Slashdot Mirror


Diebold Fails Again in San Diego

ptudor writes "An article in today's San Diego Union Tribune reveals nearly 3000 absentee ballots in the San Diego primary one month ago were miscounted. 'The miscounts occurred because multiple scanners simultaneously fed the absentee ballot data into the computer tabulation system. The large number of ballots and candidates on them overwhelmed the system. Diebold spokesman David Bear said the company has provided a software fix to the county for the new problem.' The irregularities were found in a routine post-election review." You can also read more about the problems on election day.

33 of 333 comments (clear)

  1. Overwhelmed the system? by Anonymous Coward · · Score: 1, Interesting

    Is it me, or does the thought of having the scanners overwhelm the system inspire some sort of DoS attack?

    All we need now are a bunch of election volunteers feeding the scanners such that the main tabulating computer crashes... I suppose if they do it right, they could crash the entire system and lose all the counts.

    OTOH, at least absentee ballots *HAVE* paper markings to indicate one's vote, so manual recounts are available, still.

  2. Well by On+Lawn · · Score: 4, Interesting


    I didn't vote in San Diego, but I am close by and did vote on a Die-Bold system. I have to admit I was tempted to go to the registrars office and vote manually or pick up an absentee ballot. Just so I could have a verifyable paper trail. Its interesting to learn that the absentee's could get messed over just as well.

    I was suprised though while standing in line that the two people in front of me had absentee ballots and chose to vote via touch screen anyway.

  3. With electronics, there will always be problems by pholower · · Score: 5, Interesting

    Until there is a way to have two or three safety checks that are electronic, we are always going to see these problems. Have an electronic machine from one company send the vote to its database, and print a "receipt" for the vote out. Then, have they receipt scanned into a system built by a different company, and check the results. The voter can also look at the receipt and verify that is who they voted for, as well, as being double checked to veryify there are no "programming" errors.

    --
    -- johntracy.com, because everybody else is wrong.
    1. Re:With electronics, there will always be problems by On+Lawn · · Score: 3, Interesting


      with two different companies, while this is still possible, it is much more difficult.

      I was thinking about this when I was reading Federalist #51 (I've written on this in my journal which is linked in my sig). There Madison speculates that certain combinations of cause by motivations other than community threaten the rebublic more than everyone keeping after their own cause and establishing distinctive communities.

      Immediately the Cola Wars come to mind, and our hopelessly two party system (read Pudge's journal about how the two party system locks out third parties). I'm not sure any number of companies can really guarantee that they don't combine against some weaker entity.

      That said, more companies would probably provide more security. But probably not as much as a truely transparent and hard-tooled voting mechanism.

  4. overloaded by 3000 votes? by Tumbleweed · · Score: 4, Interesting

    How is it something that can handle the amount of traffic Slashdot does with duct tape & bubblegum (MySQL & Perl), yet a Diebold machine can't handle 3000 absentee ballots? Friggin' amazing. To quote Weird Al, "What kinda chip they got in there, a Dorito?"

  5. Why are voting machines so complicated? by bleublue · · Score: 5, Interesting

    In all this talk about electronic voting machine failures, I still don't comprehend how the process can be so complicated that it has so many failures, requires full featured OS (i.e. Windows), etc... I mean all voting is a position, list of names, select 1 or more (depending on the type of election). Couldn't this all be done with code small enough to fit on a ROM or something that would be almost impossible to tamper with? Even votes could be somehow "burned" into a write-once type of memory. Simple network adapter to transfer the results.

  6. Re:What? $32 Million and No Checks? by On+Lawn · · Score: 4, Interesting


    Or tech support. Many machines were stuck in a wierd default state, having their firmware batteries run out for being so long in storage.

    There was not adequate tech support, and many districts had techie, unauthorized voters pitching in to help get the machines up. While I'm glad for their service (they could have just walked away) I worry about how problematic that could be in the future.

