This article is an exclusive contribution to CoinDesk's 2017 in Review opinion series.
Years, and even centuries, never really begin on Jan. 1st.
You can always spot the real beginning in events that occur a bit later, or a bit earlier, than the official start. The 20th century, for example, began in 1914 with the burst of World War I and left the stage to the new millennium in 1989 when communist economies collapsed and opened to free enterprise.
For those interested in the practical future of blockchain, and of many other things, the year 2017 began on June 17, 2016. On that day, for the first time in human history, a robot "stole" more than $45 million from a group of human beings, following the instructions of an unnamed human being.
The robot has a name, instead, an infamous name: it was called The DAO. This robot had already broken a record: it was the first robot to get the control of more than $150 million, provided voluntarily by human beings by buying DAO tokens. The DAO was not a robot like those we see in the movies, that are often as emotional as human beings. The DAO was a real robot: a smart contract, a distributed piece of software designed to follow its internal code blindly and precisely.
So did he also when instructed to move $45 million to the account of the unnamed man, with instructions that were consistent with the code written for it. This code said also that this money would only become available 28 days later. And until then, the robot was going to protect the money from the unnamed man and from anyone else.
In the meanwhile human beings, as they often do, split into two opposite groups.
A minority thought that everything was fine, that there had been no theft since everything had been consistent with the code written for the robot. A majority, though, thought that what had happened was an error, a human error, a coding error, and that the unnamed man was a thief.
This could have been the beginning of a war. But we were not in the "real" world. We were on the blockchain, we were on ethereum.
The humans decided to redouble the distributed robot. One robot led the minority to protect the rights of the unnamed man, the other robot got a slightly new code and led the majority to the recovery of the stolen money. Both things happened... proportionally to the choice that human beings running the robot in a decentralized fashion made with their client software. It was the first big hard fork in the blockchain space.
In the meanwhile, The DAO had lost credibility, and its tokens were losing their value. Criticisms were made of the programmers, the community, its leaders. According to most observers, this was going to be the end of the concept of smart contracts and of tokens, or maybe even of the concept of public blockchain as a way to raise and store the value needed to carry on a project.
It was, in fact, the beginning. The explosion of ethereum, the ICO phenomenon and the forks of bitcoin that marked 2017 were the fires started by The DAO spark. Less shallow observers, in fact, in looking at that case, had spotted the beginning of a revolution.
For the first time, a blockchain had been able to collect millions on a real project. For the first time, the process had been tested, its weaknesses observed and governance solutions never seen before had been proposed, discussed and implemented.
People understood that the blockchain and the smart contracts were real. Now, ethereum was ready to become a hub allowing hundreds of projects to get funding more easily, more freely... maybe too easily.
This is what we observed in 2017, while bitcoin and ethereum ecosystem grew to the next order of magnitude. It was the year when it became obvious what the blockchain really is: a power to do things that were impossible before.
The idea that it is a technology for doing in a "faster" or "cheaper" way exactly the same things we have always done is misplaced. It is not necessarily cheap (often it is quite expensive) and when it does things faster, this happens because some fundamental principles have been changed, such as changing what we mean by "accounting," what we mean by "settlement" and what we mean by "privacy."
The new network
This misconception is still holding back the development of the so-called "private blockchains" that saw so many experiments and proofs-of-concept in 2017, but nothing yet that was even marginally disruptive.
Even private blockchains can in principle be a transforming power. The principle of open entry is important for blockchain robustness, yet open entry can also occur within a range of actors that does not include the whole humanity, but only those that accept some rules and satisfy some requirements.
What, however, does not exist, is a blockchain that really works without shifting the balance of powers toward decentralization.
Even a centralized entity can make good use of a blockchain. But this happens only if the central authority wants to change its role from the "validator of everything" to become the "validator of last resort" or to become an "invalidator" that uses a proof of authority only in case of well-defined problems.
When a business is run on a blockchain, additionally, there is an unavoidable increase in transparency and in determinism. Even without necessarily losing privacy, certainly, players have to expect to lose some accounting opacity. And some elements of vagueness that are typical of paper-based business need to be lost.
And, last but not least, there are human controls that need to be replaced by cryptographic security and distributed automation if we want a blockchain to be of any use. Most private blockchains are vague, or even agnostic, on topics such as the governance and the balance of powers. They do not define who validates transactions, or what is the consensus protocol.
But the purpose of a blockchain is to innovate and define precisely these aspects! Most private blockchains are also vague in defining what kind of value will flow through digitally signed transactions and will be stored on the shared registry.
It may be digital versions of fiat currencies - yet unseen - or the IOUs of an institution, or even no value at all, just information. But a blockchain is a web network of value, if value does not flow one risks having a lame network.
What you missed
An often overlooked truth in the noise about cryptocurrencies, blockchain and smart contracts is that they mark the entry of robots in the field of transactions and contracts.
This means that, as robots start driving cars, they will also start managing our money, our transactions and our agreements in a decentralized fashion, with the purpose of making them less reliant on centralized control or multiplication of intermediaries. This is happening already in the public blockchain, and this is what the public blockchain has shown, also in 2017.
Not everyone, unfortunately, has been able to see what was really going on.
Even the name smart contracts may have been a misnomer, hiding the fact that ethereum smart contracts are more "robot counterparties" than "contracts" in the classic sense. But we may remind that Nick Szabo, in the paper that introduced the concept of smart contracts, described the "humble vending machines" as the "primitive ancestors of smart contracts." (So were the old juke-boxes. The first reassuring prehistoric robots we got accustomed to giving our money to.)
Of course, many will be frightened of obtaining decentralization, transparency and higher guarantees through giving part of control to that special kind of immaterial robots that are called smart contracts.
I feel the need to make two remarks here. First, the DAO showed clearly that such a fear is not irrational. Yet, we will probably get over it. (Self-driving cars are still creepy, for reasons that are surely rational.) Yet, they will grow in reliability and will drive us all, and this is likely to start sooner rather than later.
Second, there are ways to make smart contracts less frightening. Although decentralized automation is needed for a smart contract to work, smart contracts can be designed to be "state channels" that formalize rights and set constraints in a bilateral flow without the need of robot counterparties.
This is what we have seen in some proposals for private blockchains.
More importantly, state channels can be the foundation of a mixed private-public approach to smart contracts, where agreements are taken bilaterally and, as long as the two counterparties agree, the exchange of value happens off-chain in a layer-two bilateral communication.
Just in case settlement is needed, or a dispute emerges, a representation of the current state of the transaction between the parties is sent to a public blockchain to be settled, potentially following some pre-agreed code whose hash is stored on the blockchain.
This can also be a solution to the scalability problem that exists in current blockchains, where for each and every transaction all validating nodes have to run the code of the smart contract. This can also be a solution to the privacy issue that still makes many institutions wary of the blockchain. (Another solution to privacy may come from ring signatures or zero-knowledge proofs, and we may see relevant breakthroughs in 2018, probably coming from the efforts of researchers and the public community on the topic.)
All this points to the fact that the year to come, 2018, could be the year when (financially oriented) human beings and (smart contract) robots make contact.
And start understanding each other, marking the arrival point of a process started with The DAO failure.
And so I have spoken more of 2016 and 2018 rather than of this 2017 that comes to an end. This was maybe to be expected, since the present is, in the end, just an immaterial veil between the past and the future, a haze on the path where the flow of time, technology and innovation is taking us, fuelling our fears for what we have not yet fully come to understand.
Disagree? CoinDesk is looking for submissions to its 2017 in Review series. Email email@example.com to pitch your idea and make your views heard.
Astronaut with halo via Shutterstock