Amusing, but also sad:
Oops, hit the button too soon!
Now if Oprah can’t reform voting, who can?!
I don’t think a perfect voting system exists. If you use paper ballots, you risk the chance that some poll worker will lose/toss/alter them. If you use electronic ballots, you risk the IT person programming in some code to miscount/drop/add votes.
The one Oprah used (and we use here in Ohio) seems to me to be the most foolproof. It electronically counts your vote but also prints out a paper slip so that the vote can be hand counted, if necessary. I’m not sure what else counties/voters can do to help ensure the validity of an election, but if you have any ideas, I’m sure they’d love to hear them.
It seems to me that it would be much easier to surreptitiously alter a vote count on a computer system than it would be with pencil and paper ballots, where (at least in the case of Canadian elections) you have two poll clerks (nominated by different parties or hired by Elections Canada), the count can be observed by party scrutineers, and the voter places his or her own ballot into the sealed ballot box. I’ve scrutinized the count in two or three elections and I didn’t see a point at which votes could be discarded without someone around to raise a stink.
Well, the guy with that website sure has found a cause, hasn’t he? My computer’s on a partial strike right now and won’t show videos - has anyone figured out exactly why he thinks a paper verified electronic ballot isn’t verifiable and that “even if those “paper trails” were 100% hand-counted (and they are almost never counted at all, in any case) the hack would likely never be revealed and the “paper trails” would match up perfectly with the hacked internal numbers.”? He keeps saying it, but doesn’t explain *why *that is, and I’m stumped. Is it just because people won’t look at their paper slip and make sure it says what they intended?
It would be no big deal to program the paper to match the vote while throwing some percent of the electronic tally to a different candidate. If the paper is retained to be used to audit against the electronic tally, the fraud, (or, at least, the discrepancy), would be discovered, but in the cases where no paper matching is performed, the fraud would go unreported.
OTOH, any code employed to make that happen should be easily discovered by any competent auditor (although there are ways to obscure it if the auditor is tired), so it seems unlikely to be employed.
I doubt that the web page author has anything like a serious complaint, but I wanted to note that his scenario is techically (if not at all realistically) possible.
An IT person is the person at work who keeps your network and email running. A software engineer or computer scientist would be the person doing coding for a voting machine.
Yes, I get that, but what the (slightly rabid) webmaster keeps saying is that EVEN IF a paper trail audit is done, this nefarious scheme will not be revealed. That somehow the paper trail is as hacked as the electronic votes. That’s the part I’m scratching my head about.
The article linked in the OP also links to this page, which contains the following claims:
I just can’t watch the videos right now and wondered if anyone else had.
The video shows a way to make the paper printout tally with what the voter intended when the voter checks it, but then after the voter walks away from the booth, the malicious code voids that paper receipt and alters the vote and the new paper receipt matches the altered vote. So, when they examine the paper trail it tallies with the altered votes.
The changes made to the machine by the researchers were not especially hard to do and were actually quite difficult to detect (because the code “knows” when it’s being verified and when it’s actually counting proper votes so acts correctly during verification and only alters actual votes).
Huh. So, yeah, not much of a paper trail. I was envisioning something like a paper receipt, which one could read, verify and hand to an election official, who stamps it and puts in in a big box just in case they need to do a manual count or something.
Can you tell I haven’t voted in years?
WhyNot, I wrote a DailyKos diary re the Sequoia Systems voting machine hacks and the UCSB video. The problem is that any of those machines can be hacked with software small enough to fit on a USB thumb drive and the hack can be tailored to change the electronic vote while still giving a correct paper trail. Since nobody wants to go through the paper trails of millions of votes hand tabulating whether or not they all match the electronic vote and since the voters don’t get a hard copy themselves to take away, which could then be used by outside parties to spot verify whether or not the electronic results seem reasonable and form a basis for challenging the results, it’s tough to establish that fraud has taken place and it tends to get handwaved off as user error.
For being in charge of something so important, these voting machines are disgustingly easy to hack. I can say with no hyperbole whatsoever that any of our computers as well as our home wireless network have infinitely better security than one of those machines. What makes it even worse is that in many states poll workers are allowed to take the machines home over the weekend before the election, thereby placing them in totally unsecured areas for extensive period of time. All you need to hack one is about a minute with a thumb drive and bob’s your uncle.
