Home > Standards Blog

Advanced Search 

Welcome to ConsortiumInfo.org
Thursday, June 29 2017 @ 10:45 AM CDT

The following comments are owned by whomever posted them. This site is not responsible for what they say.
Showdown in Geneva: OOXML Fails to Achieve Majority Approval at BRM
Authored by: Andy Updegrove on Sunday, March 02 2008 @ 07:15 AM CST
Inigo and Doug,

First, thank you both for commenting.  I'm finding that there is a difference of opinion among delegates as to what they thought they were indicating when they were voting on the final c. 900 dispositions.  This is probably not surprising under the circumstances.  Some of the explanations that I have heard, though (I just got another by email), I'm having a great deal of trouble understanding, as they seem illogical on their face.  I'm hoping that a lot of delegates will do what you're doing in commenting here so that people can try and sort it all out.

With that by way of background, here are some responses to your comments:

1.  Inigo, on the subject of the votes on specific dispositions: I agree that what you say makes a great deal of sense, but my bet is that it will be impossible to draw conclusions in all cases based on such an analysis.  Assuming for example that the figures that someone has posted above are accurate, how would one know the following:
-  If the only vote received was from an NB other than the NB that had lodged the comment, it seems that this vote would prevail.  What exactly would that tell you?

-  If only two votes were received on that disposition, one from the NB most interested and one vote from another NB,  and the two votes were not the same, then two would cancel out, if I understand the process correctly.  What exactly would that tell you?

-  If no one voted on a single disposition (even the original NB that made the comment that generated the proposed disposition), does that mean that it's a good one, or that no one had the time to really think it through, or couldn't decide?

-  If the only vote was a disapproval by the NB that asked for the change, because they like the proposed change even less, then as I understand it,  no change will be made at all.  Is that a good result?
And so on.  What concerns me is that, according to Frank Farance, those dispositions that were discussed were almost invariably changed, often quite a bit.  So while a disposition might have been "approved" by as few as one NB in addition to all of the abstains, I'm not sure how much to be comforted by that. 

I would certainly agree that if the one vote that was received, or if one of the votes that was included in the "approve" category, was from the NB that had asked for a change, this would certainly be an important point to be taken into account (but not if that NB voted to disapprove - see above, and that carried the day), then I would agree that the 900 vote process was a useful device.  So here's the analysis that I would pursue:
  • Of all of the dispositions out of the 900 that received individual votes, how many included votes by the NB(s) that had originally asked for the change?
  • Of  those, when the votes are counted, how many went the way that those NB(s) voted?
  • Of those, how many of those NB votes were to approve rather than disapprove?
  • Of the total of c. 900,  how many were voted on at all?
Does that sound like a rational way to do the analysis?  If so, what's your guess on the number that remains after the first three bullets, and also for the last bullet?

2.  Doug, I don't think I said that 80% were "protesting."  What I was saying, and as the blog entry title indicates, 80% of the proposed dispositions failed to gain majority approval.  I'm not sure how you can call an "abstention" anything other than that.  If someone chooses not only not to vote on an individual disposition, but to abstain on the rest, then it's hard for me to see this as anything other than "no comment." 

That said, I agree that it would be fair to say that (for instance) if a proposed disposition got several votes to approve (including the original NB proponent(s), then it should be deducted from the 80% figure.

3.  Doug, I talked to a lot of delegates, both at the wine and cheese events and the evening keynote sessions and at lunch as well as at the bar at the Intercontinental.  They were from a lot of delegations, and few were affiliates of IBM.  All agreed that everyone was trying very hard, and that it was all very frustrating.  I also agree that doing the 900 vote was a good idea, if the goal was to come out of the BRM having done _something_ with everything. 

But was that actually a good idea?  My personal view is that a better result would have been to have voted something along the lines of  "this is as much as we had time to do; no opinion is expressed on the rest," or perhaps to deal with the last 900 on a different basis (e.g.,  "the indications on these 900 should be regarded as advisory, rather than authoritative"). 

It seems to me that no one's interests are well served by doing so much in so little time.  If those that want to see OOXML go through are successful, then we will have to live with a standard that could have been much better if more time had been spent.  And if it fails, then those that wanted to see it go through, and those that tried hard to try and make that happen, will have seen that work to have been in vain. 

In that case, hopefully Ecma would still make the changes and Microsoft would implement them.  If that happens, something would be salvaged, but more could have been accomplished.

4.  Doug, on the P vs. O point:  I've included this mostly because I know that it's going to be brought up by others.  I'm not personally that interested in it, because I think that the big questions are these:
  • Is OOXML now good enough to approve, leaving aside other issues like IPR, politics and so on?
  • Should OOXML be approved, given the fact that such an inappropriate vehicle was chosen to process (the Fast Track process) a specification that required so much work?
I'm not competent to judge the first, so I'll defer to others on that.  But I think I have heard enough on the second point to have an opinion.  Many delegates told me that OOXML as it came from Ecma was not ready for prime time (the words usually used were "total garbage" or similar).  Many also told me that the process was crippled at every stage by inadequate time to do a good job.  Thus, the BRM was just the latest and most clearly problematic example of an effort to cram something through in a way that is not conducive to producing a good result. 

Just because there was a "result" does not mean that it was an adequate or useful result.  Just that a lot of good people worked very hard to do the best they could in a situation that they really shouldn't have been placed in to begin with.  My concern is that it be made clear to all what the results of the BRM really were.  Needless to say, I don't think that Jason Matusow's Unqualified Success blog entry is consistent with what happened.  To the extent that my blog entry isn't either, I welcome and appreciate your assistance in tuning it up.

Again, thank you both for commenting, and responses to the above are welcome.

  -  Andy
[ # ]