Partager via

US technical committee reaches deadlock

  • Article

Last Friday, the INCITS V1 technical committee had a meeting via teleconference to try to reach consensus on a position that we could recommend to the INCITS Executive Board regarding DIS 29500, the submission of Ecma 376 "Office Open XML" for consideration as an ISO/IEC international standard. The meeting follows the face-to-face meeting we had at IBM's office on K Street in Washinton, DC two weeks ago.

The ISO process is fairly complicated, and I'm no expert, but in general terms it works like this in the US: the technical committee (V1) reviews the proposed standard and recommends a position to the INCITS EB (Executive Board), and the EB then casts a vote for the US in the September 2 ISO/IEC vote on DIS 29500.

The main focus of V1's work has been to review technical, editorial, and general comments about DIS 29500, the spec for Open XML. We've talked through 96 comments in detail during our meetings, and all but a handful of those have been processed since the beginning of June. That's when we started to really dig into the comments, because most of them were submitted in the final two months of the 5-month DIS ballot period. That's also when the membership of V1 increased significantly, as organizations in the US that have hands-on Open XML experience got more involved in the process. As of May, only two of the V1 members (Microsoft and Mindjet) had any experience implementing Open XML, but now there are many companies with hands-on Open XML experience involved in the process. 

The comments we've been processing in V1 have come from several sources. IBM submitted 234 comments, and others in V1 submitted 66 comments. There were also 207 comments submitted by the public to the INCITS web site, including 171 letters of general support, 31 letters of general opposition, 3 general cautions, and 2 substantive comments. If you want to see the original submissions of these comments and the email discussions around them, check out the public V1 email reflector -- I've included the links to it in a previous post.

In Friday's call, Patrick Durusau (the chair of V1) explained to us the five positions that we could recommend:

  1. Approval
  2. Approval with comments
  3. Disapproval with comments
  4. Abstain
  5. Abstention with comments

We then discussed these possible positions. There were many opinions and perspectives represented in the discussion, and I won't characterize any specific opinions (everyone involved can speak for themselves), but there were three general groups of members:

  • Some V1 members felt that "Approval with comments" was the position to take, because it sends a message that Open XML should become an ISO standard, but there are some things that need to be corrected or modified in the spec. Many of these members didn't support "Disapprove with comments" because they believed it would allow anti-Open XML lobbyists on V1 to obstruct the process in the future. (Personally, I fall into this camp.)
  • Some V1 members felt that "Disapprove with comments" was the position to take, because it would force the resolution of certain comments. Many of these members didn't support "Approve with comments" because they believed it would allow critical corrections to slip through the process without being handled.
  • One V1 member (as you can see in the voting totals below) felt that V1 should abstain from recommending a position, and simply submit the comments.

We agreed to discard the "edge cases" in the list of 5 options above: Approval (unconditional) and Abstain. Nobody on V1 felt that the spec is ready for ISO standards status as it stands, with no changes, and there was also no motion to abstain from the process without even submitting comments. So we then tried to reach consensus on one of the other three options.

After strawman votes to gauge where everyone was at, we took three official roll-call votes on the three possible positions. The rules require a 2/3 majority to approve a position (66.67%), and here are the results of the votes we took:

Recommended Position YES NO ABSTAIN
Approval with comments 15 (60%) 10 1
Abstention with comments 11 (42%) 15 0
Disapproval with comments 10 (40%) 15 1

So we wound up in a deadlock. This means the matter passes on to the INCITS Executive Board, where they'll try to reach a decision on how to proceed. They have until September 2 to decide on a position that the US will take.  Stay tuned. 

P.S. I'm going to take a long overdue vacation for the rest of this week. I'll be back next Monday, tanned and relaxed and ready to get back to blogging on the technical details of Open XML. See you then!

Comments

  • Anonymous
    July 16, 2007
    We would be interested to know which of the particular comments you feel would obstruct the standardization process in the future ? Were those comments for instance referring to the name of the spec, to splitting up the spec or to altering parts of the spec in a way that it would affect the compatibility goals ?

