The Standards Blog

OOXML Vote Tracker and Calculation Guide

OpenDocument and OOXML
Updated:  (8:45 AM EDT 4/1):  OOXML has  been adopted

Updated:  (1:45 OM EDT 3/31):
Reuters has just reported that ISO will not announce the results of the OOXML vote until Wednesday April 2

Updated (3:30 PM EDT 3/29): 
Unless thus-far unannounced votes that were formerly "approve" or "abstain" switch to "disapprove," it appears that OOXML will be approved.  See details in the cumulative "updates" section below


Like many I'm sure, I'm trying to keep track of the votes on OOXML as they become known.  I've set up a spreadsheet where I'm recording votes as they become known, whether they are formal and confirmed, or coming to light from other sources, and therefore to a greater or lesser extent possibly not accurate, what the sources are, and any associated comments (mostly from Pamela's articles at Groklaw, the most recent of which is being updated with new votes as news comes in to her).  You'll find the most information about specific country voting there, and at several of her prior blog posts, including this one, this one, this one, and this one.

For the benefit of those that want to get a quick look throughout the weekend, I'll post the running tally here of which votes have switched, what the net change has been, now many votes have come to light, and how many remain to be announced.  It is likely that it will not be possible to know the final vote until all votes are in, due to the complicated, double test way in which the vote is counted, which is complicated by the fact that the final number of abstentions, and whether they move from "yes" or "no" votes, can decrease the number of votes that need to switch to "yes" votes. For that reason, I also include an explanation of how the omplicated two-part test for approval will be calculated.

You may also want to read my last blog entry, which discusses the impact (or non-impact) of a vote to approve OOXML, called The Future of ODF and OOXML.

I.  Updates:

March 31

1.  France moves from disapprove to abstain
2.  Norway files protest with ISO relating to its own vote, asking that it be suspended pending an internal investigation (the suspended vote had previously changed from disapprove to approve)

March 30:

1.  Australia and Malaysia stay as abstain (Malaysia position reported at Open Malaysia)

March 29:
1. I've now learned of several other sites that are tracking the results as well.  They can be found at Open Malyasia and Command Line Warriors
2.  Canada stays in the "disapprove" column (private email)
3.  The United Kingdom and Ireland have each now switched from "disapprove" to "approve"  (private email)
4.  New Zealand will maintain its "disapprove" vote; watch for a press release at www.standards.co.nz (private email)

March 30:

1.  Confirmation of the New Zealand position per anticipated press release
2.   France stays in the "disapprove" column (Open Malaysia, with this link)
3.  Further confirmation of South Africa: (Open Malyasia, with this link)


II.  Vote Tracker

Here is the current tally, as I have it.  Please let me know of any updates, new votes, inaccuracies, etc., and I'll post them promptly.

A.   Principal Countries

Gains:

No to Yes:    Czech Republic, Denmark, Korea, United Kingdom, Ireland

Abstain to Yes:  Finland,

No to Abstain: France

Losses:

Yes to No:  Venezuela

Yes to Abstain:  Kenya

Abstain to No: None so far

Limbo:  Norway - Norway had switched from "Disapprove" to "Approve," but has now filed a formal request with ISO asking that its vote be temporarily suspended, pending completion of an investigation

Reported with no change:  11

Chile, Germany, India, Poland, South Africa, USA, Canada, New Zealand

 

Net Change: 4 (5, counting Norway) [Five more yes votes needed, subject to what happens with abstains]

Abstain:  1 net change

Unconfirmed:  Trinidad and Tobago from Abstain to "approve" (see comment below from reader); Italy maintains abstention

B.  Observer Countries

Gains:

No to Yes: None so far

Abstain to Yes: None so far

No to Abstain: None so far

Losses:

Yes to No:  Cuba (see note below)

Yes to Abstain:  None so far

Abstain to No: None so far

Reported with no change:  10

Romania, Brazil

PJ reports that Cuba was originally recorded as an approval last summer; it has now voted to disapprove, and says that its original vote was to disapprove as well, and was inaccurately recorded
C.   Total Votes Count
Last summer:  87

