Unless you have been living in a cave for the last couple of years, you will have heard about the technological hype of the moment, the blockchain! This technology is set to revolutionise every single industry under the sun, and will disrupt the way in which we do businesses, interact with each other, and communicate with data.
Or will it?
While I have been rather vocal about my Bitcoin scepticism, I thought that the were some interesting ideas about the blockchain, the underlying technology behind cryptocurrencies. My thinking was, and this was shared by many others, that even if Bitcoin failed, the blockchain technology would remain and become an important contribution to the way in which online transactions are made. But I have to admit that in the last few months even that optimistic feeling about the blockchain has started to dissipate. It seems like I am not alone, and already Gartner has placed the blockchain in the trough of disillusionment in their 2017 hype cycle.
Trouble with definitions
A salient question is that for something that has been so hyped, there is little consensus on what a blockchain actually is. The most general definition is that a blockchain is a public decentralised cryptographic database that operates as an open ledger of all of the transactions that have been recorded, and that this record is immutable and tamper-proof. In technical terms, the first definition was given by the Bitcoin white paper by Satoshi Nakamoto, which describes a cryptographic coin that is comprised of a chain of digital signatures that are added to a block and time-stamped and appended at the end of the chain; so all transactions are publicly available and verified, but also the system is tamper-free because to change a transaction you would also need to change all the transactions that came after the one you are trying to verify.
So far so good, all we need to do is to follow this definition of a blockchain coined by Satoshi himself, whoever they may be. The problem is that the Bitcoin blockchain comes with some limitations, transactions can be slow because of the need to conduct verifications, and they can also be bad for the environment because of something called proof of work. Proof of work protects the blockchain from attackers and spammers because it requires those participating to perform some work in the shape of computing time. In Bitcoin, the proof of work consists in trying numbers until the correct solution is found and then a new block is added to the blockchain. This is extremely energy wasteful as the difficulty of finding a solution increases over time.
So other blockchain solutions have arisen to make transactions faster and to use less power, which further complicates the definitions. So you may have a proof of stake blockchain, which instead of requiring computers “yelling numberwang at each other”, it chooses the allocation of the next block between those with a stake in the system without the need for large expenditure of resources. Similarly, there are private blockchains where the ledger is actually not open nor transparent.
The result of this confusion is a situation in which anything can be called a blockchain, even if it does not meet the traditional definitions. Adrianne Jeffries explains the conundrum in a must-read article on the subject:
“There are countless blockchain explainers in text, audio, and video around the web. Almost all of them are wrong because they start from a false premise. There is no universal definition of a blockchain, and there is widespread disagreement over which qualities are essential in order to call something a blockchain.”
Things become more complicated when we see that there are attempts at cementing some definitions into legislation. This can be tricky as favouring one definition over another could result in dire legal consequences. Jeffries accurately points out that some definitions that require immutability and “uncensored truth” could particularly be dangerous, as the blockchain is not an arbiter for truth, but it only provides evidence of whatever is entered into it.
One of the reasons why the blockchain hype may be on the wane is precisely because, at least so far, the technology has failed to deliver on the various promises made by enthusiasts. At the more extreme end of the spectrum we have people who have proposed the blockchain as a solution to everything from poverty to corruption. Even those who do not promise it as the end to world hunger seem to rely on the hype to make promises that seem far-fetched, or not grounded on reality.
One particularly annoying feature is that the hype and the potential are often reported as fact. If someone proposes a blockchain for the manufacture of digital widgets, this often morphs into “digital widgets are already being produced using the blockchain in country X”. Dig a little deeper, and that is almost never the case.
One of the subjects where this is happening is in elections. For the last couple of years blockchain technology has been offered as a means to ensure elections are fair, and to reduce voting fraud and certify ballot counting, or just to make online voting a reality. But while in theory voting looks like a potentially viable use of the blockchain, there are a few issues. Firstly, electronic voting does not fit well with various characteristics of traditional blockchain definitions, particularly the open ledger and the decentralisation. A robust voting system requires a central authority to validate that a person is legitimised to vote, which requires some form of registry as well. A voting system also needs to be both secret and anonymous, and a certified authority must be able to validate the voting. All of these seem to be counter to the open nature of the blockchain. While some people have proposed elliptic curve cryptography to solve part of the secrecy issue, the need for a centralised authority remains.
In the end, the solutions become convoluted, and the need for a blockchain stops making sense as a replacement for existing systems, imperfect as they may be. Perhaps evidence of this is that we are yet to find a proper implementation of blockchain for voting. A few examples are wheeled out when talking about elections and blockchain. The first is Estonia, which as Jeffries explains, is not even a blockchain, it’s more like an ID system that has retroactively been called a blockchain. Then there is a plebiscite in Colombia which is supposed to have used a blockchain system to involve citizens living abroad not registered to vote. However, when one looks at the actual implementation, this was not part of the official vote, but a pilot project to test out the technology. Plebiscito Digital was an symbolic ballot requiring self-registration, and as far as I could see, the results are not open (lacking a few of the requirements of transparency and decentralisation). In fact, the article describing the project cites as the result of the experiment the fact that a conversation was started, but it is difficult to verify the number of participants, and even if the results are stored in an open blockchain somewhere.
