[pp.int.general] Online voting versus online discussion

[Thomas Bruderer]

First:

There is a big difference between voting-machines and e-voting. As long
as we are not on the same page on this, this discussion is without any
hope. You can do e-Voting in different ways, we do not support any
simple scheme like the voting-machines. And it seems even pirates have
problems in distinguishing them - I won't try in this email.

But you can start reading here:
http://en.wikipedia.org/wiki/End-to-end_auditable_voting_systems

> If you want to have secret voting then it is not possible to check that
> your vote is taken rightly into account in the total...
That is not correct. Have you hread of blind signature or homomorphic
encryption?

What you say is simply wrong, it's not true.

> With electronic systems, whatever way you put it, everybody cannot check
> by himself that their is no fraud at least at the level of his polling
> station.
This is obviously not true either.

1.) You cannot check your own vote.  (obviously because the secrecy of
voting)
2.) You cannot count all votes. (too many, too far apart)

>  Sure there could in both case have fraud with lists of voters,
> but apart from that paper ballots are the only way to achieve secrecy of
> the vote together with the possibility for the public to control the
> full progress of the voting operations.
>
> So, yes, anything "done right" as you say is perfect. The problem is to
> build a system where everything is right even if the people who run the
> vote are desperately wanting to fraud !
Thats what cryptographic systems are for. you can check the whole system
mathematically,
obviously it must be open source. But there are systems which are great.
we use one.
> All the electronic voting systems that would reach that for the public
> would have to breach, one way or the other, the secrecy of the vote.

The keyword is cryptographic voting, you are repeating the same false
facts over and over again. Please start reading the article linked above
if you want to know how it works.

There have been great breakthroughs in the last 10 years. In theory even
fully homomorphic encryption is possible, which means any calculation
could be done secretly. However in practice it still too slow. But
Homomorphic addition can be done without a problem.

You and Stallman talk like there is a central server which does "vote++"
- this is complete nonesense, and we obviously oppse that.

Regards,
Thomas