Catch up on stories from the past week (and beyond) at the Slashdot story archive


Forgot your password?

Quebec Bans Electronic Voting 222

gfilion writes "The Chief Electoral Officer of Québec tabled an evaluation report that makes a troubling diagnosis of the problems that occurred during the municipal elections of November 6, 2005, in some of the 162 Québec municipalities that used electronic voting. He says: "Not only did the systems fail, but the corrective measure proposed were insufficient, poorly adapted and often came too late." There was a moratorium on electronic voting prior to the November 6 election, it will be extented for future elections."
This discussion has been archived. No new comments can be posted.

Quebec Bans Electronic Voting

Comments Filter:
  • *golf clap*.

    Honestly folk, count the f'int ballots by hand, stop throwing them out, etc.

    We purport to have this great democracy yet we do all in our power to screw up the vote...

  • by msobkow ( 48369 ) on Wednesday October 25, 2006 @10:08AM (#16577314) Homepage Journal

    So far I've read dozens of reports over the past 5-6 years about failed, hacked, and broken electronic voting machines.

    How many failures does it take before those providing the crap equipment are sued and forced to FIX the results of their incompetent designs and testing?

  • Wow (Score:5, Funny)

    by Anonymous Coward on Wednesday October 25, 2006 @10:09AM (#16577330)
    They must hate freedom (tm).
  • by MtViewGuy ( 197597 ) on Wednesday October 25, 2006 @10:15AM (#16577432)
    The problems with electronic voting systems--namely no paper trail--is the reason why many municipalities are switching to mark sense voting ballots.

    Since mark sense paper ballots (filled out in pen to make sure the mark is clearly seen on the ballot) can be both machine-read and hand counted, this mostly avoids the Florida 2000 fiasco of difficulties reaching punched card ballots, complaints that electronic voting machines can be biased towards one candidate, and the numerous problems of the old mechanical voting machines.
    • The problems with electronic voting systems--namely no paper trail--is the reason why many municipalities are switching to mark sense voting ballots.

      So that's what it's called.. we used exactly that in the last municipal election for my riding in St-Laurent Quebec.

      • by gavriel407 ( 897344 ) on Wednesday October 25, 2006 @11:31AM (#16578732)
        So that's what it's called.. we used exactly that in the last municipal election for my riding in St-Laurent Quebec.

        It seems that many people are assuming that Quebec used the same Diebold electronic voting machines, when they were clearly not. This press release is outlining the problems of an electronic means of counting paper ballots, which are 100% verifiable if the counting machine fails.

        Here's how the voting process went: After voting on a ballot that looked similar to previous elections, the ballot was inserted into a black cardboard holder, which was then fed into a rather simple looking machine. I was astonished when I later found out that these boxes were actually counting the votes as they went in. (And yes, I do live, eat, drink Pepsi and vote in Quebec)

        If we have problems counting regular-looking paper ballots, how are we supposed to trust our votes to a machine?
        • by neoform ( 551705 )
          You're not a true Quebecois unless you drink Pepsi and eat Maywest and Lays for breakfast.
  • by Channard ( 693317 ) on Wednesday October 25, 2006 @10:19AM (#16577522) Journal
    .. where there's a company called EDS [] that has been regularly screwing-up government and council contracts by producing flawed systems, and yet still manages to get work!
  • by Utopia ( 149375 ) on Wednesday October 25, 2006 @11:01AM (#16578212)
    I am suprised at the amount of electronic voting problems in US & Canada that I read about on Slashdot.
    A big democracy like India successfully used electronic voting in the last election.

    Indian voting systems using basic $200 machine while US machines are $3000 systems loaded with millions of lines of code.
    380 million Indians cast their votes on more than 1 million machines and the election, the largest electronic voting in the world.

    The lesson here is to simply the system. Don't make the system overtly complicated. The more complicated the system, the more bugs it will have and more difficult solutions to the problems.

    • A big democracy like India successfully used electronic voting ... The lesson here is to simply the system.

      Just for the sake of argument let me play the devil's advocate:

      How can you be sure? How about the possibility that the checks and balances to catch election fraud failed in India, so you _think_ the system worked flawlessly, whereas here in North America those deficiencies are revealed?