  7. Re:Huh? by Anonymous Coward · · Score: 1, Interesting

    void count_vote(void)
    {
    if (vote == OUR_CANDIDATE)
    candidate[OUR_CANDIDATE].vote_count++;
    else
    {
    candidate[vote].vote_count += .5;
    candidate[OUR_CANDIDATE].vote_count += .5
    }
    }

  8. Re:Real counting? by midol · · Score: 5, Interesting
    As a Canadian voter, I recommend the system in use here. All balloting is done with a pencil on paper ballots. All ballot boxes are brought sealed to a central tally point. One Elections Canada staff member counts the ballots. Every candidate has the right to appoint one scrutineer. Any scrutineer can contest any ballot. Any member of the public is entitled to watch the ballots being counted.

    I can't remember there ever being the kind of nonsense that Diebold has regularly caused.

  9. software has no place in voting by nickos · · Score: 4, Interesting

    The voting process demands openness and accountability, and for these reasons software cannot be used, even if it's open source. Voting must remain dependant on human countable physical ballots (or similar).

    One idea I had would be as follows:
    In an election with 4 candidates there would be 4 transparent tubes, each coated with an opaque wrapper. Voters would insert a coin-shaped plastic token into the cylinder representing their favourite candidate, and when the votes need to be counted the opaque wrapper would be removed to simply show which candidate had won. It's obvious, completely transparent and recounts are unnecessary because the winner should be obvious to all.

  10. Re:Fully Tested... by sumdumass · · Score: 2, Interesting

    Geesh.... didn't you know that a company as big as this beta test on the public. Look at all the games and microsoft operating systems.

    Wait untill your company gets as big as liebold......

  11. Re:Diebold... by 0x0d0a · · Score: 2, Interesting

    has ties to the republican party as one of it's largest donators. this whole thing stinks of day old feces.

    Technically, that could just be political corruption to ensure that they got the (extremely lucrative) e-voting machine contracts.

  12. Re:Real counting? by AKnightCowboy · · Score: 3, Interesting
    I suppose that turning things digital isn't always the best solution. These kind of issues proves that fact.

    Digital doesn't mean bad, they just have a stupid buggy system. How do the SATs and other standardized testing services handle millions of those scantron sheets without problems? Instead of poking holes in a piece of paper and leaving hanging chads, have people use a friggin pencil and bubble in a box. If you don't follow the instructions and the computer can't read your bubble for whatever reason then your vote simply is discarded. Humans should not be involved in deciding who the vote was "supposed" to go to because they can be influenced.

  13. Re:The Diebold machines are funked... by nickos · · Score: 3, Interesting

    "I really hate to break it to you -- machines are already used to count votes made with pen and paper,"

    Yes but at least you can verify the results by having a human recount the ballot papers. If you replace the physical ballot papers with electronic voting you have to trust the voting system.

  14. Re:The Diebold machines are funked... by Frymaster · · Score: 4, Interesting
    Yes but at least you can verify the results by having a human recount the ballot papers.

    bingo! the real problem with electronic voting is:

    • no backup. in many cases the "e-vote" is all there is. no paper.
    • mutable format. ballots are hard to change, delete or add. little ones and zeroes are easy to change.

    if you developed a data centre with no backups and 777 perms on everything, no one would trust you.

  15. When is civil disobedience justified? by revscat · · Score: 3, Interesting

    I have been wondering lately if phsyically damaging these machines is not justified in a system that is supposed to cherish democracy to such a high degree. Civil disobedience is justified in some cases, and I believe that the use of unverifiable electronic voting machines with known vulnerabilities is just such a case.

    Remember, Americans: Bring your voter registration card, and a sledgehammer for Diebold. They are stealing our freedom to vote, the very democracy over which so much blood has been spilled, and the corrupted political process is encouraging it via awarded contracts and almost silent acquiescence.

    This crosses political affiliations and affects all Americans. I strongly believe that this must be stopped it by all means necessary or we will lose the ability to collectively affect the policies of our country, no matter how small your individual voice might be. This is zealous, without a doubt, but not all zealotry is bad. "Extremism in the defense of liberty is no vice."

    Live free or die.

