Follow Slashdot stories on Twitter

 



Forgot your password?
typodupeerror
United States Government Politics

Aftermath Of Failed Electronic Voting 84

Posted by timothy
from the how-do-you-recount-an-electronic-ballot dept.
dstates writes "The Christian Science Monitor and NPR report that failed electronic voting machines lost thousands of votes in Carteret County North Carolina, and the election for state agriculture commissioner is headed to court. A combination of human error (setting the machine to record a maximum of three thousand votes when eight thousand people voted) and a software malfunction (the machine kept accepting ballots after its memory was overloaded) resulted in the loss of 4,500 votes in an election decided by only 2,300 votes."
This discussion has been archived. No new comments can be posted.

Aftermath Of Failed Electronic Voting

Comments Filter:
  • This was reported in the "Election Results Weirdness Continues" story, some time back.

  • > A combination of human error (setting the machine to record a maximum of three thousand votes when eight thousand people voted) and a software malfunction (the machine kept accepting ballots after its memory was overloaded)

    Sounds like the errors should cancel each other out.

    • Unfortunately, the errors tend to mostly favor one particular side...
  • Ok, so it only recorded 3000 votes of 8000? I'd guess that would mean 5000 votes were lost, not 4500. But let's just think about those 3000 votes for now... the election was decided by 2300 votes. So that means, of the 3000 votes that got counted, 2650 were for Candidate A and 350 were for Candidate B. If Candidate A got 88% of the vote so far, isn't it somewhat silly to think that it's going to suddenly swing to Candidate B if they count the other 60% of the ballots?
    • by over_exposed (623791) on Sunday December 12, 2004 @08:32PM (#11069438) Homepage
      I think you're missing the point... If that many votes weren't counted, there is something very very wrong with the system. Who cares who had more votes in the first XX% of voters who showed up? Fix the problem, don't argue semantics.
    • That's like saying "I'm going to decide the outcome of this election by checking 5 votes from each district."

      How do you know that the votes wouldn't have gone for Candidate B?

      You cannot know.
    • First, you're assuming that these 8000 voters were the only in the county. That seems rather unlikely; I assume this is just one polling station. Second, you're assuming randomness. Even if you ignore the very real possibility of fraud (come on, who can't count voter rolls and compare to memory space?), those who voted earlier would have their votes counted, while those that voted later - the last 4500 - would be ignored. Working class voters that don't get off work to go vote until 5 or 6 would therefo
    • Aside from the fact that we should be counting all of the votes whether they affect the election or not...

      If Candidate A got 51% of the statewide vote and Candidate B got 80% of the vote in the district where the majority of the votes were thrown out, what does that tell you?
    • You do realize that this was a statewide election and that the problems happened at a single polling station in a single county (out of 100), right? ~2.5 million North Carolinians voted for President, and I would assume that about that number voted in this race as well. so 2300 votes is a margin of less than 0.1%. That is a pretty close race.

      I don't really have a dog in this race, despite being from NC, but it seems to me ($0.02) that they should only repoll the ~4500 people whose votes were lost. Ra
    • > isn't it somewhat silly to think that it's going to suddenly swing to Candidate B if they count the other 60% of the ballots?

      No, it isn't silly. As another pointed out, those voting later usually have jobs, whereas those without jobs (and more likely to vote Democrat, not that the "side" matters too much) would vote earlier. What if the first 5000 votes were missed? The result would have probably favored Republicans slightly. However, you are right in that I don't think it would usually affect the
  • All Human Error... (Score:4, Insightful)

    by curunir (98273) * on Sunday December 12, 2004 @08:31PM (#11069435) Homepage Journal
    A combination of human error (setting the machine to record a maximum of three thousand votes when eight thousand people voted) and a software malfunction (the machine kept accepting ballots after its memory was overloaded) resulted in the loss of 4,500 votes in an election decided by only 2,300 votes."

    It was human error on the part of the those who set it up and human error on the part of election officials who decided to use a product that wasn't thoroughly tested. Someone beyond the techs that administer the machines needs to be on the hook for this. Just because the machines that failed are electronic doesn't mean that there was no negligence on the part of those that chose to use them.
    • It was human error on the part of the those who set it up and human error on the part of election officials who decided to use a product that wasn't thoroughly tested. Someone beyond the techs that administer the machines needs to be on the hook for this. Just because the machines that failed are electronic doesn't mean that there was no negligence on the part of those that chose to use them.

