The federal panel charged with writing the guidelines for the U.S. Election Assistance Commission rejected a proposal that would have required voting machines to have a paper trail or another independent means of verifying election results.
Some of the members who voted against the proposal said that they support paper trails but don’t think the requirement should be rushed into place and “overwhelm state election boards.”
I guess that 18,000 vote undercount in Florida was not enough to convince them that something needs to be done. Then again, maybe they were relying on faulty information:
Verifiable paper records are already used by many states — 27 mandate them while another 18 don’t require them but use them in all or some jurisdictions. Only five — Delaware, Georgia, Louisiana, Maryland and South Carolina — use machines without a paper record. More than half of all voters used machines with paper records during the 2006 elections.
Hmm – when did all of Virginia add machines with a paper record? The touch screens in use in Norfolk certainly didn’t have any. And according to this site, fifteen states have no paper trail requirement.
If their information was correct, how could imposing a paper trail or other independent means of verifying election results on five states “overwhelm state election boards?” If it is good enough for half the voters, shouldn’t it be good enough for the rest of us?
Why are the decision-makers so resistant to audit trails?
VIVIAN J. PAIGE | All Politics is Local 
As far as I know, there is no paper record in Fairfax County, either. Is it possible such records are kept inside the machines? If so, how does a voter verify his vote?
I think the AP writer did a CYA when he said “all or some jurisdictions.” Perhaps those areas that use optical scan machines have an audit trail. But to say that only 5 states have no paper trail is quite misleading.
Jack-
From what I know, Fairfax keeps the printouts of the total vote and verifies it with the vote totals downloaded into the master machine.
Comparing totals to totals doesn’t help if they both come from the same flawed source. The idea is to have a record a human can interpret that proves that the votes stored in the computer’s memory match the buttons the voters push. A DRE could very easily show voters the votes they think they cast on the final review screen, then store something entirely different on the memory card.
It looks to me as if the committee was playing semantic games with the term “paper record”, lumping tabulation tapes in with actual raw data records. They aren’t the same, not by a long shot.