Digital Voting
This is the long version and back ground to the summarised plan proposed in “Electronic Voting – the 3 X 3 X 3 Program



– now for the full story…

Democracy is broken!
No I don’t mean that in a anarchist or communist revolutionary way. I am after all neither a great political thinker nor particularly and active campaigner on any particular front. Someone who knows the subject in the UK however agrees, Will Moy, Chief Executive of Full Fact, the UKs leading fact checking service, says ‘Our election laws are decades out of date’ and this can probably be said of most countries.
Others in the know have noted, “(UK)General Election 2019: ‘Millions’ of votes wasted, say Electoral Reform Society.”
And yet others who spend all day looking into it, say of Oxford University on the London School of Economics blog, so not nobody, has advocated a move to the Alternative Vote method, which adds granularity even to the ultimate decision of the national election.
But not just my birth country, the Heritage Foundation has listed 1,241 cases of proven voter fraud in the USA. Even in 1861 John Stuart Mills remarked that democracies are plagued by the risk of distortion of the voting by misinformation and by ruling elites twisting the process in their favour. Other examples has been seen in the return of extremist politics in many European countries, such as Salvini in Italy.
I don’t intend however to look at this from a political perspective. Living for the last 20 + years basically doing digital transformation, my credo has been “where there is paper and stuff being stamped by hand there should be digital automation”. So it is that with that slightly technocratic cast that we approach the problem of democracy.
Let’s first set the scope, in true consulting fashion. When we’re talking about democracy, we are not restricting the definition to national elections or regional elections or national referenda but also include all kinds of day to day voting or even expressions of interest, or expressions of wish that make up the day-to-day operation of our lives. This definition will become important later when we look at a road map for implementing a new and more efficient and effective democracy. The definition of effective democracy is also for debate.
Here we are talking about the most granular possible expression of will by people and organisations within their relevant scope, be that national regional, city neighborhood, social association or social grouping. Even as small grained as a group for blokes that like Fiat Pandas.
In fact, whatever granularity we wish, as this is one of the real strengths of the new technologies (yes there is an S on the end of that, it’s not only about blockain). Even more importantly, the new ways of delivering technologies, that are the real shock to the system to the current proponents of modern technologies (any platform delivered in the last 40 years or so if we take Unix epoch of 1/1/1970 as being a milestone in ‘modern’ computing).
But let’s get back to the brokenness of the state of affairs. It should be obvious to anyone alive, apart from perhaps some dormant nuclei on Mars, that democracy has not caught up with technology. Indeed it is lagging very far behind as proved by episodes such as the Cambridge analytica scandal, which in turn then wrapped up the UK in Brexit and the US was landed with President Trump. There are many other similar episodes such as the Salvini effect in Italy.
“Aha! But here he is taking sides” I hear you think. But no. The examples are given purely as as a clear indication that technology, in this case social media, has been used to influence the democratic process in a way that no regulatory or individual could imagine or control under current regulations.
In fact, the examples go far further back than that. If I told you that a certain large technology conglomerate was allegedly instrumental in facilitating the Nazi Holocaust, you might think of me going off topic hysterically. But no again, just as today, technology was used to, in this case, circumvent democracy and and use democracy to enable a particular and very hateful agenda which ended up with many millions being massacred. The enablement in this case was the use of pre digital computing technology in the form of card counting machines to empower the 1934 census. This was particularly insidious as it introduced the obligation to declare race and other personal parameters which therefore created a very nice little list for the Nazis to use when they got around to the round up of Jews, Roma and other people’s they had decided to exterminate. Is this the proper place to be bringing up this topic? Again I am not campaigning for any particular grievance, but stating, from my historical and technical understanding, that democracy was circumvented through technology.
And by the way, I am a student of history much more so than modern politics. In the same historical vein, but much more recently, we can recall the Chad’s in Florida where the paper voting cards were found to have had a flaw which made votes illegible and potentially therefore open to fraud.
