OOXML and the future of open standards

The International Conference Centre in Geneva has just played host to a critical meeting of standards bodies, a group that Microsoft has been legitimately lobbying in the hope of getting its Office Open XML (OOXML) format ratified as a standard.

At the same time, the UK-based OpenForum Europe group hosted a complementary conference in the same building to discuss "Standards and the future of the internet". The focal interest of the conference was not the OOXML decision, but nevertheless, interest was inevitably diverted by events upstairs. OOXML, and the process by which it has reached its current impasse, is identified by many as a turning point in the debate about open standards.

OpenForum's conference featured keynote speeches from Vint Cerf, who is often dubbed "the father of the internet" for his role in developing ARPAnet, and shared the common concern that "the standardisation process has become distorted out of proprietary interests"; from Hakon Wien Lie, who shared a room with Tim Berners-Lee at CERN, coincidently based in Geneva, and came up with the concept of Cascading Style Sheets, later becoming the chief technology officer of Opera Software, creator of the world's best selling web browser "because the others are given away free."; and from Andy Updegrove and Bob Sutor, who produce the most read standards blogs on the net, and are among the leading experts on the successes and failures of the standards process.

Gaming of the system

Interestingly, many of the delegates of the "other meeting" attended the keynotes and some presentations, presumably out of a common interest in the meaning and definition of "standards", but were the soul of discretion when it came to discussing events in the other place. The structure of the conference and the tone of its presentations was discursive, not didactic, and there was a genuine interest in finding resolutions to the issues that beset standards in the computing industry. Standards affect innovation, barriers to entry, interoperability and the neutrality of data, and are fundamental to the future of computing. Technology moves at speed and the idea that protocols, APIs and data formats are "trade secrets" can be viewed as regressive, and an impediment to the transmission of ideas.

Equally it is recognised that technical standards are evolutionary and should not be cast in stone. Hence, Cerf's observation that standards can only be considered open "if there is non-proprietary opportunity for interested parties to contribute to their evolution, and demonstration that independent implementations have been made and shown to interwork. Publishing of the specifications for protocols does not automatically confer openness."

There was a common feeling among those present at the conference that OOXML represents a crisis in the standards process, heightened by widespread allegations of "ballot stuffing", "bribery", and "gaming" of the system which have been detrimental to the reputation of the ISO. Inevitably, OpenForum's conference became in part a protest against the proposed adoption of what is perceived to be a proprietary standard, and in part a positive statement that the adoption of open standards is a necessity, and not just an option, for the future health of the computing industry. Either way, participants could not disguise their interest in the decisions being made upstairs.