      That's the problem with people that don't understand computers.

      They just bow to the computer gods and chant "Com
  • Who needs more than 3000 votes?
    • by Anonymous Coward
      "Who needs more than 3000 votes?"

      I hear that! I've never voted more than 2500 times in any one election - and I'm a Republican!

    • I originally read about this story several weeks ago. The company that produced the machines sold several different models of the machine with different price tags. The people who made the decision to buy a machine with a 3000 vote cap are partially to blame. The company who designed the machine is mostly to blame. The tried to pass the buck by stating that a little light would start blinking on the machine after the memory was full.

      Okay, I sort of understand the idea behind the pricing scheme, but the
  • After all this crap, isn't it time that Diebold (and similar companies) be dropped, and another one (preferably FOSS) be chosen? Or is that solution too obvious? I wonder how many more elections fall victim to stupid and/or rigged coding....
  • by ninjagin (631183) on Sunday December 12, 2004 @08:55PM (#11069527)
    What I can't understand is:

    A. Why you would have a maximum number of votes for a machine AT ALL.

    B. Why you would have something like a memory contraint AT ALL in these days of cheaper-than-dirt storage.

    C. Why you would have either or both of A and B if you wanted a fair election.

    Can someone fill me in?

    <tinfoil_hat action="dons">
    D'you think it's because North Carolina was John Edwards' home state, mebbe
    </tinfoil_hat>

    • The machine I voted on used scanned paper ballots, so it would presumably only be physically able to hold so many before they'd have to empty it or whatever. However I don't see any legitimate reason either to make a machine have a maximum of only 3,000 when something with the power of a four function calculator would damn near be enough to count many orders of magnitude higher. And why the hell does the thing have a selectable maximum anyway? This implies that someone deliberately added this as a feature.
    • D'you think it's because North Carolina was John Edwards' home state, mebbe

      Except, of course, that the errors were in Bush's favor. That means it's a vicious Democratic conspiracy, natch.

      Look man, I'm a Democrat because the GOP has become so jaw-droppingly criminal and almost cult-like. And don't give me any of that "they're all corrupt" BS, because when it comes to corruption, no one can hold a candle to the modern Republican party. Don't believe me? Do your own research. I did [kuro5hin.org], and was as surprised as

      • by MarkusQ (450076) on Monday December 13, 2004 @12:06AM (#11070249) Journal

        I'm a Republican, as are most of the people in my family, and for that matter most of the people in my state. And I don't know anyone who approves of what was done in this last election, once they are confronted with the facts. The closest is a sort of lame "well, they probably meant well" or "it must have been overly enthusiastic grunts"--but you can see in their faces that they don't buy it.

        But none of them are happy about it. We were raised, I guess, with those "moral values" that everyone's talking about. And I don't recall cheating on that list, anywhere. No, I take that back. There was "Cheaters never prosper" and "If you cheat, you only cheat yourself" and "Better to die for the truth then live a lie."

        But to hear the media tell it, we're all a bunch of saps that aprove of doing anything to win (When in fact we were taught "The ends don't justify the means." and "It's not whether you win or lose, it's how you play the game.").

        As my brother said at Thanksgiving, "I want my party back!"

        --MarkusQ

        • But none of them are happy about it. We were raised, I guess, with those "moral values" that everyone's talking about. And I don't recall cheating on that list, anywhere. No, I take that back. There was "Cheaters never prosper" and "If you cheat, you only cheat yourself" and "Better to die for the truth then live a lie."

          I wish there were more Republicans like you, instead of Sean Hannity/Limbaugh/random GOP apologist. The "good" Republicans seem to have been almost entirely sidelined, while those in power

        • As my brother said at Thanksgiving, "I want my party back!"

          Republicans aren't what they used to be.

          GWB puts shame on the whole party.
          I don't know what is worse, that he is acting as if he has a mandate when he just barely one, and some are even contesting the legitimacy of the election, or that the media doesn't seem to be questioning this. [reclaimthemedia.org]

          Politicians have always been scumbags, but while they used to be wallowing in a Cesspool scumbags, now they are wallowing in a septic tank scumbags.

          Think about
      • Look man, I'm a Democrat because the GOP has become so jaw-droppingly criminal and almost cult-like. And don't give me any of that "they're all corrupt" BS, because when it comes to corruption, no one can hold a candle to the modern Republican party. Don't believe me? Do your own research. I did, and was as surprised as anyone at the outcome.

