Voting System Auditability and Verifiability

Kathy Dopp's Voting System Papers (many of them collaborative)

Nineteen Election Reform Groups Oppose Full Public Disclosure of Software on Voting Systems Want COTS Exemption, April 2007

Avoid Another HAVA Train Wreck: Software Disclosure Requirements are a Good Long Term Goal But Need to Be Redrafted in Current Federal Election Integrity Legislation. March 2007

How Can We Ensure the Accuracy of Vote Counts? July 2005

Electronic Voting System Best Practices March 2005

Summary: Utah Voting Equipment Selection Advice Advice Against Purchasing DRE voting machines, October 20, 2004

Concept for Proposal Utah Voting & Elections System August 2004

Response to: “American Attitudes about Electronic Voting” Survey And Advice for Utah’s Voting Equipment Selection Response to Utah Election Office Hiring a Political Scientist to promote Paperless E-voting, September 24, 2004

RESPONSE TO REQUEST FOR PROPOSAL VOTING SYSTEMS/EQUIPMENT TO: Governor Olene Walker, Lt. Governor Gayle McKeachnie, Amy Naccarato, Utah Procurement. Response to Flaws in Utah's RFP Process for Purchasing Voting Equip., July 19, 2004

Utah Count Votes SUGGESTIONS TO: Utah Voting Equipment Selection Committee, Wednesday, July 07, 2004

Resources for More Information

Verified Voting's List of State Voting Systems in Use

NJ Attorney General Denies Certification of Voter Verified Paper Record Systems for Electronic Voting Machines

The Trouble With Touch-screens Aug. 2007 Dan Rather, - Includes full transcript and links to videos

Another Rebuttal to Michael Shamos. Shamos is a professor at Carnegie Mellon University who certifies voting systems for the state of Pennsylvania. Shamos believes in using paperless electronic-ballot voting machines. This particular rebuttal to his work is by Arthur M. Keller, Ph.D., Edward Cherlin, David Mertz, Ph.D.

How Long Does it Take to Change a Voting System? by Verified Voting

Contractual Barriers to Transparency in Electronic Voting
by Joseph Lorenzo Hall, School of Information, University of California, Berkeley

See the section on Security Reports Voting System Security by Jim Soper

The US League of Women Voters' Official Position on Voting Systems

Peering Through Chinks in the Armor of High-Tech Elections May 27, 2007 - A sobering, documented review of the daunting challenges of secure electronic elections. Translated for non-computer experts.

New Jersey: Tests Find Flaws in Printer Performance, Could Jeopardize Election Accuracy

Auditability of Diebold Vote Count Accuracy - Does 2+2 = 4 ?

Are the machine vote counts produced by Diebold DRE voting machines auditable for accuracy?

On April 19, 2007, the "Collaborative Audit Committee" released an audit of the 2006 federal election (held on Diebold voting systems) in Cuyahoga County, OH. Kathy Dopp and Frank Stenger's method was used by the auditors to calculate the number of precincts to audit.

Kim Zetter of Wired Magazine reports, "The audit found that the Diebold system ... doesn't produce report results at the machine level, only at the precinct level, which makes it difficult to trace faulty results back to a specific machine."

According to the results of the Cuyahoga County Audit, Diebold DRE machine counts are Not directly auditable using Diebold's own GEMS election management system software - and may not be auditable at all at the DRE machine level.

A manual audit of machine vote counts produced by Diebold DREs with VVPATs could only be done by:

1. encouraging all voters to verify their VVPAT records, and

2. randomly selecting entire precincts to manually audit in sufficient numbers to produce a 99% probability for detecting errors sufficient to produce incorrect election outcomes.

Why would anyone design or even more surprisingly, why would anyone purchase and use, a voting systems that is this difficult (or impossible?) to check the accuracy of individual DRE machine election vote counts; and so easy for anyone with the right skills and access to wrongly alter its vote counts - even during the process of checking them?

How many voting systems, in addition to Diebold's, are similarly difficult (require using backdoors to directly access the database) to check the accuracy of its individual DRE machine vote counts?

In addition the Cuyahoga County Audit revealed other major problems with Diebold voting systems:

* Diebold optical scan hardware does not include the ability to count the number of ballots which were counted (only the number of ballot sheets). This makes it very difficult and expensive to know whether or not all the optical scan ballots were counted, or counted twice.

* Diebold uses a database that is vulnerable to corruption and is recommended AGAINST using by Microsoft in applications where "a lot of concurrent activity is happening with the database" like elections.

* Diebold uses two separate databases to count votes that should have been identical in OH and were off by 100 or more votes for each of the three races which were audited.

* Diebold DRE voting machines had duplicate serial numbers.

Election audit finds mistakes aplenty
According to Associated Press reporter Thomas Sheeran "1 in 4 Cleveland memory cards fail to match up to official totals". 37 out of 132 precincts had discrepancies between vote-memory cards and paper records of ballots cast on electronic machines.

Cost Analyses of Voting Systems Comparing DRE to Opti-Scan

Cost Analysis of Maryland’s Electronic Voting System March 2008

Verified Voting of New York - Cost Comparison Resources

Florida - Survey of Changes in Total Annual Expenditures for Florida Counties Before and After Purchase of Touch Screens and A Comparison of Total Annual Expenditures for Touch Screens and Optical Scanners

Maryland - 2008 Cost Analysis This analysis has excellent charts.

North Carolina - Fiscal Impact Legislative Study
North Carolina - Cost Comparison

New York: Facts About Ballot Printing Costs
New York: Cost Comparison of Paper Ballot Optical Scan And Direct Recording Electronic Voting Systems