Andy Updegrove has the results in detail here, including a breakdown of the votes. Basically, there were too many proposed changes to be able to cover them in the BRM, so they tried a workaround, but the upshot is ... it's a mess. Oddly, despite the rules, Alex Brown, Updegrove reports, allowed non P countries to vote, but OOXML still couldn't get a majority of the delegations to back it at the BRM. Nor is it clear that allowing non P countries to vote is even legitimate. Now it's the 30-day voting period, but Updegrove asks, if they never could discuss all the issues, which is the purpose of a BRM, what's the basis for a vote? And with the vast majority either voting to abstain or even refusing to vote as a protest, I think one may conclude this proposal didn't belong on the fast track, and it isn't getting the kind of support you would have thought it might, given all the muscle that has gone into the push to get OOXML approved.
Here's what happened next in this unusual situation, after they realized it was humanly impossible to cover all the comments and try to reach consensus:
Acknowledging the impossibility of achieving the stated goal of a BRM (e.g, to carefully review each proposed disposition and reach consensus on an appropriate resolution), a proposal was made on Wednesday to approve all proposed resolutions in a single vote before the end of the BRM, thus nominally "resolving" each remaining proposed disposition without any discussion at all. It was agreed that this was the only available option, and a written ballot with all of the c. 900 proposed dispositions that had never been discussed was accordingly issued on Thursday. Each National Body delegation was requested to complete and return on Friday. The alternatives offered were as follows:
1. Indicate "adopt," "disapprove" or "abstain" after each proposed disposition.
2. Indicate such a vote on as many proposed dispositions as desired (or none), and vote "accept," "reject" or "abstain" on all of the rest.
It is significant to note that voting to accept all dispositions that were not discussed is a less obvious choice than might be assumed. In fact, few if any of the dispositions that were individually discussed and voted upon during the week were adopted without change. In other words, adopting a proposed resolution without discussion could result in making OOXML worse, rather than better, because of dependencies.
On Friday, the ballots came back.
There were 25 P countries, and 32 voting if you add in the non P countries. The breakdown for the P countries worked like this:
Approve = 4
Disapprove = 4
Refuse to Vote = 2
Abstain = 15
Even if you add in the non P countries, which may or may not be legitimate under Directive 9.1.4, Updegrove tells us, it goes like this:
Approve = 6
Disapprove = 4
Refuse to Vote = 4
Abstain = 18
However, only approve and disapprove votes are supposedly going to count, so I'm sure you can figure out what Microsoft will spin from the numbers. They acted like the first vote was some kind of a victory, if you recall, back in September of 2007.
What next? I mean that in every possible sense. If they haven't discussed so many dispositions, how can they meaningfully vote on whether OOXML should be an ISO standard in the 30-day voting period left in this weird process? Updegrove concludes:
Many, many, people around the world have trie[d] very hard to make the OOXML adoption process work. It is very unfortunate that they were put to this predictably unsuccessful result through the self-interest of a single vendor taking advantage of a permissive process that was never intended to be abused in this fashion. It would be highly inappropriate to compound this error by approving a clearly unfinished specification in the voting period ahead.
That would be harmonious with the directions in the FAQ Alex Brown directed all the NBs and the public to read prior to the BRM:
4.5 What if there is not time in the meeting to satisfy NBs’ concerns?
If NBs find the outcome of the BRM inadequate then their recourse is to disapprove the DIS.
That's clear enough.
I have a question: were countries told in advance that only approve and disapprove responses would count in this workaround vote? I'd very much like to see the instructions given, wouldn't you? I thought it was supposed to be only approved if by consensus, technical point by point. How do you get that without discussions? I've never seen anything like this process, and I hope never to see anything like it again.
InfoWorld's Peter Sayer caught up with Vint Cerf, and in the article, Does OOXML vote really matter?, Cerf points out that if interoperability is really your goal, duplicate formats will not help you get there:
One who thinks the vote does matter is Vint Cerf, co-designer of the TCP/IP technology that underpin the Internet, and now chief Internet evangelist at Google.
"If OOXML is adopted, it leads to a problem of duplicate formats for document exchange," he said.
That duplication is bad for interoperability: In the Internet world, standards makers work hard on agreeing "one way to do things, and then evolving it," he said. "We don't reinvent the wheel."
You will find Bob Sutor's Some metrics for measuring the success of the OOXML BRM, his recent blog post, relevant in trying to put everything into some kind of context.
