Which is the subposition posed by techie folks a long time. Like where I'm from it's still a paper ballot. But the idea that a paper ballot is that much safer because it's in paper form, but in the end it's the process and the framework around it that does the heavy lifting.
Voting machines can work... just not on an x86 running Windows enterprise lol abort, abort, the milk has soured, I repeat; the milk has soured...
But let's say a well designed RISC-V processor, no accelerators or things that make CPU go vroom vroom but that also introduces the threat of speculative execution and a solid, LTS Linux system with no WiFi, Bluetooth or anything, just a NFC or USB key pair that allows for anonymous voting, whilst also ensuring the integrity of the ballot using identifying measures like cameras in the voting locales and signing in at the entrance before voting.
Again, the process and the framework is the thing here. Even the hardware. Can the current industry handle it? Nope. Will open hardware and open firmware create a new revolution within the use and implementation of computers? I do believe so, and even within voting.
But some Oracle/Microsoft type job? No. Just no. There when you return to the paper ballot.
Did you watch the video? Because he pretty much covered all that. Electronic voting will never, ever be viable for one simple reason: there is no added cost to scaling an attack on an election.
Which part of "not with the current hardware" didn't you understand? How did those hackers gain access?! WiFi and Bluetooth, as well as some enterprise system shellacked on top of it.
I respect Tom Scott, but that kind of absolutism isn't anything else than a dismissal. But yes, don't trust these IT companies. They are trash, their licenses are trash and their code is trash.
If that's the basis he's going off, he is right, but that doesn't dismiss the point I was trying to make - which completely escaped you.