E-voting company slapped with lawsuits for terminal stupidity

By Martin Bosworth

Both the city of San Francisco and California State Secretary Debra Bowen have sued electronic voting machine maker ES&S for selling uncertified and untested voting machines to California counties. The lawsuits come on the heels of Bowen ordering thousands of the machines decertified for basically being crap in a box. Bowen is seeking $15 million in damages from her own lawsuit, and the City by the Bay’s additional suit could pile millions more on top of that.

How did this ever come to pass?

Among other things, ES&S apparently not only lied about the true nature of the machines (that they were new, uncertified versions), but failed to disclose that the machines were partially manufactured overseas, and threatened to hold Bowen personally responsible if her investigations revealed any trade secrets. From the Chronicle article:

California only learned about the changes when an ES&S representative inadvertently mentioned the new version of the AutoMARK in a telephone conference call with state election officials. The company never even mentioned to the state or the five counties that changes had been made to the machines that were shipped, Bowen said.

The company already is facing a suit from San Francisco city officials angry that restrictions put on ES&S equipment used in the Nov. 6 mayor’s election cost the city an estimated $300,000 and delayed a final vote count for more than a week.

Because of concerns that the city’s aging voting machines could not read ballots marked in light-colored ink, Bowen ordered each ballot to be individually checked by an election official before it could be counted, a process that added days to the vote count.

Every time a company like ES&S fails to live up to the standards expected for running a smooth election, it sets back the otherwise worthy goal of fully electronic voting by years, if not decades. Every time what should be the responsibility of the public infrastructure is placed in the hands of a private company and they drop the ball, it makes voters that much more disenfranchised and uninterested in participating in the process. As I said in my debate with Daniel Castro over e-voting and audit technologies, these things need to work right the first time–there’s no excuse for putting out sloppily coded machines that are supposed to handle the essential task of ensuring votes are tabulated accurately.

The real kicker? California is trying to replace its ES&S contract with one from Sequoia Systems, another e-voting company that has its share of problems with defective machines and tabulation screwups that delay election results. It isn’t really much of a consumer choice when every choice amounts to eating a pile of feces or stabbing yourself in the eye with a fork, as my friend Joe would say.

We can and must do better than this. Our democracy deserves it.

ADDENDUM: I came across a great Ars Technica story about Andrew Appel’s look at balancing convenience and security in electronic voting. Excellent quote:

The solution, in Appel’s view, is to try to ensure that voting software is designed for security from the ground up, and has a voter-verifiable paper trail. He says that the biggest current problem is explaining best practices on security to the legislators that need to demand them. Unit testing, code analysis tools, and data validation tests could all help with the basic security of voting machine code, but it’s clear that the manufacturers of these machines aren’t going to adopt these practices unless they’re forced to.


4 replies »

  1. I think I must be missing something. It seems to me that the whole electronic voting issue could be solved with a simple fix.

    Surely, it’s more efficient and easier to vote electronically than to use paper systems. The problem is that the lack of a paper trail makes software mistakes or fraud much more likely.

    Why not have every electronic voting machine print out the voter’s choices on a paper record? The voter could then verify the choices and drop this printed ballot into a ballot box so the results could be checked at a future date, if necessary.

    It seems so simple, I must be missing something. Right? Right?

  2. JS,

    You would think, right? Yet e-voting proponents react to the idea of paper audit trails like you suggested they share needles with a junkie for fun. Read through my debate with Castro and his paper–these companies know that their products are defective, but are pushing junk science to debunk the idea of paper trails in order to cover up their own machines’ faults.

    It’s ludicrous.

  3. Be careful of using the “junk science” statement, Martin, lest you be associated with, the purveyors of “tobacco smoke isn’t dangerous for you” and “pesticides aren’t dangerous when you eat them” and “global warming isn’t real and there is no scientific consensus that it’s even real”.

    Besides, calling something “junk science” or for “sound science” is a frighteningly effective frame that was created by Philip Morris in the 1990s to fight tobacco legislation. Be careful when using a right-wing frame.

    Just a little foreshadowing for next week…. 😉

  4. Brian,

    Another valid point. I tend to think of that phrase for use in describing such right-wing constructs as you describe, but a frame is a frame is a frame.