Fwd: Re: [E-voting] Next wednesday's committee meeting

[David GLAUDE]

> Subject: Re: [E-voting] Next wednesday's committee meeting
> Date: Saturday 06 December 2003 19:00
> From: Dave Madden <sares at redbrick.dcu.ie>
> To: Fergal Daly <fergal at esatclear.ie>
> 
> On Sat, Dec 06, 2003 at 06:49:16PM +0000, Fergal Daly wrote:
> 
>>>"So what happens when a voter says their paper ballot printout isn't
>>>correct?"
>>
>>I'm not sure what the "best" answer here is, they may all in some way 
>> violate 
> 
>>the secrecy of the ballot so it's essential to point out that no matter what 
>>the consequences are they are much better than having the person's vote 
>>recorded incorrectly.
> 
> That plays right into Cullen's claims that VVAT breaches secrecy of the
> ballot though. Meaigs needs to be able to give an answer that protects
> both integrity and secrecy - although there is always the option of
> saying "No E-Voting", or at least no to DRE. Meaigs has pointed out
> before that Optical Scan is a kind of VVAT-system, so it's an option,
> with the only problems being that it's imperfect and will cost a lot (in
> that we'd have to junk Nedap). Retro-fitting is much politically nicer
> for the Government.
> 
> (Although I think someone - Colm MacC? - mentioned that in Japan they
> use stickers on their ballots, to increase accuracy?)

In Belgium, when they (interior minister) wanted to demonstrate the 
general public that our e-voting can be trusted... they introduced VVAT 
in two specific location.

That VVAT is called 'Ticketing' they say the system come for Brazil(?).
It is a printer add-on to our "normal" magnetic card voting system.

We (PourEva) complained that this was ridiculous.
1) A proof in those two location will proof nothing (since the program 
could be different there).
2) If we want to count paper (because the paper was to be counted and if 
there was no match between paper and magnetic result... paper result 
will be validated... that is our law). Then the best way is to let 
citizen vote on paper! Why put a computer between the voter and the 
paper ballot???
3) Their is a breach in the secrecy of the elector!!! You have to 
confirm, your vote... but if it does not match (paper and screen) then 
you have to reveal something and get assistance from someone.
4) Adding a printer on every voting machine will cost a lot of money.

So we are in favor of "scanning" (with rules to verify if the process 
work and full recount in case of problem).

After the election (18 Mai 2003) we found out that:
1) Ticketing was a failure because ticket were hard to read.
2) Ticket were also printed with a bar-code (that was not used for 
"hand" counting the vote).
3) Ticket (paper) result and magnetic card result were different. The 
law stated that the paper count must be use in that case... but official 
expert choosed to use the magnetic card result.
4) The problem with different result in paper and magnetic card was 
supposed to stay not public. The official report does not talk about the 
problem, but an extra note stamped confidential explain why ticketing 
was a success dispite the discrepancy in the result.
5) One of the e-voting vendor said on national radio that Ticketing was 
never supposed to be generalised, but was only to convince the 
population that e-voting was OK.

So to Keep it Simple Stupid, do not introduce computer in the casting of 
the vote. Maybe the counting can be accelerated by scanner, but human 
verification must be mandatory in the law.

David GLAUDE