OOXML Fails to Get Majority Approval at BRM (Groklaw)

In addition, only a few tens of the thousand or so comments were actually addressed at the
five day meeting, the rest were all batched up together for a single vote "as is".  A
perversion of process indeed, but Microsoft now crows about how an overwhelming majority the
comments were accepted without comment.  As if there was ever a chance.

It's more complicated than this simple analysis. I spent half of monday reading about this, so
let me try and save some of you some time :-) I have no particular expertise in the ISO
process or XML document standards, and am not involved at all except being an interested
observer. All the info below is gleaned from reports and comments published since the event,
many by participants. Some of those reports are utterly contradictory, and below is the best
sense I can make of it all.

Firstly remember that the BRM votes are not about accepting the standard; they are about
whether to accept amendments to it. Those amendments are new text proposed by ECMA in response
to the many comments received as part of the initial september vote on whether to accept the
ECMA standard as an ISO standard 'as -is'. It was not accepted because many countries voted
'No, with comments' - i.e. We don't accept it as is, but we might if these things are fixed'.
There were thousands of comments/objections, merged down into about 1100 for this BRM.

It was always clear that discussing 1100 (or 900) items in 5 days with 30 countries was never
going to allow proper consideration of all the comments, but the process was fixed by the ISO
rules. The problem was really caused by trying to fast-track a huge, contentious standard, and
that's a problem caused by ECMA(/Microsoft) for attempting this. Discussion of that aspect is
'politics' as opposed to 'technology standards', so I'll skip that.

The convener and all the participants did their best under these unsatisfactory circumstances.
200-odd comments were minor editorial items and were dealt with quickly. 30 of the most
substantive comments were discussed (countries got to choose which points to bring up first so
one presumes they did the ones they cared about most) - those comments nearly all got modified
before being accepted into the text. When it was clear that little time was left the remaining
900 were just voted on in their existing form. A delegation could vote on each individual one
and set a 'default' vote for any they didn't express an opinion on (This is the overall vote
for which numbers are given above). We don't have good info yet on how many were explicitly
voted on verses what the default votes were, (although there are some figures floating about
in blog comments). Whether the default applied to 5 or 900 items matters significantly when
considering the importance of the default vote.

This process was agreed by those present at the BRM, although some were not happy about it.
Proper standards people are not impressed because the whole point is that a standard should be
properly scrutinised and discussed, not voted on 'this text or nothing' in a tearing hurry.

So, the  net result seems to be that nearly all of the suggested changes have gone through, by
a narrow margin. Whether this is a good thing or not depends on your point of view. The
changes generally make the standard better, given that they are attempts to fix objection to
it, but obviously as most had no discussion they may well be less than ideal, or even
contradictory. And of course if you don't believe this _should_ be a ISO standard then having
these improvements made gives countries which voted 'no, with comments' in September the
chance to change their vote to 'Yes' next month, and that's bad.

The microsofties spinning this as a huge love-in and massive consensus success astonish me. So
far as I can tell that's pretty close to bare-faced lies, but that Jason Matusow guy really
does give the impression he actually believes it. They live in a strange world.