[geeks] I just saw...

[Lionel Peterson]

>From: Joost van de Griek <jvdg at sparcpark.net>
>Date: 2006/11/08 Wed AM 04:12:20 CST
>To: Lionel Peterson <geeks at sunhelp.org>
>Subject: Re: [geeks] I just saw...

>Lionel Peterson wrote:
>
>> All this talk of paper trail and audit receipts is a bad idea, in my
>> opinion - if a machine makes two independent records,one electronic,
>> one on paper - if they diverge, which is trusted? If one is not
>> trusted, why include it in the process?
>
>Both are trusted, but one is leading: the paper one. The electronic one
>just provides a convenient and fast way of counting the votes, but it
>is always recognised that it can be easily tampered with. If any
>suspicion arises that the electronic tally is corrupted, the ballot box
>is opened and the paper votes are counted.

Let me be clear. I favor paper ballots over "automated" solutions.

What dual-record proponents seem to be saying is:

We will collect electonic and paper-based duplicate ballots for each vote.

But we acknowledge that the electronic results are more easily tampered with (i.e. one software change is easier than altering 40,000 paper ballots).
 
We will give preference to the electronic results UNLESS we suspect a problem, THEN we will count the paper counts.

My argument is that if electronic results are so much more fragile than paper (as confirmed by the need for the paper audit trail backup), why not just go with paper?

If you craft a subtle hack that alters votes in a non-obvious way, the paper trail will not be consulted,and the fraud never found...

>True, paper votes can be tampered with, too, but the paper ballot
>system has been in place for quite some time and is easily controllable.
>The risk of fraud with paper votes is at an acceptable level; that of
>the current electronic voting systems is not.

Agreed.

>The process is laughably simple: the voting machine records the vote,
>prints a receipt indicating what was voted, the voter checks the receipt
>and deposits it into an old-fashioned ballot box.
>
>Occam's Razor dictates that this is a miles better solution than the
>myriad of security measures proposed by voting authoroties and voting
>machine manufacturers which, quite frankly, does nothing but raise
>suspicion that they are in fact unwilling to provide a proper solution
>to the problems posed by electronic voting.

That would be the "ballot generator" that Avi Rubin (and others, I'm sure) advocate.

Lionel