A loose end: Votes versus GDP

by Rick Jelliffe

I had thought that we all would be buckling down to productive work by now, but I see that there still is some attention being paid to the idea that there is a correlation between OOXML Yes votes and the corruption index. The most recent form of this (which I consider plays to racist views) replaces the corruption index with GDP (and then mentions that the corruption index is correlated to the GDP anyway, wink wink nudge nudge.)

So I thought readers might be interested in seeing a quick graph, in which the final national body votes are displayed against the per capita GDP. It is all a bit crude graphically: the horizontal axis gives 87 national bodies, and the vertical axis gives the $ per capita GDP from the World Bank figures. The green triangle gives the data points for each NB. On the 100,000 line above each green triangle is a little icon showing whether the national body voted Accept (blue square), Abstain (red diamond) or Reject (yellow triangle).

(Please ignore the icons down on the horizontal bar, that is just my general ineptness. Sorry it is PDF, I couldn't figure out how to export the long diagram any other way.)

To quickly check the distribution, lets see how the numbers are distributed when we divide into four quarters.

1 (highest GDP)1641

So I would like to make the bold interpretation: national bodies did not vote on per capita lines. A curve could be fitted to describe the relationship, but it has no explanatory power either.

What did they vote on? I have an equally daring view: all sorts of reasons, but mainly just the boring technical ones. Technical people discussed and judged. Standards people reviewed the technical people's judgments and made their judgments.

But that is no fun. Lets entertain the idea that voting was on some subterranean basis, what could it be? I'd say that the more that a national body came from a English-speaking "peripheral" country (e.g. not UK or USA) the more chance that it would not vote "accept". And the more chance that a national body came from a socialist (or previously non-aligned) government (China, India, Cuba, Venezuala) the more chance it would vote "Reject". And perhaps the more that a NB came from a member of Bush's Coallition of the Willing, the less chance the NB would vote "accept": it is a funny place for frustrations with having to follow the American lead to emerge, but I think the international mood for independence is very strong, and perhaps it affects the way people see even these standards issues.

However, the way to get standards that are less US dominated, is for non-US people to participate in the various technical groups that are developing these standards: OASIS ODF TC including OpenFormula TC, Ecma TC45, ISO SC34, the ISO work on PDF, etc.


Rick Jelliffe
2008-05-22 02:05:55
Two other aspects come to mind.

First, one of the features of National Bodies is that while most are NGOs, some are directly part of the government and under ministerial control. India, China, Cuba, Malaysia and so on. For these bodies, there may be some implication that if they vote for a technology to be on the books as an ISO standard, it also carries some idea of domestic endorsement. While I think this is a bad thing (because it is not how ISO itself sees its standards) it may certainly be a factor.

Second, even if the final vote is not skewed by GDP, it may well be that the proclivity of an NB to vote Yes with comments rather than No with comments does have a GDP angle: this is because when an NB votes No with comments they are then obliged under the ISO/IEC JTC1 Directives to participate in the BRM, and this may be outside the budget for the standards body of a poorer country (especially if there is an expectation that comments will be raised by other NBs and dealt with.) ...To go from GDP to corruption is racism, in the absence of hard facts.