E-Voting Companies Answer Critics With ... Spin

Whammy666 writes "Wired has a follow-up article which tells of how Diebold and other E-Voting machine manufacturers have enlisted the Information Technology Association of America (a trade public relations and lobbying group) to 'generate positive public perception' of the companies and to 'reduce substantially the level and amount of criticism from computer scientists and other security experts about the fallibility of electronic voting systems.' It seems the concerns about the lack of an audit trail are finally being heard as the industry is reconsidering its opposition to giving the voter a paper receipt of his vote. Of course, a paper receipt given to the voter still doesn't allow for a manual recount should an election dispute arise unless the receipts are collected and secured by election officials." Reassuring PR is Stage Two; remember that Stage One is silence your critics.

Comments

Sure... whatever... blah blah blah.

[Misch]

Dill said, however, that the design of a voter-verified paper system is not a trivial undertaking and that the usability and security aspects of such a feature need to be thought through carefully so companies design systems under standards that meet both these criteria.

Yes, trivial. Done. Completed. In use nationwide in Brazil.

Re:Sure... whatever... blah blah blah.

[ScrewMaster]

Amazing. That actually looks like a nice system, although the brochure didn't say a whole lot about it from a data integrity perspective. But still ... what do we have? Something infinitely preferable from a political point of view. That is, an electronic voting system with no independent audit trail, no intrinsic security, and that uses a Microsoft Access database for fast, efficient manip^H^H^H^H^Htabulation of election results. Me, I'm all for it. Hey, we could even make a game out of this to attract voters. Every 10,000th voter gets his or her vote actually recorded accurately and receives a hardcopy of same, along with a gift certificate good for a free Grand Slam breakfast at Denny's.

{sigh}

It's almost enough to make you want to throw up. I mean, this is the U.S. of A, the world's greatest Republic, the nation that built the first aircraft, the first atom bomb, the first nuclear reactor, invented television, the laser, the computer, the transistor, the integrated circuit, the spaceship, put Man on the Moon (repeatedly!), created the Internet itself ... heck we even invented air conditioning! And after all that, we find that we can't even deploy a reliable, accurate computer system that can count. That's just ... disturbing.

Re:Sure... whatever... blah blah blah.

[randyest]

I think the best way to insure these machines aren't used nefariously is to do rigorous exit polling and make sure your candidate (whoever he/she may be) suceeds by a margin that can't be fudged by a few hundred votes one way or another.

So, let me see if I got this right: to curb cheating in the existing poll, we add another poll? Oh, and make sure there are no close races.

How does the first help at all, and how do we do the second, exactly?

No Receipts to Voters!

[Effugas]

No!

It's not that _we_ want paper receipts!

It's that we want the voting infrastructure to maintain an audit trail.

Voters getting receipts directly allows for vote selling, which as another poster pointed out, is not limited to monetary compensation but includes anything people are willing to sell a vote for (health, job security, etc.)

The purpose of an election is not to determine a winner but to make everyone agree on who lost. If the losing side can say, "Sure, people voted for Bob, but it was under duress and thus didn't count", people fail to agree and fealty does not transfer.

Since we have elections precisely to avoid the violence that normally accompanies a transfer of power, this is not a small matter.

Yours Truly,

Dan Kaminsky
DoxPara Research
http://www.doxpara.com

The solution is so simple

[Kwelstr]

Are we missing something here? We all know the solution: print each and every vote on a paper ballot, check the ballot and deposit on a ballot box.

The votes are counted electronically but some machines at random should get audited and results compared to the paper votes.

A simple way to insure ourselves of no foul play or no computational errors.

What worries me is that the E-Vote machine vendors are pushing for PR so they do not have to change the system BEFORE the next election...

Me paranoid? hmmmmmm

Nothing is wrong with the paper ballot!

[LostCluster]

We seem to have forgotten something here. The paper ballot system isn't broken. What failed wad the punchcard system, and more specific efforts to explain proper operation of it.

The ideal ballot is one that results in a piece of paper that is both human-readable and machine readable. There hasn't been many problems with the "fill in the bubble" system of balloting, even though that system is open to a risk of users who don't understand that an X or checkmark in the bubble doesn't work.

The place for touchscreens is to help the user create a perfect ballot that is machine readable for speed counting, with the votes also in human readable terms for manual spot checks and recounting, and the most important spot check: The one the voter does before walking over to the ballot box. If the printout doesn't say what they thought it did, they hand the spoiled ballot to the officials and go try again.

The idea of having any form of electronic memory conduct counting within the in-booth devices is crazy. It opens the system into too much risk of data loss or data manipulation. There needs to be an audit trail, and that trail belongs in the ballot box.

Re:Nothing is wrong with the paper ballot!

[Anonymous Coward]

Too true! This US obsession with electronic/mechanical voting is bizarre, viewed from the rest of the world.

Here in Australia, we have a pretty complex electoral system (preferential voting, as opposed to first-past-the-post) and we count all votes by hand. All political parties have scrutineers present to observe the count take place. I can't recall any significant reports of problems with this system in the 25 years I've been voting.

And before people whine loudly about how the US is much bigger than Australia, let me point out that the voting population isn't *that* much bigger, and there are bigger countries than us that hand-count votes, as well. I don't believe India, for example, is a heavy user of mechanical voting systems.

Peer Review

[Anonymous Coward]

While the merits of OSS for many purposes are debatable, when it comes to voting machines, I think it's pretty clear that no system should be adopted that hasn't had its design and implementation thoroughly peer-reviewed. That means hardware schematics as well as source code.

Note that merely "providing the source" isn't particularly helpful. The elections standards arm of the government is going to have to contract out the review and assure that it is done by a diverse group of peers other than the implementor -- and most likely including their competitors -- and not just rely on interested citizens to happen to take a peek (welcome as that might be).

In this case, you can make your money by selling the hardware. There need be no trade secrets involved in building an voting machine.

A transition period?

[thepacketmaster]

I'm sure someone has already thought of this, but why don't they count both the e-votes and the receipts at the same time. I'm a little skeptical about these machines but they will eventually be used, no matter how much we might not want them.

I think it would be useful to have a transition period of 10 years or so, that would be used for the software to become more stable, and to help instill the trust in the system. People would cast their e-vote, get the receipt, verify it is correct and put the receipt in an old fashion ballot box. After the polls close, the e-votes are shown and the receipts are tallied. Then the discrepancies are examined and if there are any problems, the receipts are used for the final count instead.

Re:Sigh...

[Xerithane]

The student who made a fool out of the airport security system was conducting an act of civil disobedience, but the part of civil disobedience everyone seems to keep forgetting is it involves a public crime done to get attention, of course he's gonna get arrested and charged for it. He should be, he didn't just say "Somebody could.." he went out and did it.

The thing is, before September 11th you could bring a box-cutter on an airplane. Hell, I accidently brought a 5" butterfly knife through airport security in 99 or so.

The kid who did that was proving a point, and to prove that point he had to act. Merely telling them wouldn't do anything, and the facts are supportive of this.

Now, to bring this on-topic and on-base, because I believe it was a valid point.

Civil disobedience is the best way of proving a point when the masses won't listen to you. What will it take for people to realize these voting systems are flawed and dangerous? Bruce Campbell being elected President of the United States of America?

That is civil disobedience I can appreciate, just like the student, because it shows that things aren't as good as they should be and that jeopardizes my safety.