        Are you on crack? That list is all well and good, but how conveniently you ignore Bill Clinton. The guy got a blowjob and lied about it so that his wife wouldn't find out. Impeachment was too soft a punishment for such a criminal act.

        • Bill Clinton ... got a blowjob and lied about it so that his wife wouldn't find out.

          So where in the world did you grow up?

          When I was a boy growing up in the US, one of the firm lessons that was drilled into us was that a guy with any brains wouldn't "kiss and tell". We had a name for such a guy; we called him a "jerk". If you wanted the slightest chance with the chicks, you'd keep very quiet about what you did with them in private. I understood all this at an early age, perhaps because most of my good
          • One of the reasons you said Bill might have lied:

            "Talking about your sex life would get you a bad rep and cut your chances."

            So Billy just wanted to be sure he'd be able to get some from someone else when this was over? Is that a good president? Someone who cheats on his wife, then lies about it so he doesn't cut his chances with the other women he wants to cheat on his wife with?

            "As for the "lying to Congress" charge; Congress had no business asking him about his private sex life."

            So are you s
          • While I agree with most of your points. Clinton was asked, in a sworn deposition during a trial whether he had sexual relations with Lewinsky, to establish a pattern of behavior. This was during the Tripp trial (if memory serves), where the pattern of behavior would have been relevent.
            What he did was perjury, something for which you and I, mere mortals, would have spent a little time on probation for (first offense, probably wouldn't have served any time). Now, did this warrant an impechement? Hell, no
            • How would you feel if you were on trial for let's say shoplifting. Now you got your wife/girlfriend in the courtroom to give you support and out of nowhere, the prosecuting lawyer asks you if you've ever cheated on your wife/girlfriend. Now let's say that you have. I have all my money on you saying that you never cheated on your significant other. Especially since it has nothing to do with the original case at hand.

              Maybe you need to read up a little more on your history and why Clinton was actually on tria
            • by Anonymous Coward
              What he did was perjury

              No, its not, in the slightest. Wether or not Clinton had had consensual sex with Monica is irrelevant to wether or not he harrassed Paula Jones(Tripp trail? wtf?). Even the judge said so, look it up if you want to.

              something for which you and I, mere mortals, would have spent a little time on probation for

              Nonsense. You forget that Starr and Republicans in Congress didn't set out to convict him of any specific crime, they set out to convict him of *something*, by any means nece
        • Sex != Treason and theivery. Blowjobs ain't in the 10 commandments.
      • In Bush's favor?

        In Iowa, a poll supervisor had to drive hundreds of miles to find a working counting machine. Worse, the state didn't accept a federal absentee ballot for military personnel, which meant that some Iowans fighting in Iraq were not able to vote.

        Given that the majority of the Military are republicans, this would be a blatant way of tipping the results towards a Democratic candidate.
      • Except, of course, that the errors were in Bush's favor.

        The lost votes that this article is referring to were in Carteret County. Carterey County is heavily Republican (Bush had about 70% of the vote there), so lost votes in that county hurt Bush and the GOP. In other words, you are full of crap.

        Look man, I'm a Democrat because the GOP has become so jaw-droppingly criminal and almost cult-like.

        I think it is sad how irrational and hate-filled the Democrats have become. Seriously.
      • I checked out your Kuro5hin page, and one thing stands out:
        Bill Janklow incident in which he was driving intoxicated and killed a motorcyclist.

        It really has nothing to do with the republican party, or their criminal activities.

        It could have happened to anyone who drives drunk, and there's no evidence that republicans do it more often than democrats.

    • A. Why you would have a maximum number of votes for a machine AT ALL.

      You might want to limit the number of votes to the total number of registered voters in a precinct, to prevent ballot box stuffing.

      B. Why you would have something like a memory contraint AT ALL in these days of cheaper-than-dirt storage.

      (scratches head) ... you got me there.

      • Only problem is WHEN is the box 'stuffed'. Probably not 100% at the end. thus if you cap at say 5k votes and 1k vote twice somehow, unless they all vote last, you lose up to 1K real votes that count.
        A better system would be one that didn't have a cap, just big ass warning when it thought too many had voted.
        An even better sytem is one that prevents Joe Smith from voting in 2 states and three precints of each twice. Though that is much harder to do and maintain the anonymity required to prevent bought
        • Only problem is WHEN is the box 'stuffed'. Probably not 100% at the end. thus if you cap at say 5k votes and 1k vote twice somehow, unless they all vote last, you lose up to 1K real votes that count.

