More Discussion of the Open XML BRM

The Process Challenge - A Predictable Path

Matusow's Blog — Wed, 02 Apr 2008 06:49:05 GMT

The Process Challenge – A Predictable Path In light of the approval of Open XML as an international standard,

re: More Discussion of the Open XML BRM

marc — Wed, 05 Mar 2008 21:10:46 GMT

"I do understand that the JTC 1 directives, in section 3.1 and 3.2, state that “normal JTC 1 voting” would be P-member only. But I believe the decision of the convener (Alex will speak for himself I’m sure) was looking to determine “consensus” in the room."

for god sake, why do you have rules if you can't break them arbitrarely ??

re: More Discussion of the Open XML BRM

marc — Wed, 05 Mar 2008 21:09:15 GMT

"4. There were delegations that vote in favor of all 1027 resolutions outright."

yes, for example the representative of Ivory Coast (Côte d'Ivoire) , Wemba Opota [1], who is responsible for Microsoft West Africa.

He asked to be upgraded to P-member a few votes before September/2007 ballot closing just to cast an inconditional yes vote, and now approves inconditionally 1027 comments.

For god sake! this is what i call

"standardization by corporations"

[1] http://www.noooxml.org/forum/t-43510/ivory-coast-represented-by-microsoft-senegal-at-the-brm

re: More Discussion of the Open XML BRM

hAl — Wed, 05 Mar 2008 19:03:24 GMT

[quote]Andy has now added a blog entry from Antonis Christofides, of the Greek delegation. Is he also an ardent detractor of OOXML? Help me out here![/quote]

Antonis is with the FFII I believe.

They run the no ooxml site

So I would call him slightly biased yes.

Mayby Andy did not mention that on his blog though

re: More Discussion of the Open XML BRM

Luc Bollen — Wed, 05 Mar 2008 04:12:07 GMT

@Gerhard: Indeed, OOXML is a DIS.

And in (9.5), just before the sentence you quote, it is written "The voting procedure which uses simultaneous voting [...] ***on a letter ballot*** is called the combined voting procedure."

1. The text explicitly refer to the letter ballot. But in Geneva, it was a BRM meeting, not a letter ballot.

2. In the combined voting procedure, all the NBs vote but only the P votes are taken into account for the "2/3 criteria" deciding if the DIS is accepted.

3. Nevertheless, from a email exchange I just had with a JTC1 expert, who had the same question as mine, it appears that ITTF decided that all NBs will vote at the BRM because the letter ballot was a "combined voting procedure". I don't know if this rule was used in previous occurrences nor if it is published somewhere. I continue to investigate.

Now, does this really matters ? With a little more dispositions accepted or a little less dispositions accepted, the DIS text will remain at a low level of quality. Everybody (pro- and anti-OOXML) agrees that the discussions about 100 to 200 dispositions improved the text, but that much more time and discussions would have been needed to satisfactorily resolve all the comments. And I don't see how it will be possible to have a good quality control of the new text being prepared by the editor, to be issued on 29 March latest (which is also the deadline for the NBs to change their vote).

It is obvious that the Fast Track was not the right approach to get a quality result. Determining which NB was allowed to spend a night for voting on 900 dispositions will not change that.

And the main question remain : do we really need a second ISO standard (be it of a very high quality) for text, spreadsheet and presentation documents ?

re: More Discussion of the Open XML BRM

Gerhard Goeschl — Tue, 04 Mar 2008 20:28:45 GMT

@ Luc: Would you agree that the official name in ISO for Open XML is ISO/IEC DIS 29500?

(see: http://www.iso.org/iso/iso_catalogue/catalogue_tc/catalogue_detail.htm?csnumber=45515)

If you do, then please look again into the last sentence of (9.5) Combined Voting Procedure.

It clearly says: "This procedure shall be used on FDISs, DISs, FDAMs, DAMs and FDISPs."

PS: Please have a look also at Alex Browns comment (http://www.consortiuminfo.org/standardsblog/comment.php?mode=view&cid=18819), where he clearly states: "OBVIOUSLY (given the red hot controversy here) voting procedure was discussed in minute detail, and decided, in consultation with ITTF before the BRM started."

re: More Discussion of the Open XML BRM

Luc Bollen — Tue, 04 Mar 2008 15:19:55 GMT

After further investigation in the JTC1 directives, I found 3 places stating that O-members have no rights to vote :

(2.3.1.2) [...] Observing membership (O-member) having no power of vote, but options to attend meetings, make contributions and receive documents [...]

(3.2) [...] O-members have no power of vote, but have options to attend meetings, make contributions and receive documents. O members of JTC 1 subcommittees may nominate experts to participate in any working group of the subcommittee in which they have O membership.

(7.7.4) [...] O-members and other TCs and organisations in liaison may nominate representatives who have the right to attend meetings and to participate in the discussion, but do not have the right to vote.

There is a special case, called "combined voting procedure", where all ISO and IEC countries are voting (whether P-members, O-members, or none). This is the procedure used in September to vote on DIS-29500 :

(9.5) Combined Voting Procedure

The voting procedure which uses simultaneous voting (one vote per country) by the P members fo JTC 1 and by all ISO member bodies and IEC national committees on a letter ballot is called the combined voting procedure. This procedure shall be used on FDISs, DISs, FDAMs, DAMs and FDISPs.

So, I really don't know why ITTF decided to let O-members voting at the BRM. Is there are JTC1 experts reading this, thanks to let me know if I overlooked something.

re: More Discussion of the Open XML BRM

Gerhard Goeschl — Tue, 04 Mar 2008 10:52:46 GMT

@ Rob:

regarding Andy: He was not part of the BRM, and Alex Brown has TWICE!! remarked on Andy beeing inacurate and even misleading.

