2/7/05

Electronic Voting System: Brief Background and a Case for Innovation

Electronic Voting Technology has been around for several decades, notably in the form of optical scan voting machines which replaced the mechanical lever systems in the 1980s in the United States and the punch card systems in the 1960s. But they were used largely alongside paper voting systems. The major advantage of the optical scan system was the speed gained in vote counting which was done by computers, the ballot papers are fed into a computer system. The Direct Recording Electronic Systems however were the first fully computerized voting systems, first introduced in the 1970s , where ballots are either printed and posted on the DRE (using lever machine technology for counting) or displayed on a computer screen(modern DREs) and voters make their choices by pushing buttons, touching screens or using other devices[1].

Election fraud and errors are major motivation for voting machine innovations and the electronic system while promising to remove or at least reduce a certain type of fraud (ballot tampering) create the possibility of a more widespread type, result tampering on a wide scale if improperly implemented and managed.

In the last decade or so, several countries have implemented various Electronic voting technologies with India serving as the largest example of a national test bed with over 300 million people using Electronic Voting mechanism to choose the president of the largest democracy in the world in 2003 in a process that was largely considered a success.

Election is the heart of Democracy and voter confidence in the Electioneering process reflects directly on the mandate of the elected officials and as such the quality of the voting mechanism is essential for the stability of a democracy.

The close US presidential elections in 2000 where less than 600 votes separated the winner from the looser and more than ten times that number of votes were invalidated due to error in the electioneering process (hanging chads and butterfly paper chads) in an election were over 100 million voters went to the poll, largely underscore the need for a reliable, robust, secure (security is largely a localized concept where the ingredient of security is dependent on the election law in each country), easy to use (for all classes of voters) and intelligent (able to catch and report anomaly at the time of vote – under-voting and over-voting being of special interest in the United States) as well as cost effective voting mechanism.

In response to these challenges, the United States Congress in the aftermath of the 2000 presidential election enacted the Help America Vote Act (HAVA) of 2002 which created the Election Assistance Commission to work in concert with various existing Election agencies , commissions and standard bodies including the state initiative, the National Association of State Election Directors (NASED), the Federal Electoral Commission (FEC), National Institute of Science and Technology (NIST) and the IEEE [2] and help states and other election jurisdictions replace old election machines with modern systems by 2006.

In keeping with the mandates of the HAVA and to meet the new challenges head on, various states in the country turned to DRE manufactures and various DRE equipments were used in the 2002 national elections as well as the 2004 presidential elections on a wide scale. The result of the 2004 elections was more definite but there were widespread reports (at least by the losing camp) of irregularity in the conduct of the elections. Also prior to the elections, various independent consultants [3] reported in the DRE equipment and the implementation of the test-procedures. Although no fraud was reported in the election itself (except in the case of the Washington gubernatorial election which is currently being contested in court [4], the damning report and the seemingly consensus amongst computer experts [5] is enough justification for a rethink of the existing systems and a different approach.

Since the advent of the Internet, the need to allow internet voting has generated great interest in various circles, no less by the Department of defense which tested a pilot scheme in 2002 for absentee voting via the internet. The scheme was cancelled after several complaints were filled [6] regarding the security of this system.

A major issues noted by critics of the existing DRE systems is the lack of proper mechanism to assure the integrity of the system due to the proprietary nature of all the software components and the observed weakness in the codes when they were mistakenly made available on the internet. The belief in some circle is [7] is that if the software were open to scrutiny; errors in the codes could be easily identified and corrected since many experts will be able to review them. Also, several people have made a case for an open source election system as is the case in Australia [8], which will allow for contribution from several sources and enhance the quality of the system as been demonstrated by the quality of the open source operating systems, Linux and FreeBSD, which are reputably as robust as the proprietary windows operating system, if not more so for non-desktop applications.

Also the GAO[2] reported noted the possibility of a cryptographic election system which had been proposed by various cryptographic experts including Bruce Schneier[9] and of which there are currently various proposals[10].

With the success of electronic commerce which has been largely facilitated by the invention of public key cryptographic system[11] and the development of the SSL (secure socket layer) architecture largely implemented by the openssl development group[12] and Netscape [13], the prospect for a secure, reliable, robust, certifiable and open electronic voting system is a tempting possibility that deserve further investigation and modularized [2] architecture may well facilitate its development and use in the near future.







References

[1] Election Reforms and Electronic Voting Systems (DREs): Analysis of Security Issues, Congressional Research and Services (CRS) Report for Congress by Eric A. Fischer, Senior Specialist in Science and Technology, Domestic Social Policy Division, November 2003.

[2] Electronic Voting Offers Opportunities and presents challenges – GAO Testimony before the subcommittee on Technology, Information Policy, Intergovernmental Relations and the Census, Committee on Government Reform, House of Representatives, (Statement of Randolph C. Hite, Director Information Technology Architecture and Systems)July 2004

[3] 2004 U.S. Presidential Elections – A Presentation by Dr. Michael Wertheimer
Director, RABA Innovative Solutions Cell, RABA Technologies, at the Historical Electronic Museum of Baltimore, December 2004

[4]

[5]

[6]

[7]

[8] Australian Capital Territory Election ACT http://www.elections.act.gov.au/

[9] Applied Cryptography - Bruce Schneier, 2nd Edition . John Wiley 1996

[12] the Openssl Development group : http://www.openssh.org

[13] SSL 3 Specifications: http://wp.netscape.com/eng/ssl3

No comments: