If This Was a Month Ago, OOXML Would Be Over

Andy Updegrove writes "Public announcements of how Participating members of ISO have voted on OOXML are now rolling in one at a time, and the trend thus far is meaningfully weighted towards 'No with comments.' By my count, there are now four announced Yes votes, with comments, two abstentions, and seven public No with comments votes for OOXML in ISO/IEC JT1. Korea has reportedly voted no as well, and I expect at least Canada, Japan and the United Kingdom to announce 'No with comments' today or tomorrow. There will be more no votes on the roster when the final results are announced in a day or two. But even if the 11 votes I know of now were the only votes, the vote would now have failed — but for the 11 countries that upgraded their status from Observer to Participating member status in the last few weeks. Without those extra 11 'P' countries, it would only require 10 votes to block OOXML from immediate approval. If most or all of those additional 'P' members vote 'yes' as expected, it will confirm suspicions that Microsoft has promoted extra votes in favor of OOXML not only within National Bodies, but within ISO itself."

Comments

ESR and OOXML

[mmurphy000]

Eric S. Raymond (ESR) commented on Microsoft's OOXML tactics as they relate to their proposed open source licenses in a OSI blog entry.

I agree pretty much with his position. If the playing field were anything near to level, I would have no issue in evaluating Microsoft's license submissions purely on their merits, just like any other license. However, I have difficulty in reconciling Microsoft wanting to be treated fairly by OSI with Microsoft's tactics in their attempts to ram OOXML through ISO. If Microsoft can game the ISO approval process, shouldn't it be fair for us to game the OSI approval process?

Re:Help me out

[Anonymous Coward]

Yes, the collective hive-mind of /. does care and in the latest newsletter there was the quote "we want this voted down!".

You see, a few years ago governments all around the world started realizing that when they send ".doc" files to the public they're asking people to go spend money with a particular company to read that file. Governments shouldn't say "People with FIRESTONE tyres get to stay on the road!" ...or.. "People with Microsoft Office can talk to the government!". So there's been a raising of consciousness around how file formats cost countries tens or hundreds of millions of dollars each year.

What Governments should do is say "People whose cars pass certain tests can stay on the road!" or "People with an Office Suite that uses a published standard can talk to us!". That way it encourages competition, "innovation", and cut-throat pricing.

Microsoft could tell where the wind was blowing, and they began trying to get the International Standards Organisation (ISO) to rubber-stamp their 6000 page proposed standard... a standard called OOXML. An Open Standard sounded like a great idea but the question was: Had Microsoft really told everyone their secret mix of herbs and spices? Well, no, because as it turned out many things in OOXML were left undefined and the only vendor capable of implementing OOXML was Microsoft.

(and even they're having problems ... let alone the problems other vendors have)

Now although ISO haven't announced anything it looks like it's going to go "No" for Microsoft.

This doesn't affect what software individuals or the private sector choose, but people who should use standards (government and government vendors) do care about this decision. Actually, individuals and the private sector probably should care because more competition in the office suite market may lower the cost of Microsoft Office.

A country's "no" can turn into a "yes" when an issue is addressed at the ballot resolution meeting (I think) so the more "no"s the better because otherwise a single country could just swing it in favour of OOXML. The more "no"s the larger the safety net, so it'll be interesting to see what the final vote is.

So I'd expect that in the coming days there'll be a lot of analysis of whether the actual comments in the "No, with comments" from each country are fundamental problems or superficial quirks. Can any particular country be swung to vote yes easily?

Still, it's a great start. The noooxml crowd are predicting 18 "no"s.

Re:OOXML has failed, but it isn't over.

[darkonc]

From what I can tell, it started with Massachusetts's decision to limit their next office product purcase to products which support "Open" and "XML-based" Document protocols. At that time, Open Document Format (which was based on the Open Office XML document format) was the only open document format available. This meant that Microsoft would either have to bow to using a format defined and controlled by somebody else, or somehow make their own closed format into something acceptable by Microsoft.

They went about this by following two tracks:

• On one track, they made life unbearable for the CIO at Massachusetts -- and also his successor, who also understood the issues at hand ... until they finally ended up with a CIO who would swallow.
• On the other track they suddenly decided to submit their XMLish document format to ECMA to make it an 'open' standard. ... but huge chunks of the 'XML-standard' were actually embeded binary chunks who's documentation consisted of "Uhm, reverse engineer Office-97 to figure this out, OK?". Even with these huge chunks effectively MIA, their 'standard' still consisted of over 6000 pages -- in part, because OOXML eschews international standards in favour of entrenching 20 years of MS bugs (like thinking that 1900 was a leap year, and encoding language types in wonky ways)

  More problmatically, Microsoft made supporting some of these odities 'optional', which means that

  1. You could end up with a nominally OOXML implementation which didn't support these chunks,
  2. Implementations of these chunks wouldn't be protected by Microsoft's patent grants for the ECMA standard,
  3. Microsoft could (and did) then promote these undocumented and unprotected features as 'critical aspects' of OOXML -- that any competitors who wanted to produce a competing OOXML implementation have a hard time (both technically and legally) implementing.

In short, OOXML allows microsoft to claim (to technically illiterate political cheque-signers) that Office 2007 uses an Open, XML-based document format -- but do it in a way that pretty much ensures them that nobody that's not a Microsoft lap-dog will be able to legally create an implementation that can actually read most Office 2007 documents. (or -- at the very least -- it will take them years to put something legal together, by which time Ofice 2007 will have squeezed Open Office out of the market).