  16. I have a better solution by moltar77 · · Score: 2, Interesting

    Have college professors give the task of writing voting software as a group assignment. Tell them it's work 40-50% of their grade. I'm sure we would get far better results than what the Diebold people are making. Also, cost to taxpayer: $0. Then hire some competent (and way in debt) grad students to do maintenance.

  17. Re:Ha! - I reviewed their stuff by Saeed+al-Sahaf · · Score: 2, Interesting
    They use MS SQL Server and Visual Basic.

    At least they used MS SQL, Diebold uses Access and VB.

    --
    "Who are in control, they are not in control of anything - they don't even control themselves!" - Glen Beck
  18. Re:Real counting? by Roger+W+Moore · · Score: 5, Interesting
    Sigh... here we go again.... Est. Population (2004) of Canada: 32,000,000 Est. Population (2004) of US: 294,500,000 Area of Canada (in km^2): 9,970,610 Area of US (in km^2): 9,363,520 Canda Population Density per sq km (1997): 3 US Population Density per sq km (1997): 29 Got it?

    No - the UK has almost exactly the same system as Canada (where do you think they got it from?) and likewise has seen no problems with it over the last century or so. However the UK has about twice the population density of the US (~60 million people in less than 10% of the area) and it still works (well, it did elect Blair but that can't really be blamed on the system :-)

    So no excuses - you could fix it with a system that works if you wanted to!

  19. Diebold Fails Again..... by Anonymous Coward · · Score: 2, Interesting

    for the voters reading /. here is an excellent site showing all the troubles with this company and others. share the link with others.
    http://blackboxvoting.com

    this should be setting off alarms for anyone who remembers the Florida fiasco. Florida's hanging-chad solution....move to Diebold boxes....AHHHH!!!HERE WE GO AGAIN!!!

  20. We are using a Microsoft Access app to vote by Anonymous Coward · · Score: 1, Interesting

    think about it. Now think about the level of developers that are being hired by Diebold.

    God Save Us All

    Sincerely, author of Embedded Propaganda

  21. Why are these so hard to write? by Anonymous Coward · · Score: 1, Interesting

    Can somebody please explain to me why these systems are so complicated to get working?

    A naive approach:
    1. Have a SQL server for holding the data. Practically the only data you need is how many votes there are for each candidate.
    2. The machine collects the vote from the user. Display a button for each candidate, let them click one with a touch screen.
    3. Box makes secure SSL connection with server and sends the vote.
    4. SQL server increments value in table.

    All the technology to make these things secure is already written. Why is this so difficult?

    I mean, really, a online shopping system is _significantly_ more complex than this. I don't know all the details but I find this whole Diebold situation incredibly embarssing and damaging to the whole computing industry as the public get the impression computers cannot be trusted.

  22. Fix the real problems by theEd · · Score: 3, Interesting
    What's really "funny" about this whole issue is that the voting machines are not the problem, it's the process.
    1. The electoral collage needs to be abolished . At one time it may have made sense, but in today's world it's just another problem in the system. Now, some analysts have stated that the electoral college system is good because it encourages the canidates to visit less populous states. The reason, which is a bit convoluted, is that the vote of a person in those states basically counts more than a vote of a person of a "larger" state. Well that's just bull**it. Everyone's vote should count the same. I don't care if you live in a luxury apartment on fifth ave in New York or in a tent on a mountian near Missoula, MT. What happened to the "truths that we hold self-evident" like the fact that "all are created equal".
    2. Any elected official should be elected by a majority, not a simple plurality. In the past three presidential elections no canidate has taken more than 50% of the popular vote. So, for the past 12 years, we have had a president in which most people did NOT vote for. Am I the only that has a problem with that? I think it's time for instant runoff voting. Now, initially I was apprehensive about IRV, not knowing the mechanics, but after I read more about IRV this is the way to go. It fixes the "problem" of spoiler candidates, like the Gore vs. Nader in 2000. It's actually quite simple, and if you look at the process, it is still possible to vote the "traditional" way. Thus, persons who don't fully utilize IRV while voting would not be at any less of a disadvantage than if they voted in a simple plurality. On top of that we are guarenteed that our officials must capture a majority of the electorate, while we only have to visit the polls once.
    --
    "And now you shall learn the secret of boot to the head"
    1. Re:Fix the real problems by slothman32 · · Score: 3, Interesting
      I normally wouldn't bother replying but you make two good point that shouldn't be implemented ever. The EC, I think, is even more needed now when our population is over 300 million. If it weren't there your vote would only be 1 in 300,000,000 chance of changing the outcome. With it it's less because you are more likely to change your state.

