More on Belgian e-Voting

Sigh, it’s all so depressingly predictable. My last post contained some headlines on problems with the new Belgian e-Voting system made by Smartmatic. As it turns out there’s a real problem with quite a few people unknowingly voting a preference for the candidate whose button (on the second screen), happens to be right where the button for his/her party was on the first screen.

A few days after my last post there’s an article in “Het Laatste Nieuws” in which Hendrik Bogaert, deputy minister for the civil service, afraid of large political risks, would like to see e-Voting scrapped entirely. Geert Bourgeois, the Flemish interior minister, denies the seriousness of the problem:

“What is possible is that the voter touches the screen too long and thus out of lazyness just touches the closest person. Two separate actions are necessary”, Bourgeois underlines. “My agency and the company that sells the computers both contradict that this could happen by pressing too long.” Even 85.000 tests by PriceWaterhouseCoopers have “never shown this problem”.

Guess what. Just a few days later this very same minister is quoted in De Standaard complaining he has been made to look like a complete fool. It turns out PWC (“From the people that certified Diginotar”) did know about these problems after all. You didn’t see that one coming, now did you? He also fears that anyone that did not get a seat because of these problems could go to court, which would cause major problems. (Which seems like a silly fear to mention in an interview, but OK.)

More seriously: how many “incidents” does it take for people to realize that the entire e-Voting industry is corrupt and/or incompetent?

13 thoughts on “More on Belgian e-Voting”

  1. Strange. People could check their votes before submitting them to the ballot box and they could request to discard their vote and restart their vote if there was a problem. I don’t think this is a sufficient reason to discard e-voting or that this is sufficient reasoning to dispute the results in court.

  2. Anthony: No, I don’t think it’s sufficient reason to discard e-Voting either. That honor goes to the much more fundamental transparency issues with regard to e-Voting.

    There are transparency issues that are simply unsolvable with today’s technology, if you want both secrecy of the vote and observability of the count. But that doesn’t mean it’s not bad to have your voting systems supplied by an industry that manages to screw up something as simple as having a voter select a candidate on a touch screen. Do you NOT wonder what else they screwed up?

  3. @anthony
    I think a system where every voter can check for themselves if the vote is really what they want is fundamental for validating the results and making the whole process transparent for those that vote.

    So if we agree a paper trail, readable by the voters themselves (not a QR-code or something), has to exist than both a computer with paper trail or voting with pencil are both valid solutions. The only thing to debate then is the trouble and extra costs of voting with something complex as a computer versus the elegance of voting with a simple thing as a pencil.

    Ben Adida posted quite some articles about e-Voting on his blog in the recent years you might like to check out.

  4. Here’s an example of the papertrail that was generated at the Belgian election (this is a test vote used to introduce people to the system, people voted music styles and artists, I picked hiphop “party” and voted for all the people on the ballot):

    It contains both the machine readable QR code (I still need to figure out how safe the encryption on that thing is) AND the human readable vote. Every polling station in Belgium had a system where the voter could verify the contents of the QR code with a hand scanner. (OK, that could be rigged, but the text on the vote is what counts if disputes need to be resolved or recounts need to be organized)

    As in the other thread, I want to argue that subjective disputes over the validity of a vote are minimized by using these printed paper trails, which is valuable.

  5. Anthony: Not only could you rig the verification device, but even if you trust it, you still need to trust the software to count honestly. The only count that does not depend on unverifiable trust is the paper count. Right?

    (Remember all these papers 🙂 )

  6. @anthony:
    I think your point about a printed vote being sometimes less ambigues than a colored circle is a valid point (though I’m not sure how I would weight it, since I don’t know how many votes are disqualified because of being unreadable or having more than one circle colored).

    Then there’s still the point about the counting process and the ability for every voter to check if all votes are counted correctly or not. Asking a blackbox to spit out a number is not the same as looking over the shoulder of those that count and check if they are incrementing correctly.

  7. The conspiracy theorist in me sees room for a grand scheme where some layers of obvious errors were introduced (the double-tap-wrong-candidate being the most obvious one) to obstruct a clear view on mass fraud polluting the core of the system.

  8. Listening to you on Radio 1 right now. So Dutch government agáin wants to introduce voting machinery…

    Interesting times ahead.

Comments are closed.