The details are out now. Jomar Silva, a delegate from Brazil, which voted No, has now done what he said he would do and has posted what he saw and heard at the BRM. It is a deeply shocking tale of maneuvering the delegates to vote against their will by presenting a kind of Sophie's Choice of options, all designed, according to what I gather from his account, to get a positive result for Microsoft. I'd certainly like to see the Directives that support the choices as presented. Silva talks about where the numbers behind the infamous slide asserting that
"98% of issues were resolved" came from, and you'll see how it misrepresents what really happened. Best of all, he provides documents that support what he tells us. No wonder the EU Commission is investigating. What can I say? If a format can win "approval" as a "standard" only like this, is it really a standard?
Here's the full story:
To begin the history, on sunday evening, before the beginning of the BRM, at the Heads of Delegations (HoD) meeting they were warned that on Wednesday we would have to make a decision about “what to do with the ECMA’s responses that could not be discussed”.
The working methods adopted in the first two days of meeting (and at the paper it works very good), each country (in fact the so-called National Bodies or NBs and if I am not mistaken was 33) could present for discussion one of the ECMA’s responses of their interest (ie, in alphabetical order each NB presented a problem). The debate then started and if the theme was controversial and causes more than 10 or 15 minutes of discussion, the NBs involved in the discussion were invited to discuss “off-line” and the proposals was recorded in a backlog. Any discussion ended in “editing instructions to the editor” and anything that does not contain “editing instructions to the editor” was simply ignored (so resolved…).
The result of that is that until Friday, even two complete rounds were held (ie that there was NBs that only proposed one discussion)… explanation to that: lack of time (which incidentally, explains all other decisions taken there, ok?).
During the debates, some discussions had expanded and covered more than one ECMA’s responses (or sometimes different responses dealt with related themes), and this explains the high degree of items discussed (or as I prefer to call “touched”) during the BRM (withdrawn from this document, the final document of BRM [PDF]): 189 responses or 18.4% of the total (is that the amount of discussed items expected on an International Evaluation of a so important theme specification? Imagine if your country’s constitution was writ[ten] using that method, with only 18% of its laws discussed).
The decisions taken day by day can be found in this document, “edited notes of the BRM” [PDF] but I admit that this is a confusing document. One example is that we can see that there existed a voting regarding the transformation of the specification into a Multi-Part standard (page 7), in which Brazil presented an objection. This voting result (and related Brazilian objection) was not transported to the final document (I think that someone forget to copy that to the final document, or a decision with that importance doesn’t need to be recorded ?).
Furthermore, I reserve the right to not comment here what conduced to the adoption of this proposal, which actually transforms the OpenXML (or DIS 29500) in five International Standards that may have their “own life” (the only difference between those and five “traditional” standards is its shared numbering , DIS 29500-1, 29500-2… DIS 29500-5). Wonderfull, isn’t it ? (a 6.000+ pages-five-International-Standards fast-track… really amazing)…
When we reached Wednesday, we were presented four options of “destination” to the other answers (ie 81.6% of the total). They were only four options and one of them must be chosen (at that moment I’ve remembered the movie “Sofia’s Choice”):
1 - All rejected.
2 - All approved.
3 - The ITTF decides everything.
4 - Decision by vote in batch, with the possibility of declaration of each individual response vote and / or the definition of a overall vote, as “accepted everything”. (…and Mr. Barta, please note that I’m not using the “default vote” expression here, ok).
Of course, the option was the least ridiculous was option 4 and that’s why it was adopted (in the final document [PDF], page 5 there is a copy of the ballot to vote). Much has been discussed on the possibility that some NBs should cast blocking votes (as vote “Approve to all” to force the approval or “disapprove to all” to force the disapproval), but nothing could be done about it…
At that moment, there was a very interesting protest of a delegation which said that they didn’t went Switzerland to vote, which could be done from home (they were there to discuss) and the answer: “Patience…”.
Another NB protested saying that they only received the document containing the answers to their own questions and didn’t even ha[ve] the opportunity to read and discuss the whole responses document (1027 responses on a 2500+ document). For this reason the option “We do not wish to record any position” was created on the ballot. That highlighted problem also happened on other NBs and there are NBs that simply voted on proposals that they never even read! (This is really cool, right?).
A rule was made that any decision taken at the meeting overwrites the decision taken by “the ballot” vote (ie a decision by “the floor” has greater weight).
At the end of the week, the votes were delivered and the final result (which I also reserve the right not to comment or disclose), the vast majority of the proposals were approved (most by a few votes of “approval by default”…).
