7 ms·
Imagine we start putting cryprokitties technology to better use. 1. Voter registration cards are manufactured with in the US with tight supply chain controls.
by lrvick 5y ago
Imagine we start putting cryprokitties technology to better use.
1. Voter registration cards are manufactured with in the US with tight supply chain controls.
2. US based security researchers with a history of vulnerability identification are qualified and selected at random to hands-off oversee that manufacturing process is free of supply chain attacks. Firmware builds similarly are done deterministically on different platforms overseen by different security research firms, etc.
3. Electronic cards are picked up in person, at random, after traditional voting registration is complete.
4. Each card generates two subkeys via a KDF. A public "verify" keypair and unlimited private "vote" keypairs.
5. The user publishes a signed statement to a public database with their "verify" keypair including their real name and voter registration number.
5. Every election, a private "vote" keypair is allowed to sign a digital vote ballot to this public database.
6. Voters are required to view the public database at a later date to confirm their random vote keypair signed their intended values. They publish a signed statement that their verification was complete.
7. The vote is only considered valid when the the number of verified statements matches the number of cast votes, or after a cutoff time if the difference is not enough to change the outcome.
8. The vote database is forever public and can be counted and verified by anyone at any time.
---
Yes there is some handwaving here but I really feel we have the technical tools to have provably accurate digital voting.
This becomes possible when most citizens have enough education to understand and trust cryptography.
That would require a generation of much better education. We should vote on that.
- throwaway81523 5y agoThis has no obvious advantages over paper ballots, takes nerds to operate, and fails the usual requirement that digital voting be receipt-free. https://duckduckgo.com/?q=receipt-free+voting https://duckduckgo.com/?q=receipt-free+voting
- lrvick 5y agoThe advantage is people could vote remotely, thus avoiding excuses about not having time or mobility. It also makes it practical to vote much more often allowing more direct democracy.
- ksarw 5y agoDepending on how the registration is handled, the advantage could be in reduced friction, resulting in higher turnout. Even the reduced turnaround time in totaling votes could have a positive impact on turnout.
- throwaway81523 5y agoI don't see how derping around with a computer can have less friction than checking some boxes on a piece of paper. Either way, if I vote for minor offices, researching them even superficially takes longer than filling in the ballot.
- ksarw 5y agoFair point, the online scheme may likely entail more than just checking some boxes. My view of the ideal online scheme would bring the typical "instant gratification/response/feedback" attributes often associated with online stuff, though perhaps its not viable. Also true that friction with paper is minimal compared to that of doing research. In my anecdotal experience, research can be done while filling out the ballot, perhaps reducing the perceived friction from research alone or making it secondary at least.
- lrvick 5y agoYou can't prove your paper ballot was counted or that ballots were not omitted or stuffed in mass. You also end up with "hanging chad" ambiguous votes. All this nonsense about recounts last election is because provably counting paper is hard. You can however prove this in a cryptographic election with results published publicly.
- RC_ITR 5y ago>forcing voters to And >increased turnout Are somewhat or the orthogonal concepts.
- camgunz 5y agoI've said it before and I'll say it again. Pick any two: - Protection of voter identity (who is the voter) - Protection of vote integrity (what was their vote) - Protection of vote validity (is this vote the actual vote of an actual voter) This is not a technical problem. Generally, countries that take Election Day off and make it compulsory don't have anything like the problems the US does.
- TheMonarchist 5y agoI have been pondering a scheme where each ballot would have a field for an arbitrary token in addition to the vote. After the election voters could query a database of what ballots were cast with some token. So they could check that their vote was counted without revealing who they voted. While people are poor randomness generators, those who happened to pick a popular token would just use another one in the next election.
- mejutoco 5y agoAny system where you can prove who you voted for (a receipt) has de problem that it allows people to buy votes. In this case, they could pay you after verifying who you voted for.
- TheMonarchist 5y agoThey would make you use an unused token for certainty, so buying a significant number of votes could be observed statistically from unusually high numbers of new tokens.
- mejutoco 5y agoI missed the "without revealing who they voted for" bit. My bad.
- camgunz 5y agoThis works fine for verifying your own vote. It doesn't protect against ballot stuffing. If you can't map votes to people, you can't (definitively) say all the votes were cast by eligible voters. There is no tech solution to this. EDIT: Actually, reading more carefully it doesn't even do that. It's not super useful to verify simply that you voted--you want to make sure your vote was counted correctly. But even if we know that, you can't verify the final count (which is the ultimate point). To illustrate, this is the information you have: camgunz: voted for Ross Perot Totals: Perot: 1,894 | Bush: 49,762 | Clinton: 70,888 Nothing ensures that the totals reflect your actual vote.