Sweating Bullets for Democracy

This video has been making the rounds on the Internet. If you have not seen it, you should. If you have seen it and you were not affected, you might want to check to see if your heart is still beating. You could be dead. You should feel appalled and very, very scared. Nothing less than our democracy is at stake.

The video shows that a particular model of Diebold voting machine can be hacked to ignore a machine’s actual vote count and substitute its own. If that alone were not shocking enough, it also shows that the machine’s lock can easily be picked. The flash card that slips into its reader slot is easily procured commercially, and can be programmed with desktop computers. If this is not outrageous enough, it also shows how easy it would be for someone who is malicious to turn the program on the flash card into a surreptitious virus, allowing all sorts of voting machines of the same type to be hacked. One unscrupulous person in the election trust chain could circumvent the will of the people.

We can thank the Center for Information Technology at Princeton University for exposing these voting machine flaws. If you want to dig into the details, you can read their white paper. It should be a wakeup call for anyone who cares about sound voting systems. It seems a bit curious then that the White House has been silent on this matter. I am sure it has nothing to do with Diebold’s contributions to the Republican Party.

It used to be that our biggest voter fraud problem was ballot box stuffing. Those frauds were fairly easy to detect. Today, thanks to flawed electronic voting systems like this one, there is no point in cutting down trees to mark fake ballots. Now, whoever controls the voting booths (at least those that run Diebold machines) can decide who is elected. Moreover, no one will ever find out.

In many jurisdictions, the supervisor of elections is a partisan position. This was true in Ohio during the 2004 elections, for example. You may have read the article Was the 2004 Election Stolen?, by Robert F. Kennedy Jr., in Rolling Stone. It presents enough disturbing information about inconsistencies in that state’s vote to give any true democrat the willies. If partisan shenanigans occurred in Ohio in 2004, which looks indisputable (although it is disputable that it affected the outcome of the election), they sure went through a lot of hassle to do it. Now there is a simpler way: make sure all voters are using this model of Diebold machine and surreptitiously plant your vote rigging software in the machine.

The paper ballot has nearly gone the way of the milkman. Give me a punch card ballot over a Diebold voting machine any day of the week. At least there is some evidence with a punch card ballot of the voter’s intent. Thanks to Diebold’s vulnerable voting systems, there is no record at all. Even if there were a paper trail, unless the voter physically checks the duplicate paper ballot before leaving the voting booth, there is no way to know for sure that his vote was cast correctly. It is estimated that in this fall’s election, more than 80% of the votes cast will be cast on electronic voting machines.

Here in Fairfax County, Virginia we have a system with the dubious name of WinVote. I do not know about you, but its name is a marketing man’s nightmare. It implies either it works under Windows, which most of us know from experience has far more holes than a warehouse full of Swiss cheese, or that it will pick the winner of the vote, not you. I used it of course because I had no other choice. Even when I voted absentee last year, I still had to use the electronic voting machine.

Outsourcing the design of our voting machines to the private sector is a fundamentally and profoundly stupid idea. Neither politicians nor supervisors of elections are qualified to render a professional opinion on the soundness of any voting technology. If we must use electronic voting then we need not just standards, but standards that are non-partisan and represent state of the art best techniques and practices. Fortunately, such an institution already exists: the National Institute of Standards and Technology. It should be the ones writing the standards for electronic voting machines. Not one electronic voting machine should be allowed to be used which does not comply with their standards. Ideally, every voting machine would be individually tested and certified by NIST and come in a box with a NIST seal on it.

We also need a bulletproof and auditable process for voting. It should specify national procedures for voting, buying voting machines, setting up machines, and verifying accurate vote counts. It should ensure that no part of the process could be compromised. We can do it today for classified information. Why can we not give our voting process at least the same level of protection?

In my opinion, the most trustworthy form of mechanical voting was the one I used when I cast my first vote in 1976. When you pressed the lever, you could hear it click into place. When you pulled the master VOTE lever, you could hear the mechanical counters increment. The levers automatically reset to hide your vote. This kind of machine is low tech, but it has the virtue of being impossible to hack. All it takes is a simple hardware inspection to remove any ambiguity about the machine’s integrity.

Supervisors of elections need to be ruthlessly non-partisan. Ideally, a judge would oversee or appoint the supervisor of elections. This should be the last position filled because of political patronage. Only candidates with unimpeccable credentials and demonstrated organizational skills should be considered.

No voting process is foolproof, but only fools will allow voting systems to be used without the necessary integrity and process checks that ensure each voter cast exactly one vote, and that the vote was recorded accurately. This Diebold machine demonstration may be extreme. It may be more common than we think. Indisputably, it points to a larger problem that must be addressed if we are to truly call ourselves a democracy.

Leave a Reply

Your email address will not be published.