Error-prone Diebold voting machines sometimes change your vote

Oh, good:

Gary Rudolf, a voter at a polling site near Ft. Lauderdale, tried to vote for gubernatorial candidate Jim Davis (D); however, when the Diebold machine gave him the final review screen, it showed his vote was about to be cast for Charlie Crist (R). The problem took three tries to get resolved with the help of a local poll worker. Mary Cooney, a Broward County Supervisor of Elections spokeswoman, informed The Miami Herald that it’s “not uncommon for screens on heavily used machines to slip out of sync, making votes register incorrectly. Poll workers are trained to recalibrate them on the spot — essentially, to realign the video screen with the electronics inside. The 15-step process is outlined in the poll-workers manual.”

Oh, yes, because it’s such a good idea to rely on: 1) the voter to spot the error and say something before it’s too late (these are Florida voters, thousands of whom in 2000 accidentally voted for Buchanan, realized their error, voted for Gore as well, and then turned in their obviously worthless double-voted ballot, never thinking to ask for a new one); and 2) the 80-year-old ladies who work at the polls to successfully follow a 15-step process to “recalibrate” the computers. Problem solved! [/sarcasm]

Seriously, where’s Eleanor Green when you need her? I knew the new voting systems had some problems, but I never imagined anything this bad… yikes. Hopefully this problem is limited to Broward County, but why would we have any reason to believe that? (Hat tip: Briandot.)

Sounds to me like HAVA-induced madness has caused us to rush touch-screen voting machines into use before they’re ready for prime time, the result of which could be a huge election-day clusterf*** followed by a massive backlash that prevents such machines from ever being trusted again, even though — if designed properly — they have the potential to really improve* our voting system.

Yet another reason to recite the Election Administrator’s Prayer: “Lord, let this election not be close.”

*Improve, but not perfect. Nothing can perfect our voting system, because our voting system is a system of measurement, measuring the preferences of humans as expressed by humans and administered by humans; as such, it will always have a margin of (human) error, even if we could somehow eliminate all machine error. Thus, the Election Administrator’s Prayer will always be relevant.

UPDATE: George Will weighs in, brilliantly:

The lesson that should have been learned from Florida was: In Florida, as in life generally, one should pursue as much precision as is reasonable — but not more. When, as very rarely happens, a large electorate, such as that state’s 6.1 million voters in 2000, is evenly divided, the many errors and ambiguities that inevitably will occur during the marking of millions of ballots will be much more numerous than the margin of victory. That is unfortunate, but no great injustice will be done, no matter who is declared the winner in a contest that is essentially tied.

Unfortunately, the lesson the nation chose to learn from Florida was that American technological wizardry could prevent such highly unusual events, and no expense should be spared to do so. Hence HAVA, which made $3.8 billion available for states to purchase the most modern voting equipment.

On Election Day, 38 percent of the nation’s voters will use touch screens to record their choices, according to Election Data Services. Unlike optical scanners that read markings put on paper ballots, most touch-screen machines — including those that the New York Times reports will be used in about half of the 45 districts with the most closely contested House races — produce no paper that can be consulted for verification of the results if a recount is required.

Maryland’s new $106 million touch-screen system melted into a chaos of mechanical and human errors in last month’s primary election. Lawsuits have been filed in five states seeking to block the use of touch-screen machines.

Today’s political climate — hyperpartisanship leavened by paranoia and exploited by a national surplus of lawyers — makes this an unpropitious moment for introducing new voting technologies that will be administered by poll workers who often are retirees for whom the task of working a DVD player is a severe challenge. Furthermore, an election is, after all, a government program, and readers of Genesis know that new knowledge often brings trouble. So we should not be surprised if, on Nov. 7, new voting machinery does what new technologies — dams, bridges, steamships, airplanes — have done through history: malfunction.

(Hat tip: my dad.)

28 Responses to “Error-prone Diebold voting machines sometimes change your vote”

  1. Joe Mama says:

    This is retarded. Millions of computerized credit card transactions are carried out flawlessly every day, but we can’t vote electronically?!

  2. Angrier and Angrier says:

    Who says there’s a flaw.

  3. DFens says:

    Most voting machines can also be manipulated weeks in advance in as little time as 10-15 minutes. You swap memory cards and you’re done.

    People who don’t think that fanatical partisan grassroots supporters wouldn’t do something like this have never been around these people…and yes, I’m talking about both Democrats and Republicans here. Campaign offices get broken into all the time, and their networks get hacked…it’s just as easy to get at a voting machine.