Currently reported:  23

III.   How to Calculate the Final Vote

Step one: 
Count all P votes cast, and determine whether at least 50% of P members have voted 
If yes, then proceed to next step   If no, then the resolution fails
This step has already been satisfied under the vote last summer

Step two: Subtract all abstentions (in other words, throw them out)
This will stay in motion until we have the full, final vote
 
Step three: Determine whether 2/3's of the remaining P votes are to approve 
If No, then the resolution fails   If yes, then go to step four
If there is no change in abstentions, a net change five "disapprove" votes must change to "approve"

Step four: Determine whether 25% or more of the total ("yes" plus "no") votes cast by P Members, O Members, and other members of ISO/IEC JTC1 are "No" votes
If yes, then the resolution fails   If no, then the resolution passes
Note 1:  As you can see, Principal member votes are far more important than Observer Member votes, since the votes of P members count under both tests, while O members count only under the second.   Note 2:  Note also that an abstention is the next best thing to a vote to approve, since it avoids another vote in the no category, which can't exceed 25% of all votes cast.

Note 3:  A total of 87 National Bodies registered votes of approve, disapprove, or abstain in the voting period that ended last September 2; of those, 41 were P Members, 29 were O members, and 17 neither.



For further blog entries on ODF and OOXML, click here

sign up for a free subscription to Standards Today today!

 

Comments

Permalink
Denmark vote Yes (it was No) http://www.ds.dk/4225 http://www.groklaw.net/article.php?story=20080328090328998 Norway vote Yes (it was No) http://www.idg.no/computerworld/article92563.ece Trinidad and Tobago vote Yes (it was Abstention) http://ttcs.wordpress.com/2008/03/22/trinidad-and-tobago-bureau-of-standards-votes-approval-of-dis29500/ UK vote Yes (it was No) http://www.channelregister.co.uk/2008/03/26/bsi_vote_yes_ooxml/ Brazil vote No http://www.groklaw.net/article.php?story=2008032520072575 Belgium Abstention http://www.datanews.be/fr/news/90-57-17395/la-belgique-s-abstient-a-propos-d-ooxml.html Cuba vote No (ISO counted his September vote as Yes ... it was No) http://www.groklaw.net/article.php?story=20080324121844682 Poland vote ??? (vote Abstention, but ...) http://www.groklaw.net/article.php?story=20080322203811784 Netherlands Abstention http://www2.nen.nl/nen/servlet/dispatcher.Dispatcher?id=253848&parentid=000009 Italy Abstention (no link) for sure Not for sure (no link): Canada probably vote No Slovenia probably vote Yes (it was Abstention) China probably vote No Argentina probably Abstention New Zealand probably vote No Thailand probably vote No

Permalink
1) from: http://www.iso.org/iso/pressrelease.htm?refid=Ref1070
"and no more than 1/4 (i.e. 25 %) of the total number of national body votes cast negative"

so:

**
Step four: Determine whether more than 25% of the total votes cast by both P and O members are "No" votes
**

##################

2) from: http://www.iso.org/iso/pressrelease.htm?refid=Ref1070
"and 26 % of national votes cast being negative."

So:

September vote:
Yes: 51
Abstention: 18
No: 18

1800/(51+18+18) = 20.69%

1800/(51+18) = 26.09%

So:

**
Step four: Determine whether more than 25% of the total Yes and No votes cast by both P and O members are "No" votes
**

The ISO press release reads the same way my step four does, but you're right - the math would not indicate that it failed that test unless the abstentions don't count for this test, as well as for the P member test.  I'll make the change.

  -  Andy

Noted and changed. 

Cheers.

Britania, however, will Rule once again if it stays in the "disapprove" category.

Indeed has come down to each P vote being very precious.

Mailing list flamewars have already broken out here in the UK between prominent IT people so I fear the BSI has said "Yes" to the broken, not really XML, not really open, MS Office format. The BSI itself is being quiet, so it seems that people say "No" loudly and "Yes" sheepishly.

For us Brits, it is pretty sad that we do whatever big US monopoly wants, whether it makes sense or not, just like Iraq, we knew it was rubbish but we jump to our doom with Uncle Sam anyway.

