What Kind Of Voter Verified Paper Trail Is It When The Voter Doesn't Verify?
from the just-wondering dept
Ed Felten has a great writeup discussing a study done in Ohio about e-voting technologies used in a primary this past May. While most of the stories we've written about concerning e-voting machines are about the specific technology and security risks or flaws in the underlying technology, this report looked at the actual voting process to see what happened and where problems showed up. In this case, the e-voting machines were from Diebold, using the "voter verified paper trail" that Diebold finally added to some machines, after years of resistance. However, as Felten notes, the report found that the paper printers failed in about 10% of the machines: "the paper roll was totally blank, or scrunched and smeared beyond reconstruction, or broken and taped back together, or otherwise obviously wrong." Felten goes on to point out a few important points. This doesn't vindicate those who fought against the paper trail, as the point of the paper trail isn't to be better than the electronic count, but to work together, to make sure the end result is more secure and hopefully more accurate. However, it does suggest that the technology still has a long way to go, and "just adding a paper trail" is not enough. However, there's one bit of information that a commenter on Felten's post raises that deserves some attention. Considering a big part of the paper trail is that it's supposed to be "voter verified" (i.e., the voter can see the paper printout and compare it to his or her actual vote), the fact that some of the printouts were blank, scrunched or obviously wrong suggests that the "voter verification" isn't actually happening. What's not clear is if this is an education issue, a user interface design issue or something else. No matter what, it's pretty clear that we're a long way from having secure and accurate electronic voting machines.
Thank you for reading this Techdirt post. With so many things competing for everyone’s attention these days, we really appreciate you giving us your time. We work hard every day to put quality content out there for our community.
Techdirt is one of the few remaining truly independent media outlets. We do not have a giant corporation behind us, and we rely heavily on our community to support us, in an age when advertisers are increasingly uninterested in sponsoring small, independent sites — especially a site like ours that is unwilling to pull punches in its reporting and analysis.
While other websites have resorted to paywalls, registration requirements, and increasingly annoying/intrusive advertising, we have always kept Techdirt open and available to anyone. But in order to continue doing so, we need your support. We offer a variety of ways for our readers to support us, from direct donations to special subscriptions and cool merchandise — and every little bit helps. Thank you.
–The Techdirt Team
Reader Comments
Subscribe: RSS
View by: Time | Thread
Enough Already
[ link to this | view in chronology ]
Re: Enough Already
Even if voters could consistently verify the accuracy of the machine, how many of them actually would?
[ link to this | view in chronology ]
The Voter Doesn't Verify?
I wish an honest count was all we needed.
Look at the quality of elected officials at every level - sad.
Someone once said we get the government we deserve.
[ link to this | view in chronology ]
Re: The Voter Doesn't Verify?
[ link to this | view in chronology ]
Cheap machines
[ link to this | view in chronology ]
Re: Cheap machines
Move back to Russia, man. Obviously more your style how the politics are done there.
My only concern about electronic voting over the net is only that family, employers, etc, might eventually conceivably stand over a vote and pressure one to vote one way or another. Beyond that, though, I think it's an obvious idea, not too complicated, and worth the small risk I just mentioned. Hell, imagine the entire forests it'd save, or the millions of pounds of raw materials going in to rigged Diebold machines.
Hopefully they dont run unpatched, legs-wide-open Windows servers though ;) Some darkhorse candidate from Turkmenistan would astoundingly win by a landslide.
[ link to this | view in chronology ]
Re: Re: Cheap machines
But then realized, those people don't need to be voting any more than illiterate people don't need to be voting. If one cant read, or cant click a few buttons and follow on-screen prompts, they dont know about any modern issues anyway. So, not a problem there!
[ link to this | view in chronology ]
secret ballot
[ link to this | view in chronology ]
Re: secret ballot
If you can put a number into a system to confirm how your vote was counted, then you can prove to someone else that you voted in a particular way.
This will inevitably lead to "Bring your John R. Doe ballot reciept in to our office and get a $50 gift card!" vote-buying scams.
We don't want that.
[ link to this | view in chronology ]
Re: Re: secret ballot
Our desire for secrecy provides the opportunity for fraud. Knowing there is an opportunity for fraud, I won't trust the outcome of any election until I can be satisfied that my vote was counted accurately. But I can't get that verification without the secrecy of my vote being compromised.
If I must compromise, what is more important? I'm starting to think that I would prefer to know that my vote was counted accurately, and risk that some jerk could use my vote against me -- rather than know that my vote was secret, and risk that some jerk changed my vote to support the opposition candidate.