Perhaps because Andy is the legal council of OASIS, member Boards of Directors of Free Standards Group (FSG) and member Boards of Directors of Linux Foundation, where he is furthermore the Director of Standards Strategy. S he cleary has his own agenda to drive here.

Andy beeing inaccurate: http://www.consortiuminfo.org/standardsblog/comment.php?mode=view&cid=18785

Andy beeing misleading: http://www.consortiuminfo.org/standardsblog/comment.php?mode=view&cid=18819

So I rather trust people who have been there like the Danish Delegation or the Norwegian Delegation, and just have their nations interest in their minds.

re: More Discussion of the Open XML BRM

Luc Bollen — Tue, 04 Mar 2008 03:01:15 GMT

@Jim : you are right, the O-members (Observers) can vote for JTC1 ballots. But different rules have to be taken into account :

From the JTC1 directives (http://www.jtc1sc34.org/repository/0856rev.pdf):

(9.6) "For a FDIS/DIS/FDAM/DAM/FDISP to be approved, the count taken by ITTF shall meet the following criteria:

• At least two-thirds of the P-members voting shall have approved;

• Not more than one-quarter of the total number of votes cast are negative."

- So, to be accepted, a Draft International Standard (DIS) must receive at least 2/3 of the votes by P-members (Participants). For this criteria, votes by the O-members are *not* taken into account.

- But even if accepted by the P-members, a DIS is rejected if at least 25% of the total votes are negative. Here, both P and O votes are taken into account.

In a sense, one could say that "O-members are not voting for having the DIS accepted, but are voting for having the DIS rejected" (but probably this formulation will be considered incorrect by JTC1 experts).

So, my sentence above "If they choose to be O members, they know that they cannot vote" is indeed not correct. It would have been better to write : "if they choose to be O members, they know that their vote will not be taken into consideration for having the DIS passed".

The JTC1 Directives says this for voting at the BRM:

(13.8) "At the ballot resolution group meeting, decisions should be reached preferably by consensus. If a vote is unavoidable the vote of the NBs will be taken according to normal JTC 1 procedures."

(9.1.4) "In a meeting, except as otherwise specified in these directives, questions are decided by a majority of the votes cast at the meeting by P-members expressing either approval or disapproval."

The question is to know if something is "otherwise specified in these directives" for a BRM... I haven't found anything, but you have to read and understand the 188 pages of the JTC1 directives to be sure. This is indeed a quite complex matter, and only seasoned JTC1 experts can confirm what the jurisprudence is.

re: More Discussion of the Open XML BRM

Rob Brown — Tue, 04 Mar 2008 00:44:37 GMT

@Gerhard:

No, I'm not one of the people Rick Jeliffe refers to. I don't particularly even care, technically, which file format becomes predominant.

What I want is to understand what is happening, because it is going to affect my future. And I'm inclined to believe the people I quoted more than Jason and Brian, because the quoted postings provided a whole lot more substance than our MS friends. Read Andy Updegrove's blog on the event, which has its own validity because he is updating it as information becomes available. Andy has now added a blog entry from Antonis Christofides, of the Greek delegation. Is he also an ardent detractor of OOXML? Help me out here!

I'm an electrical engineer, and my main focus in all this is programmatic interaction with spreadsheets. I'm also a consumer that wants to make sure the buying decisions I make are based on merit and not lock-in.

So I have two main concerns with OOXML:

1) I don't trust Microsoft.

1a: there is a huge volume of circumstantial evidence that Microsoft has gamed this standards process to the point of ridiculousness. Whether you believe the evidence is a personal matter, as it will probably never be proven in the real world. I hope that the EU Commission's investigation into the OOXML standard process will provide illumination.

This is the point of my quotes above: it is obvious that a lot of people have put a lot of really good work into OOXML, but that it is just not suitable for the fast-track process.

Note that I'm not suggesting there were procedural problems with the BRM itself. Because the desired fully-discussed consensus result could not possibly be achieved, the attendees at the BRM agreed to a no-discussion vote result for most of the proposed dispositions. Jesper Lund Stocholm's quote is very telling: "So we basically had two choices: * Do nothing; * Do something. The BRM chose to do something."

1b) The recent Microsoft song-and-dance about openness was nothing new and changes nothing. It was Microsoft's version of "The Boy Who Cried Wolf", now titled "The Ballmer Who Cried Interoperability". Time, and real-world action, will show whether anything has changed.

1c) Jason and Brian are amazingly offhand about Microsoft's recent anti-trust woes. Maybe for those in the rarefied air of the world's most cashed-up business, a US$1.5 billion fine can be brushed off, but for those of us on the ground, it points to something very rotten indeed in the state of Redmond.

2) "Competing Standards" is a concept that has no value to me (and I don't really think that many people would value it). It's an oft-quoted example, but I work with communication protocols so it's relevant to me: would the Internet have ever happened if there was "customer choice" in wire protocols? Would there be "TCP/IP only" sites and "MegaCorp MegaProtocol only" sites? Or more likely, would every vendor of networking equipment have to carry the overhead of supporting both prototocols?

*None* of what I'm saying is related to the technical quality of OOXML as a spec. I personally think it's pretty crappy, but ODF (if we're being really honest) isn't spotless either.

If OOXML was a standard offered by a party with true interoperability and even-playing-field competition in mind, then it would be fine. Or even if the management plan specified truly independent future oversight. IMVHO, neither of these conditions is satisfied and so as a result I want to see OOXML fail.