The e-voting machines suck. period. accuracy has nothing to do with it...

[acroyear]

the problem isn't the voting environment (although that could use some help, given your brief description here).

the problem is the software and how it reports the totals to the main system.

that total is NOT auditable. there is no accountability built into it at all. the government (well, the republicans who sponsored the legislation to use these pieces of junk) wants us to just trust them that it all will work reliably and accurately and be content little sheep.

and that's garbage.

the diebold machines do not print a receipt for you to go "yeah, that's what I voted". the diebold machines to not print a receipt for the judges to go "yeah, we have paper copies of all the votes so we can validate the results should it come to a close call".

they do it all electronically, with no way for anybody independent of them to verify that it all is accurate.

would you trust the ATM that doesn't at least offer a receipt? even when you don't ask for a receipt, you can hear the ATM printing an internal receipt for its own auditing and accounting. would you trust a store that doesn't offer a receipt? would you trust online commerce if they didn't email you every time a purchase happened with your account?

would you trust your company to send the IRS the w2 forms and not send you a copy?

the banks would NEVER trust an electronic gizmo that didn't print paper references to what goes on with the money, even if they never have to look at that paper again.

The Vote is the most important aspect of our society, FAR more important than the Dollar. Do NOT trust ANYTHING that doesn't give your vote the same security and accountability you would expect and/or demand for your dollar.

Comments

[kmusser]

the diebold machines do not print a receipt for you to go "yeah, that's what I voted".

This is true, and needs to be fixed.

the diebold machines do not print a receipt for the judges to go "yeah, we have paper copies of all the votes so we can validate the results should it come to a close call".

This is not true, at least in Maryland - as an election judge I got to see a paper receipt for each machine and the totals for the precint. Both sets of totals (electronic and paper) where kept and reported to our board of elections. Maryland also did not trust sending the results via phone lines, they were hand delivered.

Diebold is still evil though.

[acroyear70.livejournal.com]

This is not true, at least in Maryland -- ok, i'll stand corrected, but acknowledging that not every state is going to use them that way, nor is that the way Diebold wants them used, IMHO.

[kmusser]

Yeah, I think we were fortunate that it seemed someone on the Board of Elections read the criticisms of the machines and tried to do what they could to make them more accountable.

I think electronic machines could be a good thing, with the rather large IF the state's demanded the accountability out of the them that would make the election secure. There is no reason that they couldn't be secure and accountable, just that Diebold has chosen not to make them that way.

[acroyear70.livejournal.com]

and my guess (just a guess) would be that diebold didn't design them with that idea in mind, so the result is that they're going to argue about it from a cost standpoint because it would involve rebuilding from the ground up...

...much like Windows can't be "fixed" anymore and simply needs to be replaced.

Did you get a chance to see this? I saw it on slashdot today, but haven't had the time to read it all the way through yet.

[javasaurus.livejournal.com]

Okay, so there is a paper receipt for the machine. But is there anyway to verify that voter input equals machine output? Only by having voters each receive a paper copy, look at it and say, "Yup! that's what I voted!" can you have a trustworthy recount.

[javasaurus.livejournal.com]

Great article! Thanks for posting the link!

I think it makes a great point about how you can't really understand what is likely/possible at an voting location until you've worked one.

[blueeowyn.livejournal.com]

This is one of the aspects that worries me the most. How difficult would it be for someone from the company to write code that states "for the first 500 votes, take as entered, then for every 25 votes for [candidate] entered, exchange 1 vote for [other candidate]. Just that would probably get the machines through the tests that they do to certify machines, the counts of voters to votes would be consistent, the swing wouldn't be THAT huge, but would probably swing the election in enough places to make a difference (esp. the states where Electoral College isn't an all or none proposition and/or states where the vote is expected to be close).

The other issues I have are the dial-in, the verification of the dial-in info to the precincts, etc.

I have a lot of faith in the integrity of the election judges (though it worries me that I have heard of multiple cases where the judge doesn't know the voter and doesn't request photo ID to verify identity).

[kmusser]

No, what was implemented in my county was definitely a partway solution - results are not voter-verified, but they aren't trusting everything to the machines either.