Guarding against vote fraud

Anyone who wants to commit fraud in an election has to have several people in on the conspiracy, especially election judges who have access to votes. That little detail hasn't kept many people from carrying out such schemes throughout the nation's history, including some verifiable instances where legions of deceased voters cast ballots in alphabetical order.

The danger with electronic voting is that such fraud could be compounded exponentially if enough people are involved in the plot. It's a possibility elections officials nationwide have to take extremely seriously. Unfortunately, well-meaning people have generated a lot of sound and fury recently with methods that have been less than satisfactory.

The latest comes from Princeton computer scientists who announced they created a computer virus that easily could infect several Diebold voting machines, changing the outcomes of elections in ways that could not easily be detected. Utah is one of many states that has begun using Diebold machines this year.

The scientists demonstrated how someone with access to a memory card could infect a machine in about a minute. Utah's system uses individual memory cards for each voting machine.

But the research left several questions unanswered. As Diebold officials were quick to note, the scientists used machines with outdated software. Diebold also uses security tape, 18 enclosure screws and numbered security tags that seal each machine. In Utah, the memory cards will not be moved from one machine to another, and the machines do not operate on a network, which would make mass infection with a virus extremely difficult — unless several election judges were in on a conspiracy at each polling location.

Also, the Princeton research was not subjected to peer review, as is common with scholarly work. The scientists said they were in a hurry to release the information before this November's election, but such review serves a valuable purpose.

This report is along the same lines as concerns raised by the Emery County clerk earlier this year, who let an independent firm study a Utah voting machine without any oversight by state officials. The results were of concern but were virtually useless.

No one should discount the possibility of voter fraud. Americans have proven through the years that politicians will try to rig elections if they feel success is possible. And in the age of computer viruses, who can doubt that attempts will be made to compromise electronic voting machines? But the evidence so far about Diebold is far from conclusive.

Utah's new system produces paper copies of all votes. The state would be wise to closely monitor and audit November's results in as many different ways as possible. Utah already has safeguards in place. Widespread fraud is extremely unlikely. But every effort should be made to verify that and to put the public's mind at ease.