3 ms·

Are voting machines too vulnerable to hacking? Georgia's having that debate

5 points by hassanahmad 5y ago

jevoten 5y ago
> A judge may soon release a sealed report which was prepared as part of a years-long lawsuit over the security of Georgia’s voting machines. Its author, Alex Halderman, was given rare access to dig through the machines Is this a joke? The voting method upon which democracy rests is kept secret from voters? That by itself should be grounds to drop machines and return to paper voting. > There's a long-running and complicated fight over how much security is necessary to protect elections against hacking. Well that's easy. As much as it would take to defend against a decades-long, focused effort by an adversary the likes of India/China/Russia/USA/CIA, possibly with the cooperation of a few compromised developers*, chip foundries, bios and microcode makers, and other manufacturers in the voting machine supply chain. Oh, and a reasonably intelligent voter should be able to verify that the defenses are indeed as robust as claimed. *Any developer anywhere in the dependency chain, including the compiler. See the classic https://cs.cmu.edu/~rdriley/487/papers/Thompson_1984_ReflectionsonTrustingTrust.pdf https://cs.cmu.edu/~rdriley/487/papers/Thompson_1984_Reflect...
- AnimalMuppet 5y ago
  > Oh, and a reasonably intelligent voter should be able to verify that the defenses are indeed as robust as claimed. This is really important. The claim that the election has been hacked/stolen can be as damaging as the reality would be. So it needs to be clearly demonstrable that the election was not in fact hacked.
  - bediger4000 5y ago
    You don't have to image the damage such a claim makes: Colorado's Mesa County clerk and record, Tina Peters, seems to have allowed someone to enter a secure room and at least take a copy of voting machine software and data. A sizeable fraction of CO Republicans now believe the 2020 election (but only the Presidential voting!) is screwed up, even though a Qanon type broke the security, and nobody seems to have found any irregularities in the leaked data. I'll caution against mentioning the Tina Peters incident on twitter: you'll draw trolls and bots, and some of the trolls have very specific inside information and views.
- bediger4000 5y ago
  > Is this a joke? The voting method upon which democracy rests is kept secret from voters? That by itself should be grounds to drop machines and return to paper voting. Yeah. The first go-round on voting machines was in the late 90s, people expressed the same kind of wonder and awe at secret software. Outside experts (those not in thrall to ES&S and Sequoia Systems) wanted some kind of receipt or non-electronic way of recounting, there was outcry, Bush won in 2000, and we swept it all under the rug. I'd put forth Colorado's paper ballots as a model. The ballots are hand-marked, and machine readable, but also human readable. If I understand the procedure correctly, "risk limiting audits" of each state wide election are made. I think they randomly select a number of ballots and hand tally. That tally has to be within a certain range of the tally of all ballots, otherwise it triggers a recount. I'm sure there's room for improvement, but CO seems to be on the right track.

5 points by hassanahmad 5y ago

> A judge may soon release a sealed report which was prepared as part of a years-long lawsuit over the security of Georgia’s voting machines. Its author, Alex Halderman, was given rare access to dig through the machines Is this a joke? The voting method upon which democracy rests is kept secret from voters? That by itself should be grounds to drop machines and return to paper voting. > There's a long-running and complicated fight over how much security is necessary to protect elections against hacking. Well that's easy. As much as it would take to defend against a decades-long, focused effort by an adversary the likes of India/China/Russia/USA/CIA, possibly with the cooperation of a few compromised developers*, chip foundries, bios and microcode makers, and other manufacturers in the voting machine supply chain. Oh, and a reasonably intelligent voter should be able to verify that the defenses are indeed as robust as claimed. *Any developer anywhere in the dependency chain, including the compiler. See the classic https://cs.cmu.edu/~rdriley/487/papers/Thompson_1984_ReflectionsonTrustingTrust.pdf https://cs.cmu.edu/~rdriley/487/papers/Thompson_1984_Reflect...

AnimalMuppet 5y ago
> Oh, and a reasonably intelligent voter should be able to verify that the defenses are indeed as robust as claimed. This is really important. The claim that the election has been hacked/stolen can be as damaging as the reality would be. So it needs to be clearly demonstrable that the election was not in fact hacked.
- bediger4000 5y ago
  You don't have to image the damage such a claim makes: Colorado's Mesa County clerk and record, Tina Peters, seems to have allowed someone to enter a secure room and at least take a copy of voting machine software and data. A sizeable fraction of CO Republicans now believe the 2020 election (but only the Presidential voting!) is screwed up, even though a Qanon type broke the security, and nobody seems to have found any irregularities in the leaked data. I'll caution against mentioning the Tina Peters incident on twitter: you'll draw trolls and bots, and some of the trolls have very specific inside information and views.
bediger4000 5y ago
> Is this a joke? The voting method upon which democracy rests is kept secret from voters? That by itself should be grounds to drop machines and return to paper voting. Yeah. The first go-round on voting machines was in the late 90s, people expressed the same kind of wonder and awe at secret software. Outside experts (those not in thrall to ES&S and Sequoia Systems) wanted some kind of receipt or non-electronic way of recounting, there was outcry, Bush won in 2000, and we swept it all under the rug. I'd put forth Colorado's paper ballots as a model. The ballots are hand-marked, and machine readable, but also human readable. If I understand the procedure correctly, "risk limiting audits" of each state wide election are made. I think they randomly select a number of ballots and hand tally. That tally has to be within a certain range of the tally of all ballots, otherwise it triggers a recount. I'm sure there's room for improvement, but CO seems to be on the right track.