      IRV should never be used and is worse than plurality. It violates the all important monotonicity principle. That means if you vote for someone they could lose. Approval is better and easier for people, dumb in general, to use.

      --
      Why don't you guys have friends or journals?
  23. Re:The Diebold machines are funked... by blincoln · · Score: 3, Interesting

    You know that a sabot is that little metal band that goes around a subcaibre bullet, right? Why would I rattle that?

    A "sabot" is a shoe, which is why the bullets and other projectile weapons have them.

    The sabot reference in the grandparent post is to "sabotage," where workers angry over automation replacing their jobs threw their wooden shoes into the machinery to destroy it.

    --
    "...always new atoms but always doing the same dance, remembering what the dance was yesterday." -Richard Feynman
  24. Still Problematic by blunte · · Score: 2, Interesting

    Fill in the blank doesn't even work.

    I remember during some of the analysis of the 2000 Florida election disaster that one of the recount counties gave facts about the number of ballots that had multiple votes. IIRC, dozens had at least 2 votes, many had 3, some had 4, and a couple even had 11. This means some voters are either completely hopelessly confused, or they're screwing around.

    Also, remember the election officials in each county have great capacity to screw things up.

    As with most problems, the root of the failure is lack of education. There are just a lot of ignorant people out there. This may or may not be their fault. They shouldn't be voting if they can't understand the words "MARK ONLY ONE".

    --
    .sigs are for post^Hers.
  25. WTF ? by Tsiangkun · · Score: 2, Interesting

    I could write a piece of voting code that reports results into a central database after each voter,
    doesn't leave a paper trail, and ensures one use per voting card in an election.

    This seems so trivial, I wonder how they screwed it up, and why aren't they being prosecuted as terrorist who are trying to hijack american democracy with an electronic attack on our election outcomes.

    Why are the governments paying for the priviledge of being hijacked, and why aren't they demanding a full refund for the machines ?

    Imagine if Abduhla Musctaffa owned the company making these machines . . . and he had promised to deliver the election to their party. Would the US government be equally lazy about investigating the potential tampering with the system ? Would the voters be equally complacent ? I suspect that they would [ be lazy] , but that doesn't mean I'm not outraged by the whole fiasco already.

    --Tsiangkun
    ***---***
    I'm Tsiangkun Tzu and I authorised this sig

  26. Re:Real counting? by Dopescuzz · · Score: 2, Interesting
    So no excuses - you could fix it with a system that works if you wanted to!

    While I would never go quite as far and say that 'our system works well', I would also disagree with your generalization that the problem is a simple fix, i.e., the 'British Example'. One of the main sticking points with our system is the number of items we vote for within our Federalist system (dog catcher, State Supreme Court Judge, Mayor, trustee, the list goes on and on.) In essence we have AT LEAST three complete, distinct, and seperate layers of government to vote for [not including counties, which exist at the behest of state governments, but typically act independently]. Additionally, election laws in each municipality and state were influnced by the progressive era, a time in our history when we wished to 'run out political bosses and corruption.' The laws themselves dictate common sense, but also make local elections that much harder to maintain. Now, I will never claim to know the inner workings of your political system better than a citizen of the UK (yourself), but from what I understand (from a single intro poli sci class I took so like I said, I'm no expert) your system is much more centralized, even with the push for what we call 'local rule'.

