OpenOffice – splits and pirouettes
by Richard Hillesley
The splits and controversies around LibreOffice and OpenOffice.org have highlighted a number of issues concerning the licensing and corporate governance of open source projects and communities.
Successful free and open source software projects such as Linux, Apache or Samba have little or no ambivalence concerning the licensing and ownership of the code or the independence of the developers. Large companies contribute to the code, and have much to gain from doing so, but the status of the developers and their relationship to the code and the governance of the project is clear and unambiguous – even though the intent and effect of the licence may vary from project to project.
OpenOffice.org never had this luxury. The project was micro-managed in-house by the original developers of StarOffice, despite the early promise of an independent non-profit foundation to run the project. The code was released under a copyleft licence but copyright was assigned to Sun, who relicensed the code to IBM under terms that have never been disclosed. A confused inheritance was further complicated by the takeover of Sun by Oracle, who had no interest in the project or the community.
On the face of it, OpenOffice.org was a success. Downloads ran into the tens of millions, but progress was slow and OpenOffice.org never fulfilled its potential or the expectations of its contributors.
So it came as no surprise when, in the wake of Oracle's purchase of Sun, the community jumped ship to form a slimmed down developer-driven alternative, LibreOffice, where the contribution of code is simple and fast, "no CWS, no hours of tagging, paperwork, no specification, no hassle", and the code is owned by everyone and no-one.
"Ten years after Sun's original promise of independence for the OpenOffice.org community", the contributors promised "we are going to create the non-profit foundation around OpenOffice that they promised on day one" – only to be sidestepped a few months later by IBM, and by Oracle, who released its grip on the code and took it into another community under another licence.
The Bill Joy font
Unusually for an open source project, OpenOffice.org had its roots in a reasonably successful proprietary office "productivity" suite, StarOffice, which was marketed by a German company, StarDivision. StarDivision was purchased by Sun Microsystems in August 1999.
An office suite was not an obvious fit for Sun's business. Sun was a hardware company. Its primary business was back room servers, and two or three years earlier had even banished the use of PowerPoint from its offices. "We had 12.9 gigabytes of PowerPoint slides on our network," Scott McNealy told the San Jose Mercury in August, 1997. "And I thought, 'What a huge waste of corporate productivity'. So we banned it. And we've had three unbelievable record-breaking fiscal quarters since."
McNealy had a cheaper and more productive alternative, which was to "give everybody plastic Mylar sheets and all the pens they need to scribble on them" and to use what he described as "the Bill Joy font. You can see where he licked his thumb and erases. It's so much faster," and leaves you time to get on with the job. (One of many ironies in this story is that the man credited with the original concept and inspiration for PowerPoint, Whitfield Diffie, later became one of the better known employees of Sun Microsystems – more famous for his adventures in the world of public key cryptography than his role in devising a program to facilitate slide presentations while working at Bell-Northern Research (BNR).)
Many commentators assumed that StarOffice was bought as a stick with which to beat Microsoft. StarOffice ran on Linux and Windows and on Sparc workstations, and Office was Microsoft's primary source of revenue. Giving an office suite away free would undermine the Microsoft hegemony on the desktop, and like Sun's sponsorship of other open source projects, StarOffice and OpenOffice.org could be seen as part of a longer term strategy to push UNIX and Linux beyond the data centre.
Simon Phipps, who was Sun's chief open source officer, gave another explanation. "The number one reason why Sun bought StarDivision in 1999", he told LUGradio, "was because, at the time, Sun had something approaching forty-two thousand employees. Pretty much every one of them had to have both a Unix workstation and a Windows laptop. And it was cheaper to go buy a company that could make a Solaris and Linux desktop productivity suite than it was to buy forty-two thousand licences from Microsoft."
An inner oligarchy
StarOffice was a loss leader and a spoiler, but was never seen as a revenue earner by Sun. StarOffice 5.2 was made available free of charge, and within six months Sun had announced that the code would be released to the open source community as OpenOffice.org. The intention was to create a non-profit foundation and a community around the code, and the code would take care of itself.
The code for OpenOffice was originally licensed jointly under the LGPL and the Sun Industry Standards Source License (SISSL). The SISSL was a liberal licence which permitted re-use of the code in proprietary products by third parties. The first release was in May 2001. The software garnered mixed reviews, and development was slow.
Critics laid the source of OpenOffice.org's shortcomings at Sun's door, blaming conflicts in the licensing, slow and cumbersome feedback procedures, lack of transparency, slow response to bug reports, and the reluctance of StarDivision developers to let go of the code. Andrew Brown put it more succinctly: "The distinction between Sun as the inner oligarchy of producers, and everyone else as more or less favoured consumers simply won't hold. The balance, difficult and unpleasant though it clearly is for Sun to understand, is not between professional engineers and happy, laughing amateurs with their intoxicating natural rhythms, but between the Sun-salaried workforce and their competitors from the real world, at IBM and the various Linux companies. The end-users are wholly irrelevant."
Much can be blamed on Sun's failure to create the developer driven non-profit foundation it promised at the outset – "a non-profit organization that will oversee the operations, technology strategy, incorporation of technology contributions, and establishment of standards in conjunction with other standards bodies and open source projects as appropriate."