![[personal profile]](https://www.dreamwidth.org/img/silk/identity/user.png){kind=small}
usernamenumberI have been thinking a lot lately about voting machines, for obvious reasons. The election seems lost, but what worries me is the apparent strength of the case being made that the results these machines deliver are simply not trustworthy.
I really, really wish I'd found this video before the election and that it had achieved wider circulation. Perhaps it wouldn't have made a difference in the outcome, but I would have at least been better able to articulate why I, a highly technical person, insisted on using a paper ballot to cast my vote. It's 30 minutes long, but please take a look if you haven't already. As I usually try to do when I find something like this, I went googling to try and find a response to the claims being made. After all, the people who made the documentary had an obvious agenda and its not that hard to make a convincing case for anything if you control what gets presented how. The only thing I could find was a message posted to Bugtraq by a diebold rep, which was subsequently ripped to shreds. On the other hand, I found even more convincing material at places like blackboxvoting.com, which have some disturbing if unverifiable anecdotal accounts of machines malfunctioning on election day and the unrelated blackboxvoting.org. The latter is run by Bev Harris, subject of the documentary linked above. They are raising money with which to make Freedom of Information Act claims to all the records from the voting machines used in the elections, which turns out to be a very expensive proposition. To me it seemed worth a contribution just to get these records, which probably should be public in the first place, out in the open. They've already made a case for fraud having occurred during an earlier election (scroll to the entry labeled "MONDAY Nov 1 2004" for more information on that) but it's important to also note the number of reports being made of machines that simply recorded the wrong candidate in the first place (ie no fraud necessary). There's also this interesting rebuttal to claims made by the voting machine makers as to why their products are trustworthy.
I've also been thinking about why I and so many other technical people have such a problem with this stuff. Well, for one thing I just don't see how putting a private company, especially one with the kinds of political connections that Diebold has (their ceo is a major Bush supported who promised to "deliver the votes of Ohio to the president" earlier in the year) in charge of a closed-source product could reasonably be trusted with the outcome of an election. As a technical person I'm all too aware of how easily this sort of thing can be manipulated and of the importance of transparency in a trustworthy product. Mounds of anecdotal evidence (as well as an some extremely damning examples of malfunctioning recorded in the documentary) bring into serious question the adequacy of whatever QA testing was done. And, especially given all the reports of mis-reported voted, it doesn't take much technical aptitude to see that a voting machine printing out the contents of its memory does not make for a valid paper trail when it's that very data's validity (whether due to tampering, voter error or malfunction) that's in question.
Basically, it all comes down to verifiability. Rather than saying "the machines are infallable", something no one with the slightest technical aptitude would ever say of a machine, the question at hand is: can errors be detected, verified and corrected easily? My feelings on that were summed up extremely well by one of the responses to Diebold on BugTraq:
That's all I have for now. Food for thought. Unfortunately, even if the machines are discovered to have been in error, I doubt it will amount to anything more than another scandal to keep the evening news interesting without actually bringing about change.
I really, really wish I'd found this video before the election and that it had achieved wider circulation. Perhaps it wouldn't have made a difference in the outcome, but I would have at least been better able to articulate why I, a highly technical person, insisted on using a paper ballot to cast my vote. It's 30 minutes long, but please take a look if you haven't already. As I usually try to do when I find something like this, I went googling to try and find a response to the claims being made. After all, the people who made the documentary had an obvious agenda and its not that hard to make a convincing case for anything if you control what gets presented how. The only thing I could find was a message posted to Bugtraq by a diebold rep, which was subsequently ripped to shreds. On the other hand, I found even more convincing material at places like blackboxvoting.com, which have some disturbing if unverifiable anecdotal accounts of machines malfunctioning on election day and the unrelated blackboxvoting.org. The latter is run by Bev Harris, subject of the documentary linked above. They are raising money with which to make Freedom of Information Act claims to all the records from the voting machines used in the elections, which turns out to be a very expensive proposition. To me it seemed worth a contribution just to get these records, which probably should be public in the first place, out in the open. They've already made a case for fraud having occurred during an earlier election (scroll to the entry labeled "MONDAY Nov 1 2004" for more information on that) but it's important to also note the number of reports being made of machines that simply recorded the wrong candidate in the first place (ie no fraud necessary). There's also this interesting rebuttal to claims made by the voting machine makers as to why their products are trustworthy.
I've also been thinking about why I and so many other technical people have such a problem with this stuff. Well, for one thing I just don't see how putting a private company, especially one with the kinds of political connections that Diebold has (their ceo is a major Bush supported who promised to "deliver the votes of Ohio to the president" earlier in the year) in charge of a closed-source product could reasonably be trusted with the outcome of an election. As a technical person I'm all too aware of how easily this sort of thing can be manipulated and of the importance of transparency in a trustworthy product. Mounds of anecdotal evidence (as well as an some extremely damning examples of malfunctioning recorded in the documentary) bring into serious question the adequacy of whatever QA testing was done. And, especially given all the reports of mis-reported voted, it doesn't take much technical aptitude to see that a voting machine printing out the contents of its memory does not make for a valid paper trail when it's that very data's validity (whether due to tampering, voter error or malfunction) that's in question.
Basically, it all comes down to verifiability. Rather than saying "the machines are infallable", something no one with the slightest technical aptitude would ever say of a machine, the question at hand is: can errors be detected, verified and corrected easily? My feelings on that were summed up extremely well by one of the responses to Diebold on BugTraq:
Paper-verified-voting is easily understood and verifiable by
_anybody_. Tricky cryptographic protocols are understood by few and
verifiable by a lot fewer. And, if the protocol is run on a computer
I don't control using code I don't control, then I have no confidence
no matter what protocol it _claims_ to use.
That's all I have for now. Food for thought. Unfortunately, even if the machines are discovered to have been in error, I doubt it will amount to anything more than another scandal to keep the evening news interesting without actually bringing about change.