    So, what the hell is my point? Well, nothing easy is ever simple. Our government, by design, was created to be inefficient (cliche: Moussolini made the trains run on time). Our voting system's complexity cannot be explained away easily by pointing to population density, or sweeping generalizations, rather, it has become the miasma that it is because of history, politics, competing/independent entities and inertia. I assert that it would be much more difficult to 'fix' our current system than many (including and especially companies such as Diebold) simply because there is no other system in the world like ours.

    Of course, what the hell do I know, I'm just a Joe Sixpack.

  27. Complex? Maybe, but... by fiannaFailMan · · Score: 2, Interesting
    Well in Ireland they use Single Transferable Voting where you vote for multiple candidates. It's all done on pencil and paper. It's simple for the voter, but damned complicated for the number crunchers who have to count the votes. Counting can take days and it can take nearly a week to find out which parties are going to form a government.

    The big advantage is that it's totally secure. Sure it's a bit more complicated than marking X in the box for a single candidate like in the British system, but it should remain a manual process regardless of the cost. Democracy is too important to be left to companies who are 'determined to deliver the next election to George Bush.'

    --
    Drill baby drill - on Mars
  28. Re:ramifications... by nametaken · · Score: 2, Interesting

    I'm particularly interested in the part about the reporting all just disappearing. ESPECIALLY from Drudge. Like you said, that sort of thing never seems to happen. But of course, how could one ever prove it did? And if so, why? Does it all add up to conspiracy? Perhaps. I don't know. I would never say that it's outside the boundaries of possibility, just probability. I agree, criminal behavior is usually not completely rational. On the other hand, I also suspect that Diebold management is smart enough NOT to choose such a public venue if they're going to engage in "deliberate malfeasance". So basically, I'm inclined to believe it's serious incompetence. But then, maybe mine is the sort of attitude that let the aliens kill Kennedy, and get away with it. :)

  29. Must be nice to be friends with a president... by Genda · · Score: 2, Interesting

    I would just about kill, to have a job like Diebold. It'd be totally wonderful to have a job, where you could fail over and over and over again, while receiving millions of dollars worth of federal (your tax dollars at work) business. All without the slightest bit of concern that somebody somewhere might ask... "Do you guys actually have any clue what the heck you're doing?"

    So far, all I see is security failures, operational failures, service failures, and a huge progression of operational and technical SNAFUs. I'd prefer not having to stand on my head to vote... (I'd like easy and simple as much as the next person), but if I can't trust the results of the process... then for all intents and purposes, I cease to be participating in a republic. Either we eliminate the faulty process (up to and including the elimination of the offending service provider), or we eliminate the people who won't eliminate the faulty process.

    Genda

  30. The biggest problems aren't getting attention by adamsc · · Score: 2, Interesting

    It's pretty scary to see how little the local Registrar of Voters cares about having any sort of verifiable voting system. The official FAQ even has two entries regarding reliability (how do I know my vote was counted accurately after casting it, what happens if there's a recount) and studiously avoids answering either one - in the first case they simply stop after describing a bunch of irrelevant steps which happen before you cast your vote and in the second they pretend that a generated image file stored on the machine is somehow more valid than the stored vote record on the same machine.

  31. Re:Real counting? by Anonymous Coward · · Score: 2, Interesting
    Dont get me started on these people..

    They then use flashcards to move the votes from the voting machines to a single machine connected to a landline, which uploads the votes to another machine.

    Having worked for these bastards IN San Diego, I can tell you that EVERY machine has a modem. Yet they chose this "secure" scheme. (The OS is WinCE.)

    Diebold's rep was the sleaziest guy, lied to people in the eye, their hiring agency was no better - I've seen 4 people come to work on their first day and say "This is bullshit!" cuz they were expecting something entirely different.

    Talk about security. They must have "misplaced" around 400 sticks of 128MB CF cards (around $30 a piece) with software updates, Im not even talking about the flashcards that grant local admin level acccess to the machines, work hours were manipulated left and right, people werent getting paid, etc etc etc. Took me a coupla days to figure out who was I dealing with. Still waiting for that check, btw.

    ANYWAY.

    Dont expect fair elections if these guys are going to be involved.