So yes it's not "perfect", but it sure looks harder to manipulate than a centralized database server that only requires a simple UPDATE sql query to completly and stealthily change an election.
So yes it's not "perfect", but it sure looks harder to manipulate than a centralized database server that only requires a simple UPDATE sql query to completly and stealthily change an election.
How? That means there is a record of my vote tied to my identity. That's not an anonymous ballot. Now you've opened up all sorts of other problems of coercion, etc. You'd have to get your key from the government, so there is a chance to get it. You could track every time someone voted and in which elections which would let you narrow down where they live pretty easily.
In our current system you can find out that I voted, but not how I voted. That should be how it's setup.
That said, We're talking about digital voting here (or at least I was) and the comparison is to a digital centralized solution, in which browser history, cookies, server logs, and not to mention the database itself has issues that can compromise the voter and the election.
The key system, wouldn't necessary have to work that way (you could have a derivative id or something like that, the system could automatically assign tokens each election without the need for manual distribution, you can throw away the temporary key, etc..). It's a open problem though, and it's good to bring these critiques up and think about solutions, I woudln't dismiss the whole thing though on the first problem found, it's easy to criticise but most technologies do have issues initially that need to be solved (and that's part of the fun). What's so interesting about decentralization technologies is that they open new possibilities and novel solutions to old problems.