more than 28 million direct downloads of OpenOffice.org 3.0 since since its release in mid-October 2008.

Unlike the better known free software projects, an office suite can claim to have much greater reliance on, and vulnerability to, quality assurance (QA) issues, affecting usability, consistency and reliability. This, for Sun developers, is a key issue, which slows development but improves the code, according to Charles-H. Schulz, a French independent OO.oo contributor. "Since we're developing an end-user software suite we cannot tolerate leaving our software at a low level of quality. Of course, there are always bugs and we have ramped up our QA teams and resources significantly over time. QA gets to register the builds, test them at various levels according to the development, localisation and QA processes. It also approves and decides whether the builds should be released or not. QA and the QA project play a central role in our development and release process."

It follows that feedback and response times will tend to be slow, which is something that Meeks feels can be better handled by the traditional free software development process. From Meeks's perspective, Sun has imposed a proprietary development logic onto a free software project, nullifying the advantages of open source development, and creating a philosophical impasse. Those developers who have worked on heavily audited government projects will recognise the paradox - audit trails which are there to provide quality assurance often get in the way of developing quality code, which is why the best code in commercial environments often comes out of skunkworks projects - as was the case with James Gosling and Java.

[pagelink Calling a cat a dog]Next: Calling a cat a dog[/pagelink]

« previous 1 2 3 4 next »