Very detailed Lemmy post I wrote about this fuckery and more
Local news article containing the quote about the voter registrar
On Homer Plessey Way, board member Daniel Milojevic stood outside the Bywater polling place in the Press Street Gallery suggesting people try the two Jefferson Parish locations.
He said the local registrar of voters gave the district only 300 ballots per location and told them they could expect about 20 people.
“We had to confirm the number of ballots weeks ago,” he said, before it was clear how high the turnout would be. Milojevic conceded that planning had clearly missed the mark.
As one astute gentleman asked while defending Reddit, and accusing me of spreading misinformation:
If hardly anybody knew, how did turnout exceed expectations within 2 hours?
Because the "expectation" provided by the registrar was literally 20 voters per location (60 voters in total) for the entire fucking city.
Surely there is a cryptographic way to count votes where someone can check that the results are correct but not how individuals voted, right?
Now you have to trust the software used to do this, the algorithm itself, and that there was no tampering before the data got stored. Which is something truly verifiable by a very tiny subset of population and even then with full cooperation from authorities. This is the opaqueness of countermeasures.
Vote counting is not a mathematical problem, but a sociological one. Any „always correct machine” is useless if people can’t reasonably trust it.
Paper ballots don’t scale - you can’t stuff ballots without someone being present - and are designed exactly in the problem space vote counting itself occupies. As an additional evidence in their favor, autocratic regimes and corrupt politicians are way too eager to switch to electronic voting.
That’s the whole point of crypto though, you publish the mathematically verifiable results, and everyone becomes a vote counter. Instead of trusting a small group of people to do it right, you can verify the counts yourself in a trustless system.
The algorithm isn’t a black box like you’re saying, it’s fully auditable and decentralized so any fuckery is immediately visible.
Now maybe I’m wrong and a mathematically verifiable algorithm can’t exist, but to my knowledge that’s never been shown to be the case.
Edit: turns out there are MANY such systems, and they are mathematically verifiable, and used in actual use cases today. The only thing stopping it is lack of political will, and arguably, the fact that even people without computers have the right to vote.
You do not have to trust the software. You could do that math for yourself if you really cared.
You're thinking about the voting problem wrong, specifically not including all of the requirements
Their point is that even in this system, there is both a small and finite number of people who are skilled enough/qualified enough to perform that audit. I'm not sure I've got the math to be able to validate a blockchain transaction by hand without referring to a (potentially tainted) source repo. There's a world where blockchain voting can solve this problem, but the competing requirements make it the less-optimal solution compared with paper voting.
Specifically, there are 3 potentially competing requirements for a secure voting system in a functioning democracy:
Blockchains, potentially, optimize the voting problem for #1 while introducing explicit exposure in #s 2 and 3, while paper ballots optimize for 2 first, then 3.
Woah woah woah, I said nothing about blockchain. That would almost certainly be the wrong, overly complex solution. The systems that exist for cryptographic voting do hit ALL your points while having the additional benefit of being perfectly auditable.
Cryptography is a much larger field than blockchain, and people use it for trusted communication every day.