[E-voting] 'Paper trail' wanted for vote
dglaude at ael.be
Mon Apr 17 12:28:00 IST 2006
What you describe is the way "ticketing" was implemented in Belgium...
* Accept the vote
* Print the ticket
* Let the voter verify the ticket
* Ask the voter to acknowledge the vote is valid
Of course, there is an option for the voter to say that the printed vote
is not the displayed vote (because the vote is also displayed on the
If the voter say the printed version is not valid, then the president of
the voting place come and confirm (or not) that fact. Once confirmed,
the voter has a last chance to vote but in the presence of the
president... The story does not say what should be done if the bug is
But what if the vote on the screen is the vote on paper but both are not
the voter intent. It is a silly kind of "what you see is what you vote".
What took place in Belgium is that many voter did call the president,
not to say that the screen and paper do not match... but to say that
this was not what they wanted to vote (just like if expressing your vote
on screen is difficult, but verifying it on paper is easy... which is
likely the case for elderly people and those without any contact with
The president accepted that those elector do vote again. But that made a
lot of invalidated ticket (not vote) and made the counting even more
difficult (also considering that two election were printed on the same
There was a barcode printed on the ticket, but it was hopefully not use
as a barcode is not VVAT.
Then there were problem because paper count and electronic count did not
match... but that was suppose to be only a consequence of the above, not
a proof that the voting machine were broken.
What it really proof is that the elector have difficulties to express
their INTENT using the computer. Once checking the printout it was easy
for them to find out about it. I don't believe that many voter did
change their mind in between those two moment. ;-)
PS: When you do paper voting and you make a "mistake" you have two option:
* Ask for another paper ballot (and the previous one to be invalidated)
* Accept your mistake and put your paper vote in the ballot box
* Nullify your vote by making a nice drawing
None of those option are possible with our ticketting system.
Michael McMahon wrote:
> Marian Beddill wrote:
>> Oakley wrote about a VVPB: "... If voters said the printout was
>> different from what they entered, chaos could ensue...."
>> Well, maybe, if the review of "the printout" from a voting machine was
>> AFTER voting! That's why I and others have always insisted that the
>> VVPB ("voter-verifiable paper ballot") must be viewable WHILE the
>> voter is still in the process, and thus, if there is any difference
>> between her INTENT and the paper, she can CANCEL before finishing, and
>> return to the process of voting, until it is right.
>> The only complication, then, is that when (not "IF") the
>> voter-verifiable paper ballots are used for a hand-count, only the
>> final one for each voter is counted -- the rejected "erroneous"
>> incomplete copies must be ignored. It will take some administrative
>> stuff and elections staff training.
> The issue here is that voters could (possibly mischievously) claim that the
> printout is different from what they have entered on the machine. For
> you could claim "I keep voting for X, but the machine keeps printing Y".
> The problem is that the poll workers are not allowed to look at the machine
> or the printout to verify what the voter is saying, and then chaos could
> My view on that is that voters should have the option to cancel the vote
> on the machine
> and fill out a paper ballot instead. In our system, these votes would
> have to be entered
> into another machine at the count centre. This wouldn't be a problem
> because it would
> be anonymous and would be done in front of several witnesses who would
> all check
> that the VVAT printout *is correct*.
More information about the E-voting