Slashdot Mirror
OOXML Vote and the CPI Corruption Index
Tapani Tarvainen writes "It turns out there's an interesting correlation between Transparency International's 'corruption perceptions index' and voting behavior in ISO's OOXML decision. Countries with a lower score (more corruption) on the 2006 CPI were more likely to vote in favor of OOXML, and those with a higher score were less likely. According to the analysis, 'This statistics supports with a P value of 0.07328 the hypothesis that the corrupted countries were more likely to vote for approval (one-tailed Fisher's Exact test). In other words, simplified a bit: the likelihood that there was no positive correlation between the corruption level and probability of an approval vote, that is, this is just a random effect, is about 7%.' Of course, correlation doesn't prove causality."
I've put up a little website with some initial thoughts, and I'd appreciate feedback from the slashdot community please.
Countries like India and China ... home to over 33% of the world's population - have voted "NO". Countires like Cote' de Ivorie and Cyprus have equal voting rights.
This population index anomaly must be rectified, before the ISO can regain any credibility as an International standards organisation.
If you keep throwing chairs, one day you'll break windows....
http://en.wikipedia.org/wiki/Image:FSM_Pirates.jp
I want to delete my account but Slashdot doesn't allow it.
The questions which are relevant for standardization can, and therefore should, be always decided in a fact-oriented, principled manner. Otherwise the corruption problem will always remain regardless of voting weights.
This is one of the best stories I've seen on Slashdot in months. Actual facts always trump FUD and jumping to conclusions.
Sorry /., but this is the sort of crap reporting that is persistent on the web and (because they're desperate to retain viewers/readers) is becoming the de facto standard in print and media journalism.
Appending "Of course, correlation doesn't prove causality." to the end of an article strongly implying causality in every sense, doesn't absolve the reporter from the false conclusions he/she implies throughout the rest of the article.
That the correlation was run at ALL implies that someone was 'looking for something' - suspect 1. The layer upon layer of dependent statistics leading to a very authoritative-sounding "the likelihood that this is a concidence is 7%" makes it sound very scientific and accurate - suspect 2
Sorry, this is FUD passed off as news supported by phony statistics.
-Styopa
First off, it just goes to show you how beholden the US government is to Microsoft money. Second of all, it shows you why the US branch of a certain worldwide NGO based in the US has been so slow in adopting the methods of its parent...because, oddly enough, International has seen fit to partner with Transparency International. Your average USian knows nothing of TI, and those that do tend to poo-poo them as being insignificant.
Brazil voted no, and we are a curruption paradise :)
Actually there was a strong fight in ABNT (brazilian standards regulation group) as happens in the government. The brazilian agencies and federal govern always used windows in a large scale from desktop to servers, while universities used most Unix and Linux. Current federal govern supports Linux, but there is a big resistance in sectors that always used windows and often there are problems with licitations (govern auctions) imposing a specific type of software.
For example, you can auction for "Microsoft Office" or "a Office Suite". Most states already fobid the first option in law, but at federal level it's not forbidden, even that there is a recommendation to avoid this kind of situation. But even with laws, often happens situations where the auction is so specific that it could only have a winner, when police does some investigation finds that someone got a "deal" to make sure company X wins.
Recently in my state they found a guy stealing mail stamps! Millions of them. He bought and kept the money for him, because the legislative house had a deal with the mail company to send letters. When found he buried the stamps in his house garden. Can you belive it?
Most research in the social sciences considers the threshold for statistical significance below .05. Since this is above, few would have confidence that this result is not random chance. So, reporting this result, while informative if you are trained in statistics, is likely misleading to the average reader...
In this particular case, there's plenty of evidence of irregulaties in how national standardization organizations handled MS-OOXML, so the causal relationship is not really at issue.
"Oh, people can come up with statistics to prove anything. 14% of people know that."
Given the inappropriate use of the Fisher's test, questionable use of a one-tailed model and p > 0.05, I'd start with worrying about having proven correlation.
What I'm listening to now on Pandora...
Come on people, we've been over this already!
If you look at the scientific studies, correlation is so closely correlated with causation that it's safe to say that one causes the other.