What I think it is interesting is how Microsoft has managed to become the flag carrier of the US IT industry; US interests are seen as aligned with MS interests. SUN, IBM, Google and Redhat are just as American, if not more (does MS even pay that much tax in the US anymore?) but they do not get the special treatment that MS gets. I suppose the difference is that Microsoft is the desktop, and among less enlightened folks, that is it when it comes to computing.

What I think is most disappointing is how divided the EU has been, unlike in some other International initiatives, the extreme hastiness of this process has not allowed us to work out a common position for Europe. When so much is done at a federal level, it would be utterly ridiculous for say France and Ireland to have governmental documents in ODF while Germany has them in OOXML. It may even be illegal for that to be the case. Anyhow, the European Commission will no doubt look into this, and it would possibly want to take control of of European representation after the national governments have hammed it up so badly.

Permalink
Please, link in the article those who have already made great efforts in trying to gather all the information. I'm thinking, in particular of  OpenMalaysia (http://www.openmalaysiablog.com/2008/03/the-last-lap.html"), but there may be others.

/Mikael

And the opposite should also be true (them linking back to me); the time stamp above updates every time I update, by the way, and this post has been up longer than it seems.  It was also created without knowledge of, or use of,  the data in the other folks that are doing the same thing that I'm only finding out about now.

  -  Andy

Permalink
Andy,<br />
<br />
OpenMalaysia has a nice table updated quite often.&nbsp; I think they have the most complete and up-to-date data currently available.<br />
http://www.openmalaysiablog.com/2008/03/the-last-lap.html<br />
<br />
Please note also that the second criteria is based on all the votes, i.e. the Yes and No (Abstain is not taken into account) by the 87 ISO/IEC &quot;members bodies&quot; that voted in Sep-07 (41&nbsp; JTC1 P-members, 29 JTC1 O-members and 17 ISO/IEC members bodies that are neither P- nor O-members of JTC1)<br />
<br />
Luc Bollen

Yes, someone pointed me to Yoon Kit's chart this afternoon after I posted this (I've added a link to it in the update above); it's quite a bit more elegantly done than mine.  There's still something to be said for having two tracking sites, each about a dozen time zones apart, as one of us can be updating one of them while the other is asleep on the other side of the world.
 
  -  Andy

Permalink
If indeed MS bribes, extorts and cheats it's way to becoming a "standard", the solution is simple.  Here's the mandate for government documents.

"We mandate that all documents must be saved in an ISO standard format.  The standard format must be VERIFYABLY certified at >95% complete by two or more independent software vendors".

There you go, since nobody is compliant, only partially compliant at best, MS wins the battle but loses the
war.  They can't buy or bribe or bully or bafflespeak their way past the second line.  Where and who certified that the application you want us to pay through the nose for is >95% compliant to the ISO "standard". 

For the yes countries, well, enjoy.  In the long run though, we can hope that if it passes, ISO is thoroughly re-organized, OOXML is suspended as a standard, and MS pays the price as the nuts and bolts of how they made this happen see the light of day.

TripleII

Permalink
Ummm,

Doing the math on the last vote. ISO state failed both criteria.  Only way that would work is if abstain votes don't count.

That means Doing the math this time. 14/55 =25.454545 = >25% No

Thats Fail, unless I am mistaken.

The operative words seem to be Votes Cast. Abstain means no vote cast.

Victory!
MrCopilot

Permalink
Andy,

it seems likely that there will be at least one serious challenge against the registered vote of a National Body.

Let us pretend that  the vote in favour of approval was determined by the margin between a single P member moving from Abstain to Yes.

What rules and procedures are there to forestall the decision by ISO until such time as any legal challenges to a NB's vote have been resolved?

 

Unfortunately, I'm not familiar with the rules relating to appeals, although it looks like I'll have the need to learn about them now.  It may be that now that the vote is done that Alex Brown may start answering questions (or at least giving his interpretations) like that again.

  -  Andy

My bet is that ISO will consider all votes final.  If the authorized representative of an NB sent in a vote and the country later wants to argue that s/he was not authorized to do so, then that's a local problem.

Permalink
I participated in OOXML process in 3 countries and closely followed all global discussions.

I can ensure the big majority of NBs simply didn't follow the voting rules so I am leaving this question that I want to see an answer.