  4. Brendan Loy says:

    As I said… not ready for prime time.

    I just hope, if there’s fraud/error with the machines, it benefits the Democrats. I say that NOT because I want the Dems to win, and certainly if there were fraud/error in favor of the Democrats, I would support it being corrected, even if that means that more Republicans would win. No, the reason I hope any fraud/error benefits the Dems isn’t because I want the Dems to win, it’s because if it benefits the Republicans, we may lose any hope of reducing the levels of cynicism on the Left to sane levels for a generation. That remains true even if it’s error, not fraud, or even if the fraud is at a purely local level, as you suggest… because if anything goes wrong and benefits the GOP, anything at all, it will be blamed on Karl Rove, Dick Cheney, etc., and the people making those accusations will never back down even once they are proven wrong.

  5. Brendan Loy says:

    (See, e.g., Ohio 2004. But that was a relatively fringe movement complaining. I fear for the health of our democracy if the appearance of impropriety causes the mainstream Left/Dems to start sounding like conspiratorial moonbats about the results of this election.)

  6. Joe Mama says:

    So, in other words, you’re hoping that any voting anomalies benefit the Left to placate the irrational cynicism they’ve heretofore shown? Outstanding. I’ve recently wondered what kind of heightened gasbaggery would emerge from the Dems’ should their much-touted electoral victories not come to pass next week, and whether it might be good for them to take back at least one House of Congress for that reason alone. But then I remembered that those who espouse such conspiracy theories are lunatics and morons, and are best not left wielding the power of gov’t.

  7. Brendan Loy says:

    I realize it’s not exactly a principled position. But I’m being pragmatic here. The crazy far-Lefties can’t be reasoned with, but we can still potentially win the “hearts and minds,” if you will, of more rational liberals, and keep them from emotionally/intellectually seceding from the political process. But if crying voter-fraud conspiracy from the mountaintop is going to be a biannual theme going forward, there needs to be a counterexample (outside of Washington state) where it happens to help the Dems (in a race that matters) rather than the GOP, so that those more reasonable libs won’t fall into the waiting arms of the Evil Hegemonic Halliburtionzation of Mother Earth crowd.

  8. dcl says:

    Brendan, you wouldn’t be able to prove those accusations wrong. ;)

  9. Brendan Loy says:

    Coming soon to a bookstore near you: “Why Do They Hate Us? A look inside the hearts and minds of the Blue-State Street” by Brendan Loy. :)

  10. Brendan Loy says:

    Conspiracy theories are inherently immune to being proven wrong… I also can’t prove that 9/11 wasn’t a collaborative effort by the Bush Administration, the Mossad and Skull & Bones, nor can I prove that JFK wasn’t shot by Gollum because he was wearing the One Ring while in that motorcade.

  11. Angrier and Angrier says:

    If machines are determining the vote, it could explain how the Terminator ended up as Governor of California.

  12. Lojo says:

    Phase Plasma Rifle in a 40 watt range.

  13. Joe Loy says:

    George F. Will ~ and by all means read the whole characteristically-elegant thing ~

    …For over two centuries before Congress passed HAVA, Americans voted. Really. Unlike today, those who were elected — Clay, Webster, Lincoln and lesser lights — often were more complex and sophisticated than the voting machinery.

    …Then came 2000 and Florida and the 36-day lawyers’ scrum about George W. Bush’s 537-vote margin of victory. In response to which, Congress passed HAVA, which in 2006 may produce fresh confirmation of the prudential axiom that the pursuit of the perfect is the enemy of the good.

    The lesson that should have been learned from Florida was: In Florida, as in life generally, one should pursue as much precision as is reasonable — but not more. When, as very rarely happens, a large electorate, such as that state’s 6.1 million voters in 2000, is evenly divided, the many errors and ambiguities that inevitably will occur during the marking of millions of ballots will be much more numerous than the margin of victory. That is unfortunate, but no great injustice will be done, no matter who is declared the winner in a contest that is essentially tied.

    Unfortunately, the lesson the nation chose to learn from Florida was that American technological wizardry could prevent such highly unusual events, and no expense should be spared to do so. Hence HAVA, which made $3.8 billion available for states to purchase the most modern voting equipment.

    …Today’s political climate — hyperpartisanship leavened by paranoia and exploited by a national surplus of lawyers — makes this an unpropitious moment for introducing new voting technologies that will be administered by poll workers who often are retirees for whom the task of working a DVD player is a severe challenge. Furthermore, an election is, after all, a government program, and readers of Genesis know that new knowledge often brings trouble. So we should not be surprised if, on Nov. 7, new voting machinery does what new technologies — dams, bridges, steamships, airplanes — have done through history: malfunction.

