May 22, 2019 — Louvain (Belgium) / Indonesia
We recently set up the largest ever blockchain application for elections to date. This was a joint effort of Effendi Gazali and SettleMint as an academic exercise. On April 17, “Democracy Anchored” was able to report on 25 million votes in the Indonesian elections within hours after the polling stations closed. This is weeks before the official results will become public (on May 21st). This shows that blockchain can be incredibly valuable in large scale elections — as long as you don’t try to replace paper voting, or maybe combine it in the digital voting.
“The most complicated elections in the world”
Indonesia’s elections are not just large, they are probably the most complicated in the world with 193 million voters spread over 17,000 islands. Counting the votes from 813350 polling stations takes weeks, and traditionally tensions rise in the weeks between the vote and the publication of the official results, with both sides accusing the other side of tampering.
Votes and vote counts not only have to travel far and wide geographically, they also need to go through five levels of aggregating before they are counted in the national election. A lot can happen to a vote on that journey, and that is where blockchain can help.
In Indonesia, NGOs have been actively trying to increase transparency in the electoral process.
They are trying to keep track of the count by crowdsourcing the local election results and aggregating them on the national level.The idea is to track the votes from the moment they are counted until they are reported at the national level.
The local vote count is easy to find. In every polling station, a committee counts the paper votes, they agree on the count and they record it on a form (called a “C1” form). This form is signed, photographed and usually finds its way to WhatsApp and other social media.
Technically it’s not a challenge to make a database of these photos.
The problem with these crowdsourced databases is spam. They get bombarded with doctored photographs and even junk photos that undermine their credibility. They can’t prove that the photos of the forms were not tampered with when they entered the database, and they can’t prove that the photos weren’t changed after they entered the database.
So while these are worthwhile initiatives, they don’t wield much authority in a debate on what the exact vote count was.
We were invited by a Professor Effendi Gazali to solve that problem of credibility by using blockchain.
“Proof of integrity”
We used the same crowdsourcing approach to create a database of C1 forms.
But we used algorithms and human moderators to filter out doctored photographs of C1 forms, which resulted in a high quality database of more than 100,000 of these forms — representing 25 million votes. (You can find the results here)
We stopped accepting new photos 2 days after the election, because at that point the risk of tampering became too big.
That’s how we ended up with 100,000 “bulletproof” photographs, good for 25 million votes. We can offer “proof of existence”, the confirmation that the photograph existed very soon after the vote. And we can offer “proof of integrity”: we can irrefutably show that a certain picture has not changed since we added it to the blockchain.
What’s more, everyone can check this for themselves on the publicly available website, and drill down to a photograph of the local C1 form: https://indonesia2019.ballotbox.eu
Says Professor Gazali: “If the Indonesian Election Commission had used this solution, Indonesia would not have fallen into the long-lasting, relatively semi-chaotic counting we are witnessing now. We saw the biggest fatalities among the workers in Indonesia’s election history ever, and according to the Election Watching Body there were several mistakes in the counting and recapitulation technology in the Election Commission system. If we can use such a blockchain database on a large scale, we can make sure that all votes are counted within 6 to 12 hours after the polling stations are closed, and thus avoid both political tension and the possibility of votes being falsified.”
That is where, in fragile democracies or simply countries with challenging geographical conditions, blockchain can truly add value at a very low cost.
The longer it takes to announce results, the more trust erodes, with propaganda undermining trust in the democratic process — or even worse, demonstrations, violence and sometimes civil unrest ending in fatalities.
The shorter we can make the window between voting and reporting of results, the fewer opportunities there are for parties keen to subvert the electoral process.
For us, this is a common sense approach to utilizing blockchain technology in elections, especially in large countries who face logistical problems in counting and reporting votes, like many democracies in Asia and Africa.
Of course, it’s interesting and fun to theorize about building an impregnable, tamper-free digital vote that enables remote or home voting.
But the reality is that paper isn’t a broken or obsolete technology for voting. Mature, respectable democracies have used it successfully for centuries.
What they lack is a fast and tamper-proof way to report and count votes — transparency and speed. And that’s exactly what blockchain can offer — even when it’s based on a paper vote.
Professor Gazali adds:” Indonesia needs to remanage its simultanoeus elections. I have long been suggesting that Indonesia divides it into two stages: the national stage (for president, national parliament, and senate) and the local stage (for local executives and legislators). Also there must be no presidential threshold to avoid the worsening division of the nation, because there were only two presidential camps since 2014. It has seriously dragged on conflicts between both camps as shown in hoaxs, fake-news, and hate speeches in social media. All of those have then raised questions on the integrity of election management and IT system. Now, by having blockchain tested in action, Indonesia has more methods to choose to solve most of the problems. I prefer the one with paper ballot and digital reader. So the ballot-reader and it box save the paper ballot, but at the same time it sends one copy to the Election Commission powered by the Blockchain Technology. We should further consider to give another copy to be saved by the voter and some other copies might be sent to some highly-credible international observers.”
SettleMint is a leading Blockchain technology company that helps organisations leverage the benefits of blockchain technology — whether those enterprises are looking to improve efficiency, to extend their current products or to completely reinvent an existing business model — by offering them our distributed middleware solution, Mint. Mint makes it easy to integrate blockchain technology. So in short, it doesn’t matter if you are a startup or a big corporation. If you think Blockchain technology can be part of your business or your vision for the future, you, or the developers working in your organisation, can start integrating blockchain technology in no time using Mint.
Commercial contact: Matthew Van Niekerk, CEO and Co-Founder - [email protected]
Press contact: Els Meyvaert, Marketing and Communications Director - [email protected]