Slashdot Mirror


Building a Better Voting Machine

edmicman writes "Wired News has an interesting article about what would make the perfect voting machine: 'With election season upon us, Wired News spoke with two of the top computer scientists in the field, UC Berkeley's David Wagner and Princeton's Ed Felten, and came up with a wish list of features we would include in a voting machine, if we were asked to create one. These recommendations can't guarantee clean results on their own. Voting machines, no matter how secure, are no remedy for poor election procedures and ill-conceived election laws. So our system would include thorough auditing and verification capabilities and require faithful adherence to good election practices, as wells as topnotch usability and security features.'"

16 of 245 comments (clear)

  1. Make it complicated please by argoff · · Score: 3, Insightful

    I'm serious. The more stupid and computer illiterate people you scare off, the better off we all will be.

  2. Random spot checks by Harmonious+Botch · · Score: 4, Insightful

    From TFA: "Random spot checks...This involves taking a random number of machines out of commission just before polls open on election morning to run a sample election on them to make sure the machines are recording and counting votes accurately.

    Before the polls open? How about during the election? At random times during the day?
    The poll workers should be required to have an extra one on hand just in case one breaks. It would be used to stand in for the one that was being checked. ( It could also be chosen for a random check. )

    1. Re:Random spot checks by Sparr0 · · Score: 3, Insightful

      $5000 per machine? Why? A $100 PC in a $50 arcade cabinet with a $20 printer could do everything that a perfect voting machine needs to do, and thats thrown together from consumer parts. If someone isn't building this 'perfect' voting machine for under $200 then something is wrong.

  3. Open source & peer review by guyjr · · Score: 4, Insightful

    I think Wired is barking up the exact same, wrong, tree, that Diebold and every other manufacturer of voting machines is barking up - namely that they have all the answers.

    The solution is very simple: require all electronic voting machines to be open source, and invite all software developers around the world to peer review the code. When that majoriy agrees that a system is secure, then it's ready for use.

    1. Re:Open source & peer review by Mab_Mass · · Score: 2, Insightful

      If you actually read the article, you'll see that they propose something just as good - requiring the full source code to be made public, which allows /. type geeks to do a complete audit.

      Essentially, though, the key requirements are simple to state: secure, transparent, auditable.

      Anything that fails any one of these is unacceptable.

    2. Re:Open source & peer review by Random+Utinni · · Score: 4, Insightful
      I think Wired is barking up the exact same, wrong, tree, that Diebold and every other manufacturer of voting machines is barking up - namely that they have all the answers.

      The solution is very simple...


      Erm? Pot... meet kettle.

      There is no simple solution to voter fraud. There always has been fraud, and there always will be. It's the nature of ingenuity. Hence the "build an idiot-proof machine, and the universe will build a better idiot". If someone wants to hack an electronic voting system, they will, open-sourced and peer-reviewed or not.

      In my view, the goal is simply to minimize the impact of such efforts, and to make it as difficult as possible to do so, as cheaply as possible. Open source *might* be a good way to go... certainly better than the closed electronic systems Diebold and their ilk are currently pushing. However, it's still an electronic system, and electronic systems are prone to making small errors very quickly (or being hacked to introduce small biases, very quickly). I'd personally prefer to return to a simple paper and pen ballot... simply check the box of the person/proposition you're voting for. Put paper in box. Let people count ballots (with observers, if desired). It scales fairly well, is difficult to introduce large errors into, and can't be hacked remotely. If it takes a little longer to get election results, so be it... there's almost two months between election day and inauguration day.
  4. Don't build anything by Anonymous Coward · · Score: 5, Insightful

    Paper and pencil. Mark your choices, put it in a cardboard box. It's the perfect solution and scales wonderfully.

    Many countries already use this advanced technique.

    1. Re:Don't build anything by From+A+Far+Away+Land · · Score: 4, Insightful

      Other famous democratic countries do use pencil and paper. Canada, one of Americas greatest neighbours to the north uses the birchbark and pinecone voting system... just pulling your leg. They, like Australians, use pencil and paper. We have about 70% voter turnout in Canada, with a voting population in the range of at least 10 million people. It takes us less than 2 hours after a poll closes to give nearly complete and meaningful results to the public.

      Telephoning the result to a central station is the extent of electrified voting in Canada. Everything else is on paper, for easy double checking if there's a court challenge. To have a system without paper that the voter marked, is an invitation for fraud.

    2. Re:Don't build anything by LiquidCoooled · · Score: 2, Insightful

      paper ballots can still be spoilt.

      Reading this from the bbc:

      Londoners had to register four votes altogether, a first and second choice for mayor, a vote for a London Assembly candidate and one for an Assembly party. While 2.9% of papers for the mayoral ballot were rejected, the London Assembly paper proved even more difficult.

      Over a hundred thousand people, 6.7% of the electorate, failed to correctly fill out the section choosing a constituency member, while 2.53% did not correctly register a London-wide party choice.


      Now that was only 2 years ago, but I think all interfaces suffer when dealing with multiple answers or pieces of information.
      A simple A B or C ballot works on paper really well, but you have to think VERY carefully about how you layout a complex set of questions.

      --
      liqbase :: faster than paper
    3. Re:Don't build anything by From+A+Far+Away+Land · · Score: 2, Insightful

      That's an interesting point. The obvious answer seems to be you choose between convenience of electronic voting - and get shafted out of your democracy that you're supposed to be getting. Or you simplify elections so that dog catchers aren't on the same paper as president - increasing the cost, but simplifying the voting. Since voting is pointless unless the voters know who they are voting for, and can easily make their mark, it's obvious that putting too much onto the ballot is actually less democracy than being able to pick every position in one shot.

