Electis is currently pushing a next-generation decentralized e-voting system together with members of the Tezos community, academics and students from currently eight different countries (and growing). Reach out to lena@electis.io if you’d like to get involved!

Even though there is a lot of research being done by many institutions, and more and more e-voting solutions are being used around the globe, e-voting appears to be a highly divisive topic — especially regarding high stake elections.

Western democracies are working with a voting system that has been improved and tested for hundreds of years and is considered trustworthy by most citizens.

If it’s not broken, don’t fix it!

This leaves us with the question this blog post is asking: why should we bother to replace a seemingly working system with a new, potentially riskier one?

Democratic principles of elections

First, let’s go to the basics of democratic election principles. There are some widely understood qualities an election requires to be considered democratic and valid.

“The will of the people shall be the basis of the authority of government; this will shall be expressed in periodic and genuine elections which shall be by universal and equal suffrage and shall be held by secret vote or by equivalent free voting procedures.” (United Nations, Universal Declaration of Human Rights 1984, Article 21, Section 3)

Any kind of voting, therefore, should solve a set of problems to secure a universal, equal, free, secret and direct suffrage election.

An election is only genuine if the ballot represents the voter’s choices and the voter is able to verify its vote was cast as intended. Also, it must be ensured there is no discrimination amongst voters. Every person only has one vote, but also, nobody is capable of deleting, changing or adding votes. This is meant by an equal, universal and direct suffrage.

Everyone should be voting after their own conscious and free will and therefore an election has to be secret. Nobody should be able to tell who a person has voted for after the election. This ensures that votes can’t be bought or forced (coercion resistance).

Finally, there are other goals that have to be factored in. For example, the goal of a high voter turnout and equal accessibility for all citizens to voting facilities (e.g. for disabled people).

Nonetheless, it appears that there are certain tensions between different requirements an election has to fulfil.

Efficiency vs. security:

Electronic solutions can help reduce the high organizational costs of paper ballot voting. A digitalized system could also save time and labour that is now done by mostly volunteer citizens, making election results ready and tallied right after the closing of the polls. Also, the high number of invalid votes that are caused by errors made in the voting booths could be reduced significantly — a machine could lead a voter through the process until he or she got it right.

Electronic voting technologies are tackling most of the human-factor risks that are common in paper ballot voting. Yet, on the other hand, it can take a hacker the same amount of effort to change one vote than it takes him to change a thousand votes (“wholesale fraud”). Compared to this, physical voting is not perfect and can be attacked, but simply doesn’t scale in the same way.

Authentication vs. a free and secret vote:

Next to the security risks, the biggest problem yet to be solved is the dilemma of the goals regarding authentication and secrecy of the vote. On the one hand, identification and authentication are important to make sure only eligible voters are taking part in an election and everyone only votes once. On the other hand, elections are supposed to be anonymous and secret.

Coercion resistance vs. accessibility:

Remote participation, like postal voting or e-voting per app, always entails the threat of coercion. Private polling booths are the only way to guarantee against fraud and forcing of votes. Considering the fact that postal voting is an option for voters in many nations already and the number of citizens choosing remote participation shows that this is not only an e-voting problem.

On the other hand, remote voting can solve the issue of accessibility. It is key to an equal and universal election, that every eligible voter has access to a polling station. Traveling citizens, people living abroad, voters who are not able to leave their house or have disabilities need to have access to voting.

In a more and more multicultural society language becomes a greater issue, as not all eligible voters speak the local language. On an electronic voting device, it would be much easier to include different languages than you can print on a piece of paper — there can even be solutions such as voice interfaces for blind citizens.

How Blockchain can help

The security problem of a centralized e-voting system lies in its nature as it is only one entity that controls the code, database and voting hardware. This is why a hack of the system risks the kind of wholesale fraud mentioned above.

Decentralized Ledger Technology, or Blockchain, is based on a peer-to-peer network where data is distributed between the members of the network and without a central authority. The data is stored in timestamped blocks of information.

Once put on the chain, it cannot be changed easily if the network is big enough.This makes a voting record, once put on the blockchain, more censorship-resistant and could be a solution to the security problem of electronic voting and solve the transparency issues many centralized e-voting systems have. Alternatively, it could also open the doors to new unforeseen issues.

Efficiency vs. security: Casted votes are linked to a public key of the voter on the blockchain, allowing each public key to cast only one vote. Since every block is connected to the previous one with an encrypted hash, no stored vote can be altered afterwards. Even if malicious action is discovered, the original data can be recovered and verified, as data cannot be deleted. The transparent nature of the networks ensures that everyone can confirm that the number of casted and counted votes are the same.

