Electronic Voting As a Standard
Overview of benefits, concerns, and requirements
Introduction
Electronic voting would be the most practical choice, preferably from home and away from others, if it were an option for the 2020 presidential election in light of the ever-growing concerns over the current pandemic, COVID-19. In April 2020, Wisconsin was even forced to continue its elections after its Supreme Court blocked the Wisconsin governor’s executive order suspending in-person voting [1].
Desktops, laptops, smartphones, and even simpler electronic devices have the means to access a network and are capable of entering data into a survey-like format on a web page. Traditionally, early voting or vote-by-mail would be the prime options to avoid the hassle of waiting in long lines to vote, but an electronic voting system could enable higher voter turnout, cause less traffic by voters going to and from the ballots, and save time for administrators and voters, among other benefits. The use of electronic voting could also improve the speed at which voters around the country can see live updates because the data could be easily retrieved.
Concerns About Electronic Voting
Important to the implementation of electronic voting are the concerns and requirements that come from it as well. Because concerns should be largely influenced by the early requirements of implementation, the two should be talked about together. The primary concerns to electronic voting would be unreusability, privacy, fairness, unforgeability, and eligibility [2].
Terms Important to Elections
- Unreusability means that eligible voters cannot vote more than once.
- Privacy means that no one can determine who voted for whom.
- Fairness means that during the voting stage, intermediate voting results that may influence the entire election cannot be obtained by the voting center.
- Unforgeability means that only eligible voters can make authorized votes.
- Eligibility means that only eligible voters can participate in the election.
Requirements
Overarching requirements to electronic voting in relation to the above concerns would be verifying voters and their voting status, protecting the voter, voter’s data, and the voting database, and ensuring networks are prepared to handle large amounts of data. It is important to note that the database should be centralized in how it is logically managed, but it should be physically decentralized to ensure that data is not tampered.
Verifying the voters can come in many forms such as how governmental websites already verify voters or physically secured methods such as fingerprints, facial recognition, ID cards, and even mobile ID [3, 4, 5, 6]. While they have varying levels of security, they are nonetheless methods to verify voters.
Protecting voters and their votes should come from the application layer and utilize strong encryption methods. There should be redundant checks with the voter and within the database before finalizing their votes to ensure that they acknowledge who/what they are voting for and to ensure that the database is retaining the correct information. The voting database which houses the data should be managed and accessed only by authenticated users as well as being in physically secure locations. One cryptosystem has already verified the possibility of and efficacy of having a secure and transparent method of electronic voting [7].
The best way to prepare networks to handle the data would be stress tests using expected data throughput before elections even begin as well as performing small-scale and simulated tests for the databases as well. Stress tests should preferably utilize the small-scale and simulated data to ensure the most accurate representation of how the network will receive and handle the data. An example below explains how data a voter accesses the website to enter their votes and how that data gets sent to the database.
Example
The process begins with the voter entering the URL of the website. For example, the URL could be https://www.voteonline.gov. This opens a secure connection to the server. The website would then be retrieved from the server and begin to load on the voter’s device. At this point, the website would ask the voter for information to verify their voting eligibility and ensure unforgeability of the votes to be completed; the verification method could be any of the ones mentioned earlier in this paper. At this point, the voter is verified and is shown the options for which election they want to vote in. This assumes that it is during a year where there are multiple elections currently ongoing. Then, the user is shown the options for who/what they can vote for in the selected election. It is necessary to separate the election options because it is likely that the databases that are storing the information are in logically and physically different locations, especially if it is a state election versus federal election. The options should be in a simple survey-style format where voters simply click on who/what they want to vote for. Voters should be able to change their votes until they are prepared to submit their final decisions.
At this point, they should be clicking a button that says “finish” or something similar to begin the final processing of their votes. This process would review each of the questions and answers on the final page for the voters to review before submitting their votes. If a voter sees an error, they should click a button to go back and correct the mistakes. This review process should only be conducted one or two times at most for the average voter. After the voter submits their final votes, the final page the user should see is a vote completion status with only time and date to verify that the voter has completed their voting and that they should close the browser. There should be nothing related to their votes on this final screen for privacy reasons. At this point, the voter has nothing more to do and the total time elapsed should have only been a few minutes of their time. Should the voter attempt to vote in that same election again, the verification would ensure that there is no reusability in the voter’s attempt to do so. The website may show that the user has already voted in that particular election, and they may have another election available to vote in, at which point they can repeat the process.
After the user has finished the voting for the election, the data should be on the server and be duplicating onto redundant servers to ensure that the database is accurate. The only people with access to the database should be those who have jobs and tasks that require them to do so. The people who are most likely to need access are technicians and “ballot counters.” Ballot counters is a tentative term for this paper and simply reuses a term that is affiliated with those who check ballot votes. These people would be the ones who provide updates to news channels as well as verifying that the data appears correct. Having people unrelated to the specific election handle these tasks would ensure fairness in the vote. Technicians are those that maintain the servers, databases, and networks to ensure that everything is running smoothly without technical problems.
Final Thoughts
Overall, the concept of electronic voting is feasible, but it requires thorough planning and analysis to check how well it would function. This would likely begin with smaller tests as previously mentioned, and eventually scale up to national levels of use. An important detail not previously mentioned would be that this whole process would likely take several years, if not one or two decades, to fully refine the process because of how much time passes between elections, especially those on a national scale. If the current capacity to conduct the 2020 census online is any indicator, it could prove the efficacy of electronic voting as well.
References
[1] Amy Gardner, Elise Viebeck, and Dan Simmons, “Wisconsin Supreme Court blocks order by governor to stop Tuesday’s elections in state’s latest whipsaw,” The Washington Post, Apr. 6, 2020. [Online]. Available: https://www.washingtonpost.com/politics/wisconsin-governor-suspends-in-person-voting-in-tuesdays-elections-amid-escalating-coronavirus-fears/2020/04/06/9d658e2a-781c-11ea-b6ff-597f170df8f8_story.html. [Accessed 6 Apr. 2020].
[2] Sung-Hyun Yun and Sung-Jin Lee, “The network based electronic voting scheme suitable for large scale election,” The 6th International Conference on Advanced Communication Technology, 2004., Phoenix Park, Korea, 2004, pp. 218–222.
[3] K. Hasta, A. Date, A. Shrivastava, P. Jhade and S. N. Shelke, “Fingerprint Based Secured Voting,” 2019 International Conference on Advances in Computing, Communication and Control (ICAC3), Mumbai, India, 2019, pp. 1–6.
[4] Shahram Najam Syed, Aamir Zeb Shaikh, and Shabbar Naqvi. “A Novel Hybrid Biometric Electronic Voting System: Integrating Finger Print and Face Recognition,” Mehran University Research Journal of Engineering and Technology, Mehran University Research Journal of Engineering and Technology, 2018, 37 (1), pp. 59–68.
[5] C. Yang, S. Tu and P. Yen, “Implementation of an Electronic Voting System with Contactless IC Cards for Small-Scale Voting,” 2009 Fifth International Conference on Information Assurance and Security, Xi’an, 2009, pp. 122–125.
[6] S. Yun and H. Lim, “The Biometric based Mobile ID and Its Application to Electronic Voting,” KSII Transactions on Internet and Information Systems, 2013, vol. 7, no. 1, pp. 166–183.
[7] E. V. Palekha et al., “Cross-Platforming Web-Application of Electronic On-line Voting System on the Elections of Any Level,” 2019 IEEE East-West Design & Test Symposium (EWDTS), Batumi, Georgia, 2019, pp. 1–4.
