In the Free TON contest, a participant challenged the theses of a group of US university professors who criticized the use of blockchain technology for voting. In addition to theoretical reasoning, the contestants formulated a practical proposal by developing various models of effective blockchain voting.
Nowadays, along with the promotion of the digital economy, the concept of e-government is a significant direction of social and technological development. Regardless of geographic location, wealth, or political system, countries around the world are incorporating digital aspects into their governments’ work, as this helps to improve the efficiency of their functions.
Many elements have already been successfully implemented; others are still under development and testing. An example of the latter is the methods of electronic voting, including the use of the Internet in voting procedures in state elections at various levels (local, parliamentary, presidential).
Electronic Voting Bonuses
The desired effect of e-voting is to simplify the election organization and vote counting, quick result, increase access and participation of voters through the option of remote voting. In addition, it is expected that after the final implementation of the electronic system, there will be a gradual reduction in the financial costs of the elections.
But there is more. The expectations are not so much of governments as of civil society. It is about ensuring the transparency of elections and the reliability of the results. Can electronic procedures help protect citizens’ choices from falsification and discrediting? In recent years, socio-political upheavals associated with the problem of state elections (Russia, Belarus, USA, Myanmar, etc.) indicate that public demand for effective electronic technologies will increase and thereby stimulate the development of this area.
Blockchain as a distributed data registry, protected from changes and high throughput is a promising tool and is already used in a number of voting systems, for example, Votem, Voatz, Horizon State, etc.
However, large-scale testing and implementation is not yet taking place. This is due to the doubts of some experts in the security of blockchain voting, which brings to the agenda the issue of the vulnerability of the new method to disruptions and cyber attacks. This is undoubtedly a serious problem, although much depends on how you perceive such an obstacle: as a dead end or as a challenge.
The Free TON community has already demonstrated the ability to meet the challenges of electronic voting. At the end of 2020, a competition was held in which they chose the best solution that provides an audit of votes in the election procedure in Latin America using the Free TON blockchain.
The next step in this direction is a contest held in February 2021 to find a theoretical and practical solution to the blockchain voting vulnerabilities identified by MIT researchers in their joint article «From Bad to Worse: From Internet Voting to Blockchain Voting.» Contestants were expected to write an analytical essay in accordance with a number of requirements.
First, to refute the theses of the article that blockchain technology not only doesn’t solve security problems, but also creates the potential for new threats, and therefore is more vulnerable than a paper system.
Second, propose a blockchain based voting solution better than existing paper based and demonstrate how it will lead to increasing voter turnout, reducing fraud, or combating disenfranchisement and coercion.
There were 49 applications, 6 of them were rejected as not satisfying the contest requirements, the rest were evaluated on a ten-point scale.
The work that took first place received 7.33 points and the rest of the submissions included in the top twenty with awards received a score of 5.44 and below.
Submission # 41 is a good example of an analytical essay with a clear structure, a consistent description of the advantages and disadvantages of different types of voting systems, as well as a reasoned positioning of the blockchain possibilities.
The author’s approach avoids naive enthusiasm and demonstrates a desire for a realistic assessment of the transition difficulties to electronic voting.
We agree that society does not have sufficient experience to trust digital electoral systems. Recently, however, there have been successful examples of digital elections, and society is beginning to gain the necessary trust in this technology.
The work describes the blockchain as a tool for solving many fundamental problems of the voting system, both independently and in combination with other solutions. There is an emphasis not so much on ease of use and cost reduction as on protecting voter choice, on the ability to resist and detect manipulation, thereby gradually changing public opinion about digital voting. Therefore, the work presents the effectiveness of blockchain technology as a system for monitoring results and a mechanism for increasing trust due to decentralized control by society.
Hybrid Voting Model
Since the paper system, despite all its shortcomings, is familiar and well-established, the authors propose to start a gradual transition to new methods. In their opinion, at this stage, a hybrid technology based on the classic «paper» system with elements of digital technologies may be a useful solution.
In the proposed model, voting takes place in a single private blockchain system built on a Proof-of-Stake (PoS) consensus. Only authorized polling stations have access to it, validators receive a special status, and access to them is also limited to voting points.
Voters are physically present and vote at the polling station in accordance with the usual procedure, and the result is recorded in two forms: paper and electronic on the blockchain. The ballot box is an automatic scanner that identifies the key, records the voter’s choice and sends the data to the blockchain.
In case of detection of the facts of interference, violation or damage, the data can be restored from the bulletins. It is possible to fully examine the entire voting process in order to confirm the purity of the procedure at any stage.
The author’s model allows testing the electronic system in real conditions, examining its capabilities and resistance to attacks, but at the same time does not jeopardize the electoral procedure, since in case of failure, the obtained copies in the form of physical media (paper ballots) can be used to establish the result.
The hybrid voting model is based on a realistic assessment of the level of trust in digital technologies and a willingness to experiment with electoral procedures. But since the information society is characterized by an avalanche-like type of development, then in the field of voting, skipping intermediate stages, a quick transition to blockchain technologies is quite possible.
In submission # 45 it is proposed to go beyond the theses proposed in the article criticizing blockchain voting, and thus expand the field for analysis and comparison of different voting systems.
..in fact, the entire global thesis of the document is limited to the minimum characterization of the blockchain voting in accordance with only 5 metrics. But this is extremely insufficient. Why not include more comparative features where blockchain voting outperforms existing systems (it really does)?