I’ve writ[te]e this post just to assure that there is a public explanation to the 98% approval that some companies are using worldwide.
See now how the 98% now doesn’t mean anything? And the 18.4%… this really need to be discussed (and explained by someone).
Finally, the documents cited in the final document [PDF]of BRM (such as an URL to the SC34’s website which is protected by password) are publicly available here and “final results table” for those who like to play with non-sense statistical data can be found here [PDF] (numbers are just numbers…).
All documents referenced here are publicly available documents of SC34’s web site. Fe[e]l free to analyze them and take your own conclusions....
If you don’t know me and would like to know how I know all this? I was in that room, member of the Brazilian Delegation.
So when we read that such and such a country voted yes, there may be a back story equally shocking as this one is. Would you like to know who was at the BRM? Here's the attendance list, which shines light on the matter also. Look for how often the affiliation is Microsoft. Then there are those that clearly support them without being directly employees, like Clever Age, et al. As you can see, Frédéric Bon, listed as representing Clever Age at the BRM, is also Chairman of the AFNOR Standards Committee [PDF], and Clever Age was retained or whatever they called it publicly to do "translator add-ins" for ODF interoperability for Microsoft Word, etc. Not that they work, from what I've heard. Connecting some dots? Why doesn't Microsoft do it itself? I think it's to preserve their litigation position regarding patents, myself. They can hardly write a translator and then sue you for using it. But if someone else does it, they can say they didn't realize, like SCO pretending not to know about the GPL. Anyway, my point is that Microsoft is approving its own "standard", I'd say. We count 20 direct Microsoft participants:
1 BELGIUM Mr. Bruno SCHRODER MICROSOFT
2 BRAZIL Mr. Fernando GEBARA Microsoft Brazil
3 CANADA Mr. Paul COTTON Microsoft Canada
4 COTE D'IVOIRE * Mr. Wemba OPOTA MICROSOFT West and
5 CZECH REPUBLIC Mr. Štepán BECHYNSKÝ Microsoft Czech
6 DENMARK Mr. Jasper Hedegaard BOJSEN Microsoft
7 FINLAND Mr. Kimmo BERGIUS Microsoft Ltd
8 GERMANY Mr. Mario WENDT Microsoft Deutschland
9 ISRAEL Mr. Shmuel YAIR Microsoft
10 ITALY Ing. Andrea VALBONI Microsoft Italy
11 JAPAN Mr. Naoki ISHIZAKA Microsoft
12 KENYA Mr. Emmanuel BIRECH Microsoft East Africa
13 NEW ZEALAND Mr. Brett ROBERTS Microsoft New Zealand
14 NORWAY Mr. Shahzad Rana Microsoft Norge AS
15 PORTUGAL * Prof. Miguel Sales DIAS MICROSOFT
16 SWITZERLAND Mr. Marc HOLITSCHER Microsoft Schweiz GmbH
17 UNITED STATES Mr. Doug MAHUGH Microsoft Corporation
18 Ecma International Mr. Brian JONES Microsoft
19 Ecma International * Mr. Jean PAOLI Microsoft Corporation
20 Assistant to Project Editor Mr. Tristan DAVIS
Is that normal? I doubt it, but if it is, could someone do something about it? How can employees of a company go against the folks that pay them, even if they privately think the format is junk? Some might, but they'd surely not be employees after that. Now might you get why the Czech Republic waxed poetic about how wonderful the BRM was?
Update: I just got this email with news from S. Korea, and you'll note we are seeing a pattern of voting yes because *that* country's objections were resolved, as if they are obligated then to vote yes:
There is some sad news...
In final voting of ISO standards process of Office Open XML, Korea will vote
it did "No with comments" in the last year. The Korean member body
confirmed "The ECMA accepted to fix most of objection
comments by a
document in voting of last year." So they changed to their position
from disapproval to approval.
The Google translation of the article about the S. Korean vote includes this funny bit: "The 'International Standardization ISO OOXML vote' voted to throw in a planned,"he explained."
And more sad news from Poland, that the strategy worked. The seven votes that never were sent by email counted as yes votes by the chairperson, and so Poland will stay with its Yes vote. The report ends:
This was a long fight and we lost it. Thanks for your support everyone. I would like to sincerely apologize for PKN, Mrs Elżbieta and all the folks who helped make this happen. I’m very sorry. Good bye.
