Playing Pac Man on a voting machine

I think this is just hilarious. Hooray for universal Turing machines!:

The story behind it is scarier, however:

Sequoia’s voting machines, used in some 20% of U.S. elections, employ Intellectual Property (IP) still owned by a Venezuelan firm tied to Hugo Chavez. Sequoia itself is now owned by a Canadian firm called Dominion. (Though Dominion, like Sequoia itself before them, lied about the continuing Venezuelan/Chavez ties in their recent announcement of the acquisition, as detailed exclusively by The BRAD BLOG, to little notice, in June.)

The Pac-Man hack onto the Sequoia/Dominion voting machine was revealed this week. It was accomplished without breaking any of the “tamper-evident” seals that voting machine companies and election officials claim are used to ensure nobody can physically hack into them without being discovered.

“We received the machine with the original tamper-evident seals intact,” the hackers from Princeton and University of Michigan report. “The software can be replaced without breaking any of these seals, simply by removing screws and opening the case.”

This particular Sequoia DRE (Direct Recording Electronic) voting machine model is known as the AVC Edge. It used to be described on the Sequoia website and promotional materials as “tamperproof”. It has been hacked previously and has failed time and again in recent elections, even though election officials continue to force voters to use them.

For example, the AVC Edge miscounted votes in New Jersey in 2008, the same election during which the systems also failed to even boot up when polls opened at a Hoboken precinct, forcing voters, including the state’s then Governor John Corzine, to wait some 45 minutes before they could cast a vote on it at all. Whether those votes were recorded accurately as per the voters’ intent, once the machines finally booted up, is scientifically impossible to know. Use of any touch-screen voting machine is the equivalent of a 100% faith-based election. No votes cast during an election — none — can be verified as having been accurately recorded on such systems. Ever.

My brother comments: “While our elections may not be safe, at least we’ll have retrogames when the hacker party takes office.”

  • Arno

    A voting machine should never, never, never be able to act as a UTM. I cannot understand how voting machines are getting used that haven’t been proven correct.

    The Chavez-connection however is irrelevant: If you have to put any trust in the manufactorer at all, you are doing it wrong anyway.

  • RW Ahrens

    Almost as scary as the video of a programmer (in Florida, I think) admitting to a judge in court that he had not only hacked into voting machines, but had actually changed the results of an election. Without being detected – and almost bragging about how easy it was!

  • James Sweet

    As a computer engineer, I have found it very odd the way that companies have approached the challenge of doing electronic voting machines. For something like that which is a) single function and b) requires very high security, I don’t understand why you wouldn’t just build it as an embedded system from the ground up, making everything as simple as possible. And of course producing a redundant paper trail.

    EVMs can be done right, I have every confidence they can… but there seems to be a very strong desire to it wrong, for some reason.

    I also agree with Arno that the Chavez connection is irrelevant conspiracy-mongering.

  • Chris Hallquist

    Wasn’t even thinking about the Chavez thing when I cut and pasted that. I just included that paragraph because of the factoid that 20% of US elections use these POSes.

  • Pteryxx

    I cannot understand how voting machines are getting used that haven’t been proven correct.

    EVMs can be done right, I have every confidence they can… but there seems to be a very strong desire to it wrong, for some reason.

    …Is this really a new concept to y’all? Ever since electronic voting machines were introduced, they’ve been ridiculously insecure, forced into use by profiteering voting-machine companies, and broadly protected from any sort of judicial or public investigation into their failures or vulnerabilities.

    See, for instance this CNN report from 2008:


    Author Ellen Theisen appeared on CNN’s Lou Dobbs show last night. Transcript:

    VOICEOVER: And many election officials across the country are outsourcing our very democracy to the private companies that have made the e- voting machines that are suspect and are not being certified by state election departments. Their interest appears to be in profit rather than in democracy. Stay with us.


    • F

      Naw, it isn’t new. All EVMs in the U.S. have been teh suck. They are prone to failure in many ways, not just through device tampering insecurity.

      Thing is, they hardly have to be that way.

    • Arno

      It’s not ignorance of the fact that this happens, but rather a mixture of confusion and despair that the public lets it happen.

  • Nomen Nescio

    as a computer programmer, i still can’t quite understand why anybody wants to do electronic voting machines. i mean, other than for to rig elections and change outcomes, that is. the hardcopy all-done-on-paper way is known to work, is known to be verifiable, and with the proper organizational set-up will report results on the same day polls close in the vast majority of cases. where’s the benefit in doing it electronically, for anybody except someone who wants to steal an election?

    • Nomen Nescio

      oh, and, if for some reason one did want to do an electronic voting machine and do it right, the effort would be non-trivial.

      all the software on the device i’d make sure would be open source by legal requirement, just for starters. the reliability of elections hinges on it, so the populace at large should have a right to see it — even to own it; making it totally public domain might be a good idea. screw intellectual property, these issues are too important for copyrights and royalty payments to matter, whoever writes this code can take a single lump sum payment for the service and be happy with it.

      i’m undecided between the “make it all embedded and hardware-closed” approach, let’s call this the “voting machine as flexible as a toaster” plan, and the “open but verifiable hardware” approach where the voting machines are turing-complete and general-purpose but can be verified to function to a known, set specification before, during, and after an election. i lean toward the latter, but that’s because i’m a software person.

      the whole verifiable, recountable record of votes cast problem is just plain a headache, though. it absolutely must be solved before the damn things can be called reliable, but it’s very, very far from simple once votes are no longer cast on hardcopy.

