Saturday, September 25, 2004
Voting and the shortcomings of the ACM
The Communication of the ACM special issue on e-voting is now out (access only for ACM digital library subscribers).
Contents:
THE PROMISE AND POTENIALS OF VOTING SYSTEMS
Introduction ~ Peter G. Neumann, Guest Editor
Voting System Standards and Certification ~ Herb Deutsch and Stephen Berger
Independent Testing of Voting Systems ~ Carolyn Coggins
Implementing Voting Systems: The Georgia Method ~ Brit J. Williams
and Merle S. King
Small Vote Manipulations Can Swing Elections ~ Anthony Di Franco, Andrew Petro, Emmett Shear, and Vladimir Vladimirov
Auditing Elections Douglas W. Jones
The Code of Elections ~ Rebecca T. Mercuri and L. Jean Camp
Analyzing Internet Voting Security ~ David Jefferson, Aviel D. Rubin, Barbara Simons, and David Wagner
Source Availability and E-Voting: An Advocate Recants ~ Jason Kitcat
Though the special issue has merits, I feel that this is yet another missed opportunity, along with the recent ACM statement of policy on voting systems. Yes, the ACM is the largest computing organisation in the world, but it also has a large and active human-computer interaction special interest group (SIG-CHI). Voting is not all about technology, and the human factor is largely ignore in this special issue.
As Whitney Quesenbery, president of the Usability Professionals' Association and founder of its Voting & Usability Project, states: As many of us have pointed out many times, the problems in Florida 2000 were not computer problems, but an information design and usability error -- one that could have been avoided through better training and usability testing. As important as issues of security and audit trails are, usability is also a critical factor in voting systems. Perhaps the only article in CACM that addresses this major concern is that by Mercuri and Camp, which adapts Lawrence Lessig's ideas from his Code and Other Laws of Cyberspace to discuss the interaction of society with technology. I think it's worth quoting some of her concluding remarks in full:Federal election initiatives, such as HAVA (written into law as a direct result of the Florida 2000 Presidential election controversy and the subsequent malfunction of new election equipment in that state during 2002), the U.K.’s remote voting project (intended to promote turnout), as well as Ireland’s attempt to computerize its national elections by 2004, have all met with resistance. These legal solutions seek to resolve organizational, economic, and political problems with technology that cannot be independently validated as unbiased. Under such conditions, the proposed technology inevitably fails, in the larger sense, to solve the underlying issues that caused the initiatives in the first place. For example, the new voting technology does not ensure usability. There is nothing in HAVA that prevents a touchscreen voting machine from displaying a butterfly ballot layout, as was used with South Florida’s earlier punch-card systems. New technology introduces new problems—such as when color and voice capabilities may be subtly persuasive. The problems of digital voting are due to a combination of technological utopianism and a lack of technology-neutral definitions of social barriers, resulting in an organizational inability to perform an unbiased examination of how and when technology can best be applied.
As the technology is incapable of meeting false hopes, experts are finding themselves in a variety of roles, from guru to skeptic. Indeed, voting technology is following the same cycle as radio broadcasting, Internet commerce, telephony, and many other significant innovations [10]. First there is hope is that the technology will solve all problems. Then comes despair at the failure of the technology to meet the impossible promises of its most zealous supporters. Finally, a regulatory response occurs in order to address technology as embedded in society.
Hopefully, the ongoing discussion between and within the scientific and election communities can be harnessed to developing consensus that can be translated into better products and procedures, rather than (as some have asserted) being harmful by raising undue fears among the electorate....
In the meantime, others are steaming ahead, addressing such issues. Whitney reports: This week, the EAC Technical Guidelines Development Committee held hearings with a full day devoted to human factors and privacy. (Agenda and list of participants: http://vote.nist.gov/TGDCAgenda3Dayfinal2.pdf)
Public testimony is being posted as it is submitted, and the complete transcript will be posted within the week. If you'd like a good sense of the scope and range of human factors and usability issues in elections, this is a good place to start:
http://vote.nist.gov/HearingsandTranscripts.htm
It's a pity that CACM is not reporting on such high profile initatives.
10:37 AM|
link to this item
|
0 comments
0 Comments:
Post a Comment
|
