Microsoft Reports Victory in Preliminary ISO Ballot for Office Open XML
3:15 pm ET September 4, 2007 - The International Organization for Standardization finally issued a statement this evening, Geneva time, saying a preliminary vote to publish Microsoft's Office Open XML as an international standard failed to meet the requisite two-thirds majority of support to attain fast-track status. The standard itself has not failed, contrary to many reports this morning.
But for OOXML to emerge beyond the draft status, Microsoft now has until February - not March, as was earlier reported here and elsewhere - to address the comments of members who voted both in the affirmative ("yes, with comments") and the negative ("no, with comments"). At that time, if members believe Microsoft has addressed those comments adequately, they may change their vote to outright approval.
As the ISO account this afternoon stated, "The objective of the meeting will be to review and seek consensus on possible modifications to the document in light of the comments received along with the votes. If the proposed modifications are such that national bodies then wish to withdraw their negative votes, and the above acceptance criteria are then met, the standard may proceed to publication."
Sadly, even the ISO's account of the vote requires us to take out our calculators - literally no one involved in this process thus far has presented a simple rundown of the tally. However, the ISO's account this afternoon does clear up a very important point about the balloting process which has been misconstrued elsewhere (and about which BetaNews hasn't said much to date until we felt we could clear it up): The phrase "participating" or "P-class voters" refers to member countries who also serve on the ISO/IEC joint committee JTC 1. There are 41 member countries who are part of JTC 1, among 104 total members. The countries who recently applied to have their voting status "upgraded to P-class" are those who urged the ISO to make their vote actually count, because despite the 104 number, it's the 41 who actually count.
Among those 41 P-class members, 22 votes were positive, or "yea," in favor of fast-track status, while 11 votes were negative or "nay." The remainder, we can assume, were for "yes, with comments," which would bring the total number of positive voters up to 74%, in line with Microsoft's victory declaration earlier this morning.
6:15 pm ET September 4, 2007 - A Microsoft spokesperson contacted BetaNews this afternoon with a somewhat different account of the vote breakdown than the ISO's own official account. What makes it noteworthy is that it actually paints a less advantageous picture for Microsoft than its own triumphant proclamation of early this morning, although the spokesperson said she would take further steps to verify the numbers she was given.
Based on the spokesperson's information, nine members of the P-class voters did abstain from the process. Of the remaining number who voted either affirmatively or negatively, 17 members (not 22) voted yea, while 15 voted nay. If that math is correct, Microsoft would need to sway four voters, not just one, in order to meet the requisite two-thirds majority among the non-abstaining voters.
But when the spokesperson agreed with information we repeated saying only one voter needed to be swayed, she agreed that both sets of information did not correlate, and that she would do further research and get back with us.
9:30 pm ET September 4, 2007 - Our Microsoft spokesperson did indeed contact us, with what she was told is a confirmation of the final vote tally, using the above math: 17 yea, 15 nay, 9 abstentions. When viewed from that perspective, it suddenly doesn't seem so much like a 74% victory.
While this isn't exactly total victory for Microsoft, it also isn't tragic defeat as has been trumpeted elsewhere today. In fact, not only is DIS 29500 very much alive and well, this could be exactly the result Microsoft wants and needs: If it had received just one more full "yea" vote, giving DIS 29500 fast-track status for approval, the outcry among both critics and lukewarm supporters over the ISO's failure to take a multitude of technical comments into account, would have been tremendous.
This way, Microsoft gets the opportunity to be seen as cooperating with the international community, with its goal to sway only a single "yes, with comments" voter its direction come February. If that happens, it will be much more difficult for critics to ridicule the standards process as defective or fraudulent.
9:14 am September 4, 2007 - Microsoft's carefully worded statement this morning appears to indicate that 51 member countries voted "Yes, with comments," with 18 countries either voting in the negative or abstaining.
The comments were alluded to by this sentence: "Along with their votes, the National Bodies also provided invaluable technical comments designed to improve the specification." It went on to say members indicated they will render their final approval in March 2008 once those comments have been adequately addressed.
Sources are reporting this morning that a 74% positive vote is just 1% short (or one vote short) of a three-fourths majority, which would have given draft specification DIS 29500 "fast track" status. Another report - slightly conflicting - says Microsoft only received 53% support among the "P" group ("participating," or fully qualified) of countries, which is short of the two-thirds requirement necessary for fast track status. There will likely be multitudes of conflicting reports throughout the day prior to a formal ISO statement, and BetaNews will report those from generally reliable sources.
Linux Foundation attorney Andrew Updegrove, who has been following the standards process, was apparently told the 74% news and proclaimed the vote a failure for OOXML, since it was not apparently not actually adopted on Sunday. Updegrove characterized Microsoft's statement this morning as "putting the best spin it can on the results," explaining that some of the 51 votes may not actually count as "participating" under ISO rules.
Similar proclamations of failure were made for OOXML's passage through the ECMA standards board <!article id="1185379787">as recently as late July. However, that US standards body did eventually conclude Microsoft had addressed its members' comments to their satisfaction.
If indeed the ISO voted on Sunday "Yes, with comments" as Microsoft has indicated, it means the company may have six months to address technical issues raised by the National Bodies (voting ISO members representing their respective countries' standards organizations), before a vote for final ratification.
Some countries' concerns may actually be quite serious, including mandates that the company provide better "mappings" between the older Office formats and OOXML, and between the already standardized OpenDocument Format and OOXML; as well as provide assurances that Microsoft will not enforce its patents on elements that play a role in OOXML, against competitive adopters.
Addressing those concerns could mean making changes to the format specification itself, which may at some point have to be retrofitted to Office 2007, as well as any other programs - such as Corel WordPerfect - that may implement OOXML support in the meantime.
Microsoft noted the "unprecedented level of participation in the standardization of a document format," which ODF supporters last week noted with some skepticism. Yesterday, Updegrove was so certain of OOXML's pending defeat that he <!external href="http://www.consortiuminfo.org/standardsblog/article.php?story=2007090315253367">already published its epitaph.
"Rather than simply voting 'no,' these National Bodies played by the rules," Updegrove wrote, believing the preliminary ballot would be defeated, "and did what they are supposed to do: They took the time (a lot of time, given the length of the OOXML specification), during an unfairly brief period, to do what Microsoft and Ecma should have done to begin with, and didn't: properly vetted the specification to make sure that if it is offered to the world with the blessing of ISO/IEC JTC1, that it has met minimum quality standards to be entitled to bear that designation."
On Sunday, ODF system developer and OOXML adoption supporter Dennis Hamilton, prior to knowing the final vote tally, urged both supporters and opponents to <!external href="http://orcmid.com/blog/2007/09/oox-odf-oh-outrage-stridence.asp">work together toward some form of interoperability, no matter the vote's outcome.
"We are seeing conversations and posturing reminiscent of the cold war, the 'axis of evil' proclamations, and other excesses of global politics," Hamilton wrote. "You'd think the stakes were civilization as we know it. I suspect many of the participants believe exactly that. All I can say is, 'Follow the money.' Then get a grip."