    • Pteryxx

      I agree… I get serious “but but voter fraud!!” non-issue vibes from everything about electronic voting machines.

      • Arno

        Let’s assume every voter has a secure private/public key pair anyway. The voter submits her vote encrypted with her private key, and receives back a copy of her encrypted vote, encrypted again with the private key of the voting station.

        That way everybody can verify the tallying of the votes, and that his personal vote is included in the official database.

        If the assumption were true, this system would be far superior to the paper thing going on now.

        • Nomen Nescio

          true, but how do you keep that from killing vote secrecy dead? it’d seem to indelibly entangle the vote with a personally identifying pubkey.

          assuming, of course, that the organizational and sociological headaches with getting everybody a keypair and establishing identity and trust for all those keys could be solved first. if it could be i’d still be using my PGP key, but i’m not because i was the only one in my social circle who knew what “PGP” was.

        • Pteryxx

          Um, but I don’t much care about verifying that the system gets MY vote right. I care about verifying that the system is getting ALL the votes right. Without a majority of the individual voters having to stand up and declare how each of them voted in order to verify the whole damn election result.

          We used to have exit polling, paper trails, chains of custody, and preliminary counting in all voting precincts (thus reducing error from any single source). A lot of those failsafes are no longer policy, if not outright illegal.

  • Jim Baerg

    I have voted in many Canadian elections & it has always been by marking a piece of paper & dropping the paper in the ballot box.

    That seems to work fine here. Why does anyone want to go to voting machines of any sort?

    • Pteryxx

      The only halfway sensible reasons I’ve ever seen are: first, that voting machines could be more accessible to disabled people than paper ballots that have to be marked by hand. Generally disabled or low-vision people need a trusted assistant, sometimes a volunteer at the polling place itself, to mark the ballot according to their wishes. They could also use absentee ballots. The other halfway sensible reason is that paper ballots are sometimes confusing and error-prone (see: butterfly ballots) but again, that’s easily addressed by sensible, transparent ballot design.

      CNN: Butterfly ballot cost Gore the White House

      Voters confused by Palm Beach County’s butterfly ballot cost Al Gore the presidency, The Palm Beach Post concluded Sunday.

      The newspaper’s review of discarded ballots found Gore lost 6,607 votes when voters marked more than one name on the county’s “butterfly ballot.” A leading Republican called the finding “speculation.”

      Voters who marked Gore’s name and that of another candidate totaled more than 10 times the winning margin Bush received to claim Florida’s 25 electoral votes and the White House, the Post concluded. The newspaper said the result was “an indictment of the butterfly ballot, political experts and partisan observers agree.”

  • Pteryxx

    Also I’ll just leave this here:

    Electronic voting machines in Florida may have awarded George W. Bush up to 260,000 more votes than he should have received, according to statistical analysis conducted by University of California, Berkeley graduate students and a professor, who released a study on Thursday.


    They discovered that in the 15 counties using touch-screen voting systems, the number of votes granted to Bush exceeded the number of votes Bush should have received — given all of the other variables — while the number of votes that Bush received in counties using other types of voting equipment lined up perfectly with what the variables would have predicted for those counties.

    The total number of excessive votes ranged between 130,000 and 260,000, depending on what kind of problem caused the excess votes. The counties most affected by the anomaly were heavily Democratic.

    Sociology professor Michael Hout, who chairs the university’s graduate Sociology and Demography group, said the chance for such a discrepancy to occur was less than 1 in 1,000.

  • P Smith

    The stupidest thing about US elections is the impatience of the media, politicians and voters. Of all democracies and pseudo-democracies (of which the latter includes the US and Russia), the US has the longest period of time between election day and inauguration day, two months.

    In parliamentary democracies (e.g. Canada, England) paper ballots are counted and recounted (i.e. counted TWICE) within days of the election and the government is sworn in. In multi-stage democracies where voting happens on more than one day, votes are counted and recounted by hand at each stage. Why is the US, a country more wealth and resources than any other, incapable of counting hand marked ballots or at least ensuring the results are correct?

    US elections run for terms of two, four or six years, yet Americans are too damned impatient to wait two days, let alone two weeks, to get correct totals. You can count votes fast, or you can count the votes correctly – you CAN’T have both. With the amount of time Americans have between election and inauguration, the only reason to have instant and final results within five minutes of the polls closing is that you don’t really want a democracy.


  • Pierce R. Butler

    If Hugo Chavez is pulling hidden strings behind the vote-tallying process, why does it always work out that Republicans win the questionable counts?

  • Pingback: Have fun in fun world