Re: Open voting platform for the 2020 election - volunteers needed

[quiliro]

Chris Franklin <code@bythepeople2020.org> writes:

>  Thank you Chris for thinking about democracy as a result of libre
>  software.
>  I also like your campaign. If I would live in the USA, I would vote for
>  you for your healthy social values.
>  Please consider that voting electronically would reveal the identity of
>  voters and direction of their votes individually. Perhaps providing a
>  token to vote and making it impossible to identify specifically with the
>  voter's choice would work to preserve privacy. But I am not clear how
>  or if that would work.

> Using a token is definitely a possibility, and, used in combination
> with other technologies as well as the incorporation of time and
> method as variables, we’d already be most of the way to a
> reasonably-secure solution.

> It's important to remember that identity and intent (the two primary
> factors here) are processed billions of times daily in the form of
> credit card transactions, so this project is absolutely feasible with
> the proper focus and effort.

A caveat: I am not a programer. I am just a newbie.

Using a token could restrict each voter to emit only one vote. The token
must be as generic as a coin is in order to keep the privacy of the
vote. The user could use it as they would want. (They could even sell
it. Obviously we don't want that. Or do we?)

The would be:
1) Every voter receives one generic token for each vote (one for each policy).
2) That token is used to vote.

That's the only way I can envision private electronic voting. Otherwise,
the voting system can know who voted for what. I am not sure if the
votation order could reveal who voted for what. But that is an easier
issue. There can be ways to mangle the data. Databases are exact,
estimates are not.

I think there are already a bunch of voting software. But I have not
heard about a privacy-respecting or auditable electronic-voting
software.

With respect to the auditability of electronic-voting, please consider
Richard Stallman's opinion on the subject (as was referenced by Adonay
Felipe Nogueira on a hyperlink):

"How can you tell if a voting machine will honestly count the votes?
You'd have to study the program that's running in it during the
election, which of course nobody can do, and most people wouldn't even
know how to do. But even the experts who might theoretically be capable
of studying the program, they can't do it while people are
voting. They'd have to do it in advance, and then how do they know that
the program they studied is the one that's running while people vote?
Maybe it's been changed."

How do you propose we contend this problem? Maybe there is a common
sense solution or maybe there is a very complicated IT solution. But I
have not seen either yet.