Every time someone confidently claims that we can cryptographically verify voting, they are deliberately or ignorantly keeping the complexity and necessity of verifying the verifier runtime, the data source, and the communication channels out of the picture.

Cryptography doesn’t solve voting verification problem, it obscures and shifts it.

OK, cool now teach your family that calls their web browser "The internet" enough computer science to adequately understand and audit this proposed open system and convince themselves that their votes are counted in a fair, verifiable and secret manner. Also that the implementation does not have obvious side channels and what is actually running is built from the published source code.

Like, If I was part of some shady powerful elite I'd love a fully automated setup. Most people will not be able to check the system deeper than "phone displays green check mark" without an unreasonable time investment.

On the other hand, "room full of people opens box full of papers and counts them while verifying each other" is intuitive enough for almost anyone to grasp and gain confidence in.

You may be able to verify that a given code is correct, or that a given device is correct, but no amount of software can conclusively prove that a given person has voted. All that cryptography prevents is a man in the middle attack, it does not prevent bad inputs from being entered by people who have stolen credentials.

Voting should be easy and convenient, but paper ballots and voting booths can be easy and convenient.

yes but 3 different exploits would result in different results. which is an indication that the voting result is irrelevant, and should be repeated

edit: I am by far no expert, yet having anykind of kniwledge in that area, so I apologize for naivity in advance.