Drenthe looks at the practical and political challenges
Adri Wischmann – Blockchain Lab Drenthe, Netherlands Niels Annema – Province of Drenthe, Netherlands
In some way our society ‘votes’ nearly every day. But these polls differ in importance, security level and auditing. Conventional voting using paper ballots is slow, inefficient, costly and does not treat everyone equally. E-voting could potentially deliver dramatic change, but there are big challenges (and a lot of scepticism) to overcome before it can be used in national elections.
To better understand the issues that were holding back wider acceptance of e-voting, the Province of Drenthe commissioned BlockchainLab Drenthe to review the technological issues and challenges that might be holding
back’ e-voting in the context of a potential e-voting pilot that used some form of blockchain solution. This analysis of the potential value of blockchain-enabled voting system complements the work done by BLING’s Swedish partner Länsstyrelsen Skåne (the County Administrative Board of Skåne) who searched Europe for an open source, GDPR compliant e-voting solution (see ‘Can Democracy become Digital?’.
There have been big changes in how we vote
A big part of our lives is spent online. We make friends online, we communicate online, we pay our bills and taxes online and during the recent pandemic we worked online. But voting still requires a pencil, a piece of paper, traveling to a crowded building, and standing in line. Granted, we have come a long way since the times where ballots were printed in the newspapers, and voting was in public – you had to say the name of your election candidate in public where everybody could hear you – and votes could be bought. But taking the next step to bring voting to the digital realm seems to be a giant leap with many benefits and many potential problems.
Social acceptance of new ways of voting
While most of the technical and organisational challenges to e-voting can eventually be addressed, we will probably not see national elections using blockchain solutions in the next decade.
It is however very important to try and push the boundaries of implementing blockchain in e-voting to local and regional levels if we are to gain traction for the idea of e-voting, and to create the broader social acceptance that would be required to support the transition to national e-voting on some future election day.
Although a 100% perfect e-voting solution might prove to be a utopian ideal, our local goals should be to improve and innovate the voting system to get to a solution that is better, faster, more efficient, cheaper, more accessible and more inclusive than the current paper based or electronic voting systems.
Principles of successful e-voting
BlockchainLab Drenthe identified a common set of principles for e-voting:
- E-voting elections must be verifiable – from end-to-end
- Ideally e-voting elections have a low operational footprint
- E-voting elections should be transparent and ‘explainable’
- E-voting elections should increase voter engagement
- E-voting elections might be more frequently held
- E-voting should be secure (and that is hard) • The goal should be to get to a better voting system, not to an absolutely infallible one
E-voting is more complicated than you might think
Our pilot created a report that thoroughly reviewed the current possibilities for e-voting, the processes and mechanisms that needed to be included, and identified possible problems and solutions. As this is a forward-looking paper, there were a number of people in our organisation who were interested in our results. However we do not expect the Province to be ready to shift towards new e-voting systems, as these systems remain difficult to explain and keeping the trust of voters is a big factor behind the successful delivery of any election. If it becomes too difficult to understand the blockchain technology underpinning e-voting, then there is a real risk that voters’ trust will be lost. So, for the near future e-voting will remain an interesting idea.
Looking at platforms
There are literally hundreds of blockchain platforms and protocols which could be used to build an e-Voting system – and each of these has pros and cons. For practical reasons we narrowed this down to look at 2 protocols in greater depth: Ethereum (the most popular and widespread blockchain protocol and the first one to implement smart contracts) and IOTA (a 3rd generation protocol working with a DAG – Directional Acyclic Graph – being one of the most energy efficient, fee-less, miner-less blockchains in the industry).
Ethereum quickly proved to be too inefficient and congested, after which the BlockchainLab explored the IOTA options. BlockchainLab has previously developed solutions on IOTA, and our experience is that IOTA is fast, free and eco-friendly.
Many challenges lie ahead
Our work identified a series of practical challenges that would need to be overcome if we were to seriously consider blockchain-enabled e-voting systems:
- the (commercially) available solutions are not open source
- the available solutions are not truly End-to-End Verifiable Internet Voting systems (E2E-VIV)
- the risk of DDOS-attacks on voting systems would need to be addressed
- we would need to be sure that there was no prospect of undetectable manipulation of thE system (leading to betrayed trust of voters) There were particular challenges facing blockchain systems that we identified, included the vulnerability to ‘51% attacks’ in blockchains that are secured by proof-of-work algorithms, the need to bring in adequate privacy controls, the economic and ecological transaction costs, the challenges of scaling blockchain systems, and the need for high transaction speeds.
The lessons learned from our review have been published and disseminated across the organisation and to other governmental bodies in the Netherlands. If Drenthe decides to move towards e-voting, then we will do this with a much more robust understanding of the advantages and challenges of blockchain-enabled approaches.