2 ms·

> First of all zero-knowledge proofs allow you to verify stuff without being able to prove it to others I doubt it, and I suspect if you try to point at a spec

by roenxi 2y ago

> First of all zero-knowledge proofs allow you to verify stuff without being able to prove it to others I doubt it, and I suspect if you try to point at a specific system to implement you will find that that none exist even in theory. I can verify I voted with zero knowledge, yes. But I can't verify who I voted for. So I can put candidate A into the machine, it switches to candidate B and we can all prove I voted in the election. Conversely, if I can prove who I voted for then the scheme is vulnerable to the well known after-election issues because I can prove it to others. If I can only prove something with plausible deniability note that I probably can't tell if the machine switched my vote around. There might be something that can be done in the space, but it is a tricky one to resolve. > But honestly, I think the whole idea of being able to prove how you voted being dangerous is overblown. If you check you may well find it in a reasonable-worse-case scenario it is a matter of life-and-death. I think maybe literally zero government electoral systems make the voter's vote public (ie, we have near universal secret ballots [0])? There is a reason for that. If we wanted people to sign their name on the vote slip that'd be great for auditing - but we don't because that would set the system up for some really horrible failures. The one that leaps to mind is "if you don't vote for me and I get in, I will do [insert blank] to you" strategies. [0] https://en.wikipedia.org/wiki/Secret_ballot#Chronology_of_introduction https://en.wikipedia.org/wiki/Secret_ballot#Chronology_of_in...

EGreg 2y ago
“Literally zero make the voter’s vote public” In most US states I can get a voter’s database and “party affiliation”. I was shocked that thus info is publicly available, and all the people’s addresses and driver’s license info are also stored there (and can be leaked) And make no mistake, these databases are regularly leaked / hacked: https://qbix.com/blog/2023/06/12/no-way-to-prevent-this-says-only-industry-where-this-regularly-happens-2/ https://qbix.com/blog/2023/06/12/no-way-to-prevent-this-says... In fact there is a law for states to create and maintain this information. https://ballotpedia.org/Availability_of_state_voter_files https://ballotpedia.org/Availability_of_state_voter_files The “party affiliation” is a very good (around 95%) proxy to how they’re going to vote when they show up, as long as the two-party system dominates, which is why I say the whole “ability to prove your vote” thing is overblown, since your party affiliation at registration is already known, even publicly: https://www.pewresearch.org/politics/2023/07/12/voting-patterns-in-the-2022-elections/ https://www.pewresearch.org/politics/2023/07/12/voting-patte... Explain how Estonia is able to reliably and securely do online elections, if only paper elections are secure: https://e-estonia.com/how-did-estonia-carry-out-the-worlds-first-mostly-online-national-elections/ https://e-estonia.com/how-did-estonia-carry-out-the-worlds-f... Many times, people claim that technology would never be able to do a good job at what humans do manually — and almost always this has been proven wrong after a while: https://www.coindesk.com/tech/2020/03/12/in-defense-of-blockchain-voting/amp/ https://www.coindesk.com/tech/2020/03/12/in-defense-of-block...
sethammons 2y ago
If everyone got a unique prime number and a running total vote product was available, I always thought this would be a neat solution. Still susceptible to the goon-with-a-wrench technique I think
- EGreg 2y ago
  BLS signatures can provide similar properties Crypto (by which I always mean cryptography) can help secure a lot of things that normally are just “trust in a middleman” https://news.ycombinator.com/item?id=29306829 https://news.ycombinator.com/item?id=29306829
- roenxi 2y ago
  That doesn't sound like it secures anything. I can't verify that my prime number is unique (I vote for A. There is already a vote for A at 3. The machine logs a vote for B at 5 and reports to me my number is 3). So it'd be a scheme where nothing can be proved to me that I didn't already know. I still have to trust all the same middlemen, and I don't gain any knowledge about the integrity of an election.
  - sethammons 2y ago
    > If everyone got a unique prime number
    - roenxi 2y ago
      That is an unreasonable assumption, there isn't a secure way to issue everyone with a guaranteed unique prime. And even if there was, what is this system supposed to be doing? It doesn't secure anything and it doesn't enable any new knowledge that wasn't already available by just signing ballots with your name (which is a bad idea, so by comparison we might expect that this prime scheme is also a bad idea). Are you sure that is an interesting thread to pull? It doesn't look at all promising to me.

by roenxi 2y ago

EGreg 2y ago

“Literally zero make the voter’s vote public” In most US states I can get a voter’s database and “party affiliation”. I was shocked that thus info is publicly available, and all the people’s addresses and driver’s license info are also stored there (and can be leaked) And make no mistake, these databases are regularly leaked / hacked: https://qbix.com/blog/2023/06/12/no-way-to-prevent-this-says-only-industry-where-this-regularly-happens-2/ https://qbix.com/blog/2023/06/12/no-way-to-prevent-this-says... In fact there is a law for states to create and maintain this information. https://ballotpedia.org/Availability_of_state_voter_files https://ballotpedia.org/Availability_of_state_voter_files The “party affiliation” is a very good (around 95%) proxy to how they’re going to vote when they show up, as long as the two-party system dominates, which is why I say the whole “ability to prove your vote” thing is overblown, since your party affiliation at registration is already known, even publicly: https://www.pewresearch.org/politics/2023/07/12/voting-patterns-in-the-2022-elections/ https://www.pewresearch.org/politics/2023/07/12/voting-patte... Explain how Estonia is able to reliably and securely do online elections, if only paper elections are secure: https://e-estonia.com/how-did-estonia-carry-out-the-worlds-first-mostly-online-national-elections/ https://e-estonia.com/how-did-estonia-carry-out-the-worlds-f... Many times, people claim that technology would never be able to do a good job at what humans do manually — and almost always this has been proven wrong after a while: https://www.coindesk.com/tech/2020/03/12/in-defense-of-blockchain-voting/amp/ https://www.coindesk.com/tech/2020/03/12/in-defense-of-block...

sethammons 2y ago

If everyone got a unique prime number and a running total vote product was available, I always thought this would be a neat solution. Still susceptible to the goon-with-a-wrench technique I think

EGreg 2y ago
BLS signatures can provide similar properties Crypto (by which I always mean cryptography) can help secure a lot of things that normally are just “trust in a middleman” https://news.ycombinator.com/item?id=29306829 https://news.ycombinator.com/item?id=29306829
roenxi 2y ago
That doesn't sound like it secures anything. I can't verify that my prime number is unique (I vote for A. There is already a vote for A at 3. The machine logs a vote for B at 5 and reports to me my number is 3). So it'd be a scheme where nothing can be proved to me that I didn't already know. I still have to trust all the same middlemen, and I don't gain any knowledge about the integrity of an election.
- sethammons 2y ago
  > If everyone got a unique prime number
  - roenxi 2y ago
    That is an unreasonable assumption, there isn't a secure way to issue everyone with a guaranteed unique prime. And even if there was, what is this system supposed to be doing? It doesn't secure anything and it doesn't enable any new knowledge that wasn't already available by just signing ballots with your name (which is a bad idea, so by comparison we might expect that this prime scheme is also a bad idea). Are you sure that is an interesting thread to pull? It doesn't look at all promising to me.