[EM] Electronic Voting Bill of Rights?

[David GLAUDE]

I think we should stop this talking about CD and DVD...

I don't want to help you convince yourself that e-voting is the way to 
go. No technical trick will make Black Box Voting (even if some source 
code is open source) acceptable to any defender of democracy.

I think a few step must be identified into an election:
0) Producing the elector list
1) Checking the elector identity
2) Recording the elector vote
3) Counting the elector votes
4) Aggregating all the elector vote and select candidate using your 
prefered advanced counting method.

0: Is hard to control since the state how the population list... making 
those list public is dangerous for privacy.
1: Should be easy, but care must be taken. In some country your finger 
get marked with semi-permanent ink (that take days to be removed) if you 
try to vote twice, you will get noticed. If step 0 or step 1 is 
difficult to archieve is some country, it could be a solution.
2 & 3: Can and should be split. If it is one step, then there is no way 
to do a recount of any kind.
4: I think the result of step 3 should be publish and display. After 
that verification that the result of 4 are valid is easy since the 
availability of all the result of 3 are available.

In the creazy discussion we have about CD and DVD, it is not clear where 
we talk about it. Is it step 2 (a CD writer in every voting machine) or 
step 3 (the central machine that count the vote and produce the result 
for that location).

I don't like DRE because it is hard to know what take place when the 
voter is alone with the machine. Maybe one can find a trick to vote 10 time.

The Belgian solution with magnetic card provide some interesting 
solution to complex problem. What we do is:
A) We replace the pre-printed paper balot by an initialised magnetic card.
B) We replace the pen by a computer to record your vote on the magnetic 
card.
C) We replace inserting the ballot into the ballot box by inserting the 
magnetic card in the magnetic card ballot box. (and the vote is recorded 
on a "special memory" at a random position to simulate anonymity)
D) The computer, at the closing of the polling station, start counting 
the vote and produce local result.

The nice thing is that you receave only one card.
All card are assumed to be anonymous and the same.
The voting machine is supposed to record your vote.
The Ballot Box is supposed to be read-only card reader.
The Ballot Box is supposed to record perfectly and exactly your vote.
Everybody at the polling place can check that you only insert one 
magnetic card.

The number of magnetic card received at the begining and left at the end 
of the election are counted.
The number of initialised card is counted.
The number magnetic card inserted in the ballot box is counted.

Now if the ballot box and the voting machine were produced by separate 
vendor. If the computer behaviour was garanteed and transparent... it 
might seems acceptable. ;-)

We even tested the printing of a voter verifiable paper audit trail. But 
that was a distaster since the magnetic result and paper result did not 
match.

>>> 2) Now you also have to fight Cosmic ray

I am talking about this: 
http://wiki.ael.be/index.php/ElectronicVotingRandomSpontaneousBitInversion

> Speaking of outer space...

> One of the things mentioned as a part of this discussion was the use of 
> Open Source to allow the checking of the inner workings of the computer 
> software that counted the votes.  I was told that NASA uses two 
> independent teams of computer programmers in order to program the software 
> that controls rockets, for example.

One thing that could help is to define interface, API, protocol, data 
format. Then all the vendor of each independant piece of hardware or 
software must implement that. Then at election time, official choose 
wich combinaison of pieces are used (mix and match). This can garantee 
that no single vendor is in control of the whole process. However it 
might still be possible for one of them to control the election result.

>>> 3) Some screen technology might be better than other...
>>> Otherwise you need to go for Tempest proof equipment that cost a lot.
> 
> Good grief...  Never thought of this.  I have basically agreed with David 
> that paper ballots are the way to, if I have read the posts correctly.  
> But now....

Once again as this list is dedicated to advanced voting/counting method. 
I would like to know wich one are possible to handle with manual 
counting on large scale.

David GLAUDE