      I agree with you in that the simpler the system, the
    • Your post entirely ignores the issue of voter fraud. Show me evidence that there wasn't any, and then we'll talk.
  • ...but they can't stop the election rigging by way of gerymandering that is prevalent in Québec.
    • Re: (Score:3, Interesting)

      by Abcd1234 ( 188840 )
      Sure they can. You do what the federal government did: place the power of drawing election boundaries in the hands of an arms-length, independant group (in this case, Elections Canada []).
  • by Maxo-Texas ( 864189 ) on Wednesday October 25, 2006 @11:18AM (#16578524)
    Voting solutions need to be pure hardware.
    Software can be patched to do whatever I want it to do-- including counting votes one way and then erasing itself after if a certain pattern of votes are entered .

    • by Teilo ( 91279 )
      The only pure hardware solutions would be mechanical, not electronic.

      Anything with a CPU is a software based solution. The software may be permanently burned into silicon, but it's still software. Swap chips. Swap bits on a storage device. Same thing. One's just a little easier.

      Technology will not solve this problem - hardware, software, or otherwise.
  • Cool!
    As a Montreal resident, not only is this possibly the first time I've seen Quebec in a Slashdot headline, but it's announcing something I agree with, too! Hurray!
  • by telso ( 924323 ) on Wednesday October 25, 2006 @11:25AM (#16578648)
    Today's Montreal Gazette says [] the electronic voting used was up to 25% more expensive than paper voting and caused delays in getting the result. (The election also resulted in more judicial recounts than normal because of the inaccuracy of the machines, causing delays in the swearing in of the winners.) The report also concluded that:

    • Machines misread ballots.
    • A backup plan covering all possible problems was missing.
    • The lack of paper ballots in some municipalities prevented judicial recounts.
    • Only partial testing of the voting machines took place in some instances.

    It's nice election officials in Quebec did what seems like a pretty successful review; I'll be happy to approve optical scan voting when these problems are addressed. Until then, it's good at least some jurisdictions in North America realise a lack of paper ballots can prevent recounts.
  • Here in Ireland we currently use manual counting, with paper ballots that are marked by the voter with a pencil or pen. The government wants to introduce e-voting, for no other reason than for its own sake. Thankfully they have not (yet) succeeded.

    As far as I can see manual counting of paper ballots should sit very comfortably with anyone who advocates open source projects - they seem to have the same advantages:
    1. Many eyes: In the Irish system at least, many people watch the official counters counting
  • I tell you, I'm not much of a programmer, but I am convinced that, given a year, I could design and program an effective voting software with: 1) A paper trail sufficient to be used for a manual recount. 2) Reasonable measures to ensure 1 voter 1 vote. 3) A barcode crypto scheme to tie #1 and #2 together so that every database record can, if required, be verified against every paper record. 4) A completely open and peer-reviewed code base.

    But as many people have already noted, this problem is not technical,
    • Re: (Score:3, Interesting)

      by Phroggy ( 441 ) *
      I tell you, I'm not much of a programmer, but I am convinced that, given a year, I could design and program an effective voting software with: 1) A paper trail sufficient to be used for a manual recount. 2) Reasonable measures to ensure 1 voter 1 vote. 3) A barcode crypto scheme to tie #1 and #2 together so that every database record can, if required, be verified against every paper record. 4) A completely open and peer-reviewed code base.

      You don't have to, it's already been done [].
  • by Pig Hogger ( 10379 ) <> on Wednesday October 25, 2006 @11:53AM (#16579152) Journal
    I have been an election official during the last two federal elections, and a candidate representative during a Québec by-election.

    The Directeur Général des Élections [] (DGE) is in charge of all elections/referendum within Québec.
    In Canada, federal elections are handled by Elections Canada []. Rules are virtually the same (exceptions listed below).

    Registration. Everyone is automagically registered. If you file an income tax report, you are registered UNLESS you specifically ask so (a part of the tax form asks for it).
    When the election comes, the DGE sends out notices to everyone on the list. If there are mistakes, or you are not listed, you can ask to be properly registered at the local election office (usually, one by riding).

    The part-time election personnel is chosed riding by riding. The incumbent hands out the "important" jobs (poll center supervisor, revision official, scrutineers) while the "less important" jobs (security, assistant revisor, poll clerk) are left to the other candidates.
    Training for the poll workers lasts about 3 hours, and happens a week before the election. It explains what are the general procedures. We are also given a book that explains special cases, which we have to read (but are not tested). The scrutineer is given the ballot box which contains the paperwork. We are to meet 2-3 days (usually in the scrutineer's home) before the election to check that the contents are okay; we are to report discrepancies so they can be fixed in time.

    We are given a list of all the people entitled to vote (about 400 per box), with those who voted in advance and those who moved-out or otherwise no longer voting there crossed-out.