Therefore, in the theoretical part of the work, the author not only answers the main theses, but also uses a broader perspective to study the object. In particular, there are reflections on how social concepts such as monetary relations or the electoral system are translated into a technical format. Since at the moment there is still no well-established theory and practice of blockchain-based voting systems, Bitcoin analysis is used as an example of such a transition.
The author notes that any encoded system can perfectly work with internal facts. These facts never leave the network, just like Bitcoin coins don’t disappear from the monetary network and in this sense, money is safe. But management is insecure because access to the wallet could be compromised at some other level. Such a system is exposed to threats from the outside, for example, due to the constant threat of hardware exploits. However, the blockchain system at least does an excellent job of handling the internal facts that it generates and processes on its own. And traditional voting systems cannot even provide this.
Cryptographic Mechanism ZK-STARK
Today, privacy in blockchain systems is ensured through various data encryption methods. One of them is «zero knowledge proof», on which the ZK-SNARK, ZK-STARK and others protocols are based. Free TON community developers are also working on implementing similar cryptographic mechanisms.
As for the blockchain voting model described by the author, it should ensure the anonymity of the voting procedure, while maintaining the secrecy of the data entered by the voters. For this, it is proposed to develop a system based on the ZK-STARK. The author justifies this choice, referring to the data on high reliability, transparency and lower cost of the ZK-STARK technology implementation.
The author believes that the proposed concept can be implemented in almost any system that includes the use of smart contracts and the generation of ZK-STARK proofs. An example of such a system is the voting system based on the Free TON blockchain.
The voting procedure includes the following stages:
- Stage 0. Deployment of the network
- Stage 1. Voter registration
- Stage 2. Obtaining creds
- Stage 3 (1). Voting interface launch
- Stage 3 (2). Understanding results
- Step 4. Disclosure of the election results. Voting audit
The author comes up with an interesting idea, but only describes its use in general terms. Given the fact that the ZK-STARK is a relatively new development, it is not yet known how realistic the practical implementation of such a system is in the near future.
Consider work # 6. It highlights positive examples of using online voting (Russia, European countries). In the author’s opinion, the advantages of the method are evidenced at least by the fact that the number of technical failures is not as great as the number of “invalid ballots”. At the same time, technical flaws are not an unsolvable problem and can be corrected with further testing of the system.
The author argues that the blockchain methodology in voting systems is able to ensure the protection of the transmission of votes to the database and access from the device of each voter in the system. A software implementation using the citizen’s cryptographic code is possible.
..the blockchain also changes the performance criteria of the election system and voting system, allows you to self-tune on the profile of the voter. This will make it possible to cope with the task of preserving the secrecy of voting and confidentiality at a sufficient level.
In addition, the idea is expressed that in order to increase voter turnout, reduce fraud and combat violations of electoral rights, neural networks should be used that learn from examples of such violations.
Using Free TON and SURF Browser
Improving voting in the blockchain, the author proposes to implement the «binding» of the ballot to the voter according to the scheme: «token (voice) — transaction (submission) — confirmation (comparison with the voter’s profile) — placement in the chain (voice accounting)».
At all stages of interaction with the Free TON smart contract (debot with a voting interface), the confidentiality of voting is guaranteed by zero-knowledge protocols. The ZK-SNARK mechanism is supposed to be used. There is also an idea of checking the hashes of a smart contract at all possible stages to prevent modification of the executable code.
When registering and confirming the right to vote (one-time actual visit to the polling station or entry through a government service), the voter is only provided with a tool to activate the public key that is generated locally. ZK-proof technology allows you to do this anonymously.
Voters get the opportunity to enter their data and confirm their right to vote, but the transparency of the votes cast is not available until the end of the voting. ZK-SNARK allows you to work only with a verifiable anonymized hash of your right to the ballot. You can check if your vote passed or not. But unlike other auditors, the voter will know that it is his or her vote. Others can see confirmation only that the vote has been received from a verified voter. The smart contract will always find and cut off duplicate votes.
The author believes that the use of one SURF browser and verified debot with an open source smart contract will increase trust in the network and increase the turnout rate and fair elections.
The described model represents a more accessible implementation using Free TON. However, this is a high-level scheme that requires further detailing to develop specific solutions for implementation.
Doomed to Progress
Contest requirements set the bar high for the participants. The request for a realistic, efficient and implementation-ready model motivated the contestants to try to surpass existing approaches and come closer to a breakthrough in this area. The ideas of the participants do not yet provide a ready-made solution, but the competitive process indicates how it can be obtained.
An appropriate product is created in an environment with strong motivation, competition and practical implementation. If you don’t test blockchain voting in real conditions, then the rates will not rise to such an extent as to spur the attack of cybercriminals, as well as stimulate developers to fight back and move the technology even further.
Yes, government agencies are afraid to take risks in matters of national elections, so experts who make reports on the current capabilities of blockchain technology at their request are careful in their conclusions. However, there is not much time left for this wariness. The world is changing pretty quickly. Even if the digitalization of political procedures suddenly slows down, the digital economy will push or the general social context will force it to move. The transition to any types of electronic voting will happen in any case. At least due to the fact that at some point the traditional method will turn out to be foreign for an automated society.
So while civil society forms the request, and political institutions think, the Free TON community is already experimenting and implementing.