“But there are lots of things wrong with democracy” I hear you cry, as you throw your toys out of the pram. “How can you possibly fix them all with one stupid technology?!” Well firstly, as I was saying earlier, I’m no Karl Marx so I’m not intending to propose a roadmap for wholesale social change, I will though boil down a lot of problems to one thing:
the granularity and reliability of voting and the expression of wishes
While being very aware that democracy is not a mushroom risotto, that you can easily boil down into a tasty dish, there is a rationale to look at the voting capabilities made available by new technologies. These can serve as a cornerstone of how not only the top level of democratic government is refreshed every few years, but how life is operated on a daily basis at all levels, down to intra personal interaction.
So what’s another example of voting that does not necessarily mean sending self opinionated chancers from one monolithic tribe…err party…to the capital city to enjoy hospitality for 4 or 5 years? There are plenty but let’s look at a concrete example.
Btw, concrete was invented by the ancient Romans and some of their examples are still in use today after centuries, where many modern counterparts have not survived decades.
Asides aside, a few years ago I was working on happiness! No not with a personal psychiatrist or marriage counselor. The city of Dubai had identified happiness as a target and we were tasked with looking into the science of it. The fantastic consultant with us told us about an experiment on happiness in the tax services in the United States. Now taxes are never something that make people happy (as in “death and taxes are the only certainties”). However, just by giving taxpayers the opportunity to express their wishes on how tax revenues should be spent, tax payments went up and were completed with greater alacrity by a few percent.
Firstly, Innovation can come in small parcels, so a few percent is a huge win. Secondly, just the ability to express their opinion, with the clear disclaimer that no commitment was being made, was enough to make people pay more tax!!! How huge is that!
So voting can be at a constitutional level, informal expression of wish and anything in between, the governance of a city or regional association for example. Unt Dan? (German for “so what?”).
Well, SO, we now have a set of mature technologies and technology implementation methodologies that can empower trustful granular voting programs. Say what? To translate that consulting technobable, we can do stuff faster, safer, cheaper and therefore, on more voting cycles that every four years.
In fact we can do voting every day.
This without breaking the bank, circle of trust nor creating a fast track for corruption of democracy. And it won’t take ages neither.
Oh and btw, going back to history, we can also fix revisionism (the only thing that would get me in the streets with a banner).
Right so how? (This is where you dear Reader, juts out your chin and sets face in a challeging manner).
The how goes to the heart of several technologies and practices that have come to maturity and mass adoption in the last ten years or so. Namely:
blockchain (obviously)open source softwareagile developmentmobile encryption and biometric capable mobile devices i.e standard mobile devices now have the power to run encrypted applications, save encrypted data and have bio metric readerscloud computing infrastructuresoftware as a servicecontinued standardisation of security on web browsersrelative ease of implementation of machine learning algorithms e.g. on cloud servicesstrong advancements in computer vision image recognition
So the corner stone is clearly blockchain, given that it is a decentralised ledger that all parties can trust. Mad? Well, US Democratic candidate Andrew Yang says he will implement blockchain-based mobile voting if he wins the 2020 United States presidential election. So other people are seeing the match.
And speaking to granularity, the Japanese City Tsukuba Trials Blockchain-Based Voting System. I myself spoke about the resilience aspects of blockchain voting in a 2017 article.
Taken together, these capabilities enable governmental and other institutions to implement voting at all levels in the same disruptive way as many other industries have done with them. Examples include the uch as the disruptions that have happened in travel, local transport, banking, supply chain and retail. As the saying goes, in all of these and in many other areas of business, “there’s an app for that”.
This combination of technology and new processes also has the potential to avoid the historic failures in IT projects at the government level that have plagued most countries. The illustrious consultancy McKinsey estimate that it is not uncommon for large IT projects run 45 per cent over budget. Recent voting app farce in the US Democratic primaries are proof of this. Our new agile world of technology reduces this risk by biting off smaller chuncks and reducing the absolute budget.
What this adds up to is that institutions no longer need to fear pharaonic costs and timelines to implement voting capabilities and, most importantly, because of that, can go step by step in rolling out voting.
Institutions can go step by step in rolling out voting. (It’s worth repeating).
The whole national elections or referenda chunck does NOT need to be swallowed in one go.
Neither however do we have to be in pilot mode forever. We can now realistically deploy micro solutions that build-up in to tackling the bigger questions.