  • Anonymous
    July 16, 2007
    With the INCITs exectuive board already containing 3 organisations that voted NO against an OOXML approval in the V1 committee (IBM Oracle and Farance Inc) and some other notable Microsoft competitors in that board, how would you fancy the chances that the US goes to the ISO meeting with an "Approval with comments" vote in mind ???

  • Anonymous
    July 17, 2007
    dough magugh said: "Some V1 members felt that 'Approval with comments' was the position to take, because it sends a message that Open XML should become an ISO standard, but there are some things that need to be corrected or modified in the spec" I've been following this process since its beginning ( BTW thanks to all the people that make open and transparent this ). Actually, the JTC1 Directives[1] don't allow a "approve with technical comments". If you have agreed about some technical modifications, then you must vote "disapprove with comments". With my respect, you, as a member of Incits/V1, should know this. dough magugh said: "In Friday's call, Patrick Durusau (the chair of V1) explained to us the five positions that we could recommend:   1. Approval   2. Approval with comments   3. Disapproval with comments   4. Abstain   5. Abstention with comments" In [2], Durusau explained "what he had confirmed with INCITS and what he thought was evident from the Directives", citing [2]: "... confirmed with INCITS

  • Approve as presented
  • Approve with comments (editorial or other)
  • Disapprove with technical comments
  • Disapprove with technical comments (conditional approval)
  • Abstain with comments /end confirmed with INCITS ... Further, note that comments are divided into three categories: Editorial Other Technical Only abstention with comments makes no distinction between the types of comments. Or to put it another way, yes with comments is accompanied only by editorial and other comments, not technical comments. (personal opinion) I suspect the reason for the difference in the type of comments is that conditional approval, no with comments, is appropriate where the proposal is favored but needs technical corrections before becoming an IS. " [1] ( JTC1 Directives, pages 112-113 ): http://www.jtc1sc34.org/repository/0828.pdf [2] http://www.ibiblio.org/bosak/v1mail/200707/2007Jul11-140715.eml

  • Anonymous
    July 17, 2007
    Doug,   Not sure about 171 letters being delivered in support.  It looks more like 1 letter sent in by 171 people.

  • Anonymous
    July 17, 2007
    hAl, I'd rather not get into specific comments here, but I'll just say there are comments that clearly improve the spec (such as correcting examples, a pet peeve of mine long before anyone brought it up in V1), and other comments that don't clearly improve the spec in my opinion.  As for what might happen in the EB, I don't know most of the players there, and have no firsthand knowledge of the process, so I couldn't speculate. Marc, there is definitely some debate about those details, not least because the JTC 1 directives and the INCITS mandate to V1 don't use the exact same jargon.  As for my reporting of events that took place Friday, I was careful to just quote the publicly available minutes, so those aren't my interpretations or opinions. Chris, I agree that many people have used a form letter, just as a large number of comments received by TCs around the world are repetitions of the same text from the same original source.  I'm more interested in seeing how many people/organizations publicly voice support or non-support in their own name. Pardon my brevity, but I'm on vacation so that's all the comments I'll have today.

  • Anonymous
    July 17, 2007
    The comment has been removed

  • Anonymous
    July 17, 2007
    You're the one who looks confused, Marc.  You should check your facts better. The same source you're quoting, the chair of V1, decided to ask them to vote on "Approve with comments."  Why would the chair have them vote on "Approve with comments" if he believes that's not allowed?  That doesn't add up. And why should Microsoft request a clearer mandate from INCITS?  If you don't like the fact the chair offered them "Approve with comments" as a choice, why don't you confront the chair about it, instead of whining to Microsoft? PS: Are you the same Marc who named all the V1 members on Rob Weir's blog? PSS: You should really get outside now then.

