RE: Left off the ballot?

From: Popkin, Laird (WMG Corp) <"Popkin,>
Date: Tue Apr 13 2004 - 18:05:13 CDT

I'm not sure that adding a copy of the ballot data would add much integrity
to the ballot, but perhaps adding a hash (e.g. MD5) of the input ballot data
file (the one that lists all races and candidates, etc., on the ballot, not
the one that records the voter's votes) in that then you'd have a paper
record guaranteeing that the votes are for the right race. That way, if
people accidentally voted on the wrong ballot (e.g. primary for the wrong
party) we'd at least have a chance of figuring it out.

-----Original Message-----
[]On Behalf Of Bob Ramstad
Sent: Tuesday, April 13, 2004 5:39 PM
Cc: ovc demo
Subject: Re: [voting-project] Left off the ballot?

The barcode is a machine readable version of the voting selections the
person made.

Would it be appropriate to add another, more dense, machine readable area
that would represent the entire ballot i.e. each position on the ballot and
all available choices for each position? Obviously this would only be
accessed in certain very specific situations -- like a challenge -- and so
the equipment required to verify this data could be more complicated than
the simple barcode reader used for reading votes.

In short, given that the software has a ballot configuration file, and
records each vote -- and the assumption is that all votes were all generated
from the same ballot file -- it seems to me that the paper ballot should
really reproduce both the vote for that person and the ballot configuration
file used.

(For those who don't recognize my email address, it's because I was just
added recently... I'm a software engineer by training living in Seattle
with a broad interest in what the OVC is doing, and a desire to help in any
way I can.)

-- Bob

Karl Auerbach wrote:


The OVC system can and must stand up to this challenge. We can confirm what

the voter saw or didn't see because we have the CD from which the voting

machine was booted...


It is quite easy for there to be glitches into code that don't show up

until after the software has been running for 6 hours and after 100 people

have voted. Such glitches could be intentional, but things like memory

leaks and counter overruns are often triggers that don't occur until a

chunk of software has been running for a while.



= The content of this message, with the exception of any external
= quotations under fair use, are released to the Public Domain
Received on Fri Apr 30 23:17:05 2004

This archive was generated by hypermail 2.1.8 : Fri Apr 30 2004 - 23:17:29 CDT