    On election day, at each poll you have the scrutineer, the poll clerk, and as many representatives as there are candidates. The representatives are there to watch that everything is done properly; they can question some aspect of the procedure, like question the identity of voters and question the admissibility of a ballot when counted (but in all respect, the scrutineer has the last word). And representatives can be expelled at will if they don't behave.

    Showing ID is not compulsory. But in Québec, anyone can demand a voter identify himself; however, in Canada, election officers are specifically prohibited by law from asking for ID. What is interesting is that many people spontaneously show their ID when they come to vote, and we have to tell them they don't need to (this shows how people accept to show their ID in order to vote).
    The situation is different in Québec because federalist parties were caught red-handed rigging elections, so when the law was put in front of parliament, they could not very well vote against it, given the huge amount of egg on their face...

    When the voting begins, the ballot boxes are sealed after everyone present agrees that they are empty. The representatives can sign the seals, and note down the serial numbers.

    The ballots are printed on stapled booklets, from which the ballots are detached. Each ballot has the list of candidates (or options for referenda), a space for the scrutineer to put his initials and two identical serial numbers.
    The serial numbers are on different tear-off stubs; the first remains in the booklet, the second is kept on the ballot when it is handled to the voter.
    Before handling the ballot to the voter (AND ONLY AT THAT TIME!!!!), the scrutineer marks the back of the ballot with his initials, with the stub with the serial number in plain view.
    The voter votes, and either tears-off the serial number stub in plain view of everyone, and shows the scrutineer's initials (this is to insure that this is the same ballot that was handed earlier - in order to avoid "telegrams"), or the scrutineer does it for him without unfolding the ballot. THE STUB WITH SERIAL NUMBER IS TO BE KEPT!!!
    The voter then puts the ballot in the box, a

  • There's a strong tendency among geeks and technology buffs to want to see e-voting, but I can't really figure out why. It's completely unnecessary and the proposed advantages of e-voting are precisely what make it unattractive. We like our voting mechanisms to have physical evidence that we actually did vote. We need it to be capable of being analyzed and understood by someone with a minimum of technical knowhow as an exigency of the fact that volunteers work polls, etc.

    Essentially, we need to rethink th
    • I am a strong supporter of paper ballots and a strong opponent of electronic voting machines. That said, there are certain people (e.g., a person who is blind) who appreciate the privacy offered by an electronic voting machine that can, for example, read the ballot to the voter.

      My solution? Have an electronic voting machine that can print out a completed ballot that looks exactly like all the other ballots that were filled out by hand. Of course, if I were to use such a machine, I would like a close friend
  • by Ignatius ( 6850 ) on Wednesday October 25, 2006 @12:07PM (#16579402)
    Obviously, when a machine with no auditing features is used in an election, there's a very high chance that whoever ordered or built said machine did so with the purpose of being able to manipulating the result. But that's not the only and not even the worst problem.

    While the manipulation problem can be fixed by generating a voter-verifyable paper trail (NOT an easy problem, but doable), this still leaves us with the harder problem of secrecy (or potential lack thereof).

    A voter who enters his decision into an electronic black box can never be sure that his vote is not secretly recorded and used against him later. The problem here is not only that this actually happens, the rumour that it MIGHT happen is already enough to make a free election impossible.

  • by kthejoker ( 931838 ) on Wednesday October 25, 2006 @12:12PM (#16579500)
    Why must all elections for all offices local, city, county, state, and nationwide take place on the same day?

    We try to saturate all of our voting into one day, and for what? Why not have 4 election days a year, instead of one. The national elections will still be in November. State elections in February. City and County elections in May. Local referendums, bonds, and other non-candidate-oriented votes in August.

    All dates above are arbitrary (so is the first Tuesday in November.) We're not stupid, we can keep up with 4 days. And then we can use paper ballots, because counting is exponentially easier. Why are we so hard on ourselves for one week in November?
    • Re: (Score:3, Insightful)

      by raddan ( 519638 )
      Not only that-- why do we have to have the results so soon? Why not wait a week or two to make sure the votes are counted correctly? I mean, we have to keep the bozos for 4 years-- what's two weeks of waiting?
  • ... to live in a place where the government cares about fair elections. I'm hoping that my district advances to the point where everyone who votes gets a blue finger.
  • The link is to the evaluation report, not to the news of the moratorium or ban.... Anyone have a link to that?

To write good code is a worthy challenge, and a source of civilized delight. -- stolen and paraphrased from William Safire