Electronic Voting: Truly, Deeply Stupid

Also in today’s L.A. Times, a frightening story in which election results are changed by electronic voting machine problems – and there’s not a damn thing we can do about it.

Although some Orange County voters cast the wrong electronic ballots in the March 2 primary, potentially altering the outcome of one race for a Democratic Party post, Registrar Steve Rodermund said he will certify the results of the election today.

In a report circulated late Monday to the Board of Supervisors, Rodermund acknowledged for the first time that his office’s failures could have affected a race … and gave ammunition to critics of electronic voting.

The report said 33 voters out of 16,655 in the 69th Assembly District received the wrong ballots and were unable to vote for six open seats on the Democratic Central Committee.

The candidate who finished seventh in that contest, Art Hoffman, trailed sixth-place candidate Jim Pantone in the final count by 13 votes. However, 99.7% of Orange County ballots were cast properly in the primary, Rodermund will tell supervisors today before certifying the election results to the secretary of state.

There are election-day issues in most elections (as we all can remember from 2000, right?) But e-voting machines are a particular problem, as presently constituted, because without a permanent paper trail, the votes – stored as records in a database – must be taken on faith.

In Florida, we could at least go back and try and figure out what happened. With paperless e-voting machines, there’s just no way.

There are a lot of things that can make e-voting work; open-source software and ISO9000 audits are two of the ones that I support.

Paper records are another, and I’d like to invite you to email California Secretary of State Kevin Shelley and suggest that he require them before electronic voting machines can damage another election. In this election, things worked out amicably.

In the case of the 69th Assembly District seats on the Democratic Central Committee, The Times analysis estimated that 19 to 38 voters had miscast ballots. Neither Hoffman, leading at one point, nor Pantone said they planned to challenge the outcome. Democratic Party Chairman Frank Barbaro said the party would resolve the inequity internally so the county wouldn’t face an expensive election.

We probably won’t be so lucky again.

21 thoughts on “Electronic Voting: Truly, Deeply Stupid”

  1. I agree. This rush to e-voting is scary. I never could figure out what was supposed to be wrong with the scantron (fill in the bubbles with a #2 pencil) method.

  2. Here’s my plan, it’s perfect as one would expect from a genius such as I….

    The company that makes these machines should be the same one that makes Powerball and other lottery machines.

    The ballot could resemble a lottery sheet, which we are all familiar with.

    Electronic screens walk you through the process.

    Upon voting, the machine will then spit out of the same kinds of printers as a lottery ticket…

    TWO receipts – A blue one tells you who you voted for.

    An identical red one that you place in a ballot box.

    You’re done. (POSSIBILITY: You might even be offered a chance to nullify your vote and re-cast…. If you are dumber than most of us.)

    Added plus, you would likely have preliminary results within an hour of poll close, in the same time frame that you can tell how many people won a given Powerball drawing.

    THEN

    The state governor or legislature or whoever, on election day, picks at random the names of four counties in the state. At poll close, the red ticket ballot boxes would be impounded in those counties and hand counted to make certain the results match (close enough) the electronic results. When they do…. Certify the election.

    Done.

    Then celebrate democracy.

  3. Oops! I knew I shouldn’t of asked as soon as I clicked send!!!!

    ;-)

    I read it, but I was processing “Shelley” as a first name. My apologies.

    Thanks for sending it on!

  4. That’s why I’m voting absentee from now on. On paper.

    A second solution would be for governments (Federal, state, local) to require that companies delivering voting machines and software be ISO-9000 certified. Defense contractors who aren’t certified can’t bid on government contracts. Why should voting machine makers, like Diebold, be allowed to have far less strict standards?

    And if there are no certified companies, we’ll wait.

    The other issue is, why do we have to base our voting methods on the least capable voters? Let us use punchcards, and anyone who wants a “Big Think and Do Ballot” could just ask for one.