[ link to this | view in chronology ]
I've been thinking
- Machine and server store a long value, which is known to be good when the machine is first switched on to vote. This value could be given during the initial machine/server welcome.
- After machine sends a vote both server and machine hash this number and use the result to verify their next transaction.
- Each of these numbers would appear on the paper trail, and can be used to match an electronic vote to a paper receipt.
- If the numbers on both the client and server don't match, the machine is 'locked' with the last (pre-hash) value displayed.
- Even if the current number is intercepted, it is known that each vote entered after the 'last good key' in the actual machine is bad, and can be erased/logged after a quick telephone call from an election official.
- User then goes to another terminal and votes.
This all assumes the machine isn't tampered with electronically or physically, and if that can happen then frankly you're screwed anyway. When you have the votes and paper receipts collected, you can then:
- Sample a statistically random set of votes from those recorded in the database against the paper receipts. The size of this sample is inversely proportional to the difference between winning/losing candidates (smaller gap, bigger sample tested as fewer fraudulant votes would be needed to sway the election).
- Sample a random set of paper votes to check they exist in the system. In all cases, unless voided, paper votes are regarded as 'correct'.
*breathes*
[ link to this | view in chronology ]
where's my tamper proof vitrispool?
[ link to this | view in chronology ]
WTF?
This shit is so ridiculuos. But, most things are these days in the mis-guided state of ours.
[ link to this | view in chronology ]
Well, balancing the two ends is a challenge...
People are retarded over "I can't tell you who I voted for." Maybe if more people did, they'd realize that Diebold and other companies have had huge electon related data-losses and inconsistencies and a representative from the Florida legislature hired a company to write a virus that would fix the election 51/49. It was delivered prior to the Gore/Bush election.
Still wanna keep your vote secret!
The truth is out there - if you keep looking.
[ link to this | view in chronology ]
Will never trust electronic voting
What is so difficult about scrathing a cross behind a name? Why do we have to have e-voting? It cost to miuch and it can NOT be trusted. You are all talking about a paper trail. Why do the job twice? A cross is easy. 1 line diagonally left to right and another that crosses it left to right. You do it once on a piece of paper and it is done. Easy. If you cannot do that, then ask a voting official to help you.
[ link to this | view in chronology ]
Re: Will never trust electronic voting
[ link to this | view in chronology ]
Who cares?
[ link to this | view in chronology ]
Re: Who cares?
If I recall properly 2 states do something different, maybe 60% for A and 40% for B, but I can't remember.
There is no federal law in regards to this because that would defeat the purpose of having state elections.
[ link to this | view in chronology ]
First rule of Politics
[ link to this | view in chronology ]
Re: First rule of Politics
[ link to this | view in chronology ]
Who? Why?
It facinates me how easily people have accepted the idea that there are tens of thousands of illegal immigrants, felons and ?? out there voting and don't question the "need" to prove our identity before voting or the switch to electronic voting machines. The fraud is in the battle over which ballots get counted - 3 million uncounted in 2004! Florida 2000 was the partially the result of mechanical failure. If simple punch cards are subject to inconclusive results, why would anyone push for even newer technology that is even more prone to mechanical failure (or outside "influence")?
[ link to this | view in chronology ]
Only some of us
[ link to this | view in chronology ]
Seems simple
I click, tap, whatever, a button that says verify paper ballot, my vote comes up in that window on paper, I verify, then click, tap, whatever, "Ballot Ok" or some other message. The ballot is cast, the paper ballot is stored inside the machine just like now and I am all done.
I realize people would have to work here, but this is the freaking vote, not ordering a half-shot double espresso mocha latte.
[ link to this | view in chronology ]
money
Electronic Voting machine companies have vested interests in keeping them vulnerable and non trustworthy. They actually GET more federal grant money every time they say OMG OMG the machines need fixing! Then the fed gov hands over a bunch of money. Thats not even touching the likely possibility that some of the people running the companies are in bed with some polititions and getting money to do so (one of the e-voting machine's CEO's was indited for fraud if I recall correctly, then there's the infamous letter promising to hand an election over).
And of course its cheaper all around if they dont bother fixing it.
[ link to this | view in chronology ]
E-voting
The votes should only tally if the receipt printer works, and an internal optical character (or magnetic ink) reader can confirm that the ballot receipt was legibly printed. In other words, if the printer is busted, or out of paper you can't use it: the machine requires service.
The NIST is working on electronic voting standards.
http://www.blackboxvoting.org/
http://vote.nist.gov/
[ link to this | view in chronology ]