Introduction

Elections, like many aspects of society in the United States, have changed dramatically over the course of history. With the dramatic growth of urban areas during the last century, and passage of various federal and state laws that specify increased electoral enfranchisement of citizens, we are placing greater and greater demands upon voting technology and election administration. In the past few decades we have started to use computers and networking to further increase the abilities of this equipment. The most fundamental act of our democracy -- the mechanics of casting and counting ballots on election day -- that initially took place in plain sight and was fully comprehensible to the franchise now takes place within machines that foreclose observation and obscure this formerly fully comprehensible act. An electoral system that was once highly transparent -- supporting public scrutiny and ease of understanding its functions and policies -- has undergone an ``enclosure of transparency.'' Computer software is one of the most opaque aspect of electronic voting. Unsurprisingly, academics, activists, election officials and commentators have called for increased access to, and heightened examination of the source code that powers election systems. Efforts to increase access and scrutiny range from source code escrow requirements,Source code escrow involves depositing the source code for a voting system with a third party and/or an election official and stipulating under what conditions the source code can be released. independent code reviews,A state election official may reserve the right to ask an independent party to do source code review on top of what is done at the federal certification level. system performance testing,Performance testing involves testing a system in conditions similar to those used on election day. required disclosure of source code to requirements that systems use open source codeA note on terminology: Open source software is software that is usually developed by a team of volunteers and released under generous licensing terms that allow users to exercise a number of rights, such as copying, modification and distribution, which traditional software licenses withhold. The distinction between open source development and releasing commercially developed code under an open source license is important as these are two modes that we see clearly in voting systems (see discussion of eVACs in §). Disclosed source code allows a much more limited use of source code usually for evaluation purposes only and without permissions to make further copies, modify works or distribute. For example, see VoteHere's license agreement: https://www.votehere.net/VoteHere_Source_Code_License_2.htm..

Efforts to broaden the number of individuals with access to the source code of election technology are part of a larger project of increasing the trustworthiness of electronic election systems. This larger project focuses on both technical improvements that increase security, accuracy, privacy, reliability, usability and reforms -- at some level independent of technical improvements -- that instill confidence in the voting public by facilitating public oversight and accountability. As such, calls for source code disclosure to the public or to a set of independent experts should be measured along a number of related but independent axes:

• What role could access to voting system source code play in increasing the transparency of voting systems?
• What are the risks and benefits of open source and disclosed source regimes for security and the market?
• If open source code offers measurable benefits, what regulatory and marketplace barriers exist to the development of open source software in the voting systems environment?
• What business methods from the landscape of open source software may translate to voting systems?
• Are there alternatives to public disclosure of code or open source code requirements that might yield similar benefits in technology performance and increased transparency, but minimize potential risks posed by source code disclosure?

This paper examines the potential role of source code disclosure and open source code requirements in promoting technical improvements and increasing transparency of voting systems. Section elaborates on the concept of the ``enclosure of transparency'' of voting technology. Section explores the level of source code access necessary to support effective evaluation and oversight of electronic voting systems -- an important component of transparency and tool for gauging system performance. Section reviews recent efforts to increase the capacity for public scrutiny of voting systems through disclosed and open source code requirements. Section examines the benefits and risks of open and disclosed source code regimes in the voting systems context and considers additional issues posed where access rules are driven by regulation rather than the market. Section considers regulatory and market barriers to disclosed or open source code voting systems and contemplates which existing open source business models sectors might translate to the voting systems context. Section reviews what transparency and trustworthy-promoting alternatives might exist outside of public disclosure of source code. We conclude that disclosure of full system source code to qualified individuals will promote technical improvements in voting systems while limiting some of the potential risks associated with full public disclosure. Acknowledging that this form of limited source code disclosure does not support general public scrutiny of source code, and therefore does not fully promote the transparency goals of public oversight and accountability, we note that in a public source code disclosure or open source code model most members of the public will be unable to engage in independent analysis of the source code and will need to rely on independent, hopefully trusted and trustworthy, experts. Given the potential risks posed by broad public disclosure of election system source code, we conclude that moving incrementally in this area is both a more realistic goal and the prudent course given that it will yield many benefits and greatly minimizes potential risks.