Wednesday, July 18, 2007

OOXML - The Votes are In!

On Friday, July 13th, INCITS V1 took a role-call vote on our recommended position regarding DIS-29500, Office-Open XML. In what I imagine as a strictly partisan vote, we were not able to arrive at any consensus (a 2/3 majority). This, in spite of extending the time limit and even a certain amount of bullying pressure (or so it seemed to me) from some of the voting members. And the partisan division pretty much seemed to be Microsoft and their business partners, and the rest of the membership.

The motions we entertained were a) a recommendation of "Yes, with comments", which means we recommend acceptance of DIS-20500, and have comments that should, but are not required, to be addressed; "No, with comments", which means we recommend acceptance of DIS-29500 only if the comments are all addressed, and "Abstain, with comments" -- and I never did clearly understand that one! We did not entertain a straight-up yes or no vote.

I'm a little surprised by the outcome. Not so much that we could not agree on either a yes or a no, but that we could not reach a compromise position on abstaining. It seemed to me that the members voting yes were completely unwilling to meet in this middle position, voting down the abstain position along the same party lines. This unwillingness to compromise reinforces (to me, anyway) the feeling of being bullied -- the pro-OOXML membership essentially saying "my way or the highway".

It is of interest that in the past few months, 19 new members have joined this committee. And of those 19, 14 voted "yes". Certainly from my viewpoint, this looks like ballot-box stuffing.

Monday, July 9, 2007

OOXML -- What's Happening Here?

Lucky me, I'm a voting member of of INCITS-V1 -- the sleepy standards group that has been working on such arcane things as TopicMaps. We're now working through the 6k+ pages of DIS29500 -- Microsoft's proposed standard for Office Open XML (OOXML). And my first question is, why?

Now, I'm not invested in or in any way partisan to the so-called-competing Open Document Format (ODF) that is already a standard and is in the process of getting a new version standardized. ODF is a part of the open-office software suite. I only just recently learned that it existed at all.

And, by the way, I don't see this (as some apparently do) as a Microsoft v. IBM fight, though I do see it as a fight of some kind. For me, a "standard" should serve a couple of purposes: to promote interoperability and to enhance portability being chief among these.

So, why do we need two competing open document standards? The answer from the proponents of OOXML is that there is no way to augment ODF (the extant standard) to include the features of OOXML. This is partly because OOXML supports the Microsoft Office binary formats (from Office 97 forward, I think) and partly because "the process model is different".

I'm trying hard to understand this. If OOXML can only be implemented with knowledge of the internals of various versions Microsoft's office serialization formats, how does that promote either portability or interoperability, except, maybe, within Microsoft's product lines. And I just don't get the "different process models" response -- I would have though that a standard, especially an XML-based standard, would be above that.

To me at least, the issue is more subtle. If we (V1) approve OOXML as a standard and it fails to pass the litmus test of promoting portability and interoperability, I think we weaken our standards process as a whole, making standards less trustworthy. Kind of like what's happened to Wikipedia; a phenomenal web resource that has been co-opted by corporate marketing efforts (try searching for "Windows").