          True. I wasn't trying to say that it was an effective mechanism, only that there could be a legitimate rationale for limiting the number of ballots issued at a precinct.

          An even better sytem is one that prevents Joe Smith from voting in 2 states and three precints of each twice. Though that is much harder t

        • An even better sytem is one that prevents Joe Smith from voting in 2 states and three precints of each twice. Though that is much harder to do and maintain the anonymity required to prevent bought or coerced votes.
          Franklin County, OH already has this--they put too few voting machines in heavily populated areas and make people wait upwards of four hours to vote--and you can't easily spend four hours in line more than one or two times per day.
      • You might want to limit the number of votes to the total number of registered voters in a precinct, to prevent ballot box stuffing.

        But why? If there's 1000 voters in a precinct and you see a recorded total of 1000, you might think "Wow, every citizen is really committed to voting!" However, if you see a recorded total of 1001 (exceeding the limit), you know for sure that fraud occurred. A cap can only serve to mask fraud, imho.
    • "Why you would have a maximum number of votes for a machine AT ALL."

      To throw up a red flag if it looks like some individuals voted more than once.

      "Why you would have something like a memory contraint AT ALL in these days of cheaper-than-dirt storage."

      Lowest bidder.

      "Why you would have either or both of A and B if you wanted a fair election."

      It's not a fair election they're after, it's the appearance of a fair election that really counts. After Florida 2000, electronic voting appeared to be more fa
    • Why you would have a maximum number of votes for a machine AT ALL.

      Because talks with the client have made it clear that they want a way to limit the number of votes in certain "problem" precincts with a history of voting "problems", and you won't get the sale unless this feature is included.

  • by belmolis (702863) <billposer AT alum DOT mit DOT edu> on Sunday December 12, 2004 @09:22PM (#11069627) Homepage

    This is just further evidence of a deeply flawed system. There is absolutely no reason that we can't have an honest and reliable election system in this country. You can do with old-fashioned paper ballots and hand-counting in the presence of scrutineers from all parties. Instead we've got a mishmash of systems, many of them untested, many with known flaws, some of them run by companies like Diebold known to be both incompetant and dishonest. We can't be sure who won this election.

    • Less common non voting impropriety issue - why do we still use plurality voting? We have computers now, there's no reason not to use a Condorcet method [wikipedia.org] or some other form of voting system [wikipedia.org] that collects more data from a voter than just their first choice of candidite. Isn't it also significant who their 2nd thru Nth choices are? In the most recent presidential election we had alot of people fanatically in favor of Bush and almost as many (some would say more) that wanted him the hell out. Under a different v
    • There is absolutely no reason that we can't have an honest and reliable election system in this country.

      Actually, there are two major reasons: the Republican Party, and the Democratic Party.

      It used to be that the Democrats were the dominant party, and also the one that was well known for widespread voting fraud. But the Republicans seem to have learned, and have taken the lead in both areas.

  • I don't see the point about making a fuss over a canidate that is decided by 8,000 votes. If they don't represent that much who really cares enough to make a cort case out of it. People should pick battles more wisely and become a hippy.
    • Don't make a fuss over it?

      I guess I shouldn't expect more, seeing as how the majority of the population doesn't even register to vote, and only 50% of people registered to vote even bother to show up..

      It doesn't matter how many people are represented by the person, votes were thrown away by these voting machines.
      • It is true though - since these races are so close, the overall view must be that they are about equal in capabilities. Or seen another way, equally pathetic.

      • Don't take me the wrong way. Voting is worth your time but I don't expect miracles to happen when I do vote. I did vote but I'm saying if 8,000 people represent a very small fraction then their votes for who is president is more important then who is a local councilmember or something.
        • That is total bullshit and you fucking know it.

          Every single election, be it for President, Mayor, or Town Dog Catcher is important and ANY fuckups cannot be accepted.
          • Ya but the problem you have to remember is funding. Its a lot easier for a small town to fuck up because they have less people watching for problems. They aren't expected to be up to par with things. You know humans are going to be humans and to expect perfect results whether its with a computer or a piece of paper isn't fair. Accept that mistakes will happen when there is over 120 million votes being counted. 8,000 counted wrong does suck but hey its bound to happen and its statistically not possible to ha
  • by torndorff (566594) <tommy AT orndorff DOT com> on Sunday December 12, 2004 @11:58PM (#11070215)
    Here is some info on the voting machines used in the county.