In our case, for voting, this translates into the ability to do dozens or hundreds of votes per year based on different levels of importance. So for example, we can start with an expression of wish such as a local council asking it’s community their thoughts on building a new swimming pool. In expression of wish, no commitments have been made and the people being polled are informed of that fact. However, as in our previous example on US taxes, the fact of being asked their opinion is already a positive, as long as there is follow through in terms of explaining a road map or explaining the results of the poll or executing on what was requested right away.
Expression of wish polling in an ‘official’ context is now strengthened by the new technologies of blockchain and a clear explanation of the trust benefits of it.
Additionally to blockchain, technologies around identity will be required to ensure uniqueness of working and avoid fraud. For this, several institutions are setting standards and implementing Self Sovereign Identity.
Self sovereign identity is based on the simple concept that control over an electronic identity and how that identity is presented is handed back to the citizen or the organisation to choose how when and with what details they present their identity. Using self sovereign identity we can establish modern techniques for establishing an individual’s identity. In financial services terms know your customer or KYC, which is a strong set of standards. These techniques ensure that an individual is, firstly that individual and secondly by making a one-way obfuscation of the electronic identity presented by the citizen we can store an audit trail that that’s identity only voted once without actually storing anything about them. Further as the self sovereign identity used for voting can be a self declared identity for expressing wishes or would be the nation state’s issued electronic identity for more formal voting, the audit trail can be either simply that one person did one vote or can go right back to to the citizen’s government and their systems of record
“But privacy” I hear you cry. As the old English saying goes, “my vote is between myself and my god”! Yet here there is no conflict as we are, in most cases, only recording the fact that a vote occurred once only for a given identity can be completely anonymous. The fact of who made that vote and that they only voted once (no multiple electronic identities) and that they had the right to vote is then dependent on the level of sensitivity of the voting type. For example, rising in importance in our scale of granularity from expression of wish, to local topic voting to Government voting. In each case the “who voted” question is either nothing to do with the blockchain, built in with the voters knowledge or a separate blockchain. This increasing level of criticality also ensures that people own their own data and can only expose themselves to view to the entities they wish to.
Let’s explain this a little. With electronic identity, a citizen or company holds its own keys to that identity. In other words the holder has under their control the ability to sign documents and transactions with that identity and no one, their government included can do that without their consent. The other key characteristic is that one user with one key can expose multiple public identities. This helps people and companies to separate the various activities of their lives and give away less information at each interaction. So, for example, you no longer have to produce you national ID to join a sports club, but still providing a guarantee that you are a unique and recognized citizen of a given country. Or produce even less, if not required, that you are just an individual who lives at a given address. Or just an individual. The crux is that the identity is being provided on a trusted network and therefore we are clear that an individual is just that, unique.
Ok so there are lots of things we can do with electronic voting, powered by new technologies like blockchain, so now what? How does this become reality.
Well the next step is actually taking a decision to open up voting on something. Not something trivial, it has to have a purpose, but not something that gives control to the nuclear missiles either. So for example a local health association could have a vote of all its membership on how to spend its budget and an expression of wish or a particular decision that entails not only the management members of the association itself but the local people it serves. Another example is to enhance the digital ’suggestion box’ that many governments run today as questionnaires or polls. These could be upgraded as expression of wish engines with a guaranteed identity, thereby making them more trustworthy for the receiving government entity.
I promised to stick to the technology but I do believe that the tech is not only about the tech. Hence finding some easy candidates for voting scenarios that can build trust.
But back to the ones and zeros, it is a relatively reasonable task to take existing open source technologies and combine them to create a really secure and anonymous voting app. Yes I said “app” not infrastructure, because we also need to leverage the super computer in your pocket to get this thing done and much of the infrastructure is available anyway through public blockchain networks and cloud services. Thus the route of relative ease is to combine existing code bases or services to create an easily distributable voting system. And before you start wining about the digital divide and those without smart phones, lets look at the advantages. By using the mobile device you are distributing voting capabilities to where people are, reducing traffic and pollution of people going to polling centers, reducing the barrier to entry to voting and delivering all the paperless benefits to the efficiency of treatment of voting administration. With smart phones now available for $100 or less, this is becoming a very widely spread communication channel. Additionally, we are not cutting off those not wishing or not able to use the smart phone as the same app can be distributed to polling stations to be used on tablet devices by anyone who wishes to walk in instead.