    Football, in its disproportionate pursuit of error-free officiating, now relies on instant replays because . . . well, because it can. This technology does indeed reduce human error. But it also reduces games to coagulation as players stand around waiting for officials to study video in the hope of achieving a degree of precision and certainty more appropriate to delicate surgery than to the violent thrashing of huge padded men in what is — lest we forget, as the judicial solemnities of instant replay cause us to forget — a game.

    Democracy is not a mere game. But — write this on a piece of paper, using a No. 2 pencil — neither is it an activity from which it is sensible to demand more precision than can reasonably be expected when, on a November Tuesday, 100 million people record billions of political choices.

    Spot On.

    :)

  14. Joe Mama says:

    “Coming soon to a bookstore near you: ‘Why Do They Hate Us? A look inside the hearts and minds of the Blue-State Street’ by Brendan Loy.”

    Heh. I heard Michael Barone speak at a Federalist Society luncheon last week and make a similar crack about an essay he would write entitled “Why Do They Hate Us?”, referring to the NYTimes :-)

  15. Aaron says:

    This will sound opportunistic in light of his recent Bush critisisms, and there’s no way I can prove otherwise. But it’s true nonetheless; I’ve always felt that George Will was the smartest big-name conservative columnist in the nation.

    He knocks this one out of the park. I only wish he had emphasized one point even more strongly. The most important lesson to be learned from Florida 2000 was that you need a procedural method for resolving statistical ties. That seems to be the one lesson practically noone actually has learned.

  16. Aaron says:

    ps.

    Regarding the “procedural method” I mentioned, it’s important that it be statisticly valid. In the case of a true statistical tie, recounts are useless.* A coin flip would be better.

    *A statisticly usefull ensemble of recounts might work, but we’re obviously not going to do that.

  17. Brendan Loy says:

    you need a procedural method for resolving statistical ties

    Hmm. Not sure I agree with that, because if there’s a different procedure for statistical ties, then you have to determine what is a statistical tie… and the arbitrary distiction between what’s a statistical tie (thus invoking the special statistical-tie procedures) and what isn’t (thus maintaining the normal procedures) will itself become a subject of hot dispute. Suppose the “statistical tie” cutoff is 0.1%; instead of litigating over who won, we would end up litigating over whether the apparent “winner” won by 0.11% or 0.09%.

    Seems to me, when there’s a statistical tie, you resolve it by the normal procedures… what’s needed is not a new set of procedures, but a new understanding on the part of the public that there really is no appreciable difference between a 537-vote Bush victory and a 200-vote Gore victory (or whatever), and we just need to understand that and accept the result that the legitimate process produces.

  18. Brendan Loy says:

    Your “coin flip” suggestion elucidates my point. Suppose any election where the margin is less than 0.1% is declared a statistical tie, and is decided by coin flip. So if Donnelly is ahead by 0.10%, he wins… but if he’s ahead by 0.09%, the results are thrown out and the election comes down to a coin flip. Problem is, the 0.01% difference between a 0.09% and a 0.10% margin is itself statistically insignificant, so how can you justify making such a distinction? And do you have any doubt that the Chocola campaign would demand a recount, and sue if necessary, if Donnelly has a 0.10% margin, since they’ll know the simply need to reduce the margin to 0.09% to suddenly have a 50-50 chance of winning?

  19. Aaron says:

    I had typed out a whole bit saying basicly: that’s a good point, but here’s why you’re still wrong. Then I read back over it and realized I wasn’t even really convincing myself. I’ll have to think on it.

  20. Joe Loy says:

    In CT we dissolve a Primary tie ~ literal tie ~ by Lot (usually a Coin-toss). An Election tie vote (literal) requires an “adjourned election”, i.e., a Re-vote between those Tied. / True story: one municipal-election year a small town had to run one of those and the Re-vote was tied too! My office (CTSOTS) made them Keep doing it until they Got it right. :)

    * * * * * * * *

    And now, with due obeisance to the incomparable Wisdom of Dr. GF Will as commentposted above & Updated on the main post:

    Oh, “Man of the Year” my tucchus :).

    “I knew the new voting systems had some problems, but I never imagined anything this bad… yikes.”

