Image:Cfp banner 2.28.jpg

Towards Trustworthy e-Voting: An Open Source Approach?

From CFPWiki

Revision as of 23:29, 20 May 2008; view current revision
â†Older revision | Newer revisionâ†’
Jump to: navigation, search


Panel Summary

The troubles with voting machines produced by the private sector are well documented. One significant cause is that today's systems amount to "black box" voting compelled by necessary private sector business interests. In an increasingly digital democracy the use of computers seems unavoidable. Can open source methods to develop trustworthy hardware and software help us move toward "glass box" voting? This Panel examines the viability of this approach and what would be required for adoption.

Detailed Description

There is a growing belief that trust in elections is sliding as a result of the well-known troubles with current election systems technology, products, and vendors. An often-touted solution to this problem is â€œopen source.â€ This panel will examine a number of refinements to this vague prescription, starting with some basic questions, such as: What are some often cited examples of problems that open source could be part of a solution to? What are some of the many definitions of â€œopen sourceâ€ that are relevant? What might be a useful meaning for â€œsolutionâ€? Why are todayâ€™s products black boxes, and are any those reasons correctable with â€œopen sourceâ€?


As a framework for these discussions, weâ€™ll start with the hypothesis that one factor behind todayâ€™s troublesome voting technology is that voting systems vendors lack motivation or market incentive to make â€œreal fixesâ€ to current products, or to â€œgo back to the drawing boardâ€ in order to â€œget it rightâ€. Weâ€™ll discuss what the these catch phrases might mean in the specific context of voting systems, and compare with the vendorsâ€™ business requirement to make a return on investment in product R&D. Fortified by the real-world experience of two of our panelists on the â€œbuy sideâ€ of the election systems markets, weâ€™ll try to make a fair assessment of what can and canâ€™t be reasonably expected of proprietary systems vendors of â€œblack boxâ€ products.

Weâ€™ll compare these accounts of the current situation and speculate on improvements that might be possible with alternatives to the current for-profit vendor model, critically including the use of open source methods, or open source technology, or open source organizations, or other forms of open source. Informed by the experience of two veteran voting technologists, weâ€™ll avoid regarding open source anything as a panacea, and instead discuss open source as an enabler for various possible improvements, including but not limited to the following recipe for mitigating the technology-driven slide in voter confidence:

  • properly engineering the required assurance, reliability, security, and trust into voting machinery, outside the framework of profitable returns on investment;
  • using open source methods to ensure transparency of election technology itself, transparency of usage of election technology, sustainable audit loops for both, and public examination of both;
  • placing the technology in the public trust, making it publicly available for any government or vendor to adopt and use to produce and deliver election systems products

This recipe -- a combination of some of the various meanings of â€œopen sourceâ€ -- will be questioned in several ways: viability as a process; issues with widespread adoption of the results of the process; and whether this approach can properly address all of the issues of assurance, security, and trust. No doubt, after the discussion, weâ€™ll have more ingredients and multiple recipes, and more questions for going forward to a cookbook that includes relevant contributions from all the players, including elections commissions, states, federal organizations, vendors, evaluation and certification organizations, standards bodies, and technologists of many stripes ranging from systems design to security to human factors.


The session will consist of 3 approximately equal segments. First, each of the panelists will make a brief statement of one or two key points about the scope for improvement and/or the role of some meaning of the term â€œopen sourceâ€, based on their personal experiences in elections technology and policy. Each of these statements will be discussed by panelists, in order to help the audience generate questions for the second segment, which will be Q&A with the panelists. The moderators will pose questions both previously compiled, and collected from the audience during the first segment. The final segment will be â€œopen mikeâ€ time in which audience members may briefly comment on one aspect of the Q&A and/or pose a follow-up or new question.

Fun (?)

Throughout the session, weâ€™ll also be collecting from the audience proposed multiple choice questions for brief election that weâ€™ll conduct towards the end of the session, in which all can vote on questions of best and worst bets for open source, likely and unlikely outcomes, and whatever other election technology prognostications we come up with in a lively 90 minutes of discussion and debate.

The Panelists

Gregory Miller and E. John Sebes, co-moderators

Greg Miller is co-Director of the Open Source Digital Voting Foundation. Gregory has worked for 27 years in technology business as a software developer, IP lawyer, entrepreneur, mgt. executive, venture adviser, and technology policy analyst. At OSDV, Gregory is responsible for Foundation development, public outreach, gov't affairs, corporate alliances, and operational management.