Well, it's hard to win when the dealer uses a stacked deck. Let's see what the EU Commission does about it. You know what they say in the US, "It ain't over 'til it's over."
Update 2: France will vote No, according to this journalist's sources:
"Nous avons répondu à toutes les demandes et nous avons montré des signes d'ouverture en février. Il serait étonnant que la France ayant obtenu toutes ses réponses ne transforme pas son «non» en «oui» », estime Marc Mossé, directeur des affaires juridiques et publiques pour Microsoft.
That's saying that a Microsoft spokesman says the company responded to all the issues and showed a willingness to be more open in February. So it would be astonishing that France having had all its issues answered would not change its vote from no to yes. This is a prediction, by the article and by Microsoft, so it's not firm yet.
I am hearing that South Africa has now voted No also.
Update 3: And a Chilean newspaper says Chile will abstain, which is what it did in September, despite a 'consulting
committee' voting 60-40 for approval. If my Spanish is not failing me, it says that some pro-Microsoft maneuvers got the head of the NB so annoyed, it backfired:
Después de una primera votación, en septiembre de 2007, adversa para la mega-compañía, Microsoft ha desplegado una intensa campaña en busca de votos a nivel global que incluso escaló al nivel de las denuncias de irregularidades, una vez que se supo que funcionarios de la empresa en Suecia ofrecieron a sus socios comerciales “incentivos” y “contribuciones de marketing” a cambio de incorporarse al comité nórdico y votar a favor de la aprobación del OOXML. Suecia primero aprobó el formato y más tarde invalidó su voto debido al escándalo que llegó a los diarios.
El tema es de alta relevancia para Microsoft debido a que si no consigue los votos suficientes para la aprobación del estándar en Ginebra, el proceso abreviado o de fast track volverá a cero y podría tomar años para que el OOXML llegue a ser visado como estándar. Perder el fast track significa tiempo y dejar de ganar mucho dinero. Este es uno de los motivos para que la compañía se haya jugado el todo por el todo por lo que pase en Ginebra y haya buscado votos donde pueda, sea entre países prioritarios como Estados Unidos y Alemania (a favor) o India y Japón (en contra); o secundarios como Chile....
Según trascendió, en la decisión de Toro de mantener el voto de abstención del año pasado influyó su molestia por la supuesta decisión unilateral del presidente de la Cámara de Comercio Electrónico y secretario ejecutivo del comité consultivo del INN, Carlos Recabarren, quien apoyó el estándar de Microsoft en una segunda reunión realizada en Ginebra en febrero de 2007, sin consultar a sus representados, según se desprende de las actas de reuniones del organismo.
En el encuentro a fines de febrero en Suiza, Recabarren fue uno de los cinco votos de minoría que favorecieron un paquete de correcciones al OOXML, sin analizarlo a fondo. El tema generó ruido en ciertos integrantes de la comisión que consideraron que Recabarren actuó como juez y parte, dado que Microsoft integra la Cámara de Comercio Electrónico.
I'll attempt a rough translation, because it mentions a meeting in February in Geneva. What is this extra meeting?
After the first vote in September of 2007 against the mega company, Microsoft, it then rolled out an intense international campaign for votes that even reached the level of
irregularities being reported, such as Microsoft employee in Sweden offering "incentives" to their commercial partners
and promises of "marketing help" in exchange for joining the
committee and voting in favor of the
approval of OOXML. Sweden first approved the format and later invalidated
its vote, after the scandal hit newspapers.
The matter is of high importance to Microsoft because if does not
obtain sufficient votes for approval of the standard in
Geneva, the fast track process is aborted, it goes back to zero and
could take years before OOXML gets to be a standard. To
lose on the fast track, then, means lost time and losing an opportunity to make a lot of money. This explains why Microsoft has thrown itself in whole hog to trying to win in Geneva, trying so hard for votes, not only from important countries like
the United States and
Germany (favoring Yes) but also India and Japan (against); and even secondary countries like
What apparently influenced Toro's decision to maintain last year's Abstain vote
was his annoyance at
what he viewed as a unilateral action by the president of la Cámara de Comercio Electrónico (the
Chamber of Electronic Commerce) and executive secretary of the consultative
committee of the INN, Carlos Recabarren, who supported the Microsoft format as a standard in a second meeting in Geneva in February of 2007,
without consulting him.
At the meeting in February in Switzerland, Recabarren was
one of the five votes by a minority that favored a package of
corrections to the OOXML, without analyzing it thoroughly. That resulted in complaints by other members, who felt he'd acted like both judge and partisan, or party, so to speak, since Microsoft is a member of la Cámara de Comercio Electrónico (Chamber of Electronic Commerce).