    Yikes, schmikes: This thing ISN’T “anything this bad.” Lay off the Lou Dobbs, old Kiddoe, that stuff’ll rot yer brain. :) [“Gimme a pint of DEMOCRACY-AT-RISK, with a shot of THE PERILS OF E-VOTING on the side.” ;]

    If I understand it correctly ~ I’m No Expert on Direct Recording Electronic voting systems ~ this sounds like isolated instances of desynchronization ~ misalignment ~ of the Video Touchscreen with the electronic components Beneath it, incrementally occurring due to Heavy Touching :> of said Screens.

    IF so, it is comparable to the occasional displacement of printed ballot-label strips, in relation to candidate Pointers, on mechanical-lever machines due to curious & meddlesome voters Pushing & Poking at the labels for No reason whatsoever ~ which Yes, such voters Do do. / Also to occasional misalignment of the Punchcard ballot [RIP :] when placed on/in the Ballot-info overlay thingie [elections technospeak :] prior to Punching.

    In any case, yes Brendan, it IS “…a good idea” ~ at least it’s a Necessary idea ~ “to rely on…the voter to spot the error and say something before it’s too late…”. That is precisely the PURPOSE of the final Voter Review page, to be scrutinized prior to pushing the “Cast Vote” button. Such review-screen would be Unnecessary if, but Only if, not only the Voter himself but also the DRE machine were absolutely guaranteed to be Error-Free in, respectively, (a) Indicating and (b) Recording such voter’s intended Vote.

    (Again with the Analogy: this is why after marking my unfamiliar new Optiscan Absentee Ballot yesterday, I Reviewed it 3 times before Sealing it in its Security Envelopes and handing it across the counter to the Town Clerk. I wanted to be Certain that whilst Bubbling-in mostly candidates on Row B, I hadn’t inadvertently marked the one in Column 2 on that row ~ since in that particular instance the fate of my Immortal Soul hung in the balance. :)

    Pertinent excerpts, Excessive emphases Tendentiously added :>, from the cited Oct. 28 Miami Herald article (not Linked here because registration required, albeit Free) ~

    After a week of early voting, a handful of glitches with electronic voting machines have drawn the ire of voters, reassurances from elections supervisors — and a caution against the careless casting of ballots.

    Several South Florida voters say the choices they touched on the electronic screens were not the ones that appeared on the review screen — the final voting step.

    Election officials say they aren’t aware of any serious voting issues…

    …Debra A. Reed voted with her boss on Wednesday at African-American Research Library and Cultural Center near Fort Lauderdale. Her vote went smoothly, but boss Gary Rudolf called her over to look at what was happening on his machine. He touched the screen for gubernatorial candidate Jim Davis, a Democrat, but the review screen repeatedly registered the Republican, Charlie Crist.

    …A poll worker then helped Rudolf, but it took three tries to get it right, Reed said.

    ”I’m shocked because I really want . . . to trust that the issues with irregularities with voting machines have been resolved,” said Reed, a paralegal. “It worries me because the races are so close.”

    Broward Supervisor of Elections spokeswoman Mary Cooney said it’s not uncommon for screens on heavily used machines to slip out of sync, making votes register incorrectly. Poll workers are trained to recalibrate them on the spot — essentially, to realign the video screen with the electronics inside. The 15-step process is outlined in the poll-workers manual.

    ”It is resolved right there at the early-voting site,” Cooney said.

    …In Miami-Dade, two machines have been taken out of service during early voting. No votes were lost, [county supervisor of elections Lester] Sola said.

    Joan Marek, 60, a Democrat from Hollywood, was also stunned to see Charlie Crist on her ballot review page after voting on Thursday. ”Am I on the voting screen again?” she wondered. “Well, this is too weird.”

    >> [But it wasn’t “after voting”. It was BEFORE casting her vote. That’s the Point of the review page. See below. / ~ the commentposter]

    Marek corrected her ballot and alerted poll workers at the Hollywood satellite courthouse, who she said told her they’d had previous problems with the same machine…

    Of course there will be E-voting (and non-E-voting :) Problems a week from today. And yes, given the Newness of various voting systems to so many voters & officials, there will be More problems than will in due course become Usual. And the nexus of many Close Contests with many New Systems ~ which people now mistrust even more than they mistrusted the Old systems which they incorrectly Blamed for 2000 ~ may well prevent the Outcome, re control of the House and/or Senate, from being known for a long time.

    But I doubt that there’ll be a Meltdown. / Well. Except in the artifically-generated red heat of the Litigation. / And possibly of Lou Dobbs’s Face. (I do look forward to that one. “I’m melllll-ting!” )

  21. gahrie says:

    Angrier and Angrier:

    Arnold won last time because the Democratic Gov. Gray was incompetent and corrupt, and even California will only put up with so much.

