Why can’t you make a true voting system first?
I believe that everyone who have a sufficient computing power in his hands (smartphone) can prove (mathematically) that globally accessed data (open ledger) is true. And that everybody put in there is preserved correctly. Political voting is a huge deal and voters must remain anonymous. This can be achieved with, again, sufficient computing power. Work in that field started a long time ago.
Suppose we solved the problem of translating anonymous votes to open ledger. There is another problem occurs: too many false voices, aka botnets. This should be relevant to Americans since 2016. Oh wow where is even a wikipedia page exists.
Another (almost the same, but relevant to Russians since forever) problem is about peoples voting on stuff they don’t belong with (like bringing unfamiliar peoples to voting stations or just throwing pre-filled ballot papers in the bin). This should be mitigated with frequent voting. Everybody vote anytime whey want, ideally once a week. The system track voters positions, so you don’t have a 146% turnout.
If somebody has no sufficient computing power, make a community one, accessible to everybody. Put a small computer in every Walmart. (The fucking ATM has it. Why can’t nobody put a raspberrypi with a touchscreen?). Casting vote once a week should eliminate attempts to bring вежливые люди to a polling place. Additionally this will solve another set of problems - verifying that the results you see on your phone or a computer or a paper is actually true and vice versa.
In the latter case there is a problem about identity. How can we say that this person today voting is the same that yesterday if the system does not have his digital identity (private key in the phone)? Maybe add a printer to rpi in Walmart to print a qr code with person’s private key? Maybe generate the digital identity on the fly (face-id, fingerprints, whatever)? Hypothetically, there is another problem about assessing digital reality. What if everything is hacked by special agencies? The answer is simple - this system should be build as open as possible with the ability to verify by anyone as easily as possible. Take, for example, the effort applied to the million dollar curve. Given that nobody today can break the elliptic curve cryptography (or take any other methods by any other country/agency and use them altogether), initially we just need some starting value that is accessible to everyone without the possibility of modifying while it gets to the recipient. In other words - just print a dozens of qr-codes and throw them on the streets before uploading the app to mobile stores. Everything else should be worked out by open ledger and sufficient computing power.
Maybe the TON network will bring some algorithms to practice? Though, I don’t like it’s closeness and obscurity.
Is there any formal proof (scientific paper) on the possibility of such system?