You can no doubt do a better translation, but I thought it important to get the information out about the head of la Cámara de Comercio Electrónico's actions. The EU Commission is likely collecting stories like this one. The article is titled,"Microsoft sufre su peor derrota a nivel local" which means that Microsoft has suffered its worst defeat ever in Chile.
Update 5: Here is a much better translation, along with details I didn't notice. For example, Not only was Microsoft a member of CCE, the organization "until last January had Microsoft-Chile’s CTO as a member of its board of directors." You will also find there the official letter from Carlos Recabarren announcing Chile's decision to stay with its September Abstain vote, due to a lack of consensus. That is, after all, what the rules say to do if there is no consensus. Yoo hoo. Denmark. Did you read the Directives on that point?
I see an anonymous comment suggest that ISO now stands for:
But isn't it good to see how many are refusing to do so? Consider the pressure, the threats, the stacking the deck, evidently some promises too, not to mention the confusing and ever changing rules, and it's remarkable. And yet all around the world, people have the courage and the integrity to say No.
Update 4: We reported on Denmark changing its vote to yes earlier, but a translation arrived of a snip of this report:
After the Ballot Resolution Group meeting, there has not been consensus in
the committee, if all Danish comments had been entered into the final standard
in a sufficient way. And there has been disagreement about the ripeness of
ISO/IEC DIS 29500 OOXML as a fully fledged ISO/IEC standard.
In the final statement from the committee to 'Dansk Standard', the question of
Denmark changing the original vote also didn't reach consensus....
On this background 'Dansk Standard' is changing Denmarks vote to 'Yes'
Um. No consensus. But they change the vote anyway? And here's an interesting viewpoint from Andy Updegrove:
Many journalists and others have asked me whether I have a prediction on what the outcome will be, and also what I think it will mean if OOXML is approved. I don’t have an answer to the first question, as there are too many countries involved, and too much may change until the last minute. But I do have an answer to the second question, and that answer is the same one that I have given every time that a new decision point has loomed in the ongoing quest for a useful format standard that can bring competition and innovation back to the desktop, as well as ensure that the history and creativity of today will remain accessible far into the future.
That answer is this: if anyone had asked me to predict in August of 2005 (the date of the initial Massachusetts decision that set the ODF ball rolling) how far ODF might go and what impact it might have, I would never have guessed that it would have gone so far, and had such impact, in so short a period of time. I think it’s safe to say that whatever happens with the OOXML vote is likely to have little true impact at all on the future success of ODF compliant products.
Here are ten reasons why I believe this prediction will be borne out.
The first is the obvious: "The quality of OOXML is not there yet." Personally, I'd leave off the 'yet' and call it done. I'd add two more to the list: first, as I've written about before, I believe the Microsoft OSP means FOSS developers can't rely on the OSP regarding OOXML, and so must avoid it, and that there are intellectual property issues for anyone so much as opening an OOXML+proprietary extension document. And two, in part a result of the first one: interoperability will never really happen. OOXML is what Microsoft offers instead of real interoperability.
Update 6: Glyn Moody has some wonderful news on ComputerWorld UK. I'll let him tell it:
John Pugh is a national treasure: an MP who actually understands technology. And he's proved it again with this very pertinent letter to the head of the BSI, which seems poised on the brink of doing something very silly. It's so good it's worth quoting in its entirety.
And so he does. Here's the letter to Mike Low, Director of British Standards:
Dear Mr Low,
I am deeply concerned that some national bodies have considered approving DIS29500 "in their national interest".
It is not in the interest of the UK or any other country for DIS 29500 to be published as an international standard in its present form as there are a significant number of unresolved issues, including incompatible licensing conditions, single vendor interest and control as well as those other
factors uncovered since the original comment period closed. There is also the issue that is no current form of DIS 29500, as yet another exception to the process was made regarding the presentation of the proposed changes.
Issues raised by the BSI and others during the original 30 day objections phase have also never been properly addressed, not least of which is the unsuitability of such a large and complex specification of an unproven format to the fast track process.
Such a standard, if approved, would govern the manipulation of all future documents and lock users into a single supplier environment. The OOXML specification contains significant Intellectual Property Rights and other traps which will:
- Diminish innovation overall
- Severely restrict consumer choice
- Block Government choice
- Limit the ability of UK SMEs to innovate and compete
I thought you would like to be aware that I have tabled a parliamentary PQ as follows:
"Can the Minister reassure me that the British Standards Institute that operates as our National Body under a memorandum of understanding from the Government will not change its stance on the currently unavailable text for the proposed standardisation of OOXML without proper consultation."