Authentication vs. a free and secret vote: Every user of the network is identified by a public key, which can only be accessed by a private key that is (hopefully) only in the hands of the voter. This is crucial to guarantee authenticity. The stored vote is encrypted on the blockchain to keep the vote secret. On the other side, the transparent structure of the blockchain, in which every node has the same information, ensures that the election results can be audited and certified.

Coercion resistance vs. Accessibility: As the only requirement to access the blockchain is an internet connection and an address in the blockchain network, votes could be cast from home or from abroad, which makes this technology highly accessible. Yet only a blockchain-based voting device in a voting booth can truly ensure coercion resistance.

The limit of Blockchain

Decentralized voting systems are not ready yet to support democracy, some even argue it would be a threat to it (“Blockchain-based elections would be a disaster for democracy” and “why blockchain could be a threat to democracy”) and centralized solutions like what the Estonian government is implementing might be the best solution for now. Being aware of these limitations, the best way forward to mitigate them is to learn by doing within a real environment such as a network of universities.

Why we are building the university voting experiment on Tezos

The Tezos blockchain is unique as its governance process is currently powered by a democratic set of rules and therefore is well-suited to support democratic projects on top of it.

The Tezos blockchain uses a democratic set of rules which is especially compelling when creating an e-voting solution with the aim to improve democratic processes. In contrast to other blockchains where changes in the rules lead to a “hard-fork” and the blockchain splits into two separate chains, Tezos is a blockchain that can evolve by self-amending its own code. Tezos’ current governance process is designed to solve “fork-based governance” that defines governance of other blockchains like Ethereum.

This self-amending cryptographic ledger is realized through an on-chain voting system. Potential amendments to Tezos have to go through a four-step process over a period of three months:

1. Proposal Period: An upgrade is being proposed. It needs a minimum of 5 % of support to go on to stage 2
2. Exploration Period: The first round of votes on the highest-ranked proposal. If it gets a supermajority (minimum of 80%) of yays, it advances to testing
3. Testing Period: A test chain is forked out and is being tested for 48 hours
4. Promotion Period: the network decides on the adaption of the amendment, a supermajority is needed to approve the final upgrade

A precondition to being eligible to vote on Tezos is to be a validator (“baker”) and for that, amongst others you need to have a minimum of 8000 tokens (1 roll) as well as a machine powerful enough to run the server. It is important to note that the governance process outlined above is just an initial mechanism meant to bootstrap more effective forms of blockchain governance. The mechanism outlined above can be changed through the process itself.

Another issue with blockchains, in general, is the high amount of computing power transactions need. The so-called “proof-of-work” system used in blockchains like Bitcoin and Ethereum is very energy-intensive. Tezos uses a “proof-of-stake” consensus algorithm, which is greener as it requires less computing power. For more information on proof of stake in Tezos, see this post by Jacob Arluck.

These mechanisms of governance and upgradeability give Tezos greater platform longevity than you see at other Blockchains (Read more about this on this post by Charlie Wiser).

Conclusion

Looking at existing solutions today, e-voting seems to be an obvious improvement in places where election fraud is common and trust in the authorities is small.

There is a long way to go before e-voting on a national level is a safe alternative to paper ballot voting. This doesn’t mean it cannot already be useful to bring more democratic structures to smaller levels of society such as regional elections, company internal votes or alternative ways to vote for citizens living abroad or those who can’t leave their homes.

Other interesting democratic novelties electronic voting solutions can offer are the possibilities of innovative voting systems that can express detailed preferences of a voter rather than just their choice between candidates, like “quadratic voting”. Here, voters express the degree of their preferences by being able to “pay” for additional votes. Each voter has a budget of vote credits they can use to spend on different matters. If they want to buy additional votes on a single decision, those votes become increasingly more expensive (x²). Another idea is “Liquid democracy” where voters can delegate their votes to other voters or participate directly.

Giving up on the idea of e-voting should not be the solution to the challenges e-voting is entailing, and blockchain technology could be a part of the puzzle when searching for trustworthy technology.

Circling back to where we started — our current voting system is not not-broken. It needs improvement (a software update, if you will) as it doesn’t give the people of the twenty-first century all the participation opportunities that could be realized in the technological age.

At the end of the day we can’t just play with our democracies, but we should develop, experiment and learn what the future of democracies will look like with centralized and decentralized e-voting systems. As the famous architect, Buckminster Fuller once said: “You never change things by fighting the existing reality. To change something, build a new model that makes the existing model obsolete. “