Check the stats for yourself.
Transparency Intl should repackage and sell the corruption index as a Price+Services catalogue for each country. Would save a lot of wasted time and effort....
Obligatory old joke:
"What kind of a woman do you think I am?"
"I thought we had established that, and were negotiating the price!"
If I don't completely understand the methodology in collecting the data and what they are doing to generate the final number, then I don't trust the article. This article had me knee deep in statistics that seemed overly complicated for what they were trying to say. This puts me in the high yellow on the BS meter.
I read the internet for the articles.
Another intersting point coming out of all this the role of ECMA International
formerly European Computer Manufactureres Association - dont see many of them around these days).
ECMA is fully accredited by ISO and in ists search for a new role as a standards body did
a nice job producing a standard for the orphaned Javascript ( except for changing the name
to the disease like ECMAscript).
However since then other "standards" developed by ECMA have been:-
-- the programming language C# ( C "sharp")
-- a Common Language Infrastructure (CLI)
-- a CLI binding for C++
-- Office Open XMLOffice
Anyone spot a pattern here?
The other problem is that ECMAInternational is essentialy a club of computer software and
hardware manufacturers and unlike national standards organisations (ASA, BS, DN etc.)
does not have any public interest mandate; it exists only to serve its members and
to join you need to be a large software or hardware manufacturer.
I have no problem with any industry forming a club to standardise things among themselves
but for an industry association to be the main sponser of an ISO standard seems plain
wrong.
Microsoft for one seems to have spotted an ideal vehicle for turning proprietary products into standards.
Old COBOL programmers never die. They just code in C.
The game.
...those with Linux bumper stickers were three times more likely to believe their choice in operating system should be defended with religious fervor. How curious! Of course, correlation doesn't prove causality.
This statistics supports with a P value of 0.07328 the hypothesis that the corrupted countries were more likely to vote for approval
Since when does a p-value of 0.07 reject the null hypothesis?
Also, I love preaching the correlation/causality mantra as much as anyone, but it seems a little superfluous here - does anyone really think that Transparency International's CPI actually caused those countries to vote in a certain way?
sic transit gloria mundi
I can just imagine the journalist's process for choosing the CPI index. 1: prove correlation between OOXML support and terrorism. Unable to find any correlation. 1: (second try) prove correlation between OOXML support and stealing candy from babies. Unable to find any correlation. 1: (third try) prove correlation between OOXML support and CPI index. Correlation found. 2: write scientific sounding article about said correlation. 3: ? 4: profit!
So what if correlation doesn't imply causlity -- the point isn't whether being corrupt causes them to support OOXML. The causality issue of interest is whether the amount of money M$ dumped into countries (above board and under the table) causes OOXML support. It's somewhat Bayesian where corruption is the bias factor.
um, in the statistics courses I took, a p less than .05 was needed to show a meaningful correlation, and p less than .01 was needed for a strong correlation.
p
and of course decreasing numbers of pirates directly correlate to global warming. http://www.seanbonner.com/blog/archives/001857.ph
or how about the one where driving drunk gives a higher probability of surviving in a an accident versus a non-drunk person in the same situation. (yes I know that the drinking may increase your chances of getting in an accident at all which could nullify the benefits of being drunk in that accident, but that goes to how how people can misuse statistics)
and how do we get less than signs in out posts?
"OOXML Vote and the Corruption Perceptions Index Corruption Index"
OK, that's it for now - I have to run down to the ATM machine and put in my PIN number...
Lets make it clear that it is a corruption perception index. Not an actual corruption index. An exemple of the difference is that fighting corruption usualy increases the index at short time.
Also, lets make it clear that the index isn't that precise, and its data isn't that reliable. Transparency International used to have a disclaimer at their FAQ that basicaly said that, unless you completely comprehend the methodology, if you're using these numbers to do anything important, you're insane. Comparations between countries aren't accurate, as aren't comparations of different years at the same country. Basicaly, the average Joe should stay away from the data.
Rethinking email
It is high time the customers band together and fund an institute that will be chartered to level the playing field and foster competition.
sed -e 's/Chuck Norris/Rajnikant/g' joke > fact
This is one of the best stories I've seen on Slashdot in months. Actual facts always trump FUD and jumping to conclusions.
In the end, the study shows a correlation between corruption in a country and a country's approval of OOXML. That's interesting, but there are more direct and useful studies and actions. Those places that voted "yes" should be embarrassed, not because "yes" was wrong, not because there's a statistical correlation between "yes" and corruption.
It was easier to study OOXML and condemn it with facts directly. Plenty of people did this and published it. Those studdies point to places where the standard was woefully incomplete, contradictory and impossible to implement in a reasonable way.
It was also easier to prove real corruption where it occurred. Vote buying and stacking was proved at all levels.
Finally, it's better to spend time on a remedy than it is on interesting studdies. OOXML was rejected, but further action is required. M$ should be punished for their corrupt practices even though they failed. ISO needs to protect itself from that kind of behavior or it will lose public confidence. M$ was blatant this time, perhaps because they would like to destroy ISO itself. Next time, they might not be so obvious but should not be given the chance. Usually, when you prove malice like that you don't get another chance to screw things up. Those who took bribes should be banned forever and M$ should not be allowed to participate in new standards for years. Their participation in the future should be dependent on all people responsible being replaced.
DMCA, Hollings, Palladium. What might have sounded like paranoia is now common sense.
Correlation doesn't prove causality... are you suggesting that it's the other way round here? That embracing M$ products may lead to sloppy ethics and corruption? Not that I find that implausible.
there is no way to fix capitalism. money can buy you influence. even here. if you have enough money that you can pay 40000ppl worldwide you can alos afford to pay a few more to subvert openiso.org if it ever becomes necessary. it will make it hardare but it will not make it impossible. most likely it will not become necessary since they just buy the governments and tell them to ignore openiso.org at all. see all the legislation that is in favor of corporations like microsoft and others that is going on in the EU, the US and everywhere... most political parties in power there are already in line with capitalist interests.
what would make such an approach useful is that in the process of finding sane standards the people involved will learn about the deficiencies of the capitalist regime. capitalist regime is corrupt per definition: it means that money can buy you what you want. which is plain corruption. the more money you have the more human labor and natural resources you can command...
greetings from vienna,
mond.
1) Ask your friends (people that think like you) what they think (call this process a diverse survey of experts)
2) Obscure the fact that it was a Popularity-Contest by using an official sounding acroynym like CPI (Contest Popularity Index)
3) Now compare CPI numbers to other numbers, and produce percentage numbers (percentages also sound very official)
4) Profit!!! as you have now created Real-Proof(TM) to support your preconceived beliefs (I knew it all along, and feel so superior)
Some people use statistics like a drunk uses a lamp post; for support rather than illumination.
While I've been thinking for quite some time about launching a competitor to ISO, it's true that you inspired me with your humorous posting to name it "OpenISO.org"... if you like I'd be happy to give you credit in some way on the OpenISO.org website.
In other news: A recent survey of 1200 Linux users confirms that more than 80% are pasty-faced, overweight men. "This is a significant overrepresentation" an anonymous scientist reports. The only possible conclusion to be drawn is that Linux is the single most significant factor in obesity and albinoism. Also, it seems using Linux is presumable linked to gender chromosome changes. An information campaign is being considered by the Department of Health.
A two-tailed test should be used, not a one-tailed test, doubling the p value to .14, and thus there is a 14% probability that support for the said association in the data is pure coincidence.
The one-tailed test assumes as null hypothesis that corrupted countries are more likely to vote no. For sure this is not what we consider the default ? The default is that there is no association between corruption and voting behavior, and thus a two-tailed test is required to reject this null hypothesis.
If you read how that "index" is created basically they go out and ask a bunch of people who likely have no idea where Iraq is which countries they think are more likely to be corrupt.
So if they haven't been told good things about your country on their TV in the last few years you're going to get a bad score.
I love how it rates Japan as one of the worst places, when it is actually a richer country than the US / Aus / UK. Assuming of course that you believe wealth solves corruption.
I can see a lot of statistics 101 graduates here, eager to show their mad skills by homing in on that p value.
As said above, 0.05 is a common treshold -- in academic discourse -- of whether it is worth continuing the discussion or whether the theory should be scrapped outright. In this case, the relevant question is not how much p exceeds that arbitrary value, but whether it is worth continuing the discussion. Should the theory be scrapped outright? Or should we be concerned that corruption had something to do with the process?
Given this article together with facts that have come up (case Sweden), I personally think it's very much worth looking deeper into possible corruption in the process. It would be surprising if national organisations were affected in strict correlation with their country's CPI index, but it provides a starting point for more detailed investigation.
Btw if you wanted to, you could massage that result to show a much more dramatic correlation by e.g. including unconditional approvals only, which I think wouldn't be completely unjustified either.
I love the entire premise: if you vote FOR anything MS related... your country is corrupt. If you vote against it... your country isn't corrupt.
Pretty... um... "interesting" worldview you FOSSies have going on there.
Teh Lunix is all about choice... you can choose any application teh FOSSies want you to. But don't you DARE choose Microsoft.
"Correlation does not imply causation" is the phrase you're looking for. Causality is the belief that world state derives from repeatable, understandable rules and prior state. Make note: causality is a belief, not the description of physical law; it was the stance taken in ancient times as opposed to the idea that various things (for example, lightning) were holy. This was a significant stance at the time, and from it evolved beliefs like Determinism. Causality was a position held by certain Greek and Hindi branches, and indeed started several wars.
StoneCypher is Full of BS
"residents' viewpoints correlate well with those of experts abroad. In the past, the experts surveyed in the CPI sources were often business people from industrialised countries; the viewpoint of less developed countries was underrepresented. This has changed over time, giving increasingly voice to respondents from emerging market economies." http://www.icgg.org/corruption.cpi_2005_faq.html
It's not like they just ask a panel of Americans how corrupted Iraq is.
As for Japan, they have some 80 000 yakuza and the government and municipal governments are well known for rigging construction deals.
It's not a perfect system but it's not bad either.
"Of course, correlation doesn't prove causality." - Except when we're talking about global warming..... right?
Je fume. Tu fumes. Nous fûmes!
While you may have hit a sore spot about the open source movement (which seems to have no problem with proprietary software when that software is reliable and powerful), the free software movement has never supported "choice" where any of the choices are non-free software. Prioritizing "choice" makes equals out of all possible choices which, in this context, makes it impossible to distinguish between software that denies and supports users' software freedom. Proprietary software always denies users software freedom and is therefore always objectionable. The FSF is quite clear about this; consider what they say about the "Sincere Choice" program which is a response to Microsoft's "Software Choice" program:
Digital Citizen
I think you might win an award for missing the point not only of the story, but also of the summary and the headline. There is no premise... there was a correlation (not causation) found between countries found to be corrupt and countries backing MS on OOXML. The vote has absolutely no bearing (I hope) on the corruptness index.
haha
Copyright infringement is "piracy" in the same way DRM is "consumer rape"
Thing is that most of those 53 countries that voted for OOXML rarely vote at all, and -- as this article points out -- many of those who suddenly decided to join the conversation and vote yes have a high perceived corruption index.
As a number of people have pointed out, 7% is above the 5% normally needed to consider the correlation "proven", but it's still low enough that I'm willing to let the fog of possible corruption hang over those countries that voted yes.
OS Software is like love: The best way to make it grow is to give it away.
Not really. It comes from a rounding of 4.6%. The 4.6% (actually 100-95.4%) comes from 2 standard deviations, in a normal distribution.
Someone had a causal theory that corruption was responsible for certain voting patterns. They invented a test that could perhaps falsify the theory. The test did not falsify the theory.
Is it proved? No. But the correlation (failure to falsify) makes it a reasonable thing to believe unless/until it goes down to a better theory or a better test.
More recently Japan (very pro-whaling) went to the South Pacific and Carribean with a briefcase full of cash and bought votes to get control of IWC. Little country votes come cheap.
Engineering is the art of compromise.
Are you seriously claiming that it's actually good that nobody can get ahead in socialist societies, because it prevents corruption? That's absurd. When people can't get ahead by playing fair, like they can with a capitalist economy, they'll start playing dirty.
How many times will socialism have to fail horribly before people realize that it just doesn't work? I'd rather have a working economy than a broken economy with high ideals.
Ok I had a look at TFA and I decided they were running the wrong test. What we're actually looking at is whether the corruption index is dependent on the vote. I fired up R and ran a quick kruskal-wallis test (1 way non parametric analysis of variance): > ct kruskal.test ( ct$CPI ~ ct$Vote) Kruskal-Wallis rank sum test data: ct$CPI by ct$Vote Kruskal-Wallis chi-squared = 21.531, df = 3, p-value = 8.166e-05 So there is a significant difference between voting behaviour by corruption index. However this doesn't tell us the size and direction of the difference, so I ran a quick notched boxplot of vote against CPI: plot(ct$CPI ~ ct$Vote, notch=T) The graph is here. If you look at where the notches in the boxplots don't overlap, that will tell you where the significant difference lies. It turns out that the median corruption index score for the yes votes is significantly lower than for all the other groups. This indicates that there is an association between corruption perception and voting yes (i.e. yes voters are seen as more corrupt). This appears to be a very strong significant difference and not the marginally significant difference as in TFA at all.
When looking at that graph, remember to group "Approval with comments" with "Approval", since their effect is the same.
"Nine times out of ten, starting a fire is not the best way to solve the problem." - my wife
GDP has nothing to do with corruption, for example, New Zealand is one of the least corrupt nations and has a low level of GDP compared with the USA, which is slightly more corrupt, yet has a very large GDP.
- Kaos games and encryption systems developer
A "working system" is one that works in spite of people of bad character and one that does not depend on people of good character.
- you earn money by not working and letting others work for you. (e.g. you own a lot of shares of some companies)
- you are in the upper management of a company and let the shareholders pay you a lot of money for finding ways of exploiting the working people. (accomplice)
- creating artificial demand for something that people do not need by means of marketing
- deliberately breaking thing so you get payed to fix it. (e.g. lobbying for yet another war)
- creating an artificial shortage of knowledge and information by using so called "intellectual property"
- micr$oft
- etc...
if you look close enough: not playing fair is the only way to make money within capitalism except maybe for winning the lottery and maybe getting the money from your parents who got it by not playing fair...greetings mond.
Thanks a lot for pointing out this potential issue. I've just registered open-iso.org in order to be on the safe side in case of such censorship becoming an issue, so that OpenISO.org can change its primary domain name to that if necessary, without risk of domain-squatter trouble.
With this kind of set-up, I believe that all the right incentives for long-term stability of integrity are provided.
First of all, the problem of corruption is not specific to capitalism. Corruption is older than capitalism, and there is no country with a non-captitalistic economic system where corruption is not a very significant problem. Therefore I believe that it is fair to say that corruption is fundamentally a problem of human nature, not a problem of capitalism.
However there is a relationship to the ideology of capitalism in that this ideology is based on the assumption that human nature cannot be changed. In addition, much of the scientific research which keeps capitalism going in practice is also based on this assumption.
I this assumption true? I personally believe that this is not the case. There are fascinating historic accounts of so-called "revivals" of Christian religion which show that faith in Jesus Christ, when combined with the right kind of prayer, does in fact have the effect of very dramatically changing human nature for the better.
Now I'm not suggesting that it would be appropriate to pray without also taking intelligent action by designing OpenISO.org so that the organization can be expected to act with integrity independently of any assumptions on whether the people in the organization personally have a high standard of morals or not. My goal with the above argument is just to explain why I consider all arguments to be invalid which are based on the assumption that corruption is an unfixable problem.
Note that the goal of creating a corruption-free standardization organization is not about denying corporations any influence on standardization processes and the resulting standards. The organizational goal is just to have a standardization organization which provides an appeals path for every decision with a guarantee that the appeal will be decided by means of a process with is fact-oriented and as corruption-proof as possible.
I admit that "as corruption-proof as possible" is not the same as "perfectly corruption-proof", but I claim that "as corruption-proof as possible" is much, much better than what is provided by the currently-existing ISO/IEC processes.