I attended the SABS technical committee's meeting today, which was tasked with coming up with SABS position vis-a-vis recommending OOXML as an ISO standard. In case you are waiting with baited breath for the outcome, after much debate, by a vote of 13 to 4 it was decided to "disapprove with recommendations".
An unusual fast-tracked process
This is the first such technical committee I have participated in so it was quiet a learning experience for me. The meeting started with presentations for and against OOXML as a standard. Of the 6 presentation 5 were "against" and 1 was "for". No prize for guessing who was presenting the "for" argument.
Microsoft bought out a heavy hitter from corporate in the US to assist with their presentation, the essence of which was that there was nothing unusual about a 6000 page odd specification which has been ratified in less than a year by ECMA and which they want to have ratified as an ISO standard in 18 months. As the IBM representative, also a heavy hitter from head office, pointed out the process is highly unusual. Most standards take 4-5 years to produce and are, on average only 32 pages long. (The ODF standard is 1000 pages long.) Why there is a rush to get OOXML ratified as a standard was not answered, well not out loud anyway.
Microsoft Office document's binary format available for free!
An astounding claim made by the US representative was that the binary specifications for office documents are available for free from Microsoft for anyone to use. We naturally expressed our amazement at hearing this news. He promised to email it to all of us in the room. I am waiting in anticipation for the email. The accompanying license agreement is only a mere 3 pages long he said.
Developers want technologies that reuse existing standards!
When asked why they didn't build on existing standards such as XForm and MathML the claim was put forward that they were inadequate for OOXML's needs. No one asked why they don't help "improve" the existing standard to meet their needs instead of going and making their own. From a developer point of view this is annoying. Why must I relearn how to do something when there is already a recognised way to do it?
A long road ahead
The arguments against OOXML were numerous and are already documented on groklaw so I won't go into them here. Suffice to say it centered around the concerns of vendor lockin, patents and a whole host of technical deficiencies.
A last minute "legal" argument was put forward, by whom I assume is the Microsoft legal representative, as to the legality of the meeting. By this time it was obvious which way the vote was going to go.
Although this vote is a victory for open standards and transparent processes I am under no illusion that the process is over. I am sure there will be another vote and next time Microsoft will be more prepared.