May I urge that the BSI uphold its admirable principles and does not approve DIS29500 in its present form.
Update 7: Now Norway says yes. Translation would be appreciated. We have a comment from Groklaw member kattemann, who says that it says this:
The meeting (of the NB) was supposed to last from 11:00 to 13:00 on
Friday, but dragged on due to the massive resistance from several members of the
"In the end there were three members from Standard Norge who made the
Three members. Connected to Microsoft, you think? A research project. He reports that there is another article with a quotation from Opera's CTO:
In a related article on computerworld Norway, Håkon Wium Lie is furious:
"21 committee members against, and Standard Norge still does Microsoft's
More scandal, I gather.
Update 8: We have a translation now, and also a translation of the press release from Standards Norway announcing their decision. It is, indeed, disturbing. First the translation of the article:
The meeting, which started at 11:00 am and was scheduled to run until 1:00 pm,
dragged on, especially because of the massive opposition of several members
of the committee. Because there was no consensus in the group, key players
from Standards Norway and those who had attended the Geneva ISO conference
enumerated the meeting. [enumerate is a direct translation...perhaps
summarized would be better].
According to Chief Information Officer Marit Sæter of Standards Norway,
these key people were committee chair Steve Pepper (independent consultant),
Ivar Jackwitz and Bjørnhild Sæterøy (Standards Norway), Shahzad Rana
(Microsoft), Knut Tungland (Statoilhydro) [Norweigian State oil and
hydroelectric company] and Erlend Øverby (Hypatia).
Three Made Decision
At the end it was only three people from Standards Norway, led by Jackwitz,
who made the decision.
"The committee could not come to an agreement. Standards Norway's decision
has therefore been a difficult one, but we feel that it is important that
the standard becomes available as an open international standard.
When it becomes an ISO/IEC standard, it will be improved to better meet the
needs of users, is the reasoning of Standards Norway.
Several of the committee's members believed that comments regarding concerns
about OOXML made during the first round of ISO validation were ignored in
the revision. This did not convince Standards Norway. Rana believes the
disagreement was mainly quibbling. [could be translated as nitpicking]
"The debate revolved too much about syntax quibbling without focusing on the
solution. The ECMA does not use the same words as the Norwegian comments
[concerns] so some may have perceived this as wrong", states Rana.
Now, class, what do the JTC1/SC34 Directives say to do if there is no consensus? Just approve it anyway? This is amazing. And as for improving it later, that means they knew it isn't in good shape now, but they hope for better in the great by and by.
Here's the press release translation:
Standard Norge earlier voted "No, with comments" and gave suggestions
for changes to the format. Comments were sent to ISO/IEC and treated in
a so-called "ballot resolution" meeting February this year together with
comments from numerous other countries. The purpose of today's meeting
was to discuss to what degree the Norwegian comments have been attended
to and if Norway should change its vote from its original No vote.
There was disagreement within the committee whether the Norwegian
comments had been sufficiently taken care of. Standard Norge however
thought it was important that if this becomes an ISO/IEC standard it will
be improved to accomodate the users' needs in a better way....
Undoubtedly there has been strong conflicts of interest in the Norwegian
committee. This, however, is a natural part of the standardization
process and something that Standard Norge is not unused too.
Update 9: This Blog de Jose Miguel Parrella Romero is reporting that Venezuela has changed its vote to No.
And FOSSFA is urging African nations to vote Abstain:
The Free Software and Open Source Foundation for Africa (FOSSFA) has called on African governments and standards bodies to abstain in the balloting, due to close Saturday, on whether a file format based on Microsoft's Office Open XML (OOXML) should be an international standard.
The paper-ballot process is being undertaken under the auspices of the International Organization for Standardization (ISO). FOSSFA is urging African countries to abstain, as part of a wider effort to participate more effectively in the global debate on open standards....
Recalling that the ISO meeting on OOXML in February in Geneva had little representation from Africa, and that some African countries have voted in the past to approve OOXML without thorough and more inclusive discussions on the issues, the organization is calling for more active public engagement of the issues in the future.
African countries including Cote d'Ivoire, Ghana, Kenya, Mauritius, Morocco, South Africa, Tunisia and Zimbabwe have the right to participate in the ISO's OOXML balloting.