PACKET EDITORIAL, Sept. 19
By: Packet Editorial
Every year, U.S. Rep. Rush Holt introduces a bill in Congress the Voter Confidence and Increased Accessibility Act of whatever year it happens to be and watches it die a slow, excruciating death.
The 12th District congressman has the misfortune of being a relatively junior Democrat in a House of Representatives controlled by Republicans, but that’s only part of the reason his legislation, which would require that all electronic voting machines produce a backup paper record of every vote recorded, never goes anywhere.
The companies that manufacture the electronic voting machines have proven to be formidable foes of Rep. Holt’s initiative. So have local election officials across the country of both parties who rather like having virtual autonomy when it comes to counting the votes cast in their jurisdictions.
These defenders of the status quo, supported by the best expert witnesses and lobbyists money can buy, contend that creating a paper trail would be an enormously costly exercise. Moreover, they assert, it’s entirely unnecessary. The new touch-screen machines, they say, are more secure, more reliable and more accurate than the old clunkers they’ve replaced.
There’s no denying that backing up the high-tech vote count with a verifiable paper trail would be an expensive proposition. But Rep. Holt has argued persuasively, in our view that assuring the accuracy of election returns, especially with the memory of hanging chads in Florida still fresh in voters’ minds, is worth the cost. What he has been unable to do, however, is effectively counter the contention of his bill’s opponents that the electronic machines are virtually foolproof.
Until now.
A study released last week by Princeton University computer scientists raises serious questions about the security and accuracy of electronic voting machines. The scientists didn’t just successfully hack into a Diebold AccuVote-TS unit, proving the point that electronic voting machines are anything but tamper-resistant. They did it in less than 60 seconds, thereby demonstrating that manipulating the results of an election may be mere child’s play.
"We found that the machine is vulnerable to any number of extremely serious attacks that undermine the accuracy and credibility of the vote count it produces," wrote the team of researchers, headed by Center for Information Technology Policy Director Edward Felten. And, they noted, the same concern extends to other electronic voting machines, as well.
Rep. Holt, emboldened by this finding, vowed to redouble his efforts to get his paper-trail bill passed. "This report," he quipped, "shows that stealing electronic votes takes less time than making five-minute rice."
Five minutes is about all the time it took for Diebold representatives to pooh-pooh Professor Felten’s research. Calling it "unrealistic" and "inaccurate," the company’s president said the unit the Princeton team studied featured out-of-date software; that newer models employ state-of-the-art security, including advanced data-encryption technology; and that electronic and physical security measures the company routinely uses were ignored.
Frankly, when it comes to judging the security of electronic voting machines, we’re more inclined to listen to a Princeton University computer scientist with expertise in electronic encryption than the president of the company that makes and sells the machines. But even if we weren’t, we would still find in the brave new world of electronic voting enough uncharted territory to warrant a road map preferably one that is printed. We encourage Rep. Holt to press on toward the paper trail, and we hope his colleagues on both sides of the aisle will be persuaded by both common sense and the fresh, credible evidence provided by Professor Felten’s team to join him.