People can wibble about paper ballots all they like, but it’s goddamned hard to either “lose” hundreds of thousands of pieces of paper (all you have to do to prevent this is to keep an eye on the ballot collection areas and look for people carrying suspicious truckloads of boxes out) or to mysteriously add in more votes. In Oregon we do all elections as vote by mail–the Secretary of State’s office sends out our ballots, which are simple Scantron forms that even morons know how to use, with a secrecy envelope and a bar coded outer envelope with the return address printed on the back. The SoS knows how many ballots are sent out and how many come back. We mark the Scantrons, put them in the inner envelopes and put those in the mailers, with our signatures on the back–no signature, vote doesn’t get counted. We can either mail them in or place them into secure drop boxes at any election office or public library–I always drop mine off because I don’t trust the Post Orifice, nor do I keep stamps around. Most of us smart folk make photocopies of our ballots and signed envelopes, just for our records.
At the election offices, the envelopes with signatures are scanned and verified and an electronic record is made of them and they’re tallied against the records for the mailed out ballots. The Scantrons are removed from the inner envelopes and placed in trays which are fed into the tabulating machines at poll close on Election Day–results are done within a couple hours. The only downcheck to the process is the large amount of paper used, but I can live with that. It’s simple, it’s very hard to gimmick and in ten years we’ve had no significant problems with it. We don’t have issues with opposing parties restricting access to polling places, long lines, restricted early voting hours, inadequate numbers of voting machines, malfunctioning machines, none of it. Why this hasn’t been adopted nationwide I’ll never know.
Small enough to fit on a USB thumb drive? That’s like saying a person small enough to fit in a warehouse.
Well, true dat, but I keep thinking of the darned things as having a fairly small amount of storage on them, but I’ve seen them up to 12GB so I really should rethink my mode of expression! Mostly what I’m trying to convey is “requires no expensive or bulky equipment, one guy in shorts and a t-shirt with ten bucks worth of thumb drive could screw over multiple voting machines in a short period of time.”
And, thanks to the wonder of the internet - you don’t even need to be able to program the hack yourself. You can just find one someone else has written…
If the voting machine has a user-accessible USB port (and where “user” does not equal “engineer with special equipment and a key to the safe”) then you’ve got a pretty big problem I think.
Every body knows you can make mistakes with pencil and paper,but to really fuck things up, it requires a computer.
I’ve heard this before and refuse to believe it. I have no proof, but I have a feeling that the block is being done by public companies who actually make the electronic voting machines. Otherwise, we would have one perfect machine that is reliable. Intead, we have handful of companies who engineer and program these machines, with each wanting their system and software to be proprietary. This would leave a lot of room for “error” if you ask me. Or a backdoor “option” for a malicious software engineer.
We need one machine, and one machine only, developed by the government. Trust that I hear what you’re saying, “trust the government” you ask? Yes. The trust will be earned with the software used, to be open source, for all to view and scrutinize for all aspects of the machine. It cannot be that difficult to implement something like this, but it can be difficult if you’re infringing on the right of the “people” to develop their own machines to sell for a profit in a “free market”. Hence, no real government voting device.
Again, not sure if this is what’s going on, but there’s nothing else to explain why there doesn’t exist one reliable machine that people trust to use. I mean, what system do they use for people to file their taxes? I bet theirs not one systematic error in that process (plenty of room for human error though), yet we all trust that it does what it’s supposed do, and we eFile billions of dollars along with individual circumstances involving complex math on our tax forms, and somehow it works. And I’m sure that the IRS trusts that it’s accurate too.
We need one eVote machine for all fifty states. It might cost money, but better than costing someone an election they rightfully won.
Ahh, the old right wing conspiracy theory.
The ones in the PA precincts I have voted in are just tablet PCs running the Diebold voting software and mounted on a portable podium that can be folded up for storage. Since they aren’t really anything special in terms of hardware, they have the same I/O as any other tablet PC.
You know, people scoff and mutter about tinfoil hats, but it has been confirmed from reliable sources within the company that Sequoia Systems deliberately sent a batch of defective paper ballots to Florida for the 2000 election precisely because they were trying to convince the county governments to invest in electronic voting machines and it was in their interests to “prove” that the electronic (and expensive) machines were “better” and “more reliable” than paper ballots.
It is absolutely NOT in the best interest of the American people that for-profit companies whose partisan loyalties may be unknown and undisclosed be left in charge of making, programming and selling voting machines. Period, end of story. Remember Diebold and Ohio in 2004?
I like open source, too–there’s something about having your code reviewed and modified by thousands of programmers all over the world who don’t have personal agendas and whose only stake in the project is to make it work better that tends to produce good, clean, strong, virus resistant, efficient software. Any time I’m given a choice between proprietary and open source software I go with the open source every time and I’ve never been disappointed. There’s something about paying fucktons of money to get back buggy, backdoor riddled compromised systems that puts me in mind of being forced to kiss the rapist afterward…