Until now, it's been easy to dismiss cryptographic voting systems as academic exercises, but the fact that the new system is designed to work with optical scanning gives Chaum hope that it will catch on. "We're ready," he says. "There's no risk. If you add it on, it doesn't interfere with what you had, and if there's a problem with it, you can just ignore it."
How it works:
I'm very skeptical about that. The history taught us that nothing is safe and nothing is secure, I really think these words are too strong and maybe a "garden path" to the failure. For instance, what happens if the invisible code for BO on the ballot is "KQ" but then, after you vote KQ, the ballot posted on the web site is scored for JohnMc, but with the correct code KQ ? Again, what happens if the print fails to write the " hidden code" believing to write the correct one ? Again what happens if the counting machines counts whatever it wants chiseling about real code associations ?
Some researchers are working on it, it's still an open discussion on our research field, so actually we cannot say that exist an total secure way to vote, we need ever some basic assumptions which, maybe, could be false. Please don't believe to the "perfect secure and untouchable" voting machine, it's still a good dream.
Some more Interesting reading:
2) Range Voting