  5. Why electronic voting? Because it sharply reduces the chances for vote fraud. Last election, for instance, in one Southern state, a change to electronic voting was accompanied by the disappearance of a decades-old Democrat majority.

    No more disappearing ballot boxes. No more multiple votes. No more manual counts behind closed doors by politically appointed supervisors. No more Chicago’s.

    When you read a criticism of electronic voting, you may very well be reading disinformation by people who have been stealing your constitutional rights.

  6. Votesnotes, are you kidding or just ignorant?

    Vanishing ballot boxes are accountable; you know they are missing, because they have serial numbers and are signed in and out.

    Multiple votes? I could vote once in every precinct in Los Angeles, by simply looking at the voter list online and assuming the identity of someone who is registered and hasn’t voted in the last two elections – what are the odds they will vote this time? Electronic voting protects against this how, exactly?

    No more manual counts behind closed doors…no, instead we get unduplicatable electronic counts by unaudited software of database records that could have been created by any number of illicit means.

    puh-leeze.

    A.L.

  7. I’m glad to see that this issue is getting aired in yet another venue. It’s not news; the implicit problems with electronic voting have been a matter of concern for two or three years but it’s awfully difficult to get the mainstream press to speak up about it, for whatever reason.

    Yes, the primary problem is that there’s no possibility of audit. If the machines goof up, someone hacks them, or an employee of the company includes hidden code there’s absolutely no way for election officials to check or find out.

    Here are some further examples of the problem. In early 2003, the source code for one of Diebold’s voting machines was leaked to the internet. Four computer scientists analyzed it and found dozens of security flaws in the system that would allow hackers to reprogram the results of any election: here’s their analysis. That kind of bad software certainly argues for open source.

    A hacker attack may have already happened: in the 2002 general election, the Comal County, Texas precinct, using electronic machines saw three local GOP candidates all receive exactly the same number of votes: 18,181 votes. (Yes, it could be a coincidince, and probably didn’t change the result of that election, as the GOP generally wins in that district.)

    In another instance – probably a glitch/bug – Boone County Ind. results showed more than 140,000 votes in the 2003 November elections. The district has only 50,000 residents. They caught this one because the numbers were impossible. If a glitch or attack resulted in numbers that were plausible, nobody would ever know, because the systems as they are are totally nonauditable.

    Open source and ISO 9000 are not bad ideas, but they don’t fix the problem. Paper receipts are better, and Shelley is to be praised pioneering the mandate while officials are still showing blind faith in “electronic is better”. As for Andrew’s suggestion, IMHO having a second receipt that people take with them is a very bad idea – you risk abusive situations like “bring me a receipt showing you voted for XXX or you don’t get your raise” or “We’re buying receipts showing votes for XXX, $5 each”.

    This isn’t, or shouldn’t be, a partisan issue. For democracy to work, we need to be able to trust that our votes are counted.

    Personally, I think we should just stick with paper, preferably marked with a pen or pencil rather than punched out. Canadians just mark inside a large circle with a pencil, and the ballots are counted by hand. Yet they return results just as quickly as the US systems, rarely with any controversy.

  8. I’d stick with the Scan-Tron forms, even if it does produce GRE flashbacks. You do at least get feedback when its read in that you didn’t double-vote, and there is a paper trail that can be verified.

    The electronic receipt thingie might work if the receipt were encrypted (with a random IV also, so no dictionary attacks) so that it takes a court order to release the key or something. That way, the union can’t require all its member to show their receipts.

    Of course, as long as you have the kind of shenanigans like precincts with 100.0% voting turnout (no one moved away or died?) or long lists of last-minute votes in identical handwriting, this is only a partial solution.

  9. ISO-9000 Certified? Does anyone who’s promoting that actually have any idea what it really means? It means only that the company

    1) Has procedures
    2) Follows them

    A company with the code development process “The programmer is given a specification, he writes the code, if it builds we ship it” is quite capable of getting ISO-9000 Certified. ISO-9000 is only about auditability, not quality. Certification means that what your company actually does is reasonably close to its official procedures. Claiming that ISO-9000 certification means quality software is the same as claiming that a successful financial audit means the company is profitable.

