Computer technology has enabled humanity to make great advances in aircraft design, communications, and countless other fields.
But when it comes to recording the results of elections, it’s often unclear to election officials in towns across the country whether the introduction of computers has been a help or a hindrance.
That’s where Douglas W. Jones, associate professor of computer science in the University of Iowa College of Liberal Arts and Sciences, comes in.
In 2010, he was appointed by the U.S. Election Assistance Commission (EAC) as one of four new technical and scientific experts to its Technical Guidelines Development Committee (TGDC). The TGDC is charged under the Help America Vote Act (HAVA) with assisting the EAC in developing federal voluntary voting system guidelines that are used to test and certify voting systems.
Previously, Jones served on the Iowa Board of Examiners for Voting Machines and Electronic Voting Systems for 10 years, where he helped examine and approve voting systems before they were sold to the state’s county governments. He testified at the U.S. Civil Rights Commission hearings in Tallahassee, Fla., on Jan. 11, 2001, and was involved in reviewing the federal 2002 Voting System Standards.
Also, in August 2005, he was awarded a five-year, $800,000 National Science Foundation (NSF) grant to investigate the use of electronic voting systems in U.S. elections. The grant is part of a $7.5 million NSF project called ACCURATE (A Center for Correct, Usable, Reliable, Auditable, and Transparent Elections).
Jones recently shared his thoughts with fyi on present and future voting technology.
What is the EAC?
The U.S. Election Assistance Commission maintains a clearinghouse of information on election administration, ranging from voting system testing and certification to data on how Americans voted in recent federal elections. The technical guidelines committee examines the rules governing voting systems used in the United States to ensure that the election choices of individual voters are accurately registered and tallied. This is important because the basis of democratic government is the ability to conduct fair and accurate elections.
To be viewed as fair, an election system must be transparent. To borrow a phrase from Dan Wallach, associate director of ACCURATE, the system must convince the losers that they lost. However, the losers and their supporters often have no required technical qualifications. This means that the entire election system must be open and comprehensible so that nontechnical observers can believe the results.
What do you and your colleagues on the technical guidelines committee do?
We look at all facets of computer use in voting. As the technology used in elections has become more advanced, the opportunity to make subtle errors in design or implementation has increased. Voter confidentiality and privacy, computer system integrity, accountability, usability, resolution of irregularities, and ease of administration by election officials are some of the issues. Also, individuals with various disabilities encounter additional problems in using some computer voting systems.
When and where was the most interesting election system you were ever asked to investigate?
Two trips to Kazakhstan were real eye-openers. On the surface, they regulate voting systems much as we do, so the problems I found illustrate risks we must take seriously in this country.
In November 2005 and again in 2007, I participated in the Organization for Security and Cooperation in Europe election-observing mission in Kazakhstan. The Kazakhstan electronic voting system includes interesting innovations involving stateless voting machines and using smart card technology. However, it also has some severe defects.
As in the United States, an independent testing authority must approve Kazakh voting machines, and detailed reports by the authority on the system are delivered only to the voting-system developers, with no intention that these ever become public documents. This is dangerous. But in Kazakhstan, even some of the standards were state secrets, so we can’t even tell what the machines wee supposed to do.
So which is more accurate and efficient — paper or electronic ballots?
The best and most efficient system is one that uses a combination of computers and paper ballots. Voting on paper ballots — which can be stored and checked if irregularities in the vote count later develop — is important. For speed, efficiency, and ease of counting, computers are more accurate than humans.
It is important to keep in mind that the term “computer security” is almost an oxymoron. We can make computers more secure, but we cannot guarantee security. Therefore, if we want trustworthy elections, there must be a way to check the results without relying on the computers involved.
What other research areas interest you?
In the 1990s, I was heavily involved in real-time control of stepping motors. My Web pages on that subject are still widely cited, and designers of computer-controlled mechanisms still contact me on a regular basis for advice about motor control.
I am generally interested in the history of technology. As a result, my forthcoming book on electronic voting systems begins in the 19th century with a discussion of the origins of mechanical voting machines.
What courses do you teach?
One of my favorite courses is 22c:060, Computer Organization, the undergraduate introduction to how computers actually work. This is a course where I get to see students experience “Aha!” moments when they finally understand how it is that computers are able to execute their programs. I also enjoy teaching 22c:112, Operating Systems, and 22c:169, Computer Security. All of these courses focus on what is actually going on inside the box, as opposed to focusing on how the box is used.
How do you see preparations for the 2011 and 2012 elections developing?
As in most other U.S. elections, there will be a wide variation in how the election progresses, partly because there are wide variations in the systems and machines used in different states. Despite the fact that the U.S. Constitution puts authority over congressional elections in the hands of Congress, we leave most of the details to the states, so there are significant differences from state to state in how elections are conducted. In many states, counties independently make the final decisions about what technology is used. The benefit of this is a competitive, resilient marketplace for voting equipment, but the downside is that states and counties occasionally make significant errors.
What risks do you see in the conduct of elections in the coming years?
We conduct elections on a shoestring, using more and more technically sophisticated machinery. As a consequence, many election administrators are using equipment they hardly understand. The result is a strong temptation to outsource election administration. This is dangerous.
There is rising pressure from many sectors to use the Internet or cellphone network for voting. If we move in this direction, we risk creating a new praetorian guard — the keepers of the network. If the keepers of the network are honest, the elections will be honest, but if they are not, just as in ancient Rome, they — not the people — will determine who is elected.