NYT editorial: The Road to Reliable Elections

[DeepModem Mom]

The Road to Reliable Elections
Published: December 11, 2006

Two influential federal advisory groups have added their voices to an emerging national consensus that voting machines must produce a voter-verified paper record if they are to be trusted. One of those groups, the one dominated not by scientists but by election officials, was more grudging than it should have been. But their analyses should give further support to members of Congress who plan to push next month for a strong federal law requiring voter-verified paper records.

More than half of the states now have laws requiring electronic voting machines to produce paper records that voters can review to ensure that their votes were correctly recorded. Voters understand that without this paper, there is no way of knowing if the software registered their choices incorrectly, either by accident or by design.

The National Institute of Standards and Technology, an agency that promotes good standards in everything from medical devices to smoke detectors, recently concluded that paperless electronic voting is unacceptable. The agency’s scientists said that for electronic voting to be trustworthy, it must be “software independent,” meaning there has to be a means apart from the machines’ own software to prove that the vote tallies are correct.

The obvious way to do this is a voter-verified paper record....NIST’s report went to the technical guidelines committee of the Election Assistance Commission, the federal agency created after the 2000 meltdown to help improve elections. The guidelines committee voted down a resolution recommending that electronic voting be used only if the results are independently verifiable. Instead, it affirmed the need for paper trails or other independent verification going forward, but recommended giving a pass to localities already using flawed machines.

It is unfortunate that the stronger resolution did not pass....

http://www.nytimes.com/2006/12/11/opinion/11mon1.html?hp

[bleever]

1. Another Inconvenient Truth: We've Been Had.

Al Gore: right, again.

[EFerrari]

2. I wanted to say that but not in those words.

:evilgrin:

[enough]

3. I have a basic question that has probably been answered many times on DU:

With a "voter-verified" system, what is to prevent the machine from printing one thing for the paper receipt and then recording a different thing on the computer tally?

Are the paper receipts collected at the polling place after being verified by the voter, so they can be used for a possible recount?

Wouldn't the optimal solution be simply to count the actual paper receipts that had been verified by the voter? (In other words, back to paper ballots.)

Sorry to be behind on these issues -- any clarification is welcome.

[w4rma]

5. Your logic is correct. That's what will happen. It must be paper *ballots*, not "trails".

and certainly not "receipts". That's even worse.

And of course those paper ballots must be "voter-verified".

[w4rma]

4. If they simply implement paper "trails" then there will be another round of chaos.

It absolutely must be voter verified paper **ballots**.

[tbyg52]

6. And none of this

"let's trust the machines except for a small percentage that we audit" crap, either.

Edited when I realized that I did so need the quotes....

[dlaliberte]

8. Paper trails better than totally insane paperless machines

While I agree fully that we need paper ballots, not just paper trails, I do believe we would be better off at least requiring paper trails. We have many parts of the country still dominated by those who would do anything to subvert the real will of the people. Making it a bit harder to commit fraud is better than not doing so.

With a paper trail, it is possible to recount if there is a discrepancy between (truly) random samples and the actual vote total. Not ideal, but better than nothing.

However, I would definitely not want to see more money shoveled into adding paper trails to electronic voting machines when that same money could go into an even cheaper paper ballot system. And it would be insane to "upgrade" a paper ballot system with a paper trail system just because it is allowed. We should somehow require upgrades to actually be better regarding reliability, verification, etc.

[byronius]

7. It's a tough road in a democracy when half don't really believe in it.

That 'freeper irony' thread still rings in my ears -- the Allen race in Virignia prompting right-wing bloggers to complain about the inability to conduct a recount.

Paper all the way, I guess, and not a single corporation involved.

[WillYourVoteBCounted]

9. independent not same as paper

I hope people realize that if NIST meant for paper to be the
independent verification source, NIST would have said PAPER.

Instead, NIST leaves the door open for end to end systems,
encryption, black box on top of a black box.....

Independent means - independent of the DRE's software.

If Independent meant paper, they would have said paper.