    He’s going to win next week because the democrats are running an empty suit with no ideas and no charisma. (no, it’s not Kerry or Gore…it’s their triplet brother Angelides)

  22. Andrew says:

    Are you sure Ahnold is running on the Republican ticket? I could’ve sworn he was the Dem nominee….

  23. gahrie says:

    Andrew: Believe me..I will be holding my nose next week. I voted for him last time because it was simply a matter of throwing Gray out. Hell I might even have voted for Meathead (Reiner) if he had been running.

    It’s really disappointing how awful the Republican candidates in Calif. tend to be.

  24. Anonymous says:

    Problems with the flat screen are not limited to Florida. During early voting last week, as I finished and touched “ready to vote”, I was informed I had not voted in every category, and would I like to review my ballot. I did–and every category, even those with only one choice, had been “marked”. So, I tried to register it again, and got the same message. So I again reviewed. Third attampt, same result. I flag my age 70+ poll worker to inform her of the error, to be told, “oh, it does that all the time. Just go ahead and hit VOTE.”

    Boy, does that make you feel like your vote is worth something…

    We have a serious, serious problem here.

    Charlotte, NC

  25. Andrew says:

    I for one welcome our Chavez-backed electronic overlords.

  26. Joe Loy says:

    LOL :). Andrew tu eres el Diablo. (Apologies to Bushmills Distillery :) Happy Halloween ;>

    And Now: “DEMOCRACY ON LIFE SUPPORT: THE COMPUTER MENACE! ~ Connecticut Edition(for transcripts contact loudobbs@CrappyNewsNetwork/theskyisfallingdept.com :) ~

    The new voting machines that will be used in 25 Connecticut cities and towns next week are vulnerable to tampering, but state officials are taking steps to make sure they’re not compromised, according to a report released Tuesday by the University of Connecticut.

    The optical scan devices, which automatically read paper ballots filled out by voters, can be compromised in a matter of minutes by tactics such as neutralizing one candidate so his or her votes aren’t counted or swapping the votes of two candidates, the report said.

    “Such tabulation corruptions can lay dormant until the Election Day, thus avoiding detection through pre-election tests,” according to the report.

    But the report credits the secretary of the state’s office for implementing new security procedures to protect the machines.

    Alex Shvartsman is the UConn computer science and engineering professor who heads up a team of professors known as the Voting Technology Research Center, which is advising the secretary of the state’s office. He said the state has taken important steps, such as strict rules for how the machines get from the supplier to polling places, tamper-resistant packaging of the machines and planned postelection audits.

    “If nobody touches the devices, if there is an unbroken chain of custody from the supplier to the polling place, then we’re very confident that nothing can go wrong with them — short of a mechanical malfunction,” Shvartsman said…

    I like the Tamper-Resistant Packaging part :). Gee, in My Day Sonny we never even Thought of encasing the Mechanical Lever machines in Shrinkwrap :>. [Theoretically they CAN be rigged y’know. Touch of artificially-induced Counter-dial Spin. Bit of Enhanced Early 000 Rollover. Then there’s my personal favorite: Candidate’s Pointerbounce. :] I think They should put a little explosive thingie inside the Packaging which will Splat forth indelible Iraqi purple ink onto the Tamperer’s ass. Sort of a Liquid Trail y’see. ;}

    Yawn. ;>

  27. A Nun Mouse says:

    Brendan,

    I’m just throwing several issues out to get your reaction, but when does the “appearance of impropriety” become actual impropriety?

    I mean, isn’t that part of this problem? You can’t TELL when the machine fucks up, unless you have an objectively verifiable outside source. For example, if you run tests on the machines while having outside observers, then you can confirm the miscount.

    You can have the machine spit out dual paper receipts: the voter gets one receipt and one receipt goes into a locked box linked to that particular electronic machine.

    But that raises problems, too. When do you recount the votes from any given machine to determine if that machine was acting up? What if the receipt shows you voted for the wrong person/people, even though your ‘confirmation screen’ showed the right person/people? How do they redo just your vote? Or worse yet, what if the receipt shows the right person, yet the vote registered on the machine was wrong? When do people get suspicious of any particular machine?

    My state and city uses what appears to me to be the simplest and safest method for voting: scantron ballots. You have the benefit of having an easily countable vote. You also have a hard copy of paper ballots with pens that can be used in any recount.

    I just don’t see any real advantage to electronic voting.