John Sebes is co-Director and chief technology officer of OSDV. He has worked in information technology business for over 25 years as a software developer, program manager, Silicon Valley entrepreneur, and information security consultant. At OSDV, John sets the agenda for technical work, and occasionally gets to do some himself.

Rebecca Mercuri

Rebecca Mercuri is well-established as one of the leading international experts on electronic balloting and vote tabulation. With two decades of research, over 20 technical papers, and a doctoral dissertation at the University of Pennsylvania's School of Engineering and Applied Science on this subject, her opinions have been sought by equipment manufacturers, candidates involved in recounts (most notably for Bush v. Gore), federal organizations (such as the House Science Committee, the U.S. Commission on Civil Rights, the National Institute of Standards and Technology, and the U.K. Cabinet's Office of the e-Envoy), as well as numerous state and local legislative bodies.

She has been quoted in the U.S. Congressional Record, on the floor of the Irish Parliament, and frequently by the media (including the BBC, NPR, NBC, ABC and CNN, the New York Times, Wall Street Journal, and the Economist). Following fellowships (during 2003-2005) at Harvard's Kennedy School of Government and the Radcliffe Institute, Dr. Mercuri returned to the consulting company she founded, Notable Software, Inc. <>, where she and her forensic team perform computer-related investigations and provide expert witness services on a wide range of civil, municipal and criminal cases.

Jeremy Epstein

Jeremy Epstein runs the security group for Software AG. Electronic voting is a hobby that threatens to overtake his life. In recent years he's served on two Virginia legislative panels investigating what the Commonwealth should be doing, co-founded Virginia Verified Voting (a lobbying group for safer electronic voting), been an expert reviewer for the nationally recognized Brennan Report, consulted to the Attorney General of Kentucky, and was the expert witness for voting technology in a case against the State of Maryland. He recently published an article in IEEE Computer on how electronic voting machines work. Jeremy has been involved in security research, development, and analysis for 20 years. He holds a BS in Computer Science from New Mexico Tech, an MS in Computer Sciences from Purdue University, and an ABD in Information Security from George Mason University.

Lesley D. Mara

Lesley Mara has served as Deputy Secretary of the State since January, 2006. As Deputy, she is responsible for overseeing day-to-day operations of the Secretary of the Stateâ€™s office, including its elections and commercial recordings divisions.

Prior to joining the Secretary of the Stateâ€™s office, Attorney Mara served as the first Superintendent of the Connecticut Juvenile Training School, Connecticutâ€™s single correctional facility for youth in Middletown. She was responsible for aspects of planning for the opening of that new facility, including budget preparation, staffing, equipment, program development and training.

Between 1995 and 2000, Attorney Mara served as national Executive Director of Lawyers for Children America (LFCA), a non-profit legal program chaired by ZoÃ« Baird. LFCA leverages pro bono services from major law firms and corporate legal departments to provide advocacy for abused and neglected children and to teach conflict resolution and mediation skills in schools. Under Maraâ€™s leadership, Lawyers for Children established programs in Hartford, Miami and Washington, D.C. and when she left Lawyers for Children, the program leveraged more than $1.4 million in pro bono services on behalf of the most vulnerable children.

From 1990 until 1995, Attorney Mara worked at The Department of Children and Families, first as in-house counsel to the Commissioner and then as Deputy Commissioner for Administrative Services. As Deputy Commissioner, she assumed day-to-day responsibility for approximately 200 staff in the fiscal services, engineering, quality assurance and Consent Decree implementation divisions.

Douglas A. Kellner

Douglas A. Kellner has served as Co-Chair of the New York State Board of Elections since 2005 and was a commissioner on the New York City Board of Elections from 1993 to 2005.

Commissioner Kellner has been an outspoken advocate for improving the voting process in New York while insisting on transparency, verifiability, accuracy and uniformity in voting procedures. When he was first appointed to the New York City Board, he was the very first election official to call for a voter verifiable paper audit trail for electronic voting machines, a principle now enshrined not only in New York law, but in the election codes of a majority of the states throughout the nation. While leading the opposition to unverifiable electronic machines, Doug was instrumental in promoting new technology for scanning absentee and provisional ballots. He drafted model procedures to open the process of canvassing ballots to public scrutiny and convinced his fellow commissioners to adopt rules that provided meaningful due process in ballot challenges.

Commissioner Kellner has served as the Chairman of the New York County Lawyers Election Law Committee and is a member of the Election Law Committee of the Association of the Bar of the City of New York. He serves on the Advisory Boards of the Accurate Voting Foundation and the Verified Voting Foundation.