Update: You will not believe how Microsoft's Brian Jones describes it:
There were many technical changes the delegates made to really get consensus on some of the more challenging issues, but all of these passed overwhelmingly once they were updated. The process really worked (it was very cool).
The meeting closed with clapping and cheering, folks were really happy about the improved proposals for the specification and it was a very positive experience for me personally.
Is he kidding? Overwhelming approval? I'm sorry, but 4 out of 25 P countries is not overwhelming approval. Neither is 6 out of 32.
Update 2: Tim Bray has now written about the BRM also. [Language warning.] He also writes "the process is irretrievably broken".
And here's Reuters' account:
A meeting to hammer out a consensus on whether a Microsoft document format should become an international standard descended into near chaos this week, people close to the meeting told Reuters....
The Free Software Foundation's Coughlan said the difficulties of resolving the issue within the ISO framework illustrated a growing awareness that the implications of such decisions went far beyond the software industry.
"In the last few years, we've seen a lot more awareness of the digital world as a key aspect of society ... We're not just looking at the traditional economic powers asking questions or being involved. This is very broad," he said. "The standardization process itself has been tested quite extensively by this."
And here's ComputerWorld's Eric Lai's report, "Weak ISO support for changes to Open XML throws shadow over final approval":
"Eighty percent of the changes were not discussed," said Frank Farance, head of the U.S. delegation to this week's ballot resolution meeting (BRM) in ISO, which voted against the changes. "It's like if you had a massive software project and 80% of it was not run through QA.
"It's a big problem," Farance continued. "I've never seen anything like this, and I've been doing this for 25 years."...
Four countries, including the U.S. and Malaysia, according to Farance, voted not to approve those 900 changes.
That, according to critics, indicates a lack of actual support for Open XML....Microsoft hopes that enough countries will change their minds in the next 30 days because of the BRM result.
Another quotation from Farance, in CIO:
"Virtually every comment we processed did not survive unedited," he said.
The 80 percent of comments that were not discussed during the meeting were put to a "default vote," resulting in the automatic adoption of ECMA's recommendations without modification by delegates, he said.
And there you have it. ECMA Fast Track. The real question is, whose idea was it to do the "default vote" and were NBs asked to vote if the Ecma proposals were "improved" or "not improved"? Or "approved" or "disapproved" or was there a melding of the two, resulting in confusion and chaos and a default that in no way reflected true feelings? For example, if something was improved but not enough to approve, how do you vote? If an NB said it was improved, but insufficiently, how was such a vote counted? Or was it counted?
Update 3: More from Charles-H. Schultz:
I think the article from CIO says it all. The Head of Delegation of the ANSI (USA) explains what went wrong. I think it’s a pity that the BRM ended up like this. In a nutshell, the whole idea of the BRM was to discuss the proposals from Ecma and the comments made by the delegations, and it just didn’t turn out it was possible. Delegates were rushed to vote on hundreds of comments in bulk , were told new rules had to be applied, and when many of them tried to propose solutions to technical or legal issues they were simply dismissed.
We’ll talk about this more in detail later, but as it stands today, the BRM has failed - failed to work, failed to impress, failed to create consensus and failed to succeed. Rules that were not part of the existing JTC-1 corpus had to be invented to come up with the astounding result of 6 countries approving the bulk voting versus 4 countries formally disapproving them, 18 others abstaining, while four others even refusing to vote as a way to show their complete disapproval of the way the BRM was being handled.Only committees of countries that were present in Geneva could vote, so they do not speak for the rest of the world. It is unfortunately likely that Microsoft and the European Committee of Microsoft Advocates (Ecma) will declare victory, based on what is a pathetically weak relative majority and on the set of rules that go against both the letter and spirit of the JTC-1 legislation. Would this then be a pyrrhic victory? Hardly. It has yet to to be shown if the Ecma and ISO can actually do anything with that result and the growing resentment of national standardization committees.
In any case, we’re now back on for a month of national ballot in each national committee, and I believe our discussions will be interesting as many of them will not find the results of this week satisfying.
And Yoon Kit Yong of OpenMalaysia Blog now tells us more:
The final day was absolute mayhem. We had to submit decisions on over 500 items which we hadn't have the time to review. All the important issues which have been worked on repeatedly happened to appear on this final day. So it was non-stop important matters. Unfortunately I was caught up in a change from Malaysia, so I must have missed deliberating on a few important matters.