  5. Ive been saying it all along by The+Cisco+Kid · · Score: 3, Insightful

    Have one machine with fancy GUI's that are easy for people to use, which PRINTS a clear paper ballot on which the marks are both human and computer-readable (think of the little ovals you used to fill in with #2 pencil, only bigger ovals) and then a *seperate machine* which does nothing but scan and count the ovals.

    The marking machines could be of any complexity, wouldnt require auditing (the names on the ballots would be pre-printed, the machine would only mark in the ovals). Voters could choose to use the machine, or to mark the paper ballots themselves, and in all cases would be able to *look* at the paper ballot and verify their selections before submitting it to be counted. The specs for filling in the ballots could be released (and in fact the ballot specs would be part of the specs for the counting machine), and anyone under the sun could make marking machines, of any design that they wanted. The key is that these machines would record votes only on the paper ballot.

    The scanning/counting machine would have to be absolutely auditable, as simple and as transparent as possible. Every aspect of its operation would be required to be public domain, and available to any citizen upon request.

  6. MY Perfect Voting Machine by AeroIllini · · Score: 5, Insightful

    The true perfect voting machine consists of the following four components:

    - Paper
    - Pencil
    - Locked box with slot
    - Election official who can count

    Anything else is a solution in search of a problem, and a way for partisan election officials to send some contract money to their buddies in the tech industry.

    Seriously, who the hell cares about digital records or fast counts? I don't care how fast the results come in, I want them to be RIGHT. A voting system needs to enforce two basic principles: private votes and public counts. The voters need to know that their votes are private and anonymous, and the counting process needs to be simple and transparent enough that it can be understood, audited, and repeated. Computers, for the majority of people, are magical black boxes. They don't trust them as far as they can throw them, and that means there will always be suspicion of fraud, no matter how open the source and how impenetrable the outer casing. When we go to paper ballots, we guarantee that the process is easily understood, auditible, difficult to rig, and that counting is repeatable. There is no electronic system that satisfies all those conditions, and therefore electronic systems should not be used.

    However, if we wanted to use touch screen systems to print out ballots instead of marking them, that's fine with me (it would make voting more accessible, with a well-designed UI). The voter can verify their votes before dropping them in the box. But the printed paper ballots need to be counted by hand.

    --
    For security, the MD5 hash of this message and sig is 09f911029d74e35bd84156c5635688c0.
  7. Re:Don't get too upset over this, it isn't importa by geekoid · · Score: 3, Insightful

    "And I'll even give a pass on FL in 2000 even though the recount conducted by the press gave the state to the Republicans."

    There WAS NO COMPLETE RECOUNT!
    Shit I am tired of this fucking false rumor. There were thousands og votes not even counted, as well as hundreds of prople being turned away.
    People involved with and running varias republican campahains were filmed interruptting the election process.

    Plus there was an ever BIGGER problem in Ohio.

    Personaly, I don't give a damn about election before I was born, I am concerned about elections that happen while I am a voter.
    2000 was a shame, as was 2004, and you should be pissed about it no mattter what party did it.

    Yes, if democrates had done that I would be just as pissed.
    OTOH we wouldn't have gotten rid of habias corpas.

    "But like I said, there is enough transparency that in any national election fraud can't swing the totals more than a point or two and the Electoral College minimizes the damage in Presidential elections.

    laughable, giving people more weight then others is an imbalanced system and it needs to be changed.

    --
    The Kruger Dunning explains most post on /. http://en.wikipedia.org/wiki/Dunning%E2%80%93Kruger_effect
  8. Look at India and learn by Anonymous Coward · · Score: 1, Insightful

    Watch and learn :)

    http://www.slate.com/id/2107388/

    http://www.eci.gov.in/faq/evm.asp

    No vindaloo or call center jokes please!

  9. Trollish but valid point by joggle · · Score: 4, Insightful

    I would just like to point out that while the parent post is trollish in nature, it is a sentiment similar to what nearly all (if not all) of the founding fathers believed. That being certain qualifications are needed in order to cast a ballot. Their fear was some rogue could convince less educated people to vote for him so that he could, in turn, pillage the government and/or be a tyrant. I'll grant it's a thorny issue, but the problem of attempting to intentionally limit people who vote is that inevitably some racial groups will be disenfranchised (as well as other categories of population, such as the elderly in this case). Also, some local officials will try to exacerbate the situation to their favor (as happened-- and is still happening--in the South).

  10. Re:Don't get too upset over this, it isn't importa by Spasmodeus · · Score: 2, Insightful
    There WAS NO COMPLETE RECOUNT!
    No, but the recount for which Gore asked, and the Florida Supreme Court ordered, was finished by the press, and it showed Bush as the winner.

    Gore never asked for a statewide recount because he was afraid a statewide recount wouldn't have given him the net votes he needed. He only asked for counts of the undervotes in Democratic-leaning counties because he thought he could get just enough votes to be elected. His rhetoric about "count every vote" was total hooey.

    So if the U.S. Supreme Court hadn't put a stop to the recount, Gore would have still lost, only he would have been hoisted by his own petard instead of having the convenient "judicial coup" excuse.

    There's an interesting article on the whole despicable affair HERE.