OOXML Needs Work

Saturday, 2007-September-08 at 12:47 3 comments

I recognize that Rick Jelliffe is an XML expert.  I, on the other hand, am not.  So I try to consider his opinions carefully, when I can even understand what he is saying.

In this case, his opinion is that OOXML has been thoroughly reviewed during the ISO-ization process, so the Ballot Resolution Meeting in a few months should be smooth sailing for DIS 29500.  He implies this in response to a post by Kurt Cagle, another expert XML guy who comments on O’Reilly’s XML.com blog.

In fact, many of the comments in the first phase were of the nature of “this is too long of a document to review in such a short period of time,” especially given that the specification needs so much revision before it is acceptable.

Let us take a look at some of his points and compare them with reality:

  • Because the draft has already been submitted to ISO, Ecma is not free to edit it in response to comments part-way through the process.

This ignores the fact that people were bringing these things up during the Ecma process as well.  Ecma’s mandate is to produce a standard that is compliant with Microsoft’s pre-existing implementation.  In other words, do not perform technical edits or alterations to the specification, but pretty it up so it can win approval.

  • OOXML has had a number of national standards bodies review it during this process (twice already, with a third chance coming up), therefore the review is not limited.

These are after-the-fact reviews, sort of like trying to discuss the recipe after the cake has already been baked.  Where review really counts is before the batter is stuck into the oven.  That is to say that after the fact review is mostly quality-checking of the review that went into the process beforehand.  On this account, Ecma International appears to be a “buy yourself a standard” group.  Not that Microsoft is at fault for this. 

Back in the days of Netscape, it was rumored that JavaScript went to the group then known as the European Computer Manufacturers Association (ECMA) for the same reason: little or no alteration of the pre-existing (Netscape) implementation was expected in the final result.  As I recall, there were some surprises in the output of that process.  It has been a long time, and I was quite far outside of it, so my memory could be faulty.

  • There is no evidence that Microsoft paid people to get on their national standards bodies and vote for approval of OOXML as-is or with non-binding comments.

Assuming that such a thing actually did happen, I would think Redmond was seriously incompetent if they did get caught.  The “Swedish mix-up” would be a good example of this, but standing alone, it is too easy to blame on one employee’s mistake. Thus, while I think it was a good idea for Rob Weir to ask if anyone would produce more evidence, I did not expect to find any, regardless of whether it was happening.

Unless one has legal powers to go into companies and seize documents and compel testimony on penalty of imprisonment, only a fool would expect to find much in the way of objective evidence (proof) of this to get out.  Instead, one must infer things from the external behavior and communications of the parties involved. This necessarily involves imputing motives and ethics (or lack thereof), as inference is subjective.

When Jason Matusow says that they will use all tactics at their disposal, it is clear that they have swayed into a political campaign, rather than a technical process. This makes it easy to imagine the same kind of activities that go into political campaigns happening in this process.

Let me put it this way. There is enough going on there that it smells bad, even if no one can see any external proof yet. More than one accounting professor has warned his students that a bad aroma means that one should look deeper and be cautious about signing off on management’s representations. The scandals we have seen are largely a result of ignoring that advice.

As for M. David’s claim that this was expected, I have to disagree. You don’t spend that much effort packing the court in order to get a negative result. And you don’t have to write a spin-filled press release in order to camouflage an expected result in new clothes. In other words, Microsoft’s behavior tells me that this was not the result that they expected to get.

Let us remember that "yes with comments" and "no with comments" can both be changed at the BRM. We shall see if Microsoft and Ecma are willing and able to make the changes needed, or whether they will continue to play the same games they did after the contradictions period.

I think a some people have the idea that this is about being anti-Microsoft. I think it is more about being pro-user and pro-buyer. The object, in my eyes, is to take the focus off of any particular product or vendor and put it back on the user. This should contribute to ending the rampant stagnation that office suites have been in since WordPerfect fell from grace.

Blogged with Flock

Tags: ,

Entry filed under: ODF, Open Standards.

Office Not-so-open XML Loses This Round [Reprint] Microsoft’s Ugly Daughter


  • 1. Darius Damalakas  |  Friday, 2007-November-23 at 04:13

    pro-user 😀

  • 2. Darius Damalakas  |  Friday, 2007-November-23 at 04:15

    Personal message to author:

    Why do you have comment moderation ON?

    It discourages people to post comments, even though your topics are very interesting.

  • 3. lnxwalt  |  Friday, 2007-November-23 at 18:15

    Comment moderation is on because of spam.

    * In almost a year of having this blog, there have been about 47,000 visits, and about 19,000 spam messages caught by Akismet. Several more got past Akismet and would have been cluttering up the site.

RSS Slingshot

  • An error has occurred; the feed is probably down. Try again later.

RSS Unknown Feed

  • An error has occurred; the feed is probably down. Try again later.

RSS Unknown Feed

  • An error has occurred; the feed is probably down. Try again later.

RSS Owner Managed Business

  • An error has occurred; the feed is probably down. Try again later.


Recent Posts

Blog Stats

  • 598,611 hits

Top Clicks

  • None


%d bloggers like this: