For some time now, a lot of quite well educated people (as in university degrees and such) have been saying to me that the issues surrounding voting machines aren’t that serious. Sometimes they even accuse me of being overly sensitive or even paranoid about these issues.

Usually my reply consists of explaining them that using a black-box for a core fundament of a democratic society e.g. voting is by definition an unhealthy and potentially dangerous situation. However, I do understand the naive point from where these people are coming from. Heck, it took me almost ten years after I first used a voting machine before I realized it and I’m a geek who’s supposed to be interested in this kind of stuff. But even if you’re not a geek or a ‘paranoid’ it should be clear that voting using a proprietary ‘black box’ is a bad idea.

I’ll give you my core argument:

The hardware and software used in the voting process is proprietary and cannot be reviewed by any (interested) citizen nor can anyone be certain a vote was cast upon the chosen candidate due to a lack of physical confirmation of the casted vote

In a few bulletpoints:

• We do not know the software nor the hardware which also means we cannot verify the security of it.
• Without the source code of the software we cannot be certain it has not been tampered with. It takes only one malicious person within one of the two (Nedap/Groenedaal and SDU) voting companies to tamper with the proprietary software to ‘crack’ the voting process.
• Any errors or bugs in the system could go unnoticed unless the company or its programmers have found it and fixed it.
• There is no ‘paper trail’ left behind after a vote has been cast and no way to tell if the casted vote was really set to the chosen party nor a way to confirm the votes casted by machines. Therefor there is no way to check the machine is functioning properly. The machine could display Party A, but set the vote on disk to be for party B instead.

So why did I write this post?

Because today we’ve got another election in the Netherlands: Provinciale Statenverkiezingen (rough translation: provincial state elections). It’s not a very popular one. The turnout is usually quite low. My guess would be because most people (including me until yesterday) aren’t aware of what it really means for citizens in general. It does not seem to be that important, however voting has an implicit effect on the Eerste Kamer (Senate) because the chosen representatives of the provincial state elections may choose the Eerste Kamer representatives. They are supposed to audit the proposed laws by de Tweede Kamer (Second Chamber) and test the reality of the proposals. Therefor the -at first not so important- election might be more important than one would think. See more on the Dutch political system on Wikipedia.

Anyway, in this election today we’ve encountered a perfect example of why we should NOT be using voting machines in the current form. In the Northern part of the Netherlands in a city called Emmen there was a problem with the voting machines (webwereld.nl, volkskrant.nl) Until 10:30 this morning it was not possible to use the voting machines as the machines were not configured correctly. The result of this error in configuration:

When casting a vote for candidate A of political party A the result could be that one had voted for candidate X of political party X.

Luckily this error was found out (after people had been casting votes since 08:00 in the morning), but what if the display had not shown this error? What if the display had not shown the wrong candidate after selecting a candidate? What if some less obvious error occured somewhere else in the software? What if this misstake had gone unnoticed? Then nobody would have known about this mistake and the results were completely out of touch with the real votes.

That’s why voting machines should at least leave a ‘paper trail’, so the machines can be tested by sampling at random. That’s why citizens should be given the possibility to review the hardware and software.

Now consider the example of Emmen and the press release of a recent study done by the CBS (Statistics Netherlands) and NKO in which most [1] Dutch people (presumable only those eligible to vote) state to have much or very much faith in voting by machines. Keep in mind that there was quite a lot of media attention in the Netherlands about the flaws of the currently used votingmachines.

If you combine all this information, one becomes quite cynical about politics and the current reality of voting in this democratic society. I wonder how can we (basically anyone who cares about democracy and understands the flaws of the current votingmachines) inform all of these people without being held for a nutcase with too much spare time on their hands? How can we make people aware of the importance of a fair and open voting system while keeping it secure?

I certainly don’t have the answers, but I’ll keep on informing, discussing and explaining why the current voting machines are a (potential) danger to our democratic society and I hope you will do the same.

For more information and some good starting points:

• Wijverstrouwenstemcomputersniet.nl is a Dutch action group demanding a ‘paper trail’. They also have an English and German section
• Holland Open writes open letter to the new minister of Internal Affairs requesting voting machines to be outfitted with open source software.
• Blackboxvoting.org is an USA based initiative and call themselves an ‘elections watchdog group’. They also have a book online about black box voting which I just discovered.

[1] According to the researchers (CBS and NKO) they consulted 65% of a selected group (no word on how the selection was made in this pdf NOTE: I use the dutch version, there is also a shorter English version without tables) of 4000 people between 10 October 2006 and 12 January 2007.Which means they got responses from 4000 * 0,65 = 2600 people.

Of these 2600 people 84% (according to the study, see the table above) has very much faith or much faith in votingmachines. So of these 2600 people there are 2600 * 0,84 = 2184 people with much or very much faith in using voting machines. According to the same CBS there were 11.800.00 people allowed to vote (I presume the researchers only took those allowed to vote into account) during the last municipality elections in March 2006. So if I still remember my math correctly the total percentage of Dutch people allowed to vote and having much or very much faith in using voting machines would be:(2184 / 11.800.00) * 100 = 0,185%. Personally I find it questionable to call this ‘most Dutch people’, but since I’m no math wonder nor a statistics expert let’s presume these researchers know what they are doing and we consider 0,185% to be ‘most Dutch people’.

NOTE: The title of this post is a (supposed) quote by the remarkable and famous feminist, anarchist, atheist Emma Goldman. Although one always has to take quotes and history with a grain of salt, I really like this quote and find it somehow fitting for the subject of this post.

UPDATE: Some reporter from METRO tested the security and could vote multiple times. Personally I was also suprised that I did not have to ID myself.

This entry was posted on Thursday, March 8th, 2007 at 1:48 am and is filed under geek, politics, rants. You can follow any responses to this entry through the RSS 2.0 feed. You can leave a response, or trackback from your own site.