[EM] Why We Shouldn't Count Votes with Machines

[Terry Bouricius]

Ralph wrote:
"This can be solved by just publishing the ballot images.  This way 
everyone can work out their own result."

Note that in Burlington (Vermont, USA), all of the ranked ballot images 
(text file, not graphical images, unfortunately) are posted on the 
Internet after the election, along with tallying software and instructions 
on how to conduct your own IRV tally using any spread sheet software. You 
can see the actual city web site here:
http://www.burlingtonvotes.org/20060307/

Terry Bouricius
----- Original Message ----- 
From: "Raph Frank" <raphfrk at gmail.com>
To: <jgilmour at globalnet.co.uk>
Cc: "Dave Ketchum" <davek at clarityconnect.com>; 
<election-methods at lists.electorama.com>
Sent: Thursday, October 02, 2008 6:26 PM
Subject: Re: [EM] Why We Shouldn't Count Votes with Machines


On Thu, Aug 28, 2008 at 12:00 PM, James Gilmour
<jgilmour at globalnet.co.uk> wrote:
> Here in Scotland there is a somewhat "hidden" debate that must be had. 
> STV-PR was introduced for local government elections in
> 2007.  The counting rules adopted (Weighted Inclusive Gregory Method for 
> consequential transfers) make electronic counting almost
> obligatory.  (Manual counting to WIGM rules is possible, but long and 
> tedious because so many ballot papers have to be sorted and
> counted again and again.)  So we used scanners, OCR conversion and 
> e-counting.

That is similar to Abd's ballot imaging suggestion.

I assume that the images used for the OCR aren't made available to the 
public?

> The Scottish Government is promoting further use of
> STV-PR for various directly elected bodies.  This is raising issues 
> about the long-term provision of the equipment necessary for
> e-processing of the ballot papers for all these different public 
> elections and about the software that will be used for scanning,
> OCR and counting.

This can be solved by just publishing the ballot images.  This way
everyone can work out their own result.

> Concerns about "black box" processing have been somewhat muted so far, 
> but there have been calls for all blank
> ballot papers to be subject to individual adjudication by the Returning 
> Officer under scrutiny of the candidates and their agents.
> This is an example of the ridiculous double-standards that are being 
> applied to e-processing, because straightforward blank ballot
> papers would never be subject to Returning Officer adjudication in a 
> manual count.

A blank ballot is one that has no writing on it, or one that is not used?
----
Election-Methods mailing list - see http://electorama.com/em for list info