The latest experiment being widely publicised as I write is the alleged use of blockchain technology in Sierra Leone. I was extremely intrigued by this. Finally, a proper voting experiment using decentralised technologies! But once again, looking beyond the headlines proved disappointing. According to the article, tech startup Agora used a “private permissioned blockchain” to verify the results behind the scenes. But if you look at what actually took place, Agora employees and the CEO attended the paper ballot counting in an undisclosed number of locations (it seems like only one, but it’s difficult to tell), and they entered the tally into a blockchain. Yes, the only use of the blockchain was to enter paper voting results to make them immutable, so all of the problems of election verification, including legitimacy, privacy, and corruption were still there, instead of writing the results in a paper, they wrote them on the blockchain. You’ve just added one more failure point.
The amazing thing, which is testament of the level of hype, is that this small-scale experiment has somehow mutated into the “World’s First Blockchain-Powered Election“, making it look like the entire election was held using blockchain technology. As someone wrote on Twitter: “How is this anything other than having a corruptible 3rd party manually enter votes into a database?”
Of all the uses for the blockchain that are supposed to become more widespread, proof of provenance is often mentioned as one of the most promising. Imagine any walk of life where you would like to have proof of provenance: ethical diamonds, organic food, environment-friendly resources, etc. This is perhaps one of the more hyped blockchain applications out there. Take this example:
“… blockchains in general can be used for anything, to track everything from the food that we eat: “Cows can become blockchain appliances, enabling farmers to track what the cows eat, which medications they’ve had, and their complete health history.” Seriously- we could follow the provenance of every ingredient in every product.”
Blockchain cows, you heard it here first folks.
Seriously though, this line of thought posits that many supply chains are in need of some open, immutable and tamper-free record-keeping technology that will ensure us that the locally-sourced asparagus you are eating is indeed local. These proposals seek to verify everything from fine art to bananas. While the specifics of each proposal vary, they seem to share a common thread, the idea is to use the blockchain as an immutable provenance record where data cannot be destroyed or falsified, and which can be used to identify a product’s origin and trajectory.
Let’s say that I want to start a llama socks business with ethically-obtained llama products from Peruvian llamas. The idea is to have some sort of token that can be stored on the blockchain that will provide evidence of provenance. So I travel to Peru, obtain the llama wool at the ethical llama farm, put it on a box which has stamped an address where the origin can be verified by everyone in the world via a tamper-free and immutable system.
You may already have caught the problem with this scheme. The socks need not be there at the start, an unscrupulous actor could source the llama wool unethically, and just take the box to the Peruvian llama farm with nothing in it, or with the illegitimate products already inside. The problem with proposing a blockchain for real goods is that these are physical and not digital, and the data in the blockchain only verifies that someone entered a record at some point in the supply process. You can add all sort of data, a picture, or even GPS coordinates, and that doesn’t prove that the product that you have, be it a painting, socks or a banana, came from the alleged origin. It only proves that someone at the source entered the data into the blockchain. So even if you somehow tag the item with the adequate address, QR code, or whatever, that could have been added later. I could add a label to my socks with the verification code obtained earlier.
Reality once again gets in the way of a brilliant scheme.
We are in the midst of a technological revolution, there is no doubt about it. The blockchain can be seen either as a footnote to the cryptocurrency craze, or as the most disruptive innovation of our lifetimes. The truth may lie somewhere in between. It is possible that blockchains will become prevalent in all walks of life, but to me the most likely scenario is that some functions may be useful in a few circumstances. As of today, I do not see many uses that cannot be achieved with existing technologies. Moreover, many solutions that are touted as a blockchain may not actually meet the definition, be it because they are private, because they are not decentralised, or because there is only one institution using it. If you only have one user, then you don’t have a blockchain, you have an expensive database.
So why the relentless push towards the blockchain? An excellent article by Oliver Leistert postulates an intriguing hypothesis. The blockchain and Bitcoin should be seen not in isolation, but as part of a wider political phenomenon that includes other technologies, such as big data, surveillance, and the Internet of Things. Leistert explains that the registration of digital objects creates a new regime of control where non-verified uses are not only forbidden, but are made technologically impossible. Smart contracts become blind code.
A techno-dystopia of endless verification and commodification.
We should rebel against the creation of a system of immutable records where reality is irrelevant, and what matters is the truth as told by whoever entered the data into the blockchain. History is no longer written by the victors, but by the coders.