The reason that paper voting is so good in this regard is that everybody can fully understand the entire process. It is so very, very simple. And if you need proof, you can go see the counting for yourself.
The issue with electronic voting is that there is far greater complexity. There are many valid reasons that someone could distrust it, for example:
- You might not trust the cryptography experts that claim the algorithms are secure.
- You might not trust the algorithms to be implemented correctly.
- You might not trust the computer manufacturer to have designed a secure machine.
- You might not trust the computer manufacturer to have built a defect-free machine.
- You might not trust the machine hasn't been compromised by some bad actor.
- You might not trust that there hasn't been some random bit-flips.
- You might just not understand how computers work.
- ect. ect.
Note that it is not important whether it can be proved to be correct and secure. The unique goal here is that everyone can prove to themselves that it is correct and secure. It must be obvious to everyone that they can trust it.
In my opinion, this is not possible to achieve with an electronic system.
Do you trust the system now enough to say your views are fairly represented (looking at the war and ... all the other things) ?
I volunteered as a scrutineer for a major Canadian political party as a teenager. You show up and watch the electionworkers open the ballot box and count the ballots. The ballots were counted fairly although some people couldn't tick the boxes correctly.
It's unclear how such a system would work in the United States, though, because you've merged all elections into a single voting day. If people struggle with ticking a single box from 5 options I can't imagine what a multipage ballot binder would be.
- We already trust computers to run our markets, banks, cars, energy infrastructure, etc. Is a computer popularly untrustworthy?
- Do low-tech physical ballot systems offer good guarantees? See 2024 Russian elections [0], for an extreme counter-example.
I'd say cryptography or smart algorithms can go a long way in upholding certain invariants, but you need some infrastructure for that: e.g. key pairs per voter and a trustless counting system. If you can't get that, then you're relying on the good will of others: in some cases it's the volunteer counters, in others it's whoever deploys and operates the trust-based black box e-voting system. I think that cryptocurrencies alone should be proof to anyone observant that a trustless voting system is doable, though I'm honestly surprised by this thread, because it alludes to the opposite.
[0] https://www.lemonde.fr/en/international/article/2024/03/20/t...