Off the Beat: Bruce Byfield's Blog

October 13, 2010 marks the tenth anniversary of the OpenOffice.org project. It's a significant landmark, both for me personally and for free software in general.

For me, OpenOffice.org was the wedge I used to find a niche in computer journalism. In the early years of the millennium, few people were writing about OpenOffice.org. Almost by accident, I started comparing it with Microsoft Office, and writing How-Tos in my spare time. Before I knew it, I was writing full-time. For a while, I was worried that I would be too closely identified with OpenOffice.org to sell stories on any other subject, but, with that worry long behind me, today I can thank (or blame) OpenOffice.org for what I am today.

For the community at large, the importance of OpenOffice.org is hard to over-estimate. Ten years ago, despite all the efforts to market it, GNU/ Linux was a developers' operating system. KOffice was in its infancy, and AbiWord a glorified text editor. Applixware was proprietary and still basic. StarOffice was available for download, but not free software. People were questioning whether GNU/Linux was ready for average users, or ever could be. The lack of a fully-featured office suite might not matter much to developers, but it definitely mattered to GNU/Linux's long-term acceptance.

Then, after months of alternating silence and waffling, Sun Microsystems announced on 19 July, 2000, that it would be release the StarOffice code. At the O'Reilly Open Source Convention in Monterey that year, Sun was giving out T-shirts with an OpenOffice.org logo on the back and "Freedom" on the left chest. It sounds dramatic now, but at least once at the convention, Sun employees were throwing T-shirts to the crowd as fast as they could, and their booth seemed about to be overrun. Suddenly, the free desktop had an office suite and had taken a giant leap forward in credibility.

After that initial excitement, the actual start of the program was a disappointment. In order to release the code, Sun had removed large chunks of proprietary code, leaving the remainder gutted. In particular, the printing sub-system and grammar checker were removed. Another nineteen months would pass before OpenOffice.org would be released in a form that was actually usable, and, at times it seemed that moment would never arrive.

But it did arrive, of course, and OpenOffice.org soon became a standard part of the free desktop. If it hadn't, I don't think that free software would be anywhere near where it is now. Instead, with the last release of KOffice, free software would just now be getting an office suite good enough to make switching to it seem credible.

Not that OpenOffice.org has fulfilled anything like its complete potential. Under the guidance of Sun Microsystems, the project noticeably failed to develop an active community. Instead, the fact that most design decisions were made internally in Sun has drove dozens of enthusiastic volunteers away. Only in a few areas like marketing, documentation or localization has OpenOffice.org ever managed to create even a semblance of a community -- and, when it has, the resulting community has often been insular, temporary, or both.

Perhaps as a result, OpenOffice.org as an application as made very slow advances. Notoriously, its code is so cryptic that it has been easier to build a framework for extensions rather than undertaking any serious rewriting. Despite some improvements, its interface remains inconsistent and stuck in the mid-1990s, as some critics are quick to point out. Moreover, such publicity as the office suite receives tends to emphasize it as merely a substitute for MS Office, even though in some aspects -- notably the word processor -- it is actually more advanced than its ubiquitous proprietary rival.

Yet somehow, despite these failures, OpenOffice.org remains a standard part of the free desktop. It may not a perfect office suite, but it is the main office suite that the free desktop has, and, for all its flaws, it is more than adequate for any user willing to spend the time to learn it properly.

Where OpenOffice.org will go in its next decade remains uncertain. Many people feel that, with Oracle's purchase of Sun and all its assets, OpenOffice.org has fallen under the control of a company that makes the closed-shop Sun seem like a paragon of democracy. Yet so far it remains uncertain whether the recently announced LibreOffice spinoff and its governing Document Foundation can keep development even at the level that Sun managed -- much less pick up the pace or institute the thorough changes that are needed.

Fortunately, proprietary office suites are initiating changes just as slowly as the OpenOffice.org is likely to do under either Oracle or The Document Foundation. Under these circumstances, the OpenOffice.org code seems likely to retain its importance for some years to come --not because it is the best of all possible code, but because it the code that free software has.

Meanwhle, on the persoal side, I can only hope that I will be there to document the coming changes the way I was to record their beginnings.

Comments