    Unilect Corportation [unilect.com] is the manufactorer of the "Patriot Voting System" (because losing votes = being patriotic).

    Interactive demo [unilect.com] of their voting system!

    Verified Voting [verifiedvoting.org] has a Voter Information Sheet on the machine.

    Disinfopedia has an article [disinfopedia.org] about Unilect Corporation. From this article:

    The President of UniLect Corporation is Jack Gerbel, who has been actively involved in the election equipment industry since 1965. His career began in elections with IBM Corporation and then as a founder, Vice-President and Board of Directors member of Computer Election Systems (CES).

    Mr. Gerbel had the distinction of personally selling and installing more election systems than any other person in the U.S.

    Two major accounts that he sold and successfully installed were Cook County, Illinois and the City of Chicago.

    Mr. Gerbel became Vice-President of Sales for Business Records Corporation (BRC).


    So, there you have it. Background info. Side note: I live in NC and this is not the same machines that were being (these are the literal words of the poll workers) "tested" in Watauga County. And although they officially said these machines were only experimental and being tested, paper ballots were often withheld upon request and their availability was NOT posted. The Republic Party in Watauga County also refused to move polling locations onto Appalachian State University's campus, proposed by the Dem Party, although 22,000 of the 25,000 residents are students.
    • First screen of the demo:

      "To begin voting, press anywhere on the screen."
      "PRESS HERE TO BEGIN VOTING"

      I guess it's good that they think they're simple enough to use without instructions, since the instructions seem to have nothing to do with the actual software.
      • Ooh, ooh, it gets better:

        START OVER
        PRESS HERE WHEN FINISHED WRITING IN WRITE-IN COMPLETE
        CANCEL WRITE-IN

        Let's say you're an interface designer, and you have to assign colors to these... let's say your choices are green, yellow, and red (why!?)... how can you assign them to maximize confusion... let's make canceling red, confirming be green, and being unsure (restarting) be yellow... that's logical. Wait, let's flip them! much better!
    • Appalacian State University has less than 15,000 enrolled students.

      http://www.web.appstate.edu/news/glance.html [appstate.edu]

      Where did you get the 22,000 residents are students number?

      Are you counting High School and Grade School students also?

  • I think I can safely speak for /., when I say that the suspense in knowing who will be NC's Agriculture Commissioner has me stuck to the edge of my seat with excitement.
  • Corporate Freedom (Score:4, Insightful)

    by Doc Ruby (173196) on Monday December 13, 2004 @09:09AM (#11071816) Homepage Journal
    In that entire story, there's no mention of which corporation made the failed machines. There's talk of "human error", talk of "software error", but no talk of "corporate error", where a corporation sold machines that wouldn't work, didn't work, and have sent North Carolina at least into a constitutional crisis. Less than zero accountability for the corporations getting rich off the destruction of democracy.

    There's also no mention of the joke that is government testing and certification of these machines. Unless the elections controllers have demonstrated proof that the machines have been tested without failure or serious vulnerability, they must assume the machines will fail. And they can't claim ignorance of the risk, compounded by the absence of mitigation in a fallback auditable record, like a paper log. So these government officials, representing the people of North Carolina, are also unaccountable for their gross malfeasance.

    These people have violated the public trust in North Carolina most seriously. It's not necessary to prove they colluded to design a failed election, for their political or economic benefit. Their gross malfeasance has deprived thousands of North Carolinans their fundamental right to vote, regardless of its effect on the election, though there seems to be at least one office, Secretary of Agriculture, which is seriously damaged. The irresponsible people must be unmasked, and sent to jail for these serious crimes against the people.
  • by jc42 (318812) on Monday December 13, 2004 @10:25AM (#11072406) Homepage Journal
    All the reports just after the election claimed that these votes were lost because the machine ran out of memory. Now we're reading the explanation that someone set a max-votes limit to 3000.

    Are the reporters really so clueless that they don't understand the difference? Or maybe they do understand, but half of them are trying to put something over on us?

    I notice that TFA's explanation is "... an exhausted poll worker failed to notice a "memory full" caption on a machine, ...". But the slashdot abstract says "... setting the machine to record a maximum of three thousand votes ...", which conflicts with the article's explanation.

    So which is it? Inquiring minds want to know ...

It was kinda like stuffing the wrong card in a computer, when you're stickin' those artificial stimulants in your arm. -- Dion, noted computer scientist

Working...