WORKSHOP ON DEVELOPING A RESEARCH AGENDA FOR ELECTRONIC
VOTING TECHNOLOGIES
September 17-18, 2004
Convened by the American Association for the Advancement of Science
Main | Program | Participants | Synopses
Peter Maggs
Professor of Law, Clifford M. & Bette A. Carney Chair in
Law
University of Illinois at Urbana-Champaign,
Member, Board of Directors, Open Voting Consortium
http://www.openvoting.org
WHAT I BELIEVE ARE THE MOST PRESSING RESEARCH QUESTIONS RELATED TO ELETRONIC VOTING TECHNOLOGIES
The most important research question is that of determining an optimum design
for the system of procurement of electronic voting equipment. A great deal of
research has been done on government procurement systems, particularly on the
systems used by the Department of Defense to procure advanced military technology.
Suppose, for instance, that the United States faced the threat of missile attacks
by rogue states and that effective anti-missile technology was possible. One
possible procurement system would be for the federal government to allow each
state to file a grant application for a missile defense system and then to use
the grant money to buy a system for the state. Such a system would appear to
be absurd, yet that is the system that is being used for the procurement of
voting technology under the "Help America Vote Act." The design of
a system has important side effects. In particular the system of grants to states
heavily favors the use of private, proprietary voting systems over public, open-source,
open technology voting systems. This is because individual states cannot afford
the research necessary to create open-source systems. Research should consider
the effect of alternative models of procurement.
The second most important research question is that of determining an optimum
design for the setting of federal and state standards for electronic voting
systems. At present the United States Election Assistance Commission (with the
advice of the Standards Board created by the Help America Vote Act), state officials,
courts, and the private sector (particularly the IEEE) are all engaged in creating
electronic voting system standards. There is always a potential problem of bodies
engaging in setting regulatory standards being "captured" by the private
or public interests the standards are supposed to regulate. Research should
examine the present standard-setting process and if it is found wanting, should
propose alternatives.
The third most important research question is that of examining the effect
of lobbying and politics on system design. It should consider the effects, for
instance, of large donations by voting machine makers to advocacy organizations
for persons with disabilities, and of lobbying by state officials to prevent
unfunded federal mandates. It should also consider political issues raised by
pressure designs that make it easier or harder for less educated citizens to
vote.
The remaining issues are technical ones, related to such matters as user interfaces
and training, ballot confidentiality, software reliability, security against
tampering, and protection against accidental data corruption or loss. All these
issues are important. However, the way these issues are resolved will depend
more on the political factors discussed above than on the findings of scientific
research.