Name'n and Shame'n Where the Credit is Due: Setting the Record Straight on (Non)Voting in SC 34
Slashdot It!
DiggThis
Add to Del.icio.us
Monday, October 29 2007 @ 12:01 AM EDT
Contributed by: Andy Updegrove
Views: 5,316
About ten days ago I reported that SC 34, the ISO/IEC JTC1 committee responsible for evaluating OOXML, has been unable to make progress on any of its other important initiatives since the OOXML vote. Why? Because the eleven Observer (O) Members that had upgraded to Principal (P) member status in the run up to the OOXML vote have not bothered to cast a vote (even to abstain) ever since. P Members, you may recall, have more influence over the outcome than do O Members.
In the same piece, I observed that this further confirmed the assumptions of those (myself included) that those National Bodies that had upgraded did so solely for the purpose of voting "Yes" for OOXML, as earlier demonstrated by the fact that of the 11 upgrades had in fact done exactly that. What I had not anticipated was that a key standards committee would now be suffering serious collateral damage when these new members have shown no willingness to vote – even to the extend of simply casting an "abstention," which would suffice to meet the requisite 50% participation among P Members for a vote to pass.
There is more than one way to look at the voting, of course, and Rick Jelliffe thinks that both sides are equally to blame. I don't think that conclusion can stand up, though, once you really look at the numbers.
Rick wrote a piece of his own at O'Reilly.net on the same topic, which he called Name 'em and shame' em: non-voting on SC34 ballots. There, he noted that 12 of the "P" Members that had voted to adopt OOXML failed to vote on a recent proposal, and that 9 who had voted against adoption also failed to participate (two countries that had abstained on the OOXML vote also failed to vote). From this, he concluded:
So both sides are equally slack. What did you expect the result to be? If you expected it to show that the MS stooge countries were pretty bad, while the valiant anti-OOXML forces were pretty good it shows you have drunk the Kool-aid, with all gentle respect. Some-one says something based on no objective evidence, but if it accords with what has been said enough times before, people think “That sounds about right”: but what if what was said before also had no objective evidence, that there is a chain or ripple of make-believe and demonization that merely emotifies foregone conclusions?
Here's where I think that Rick goes off track. He begins his piece by stating that the problem is clearly the new members. But then he shifts the focus, and looks at the voting of all of the members, and not just the new ones. And therein lies the problem.
Here's why: I'm told that the historical success rate for achieving the necessary 50% of P votes has been about 90%. That doesn't mean that every P member votes in every one of the very frequent ballots (several a month), since not every P member is interested in every initiative. That's why the participation requirement is set at 50% and not (say) 75%. It just means that 90% of the time at least 50% did respond.
The problem with SC 34, then, isn't whether those that voted against OOXML (old and new) are failing to vote as much as those (old and new) that voted for it, but whether the new P upgrades are voting at all. It's also instructive to look at how those same upgrades voted on OOXML. As you'd expect, what you see is not that long-term P members have suddenly quit voting, but that the influx of a large number of new, non-voting P members simply changes the math.
When we look at the data from this perspective, we see a very different picture. Here’s how the eleven countries that upgraded from O to P membership in the months (and often just days) before the OOXML voting period closed on OOXML, and also whether or not they voted in the more recent ballot (all data is from Rick's analysis of the voting record):
Upgrades that voted to adopt OOXML and didn't vote later: 7(Côte d’Ivoire, Cyprus, Lebanon, Malta, Pakistan, Turkey, Venezuela)Upgrades that abstained on OOXML and didn't vote later: 1(Trinidad and Tobago)Upgrades that voted against OOXML and didn't vote later: 0
[Rick doesn't mention the other three upgrades, so I assume that they did in fact vote on the ballot he examined. They, and their votes on OOXML, were as follows: Ecuador (disapprove), Jamaica (approve) and Uruguay (approve, with comments)]
That tells a rather different tale, doesn't it? In fact, 7 out of 8 upgrades that voted "yes" without comments didn't vote, while the only upgrade that voted against OOXML apparently did participate in the ballot Rick selected for examination. An abstention, by the way, is a next best thing to an approval vote under the complex ISO/IEC rules.
I'll not express any opinion on whether the above (non)voting record indicates that the National Bodies showing such bad committee citizenship are therefore "Microsoft stooges," but I would submit that the information above does clearly qualify as "objective evidence."
And I'd submit it also clearly indicates the real reason why SC 34 was able to function before the OOXML vote, and now is stuck in the quicksand.
On a more hopeful note, Rick reports that there are ongoing discussions on how the situation can be addressed, and I also agree with his point that there is nothing inherently inappropriate about a National Body wanting to take a greater interest in a single standard. What I objected to in my original posts on this topic was the curiousness of so many countries upgrading at the last minute and voting so unanimously.
And yes, the timing does matter - very much. In a large number of P countries around the world, vigorous public debates were held in open meetings between those that favored, and those that opposed, OOXML. As a result, the long-term members of those National Bodies, even if they knew nothing about document formats before the vote, were able to become well informed by the time their votes were cast.
In contrast, most of the P upgrades occurred at the very last minute – and before anyone who opposed OOXML had a chance to make their opinions known, or to bring any information to the attention of their National Body.
In the normal case, we would assume that an upgrade by a JTC1 member had no significance at all. But when we see stacking and other irregularities in countries that are already P countries, and then see what can only be regarded as the stealth launch of a substantial number of submarine "yes" votes from countries that until then weren't P countries at all, it is not unreasonable to reach a different conclusion.
Even without any Kool-aid at all. Fact of the matter is, I've always hated the stuff.
For further blog entries on , click here