Apart from my visit to UBC’s CIRS, I also participated in an interactive dialogue on “Online voting” at the HiVE Vancouver last weekend. In this first session of the “E-volving Democracy” dialogue series, people with different backgrounds from computer scientists to political activists made a clear case against online voting.
The session started with a short introduction by a member of Fair Voting BC as well as presentations by Andrew MacLeod (The Tyee), Steve Wolfman (Computer Science, SFU) and Fathima Cadre (UBC Law). These introductory remarks made clear that while there may be potential benefits of online voting with regards to convenience, a higher voter turnout or lower costs, there are considerable security issues and other important obstacles.
In fact, online voting experiences in Ontario, Estonia or Switzerland showed that it only had a small effect on voter turnout (12-25% of online voters in Switzerland usually do not vote or only vote occasionally). So, while the impact of online voting on the overall result is considered small, there might be a higher impact in a majority political system like in Canada or the United States due to the impact of even a small amount of votes on swing seats.
According to computer scientist Steve Wolfman, it is impossible to build a perfectly secure system. In fact, the hard part is client-side security as end users’ computers and mobile devices are not appropriate platforms for online voting due to malware and other security threats. Moreover, there are also security issues with regards to the internet in general (DoS attacks) and encryption. So, it is no surprise, that even online voting software developer Helios recommends not to use its Voting Booth software for national or federal elections.
With regards to legal issues, Fathima Cadre talked about the complicated security and privacy implications of online voting. In fact, current laws in BC would have to be amended to enable online voting and protect privacy. Moreover, the public not only has to know how to vote but also what to do when security breaches become apparent.
In addition, extensive pre-deployment testing would be necessary and it should be clarified what happens when elections organizations outsource tasks to other companies like online voting software developers. In her opinion, the BC government currently does not have the regulatory power and technological knowledge for implementing online voting. Finally, journalist Andrew MacLeod also made clear that security breaches and hacks would probably be just what the media is waiting for in the discussion on online voting.
In small group discussions, participants later identified conditions they believed a proposed online voting system would have to satisfy before it could be used in good conscience in a public election. Despite the widespread doubts against online voting, it became clear that anonymity, transparency, security, accuracy and independent auditability were key values that an online voting system would need to have.
Moreover, open design including the release of an open source code that can be tested extensively was highlighted as a condition for public acceptance of online voting. Nevertheless, it also became clear that security and encryption issues as well as a computer’s intrinsic motivation to copy data are actually working against the stated values and conditions.
Finally, while there is widespread scepticism against online voting, it should also be stated that computers and the internet create vast opportunities for citizen participation, political transparency, government control/feedback and community development apart from the regular elections every four or five years. I hope that the next two sessions in the “E-volving Democracy” series on “Citizen initiatives” and the “Enbridge Pipeline” will further explore this important issue.