Is it time for us to have paper trails for electronic voting mechine?

Since it is said that virus can "steal" the votes in an unnoticed way. Even worse, the virus can spread from mechine to mechine.

Researchers at Princeton University said today that they have proven a popular electronic voting machine can be rigged to steal votes - in ways that are virtually undectable - in less than a minute.

What's more, they said, the virus could be passed from machine to machine through the course of normal procedures - and leave virtually no clues that anything was amiss.

The corresponding company surely don't feel the problem though

But a Diebold spokesman said Princeton had examined voting software that is 3 to 4 years old. Systems in use today deploy encryption, passwords and digitally signed certificates, in addition to random tests by poll-workers, designed to minimize risks of tampering.

"It's interesting they didn't go through a peer review," said Mark Radke, director of marketing for Diebold in Allen, Texas. "It's interesting timing of the release of the study. The fact that they did not communicate with Diebold before releasing the study also is interesting."

Well, I don't actually see the academic value to publish this in a journal or some. But this is not how they counter-argue the way they made it public.

Felten, famed for cracking a copyright protection scheme on a challenge from the music industry in 2000, said he felt the voting machine results were too important to wait for publication in a peer-reviewed journal. He wanted public officials to have enough time to take action before November's elections, he said.

I respect that.