  • Anonymous
    July 17, 2007
    Here are a few interesting links related to the Open XML formats US technical committee reaches deadlock

  • Anonymous
    July 17, 2007
    dough mahugh said: "That's also when the membership of V1 increased significantly, as organizations in the US that have hands-on Open XML experience got more involved in the process. As of May, only two of the V1 members (Microsoft and Mindjet) had any experience implementing Open XML, but now there are many companies with hands-on Open XML experience involved in the process. " This is what the chair of JTC1 SC34, a man with 30 years of background/record in standardization says about the abrupt influx of some new members of V1, citing: "A lot of time has been wasted recently by people who understand neither the ISO process nor formal operation of a meeting. I, for one, would like for the INCITS rules to say that one had to be a member of a committee for some period of time -- say six months -- before voting. In the days when V1 was started, of course, all meetings were face to face, and two meetings usually meant six months. V1, currently relying on teleconferences, is having meetings far more frequently than those who drafted the rules envisioned, and so people are becoming eligible to vote before they're really seasoned in the process." ( http://www.ibiblio.org/bosak/v1mail/200707/2007Jul16-202519.eml ) IMHO, this people are playing "the standard game". They should respect the standardization process. dough mahugh said: "If you want to see the original submissions of these comments and the email discussions around them, check out the public V1 email reflector" A better and more direct link for people interested in this process is the file "v1 comments.zip" at http://www.ibiblio.org/bosak/v1mail/200706/2007Jun30-173123.eml ( note: this is a mail file, save it and open it with an email reader [ mozilla thunderbird, outlook, etc. ] ) chris clark said: "Not sure about 171 letters being delivered in support.  It looks more like 1 letter sent in by 171 people." If you are interested, this is one of the 171 letters: http://www.incits.org/DIS29500/in070773.htm It reads: "...Even though this is a form letter from Microsoft I thought I would add this personal touch..." Interesting, uh?

  • Anonymous
    July 17, 2007
    You mean you are paid to be in this camp? Microsoft should have flooded the committee with more puppets in order to be sure to get a majority.

  • Anonymous
    July 17, 2007
    How do you feel about Jon Bosak of Sun stating that: "We wish to make it completely clear that we support DIS 29500 becoming an ISO Standard and are in complete agreement with its stated purposes of enabling interoperability among different implementations and providing interoperable access to the legacy of Microsoft Office documents" (see http://www.ibiblio.org/bosak/v1mail/200707/2007Jul16-081558.eml)

  • Anonymous
    July 18, 2007
    The comment has been removed

  • Anonymous
    July 18, 2007
    Stephane.... Is it really necessary for you to swear and make accusations of bribery just because somebody disagrees with you?  I doubt that adds credibility to whatever the point is that you're trying to make.  Are you childishly confrontational in all aspects of your life or simply emboldened by the great Internet security fence? Just a thought: Decaf

  • Anonymous
    July 18, 2007
    RogersParking, Regardless whether or not you are a Microsoft employee, you owe it to yourself to check out hAL's comment before you address mine. As for hAL itself, he's been pretty much the only person that is posting for MONTHS on MSDN blogs and blogs from people criticizing OOXML (mostly constructive criticism), always trying to tear down any criticism of OOXML with 1) zero argument 2) never providing credentials : who he is, what are his interests. To set the record straight, here is what hAL intentionally cut off Bosak's comment : "Sun voted No on Approval because it is our expert finding, based on the analysis so far accomplished in V1, that DIS 29500 as presently written is technically incapable of achieving those goals, not because we disagree with the goals or are opposed to an ISO Standard that would enable them.". Pretty minor, heh?

  • Anonymous
    July 18, 2007
    The comment has been removed

  • Anonymous
    July 18, 2007
    The comment has been removed

  • Anonymous
    July 20, 2007
    The comment has been removed

  • Anonymous
    July 20, 2007
    Suite à la recommandation d'abstention du comité technique INCITS V1, le bureau exécutif de l'INCITS

  • Anonymous
    July 24, 2007
    The various members of INCITS here in the US have been busy the last two weeks, with the technical committee

  • Anonymous
    July 24, 2007
    The various members of INCITS here in the US have been busy the last two weeks, with the technical committee

  • Anonymous
    August 21, 2007
    The September 2 vote on DIS 29500 is rapidly approaching, and Jason Matusow has posted a wrap-up of where

  • Anonymous
    March 07, 2008
    An article in Computerworld yesterday entitled "ISO officials, others dispute claims that Open XML meeting

  • Anonymous
    March 07, 2008
    An article in Computerworld yesterday entitled "ISO officials, others dispute claims that Open XML