But it's all over now. Due to the quirks in the voting mechanisms, a reported 98.4% of Ecma resolutions were approved. This on the surface projects an impression that the BRM is a resounding success. Unfortunately this is not the sentiment of the majority of participants...
This is not in criticism of the Convener Mr Alex Brown. He had a monumental task ahead of him at the beginning of the BRM. The chips were stacked high against him. It was not the failure of the National Bodies which attended. It was merely a failure of the process. And it may not be the failure of ISO as a process for creating standards, but mainly because a client chose the wrong method in forcefeeding a large draft standard in the conservative process of the ISO.
It was a failure of the Fast Track process, and Ecma for choosing it. It should have been obvious to the administrators that submitting a 6000+ page document which failed the contradiction period, the 5 month ballot vote and poor resolution dispositions, should be pulled from the process. It should have been blatantly obvious that if you force National Bodies to contribute in the BRM and end up not deliberating on over 80% of their concerns, you will make a lot of people very unhappy.
I think coming into the BRM at the beginning of the week, some people were optimistic that this could make a positive difference. But judging from the reactions from the National Bodies who truly tried to contribute on a positive manner, without having their concerns heard let alone resolved, they leave the BRM with only one decision in their mind come March 29th.
The Fast Tracking process is NOT suitable for ISO/IEC DIS 29500. It will fail yet again. And this time it will be final.
And please now, don't say we didn't try ...
Update: Updegrove has posted on March 2 a comment on Jason Matusow's blog, and here's a bit of it:
2. My figures come from the SC 34 Web site and were given to me by a delegate. We will have to wait and see whether ISO/IEC JTC1 releases the exact numbers and countries or not (I had, but did not release the countries). I totally reject that an organization that purports to set "de jure global standards" for preferential adoption by governments has any business keeping the facts about how their "open" standards are determined secret from those that are affected by them. Fortunately, many delegates agree, and are therefore sharing their stories....
9. I find it incredible how often you, Doug Mahugh and others rave about "how much better OOXML is now!" as if this was some great thing. ODF sailed through with _no_ significant comments, and did not need a BRM at all. To have submitted a 6,000 page specification under Fast Track at all was regrettable. To submit somethig in this poor shape was unconscionable. To subject the BRM delegates to this miserable process and then take advantage of their efforts by calling it a huge success is pathetic.
In conclusion: It would have likely taken Microsoft fewer engineer-years of effort to implement ODF in Office 2007 than it took engineering years for all these delegates to prepare for and attend the BRM,and certainly been far more beneficial. And your customers would have been better off and happier if they did.
Speaking of which: I'll close with a story from Geneva last week. You mentioned above, i think, that Microsoft did not attend the simultaneous OpenForum Europe event. In fact, there were one or more Microsoft representatives in attendance all the time. In one panel presentation, Nick Tsilas from your legal department, and someone I've known for years, tried once more to say how Microsoft's customers haven't yet asked for Microsoft to implement ODF.
Unfortunately, he picked the wrong panel to make that claim, because one of the panelists on stage just then was Dr. Rolf Theodor Schuster, the CIO of the German Federal Foreign Office. Rolf lit into Nick in no uncertain terms, saying that his office had demanded that Microsoft implement ODF and not OOXML. After "listening to its customer," Microsoft responded by going over his head to his superiors and getting these bureaucrats to tell them not to worry about it. Nick tried to make the same claim again, and Rolf really ripped him a new one.
Jason, people generally think you're a good guy. But this blog entry is so far out in left field that I don't think that I'll be hearing that as often as I used to. Go give a read to the first hand accounts I've compiled at my blog, and then let us know whether your blog entry requires modification, will you?
A final thought: I heard many, many first hand horror stories from National Body delegates while I was in Geneva. Here's a sample: inone NB, the vote was 8 to 1 against OOXML last summer. The 1 vote to approve? The Microsoft employee. The result? An entire country had to abstain in the vote, because one conflicted (new) committee member took advantage of the unanimous voting rules of that country.
Of course, I also heard from many delegates that each European country has received a formal request letter from the European Commission calling for the disclosure of all bad acts that Microsoft had performed in the process so far as part of the EC's new antitrust investigations. This blog entry will, I expect, make them be even more attentive to what they hear in response.
In over twenty years of repsresenting standards bodies, I have never seen anything like this before, and I certainly hope that I never do again.