  10. As a conservative, I fully agree that this should not be a partisan issue. I don’t even understand why it is. This electonic voting stuff scares me, not because I think it’s a big right-wing plot, but because it’s so dang unreliable.

  11. Patterico, it’s a partisan issue because – all around the country – the Democrats depend on manual voting systems to steal elections. They have been doing this for decades, and they’ve been doing it everywhere they can. That’s why they are fighting like hell against it.

    Take away vote fraud and the Democrats are in real trouble.

  12. I don’t have up-to-the-minute coverage, but as of two months ago, of the four lawsuits brought over electronic voting, three were filed by Republicans. I don’t know why anyone who loses a race by less than 5% doesn’t file suit. The margin for error isn’t any better than that. Unfortunately, many races are closer.

    This is not a partisan issue. Sure, really bad things could be perpetrated by anyone who buys a $3 smart card and has had a few weeks of a programming class, but the possibility of things going wrong just because the code stinks is even higher.

    All we’re asking for is that our votes be as secure, private, and auditable as our bank accounts. I’m not sure why that’s a hard concept. And if it is, maybe we ought to look into the software our banks use. “Trust us” doesn’t work when it’s our money; why should it work for our votes.

    And, no, ISO 9000 is not the answer.

  13. It’s charming that votesnotes assumes Republicans win all elections, and any Democratic wins came from vote fraud. Couldn’t possibly be the other way around. Real sweethearts, the fellows you moderates get to run with.

    I suppose evidence would be too much to ask for? I mean, we have evidence that electronic machines don’t work (e.g., more votes than voters). Do you have any evidence of Democrats stuffing manual ballot boxes any time in the last 20 years? If not, could you go back to Free Republic and leave us alone?

  14. Lazarus, you can start by googling “the vote fraud archive”

    It shouldn’t take you long to find plenty of sources and examples. By the way, comrad, what’s your current project…finding ways to make sure our conservative-leaning overseas troops have trouble voting and getting their ballots counted?

    Here’s a teaser for you, off the web –

    Today the Senate will approve and send to President Bush a landmark bill that will upgrade voting machines and begin to curb the voter fraud that is creeping into too many close elections. It can’t come soon enough. Last week, a massive vote-fraud scandal broke out in a U.S. Senate race in Tom Daschle’s home state of South Dakota that could determine control of that body.

    The FBI and state authorities are investigating hundreds of possible cases of voter registration and absentee ballot fraud. Attorney General Mark Barnett, a Republican, says the probe centers on or near Indian reservations. “All of those counties are being flooded with new voters,” says Adele Enright, the Democratic auditor of Dewey County. “We just got a huge envelope of 350 absentee ballot applications postmarked from the Sioux Falls office of the Democratic Party.”

  15. Votesnotes: the Republican Atty General of South Dakota investigated the SD election and decided it was all a tempest in a teapot. The “fraud” charge was invented by a Republican journalist and never had much evidence. Google is good, but reading the papers is better.

    You probably don’t realize the irony that the overseas ballots you accuse me of wanting to get rid of are very easy to use fraudulently? (For instance, it’s much less likely that an American permanently resident abroad would be removed from the voter rolls upon death.) WTF? In America you want the new electronic machines, but for overseas votes you expect to lean Republican, you want the system you claim is easiest to tamper with. Well, that says it all, doesn’t it?

    Incidentally doesn’t hiring a Republican-run company to remove Democrats from the voter rolls because they shared very common names with felons qualify as vote fraud? Oh, no, ’cause the good guys are doing it.

Leave a Reply

Your email address will not be published. Required fields are marked *

You may use these HTML tags and attributes: <a href="" title=""> <abbr title=""> <acronym title=""> <b> <blockquote cite=""> <cite> <code> <del datetime=""> <em> <i> <q cite=""> <strike> <strong>