Government organizations do not need to look for complete solution either. Solutions can be custom built based on the wide variety of open source platforms to give comfort on their inner workings, or encourage local providers to incentivize the digital economy. In fact open source is the best driver for local economic growth as it gives a massive booster to level the playing field with many of the more advanced digital hubs. Robust blockchains, databases, application servers, mobile app frameworks are all available open source. Even voting applications are available, although most are still nascent, yet proving their robustness is part of there process.
So the steps to realizing an electronic voting platform and engaging citizens and companies are multi level but can end up with a simple ecosystem that voters an adopt and trust. The first step is not to up an buy a solution and rush it into a sensitive area, such as the US Democratic party primary voting app fail.
The first step has nothing to do with the tech and is to identify a set of voting situations that start with useful but low criticality situations and grow in complexity, reach and criticality. Examples of these can be to start with expressions of wish, growing into formal opinion polling and onto elections of growing importance. For example, a local or national government could create a roadmap, with increasing levels of importance, such as:
Expressions of wish for use of particular budgets such as recreationFormal polling for preference on medium budget proposals such as solutions for re-use of regeneration landVoting for specific posts such as local associationsVoting for council rolesLocal elections
These are just examples but show the growing importance of the uses of the system, in a classic conservative approach to implementing sensitive systems. The secret of this approach is that it leverages exactly the same technology for all steps and allows for full evaluation each round against the same environment that will be ultimately used in the highly sensitive voting campaigns later.
Secondly we get to the tech. This is the establishment of a platform to carry out testing on for the lower levels of criticality of voting and then goes on to be scaled for critical voting applications. The platform, as mentioned above, should use a number of open source technologies. This includes the blockchain protocol itself, all of which are open source. Protocols such as Ethereum, EOS, Stellar or Hyperledger Fabric can all show robustness for voting applications. They vary in their openness as relates to their network and so the only real decision to take is to decide if the governmental ecosystem wants to go for a global blockchain to gain, arguably, in resilience and immutability or a more contained public chain with private consensus. A great compromise is to have a consortium chain that achieve more resilience and immutability than a local area can achieve but does so in a controlled environment. Consortia are occurring in the private sector, such as in supply chain, and increasingly at the government level. The European Blockchain Services Infrastructure (EBSI) is such a consortia, being a collaboration of 20 European Union and European Economic Area countries. A consortium gives the advantage of a fast start on all the decisions on blockchain protocol set up and allows for reduced costs as well as increased resilience and immutability.
But the blockchain is not all as we mentioned before. Putting together an application, the flows of identity and voting, as well as deciding how to deploy them take some effort, but again can lean on existing resilient work. As an example, there are very solid code bases for creating the user’s private keys, the application to store and read their transactions on the blockchain (voting history) and more. These are the lego bricks that come together to create the voting solution, including taking on elements from associated parties that have the same goals, such as blockchain consortia. Putting these together in a way that suits the government body, through local developers, can be as fast, practical, secure and robust as buying in a prefabricated solution which may be inflexible on certain important aspects.
Finally we come to the menu for deploying the voting application. As discussed above, the roadmap of possible uses for digitizing voting has to be defined before or at the very least at the same time as the platform is being built. Therefore we have a plan to test out the voting application and roll it out, while testing and reviewing results at every stage. The type of KPIs that we need to observe to ensure that the voting app can pass on to the next stage will include the adoption of the system in the given ecosystem, security reviews, resilience review, trust in the system and operational factors.
So with this recipe, which may not be simple but is accented on success rather than a particular technology, we can get electronic voting accepted and deployed in a government environment of any scale. By so doing governments can bring the benefits of digital transformation to one of their most critical roles.
The key is to get going and grow in criticality over time.
Andrew Rippon is a blockchain consultant specialized in government digital transformation and founder of Thrupny, a startup providing regulated fintech solutions. Contact is