There Is No Such Thing as “the” Blockchain

We’re using one word to describe a bunch of different technologies.

No such thing as 'the' blockchain
The word blockchain is often used to describe two broad types of computer systems.

Thinkstock

How do you quadruple your stock price without lifting a finger? In December 2017, the Long Island Ice Tea Corp. found an answer: Add the word blockchain to your company name. The beverage business made this move before doing much to justify its new branding as Long Blockchain Corp. Still, investors rewarded the company: Its stock rose 432 percent. That’s because blockchain, a piece of bitcoin’s technological plumbing, has recently become the buzzword of all buzzwords in the business world.

Already, blockchain has been hailed as likely to revolutionize … well … everything. Banks, health care, voting, supply chains, fantasy football, Airbnb, coffee: Nothing is beyond the hypothetical reach of blockchain as a revolutionary force. These predictions are easy to sell because blockchain is still little-understood. If you don’t quite know what blockchain is, it’s easier to imagine that it is whatever you want it to be. But before we can begin to search for the real potential amid the mass of blockchain conjecture and hype, we need to clear up what exactly we mean when we say blockchain.

One cause of confusion is the phrase the blockchain, which makes it sound like blockchain is one specific thing. In reality, the word blockchain is commonly used to describe two broad types of computer systems. Both use similar underlying protocols, but they have other important differences. Bitcoin represents one approach to using blockchain, one wedded to principles of radical decentralization. The second approach—pioneered by more business-minded players—puts blockchain to use without adopting bitcoin’s revolutionary, decentralized governance. Both of these designs are short-handed as blockchains, so it’s easy to miss the crucial differences. Without grasping these differences, it’s hard to understand where we are today in the development of this promising technology, which blockchain ventures are worth your attention, and what might happen next.

To understand the tangled blockchain ecosystem of today, we have to go back to a time when blockchain was easy to define. This was around 2009, when a block chain was a specific data structure designed to solve a specific problem for bitcoin. In their mission to cut the middlemen out of currency exchange, bitcoin’s still-unknown creators proposed to put copies of its complete financial ledger on every computer using the system. For that to work, bitcoin’s design had to prevent forgery. If anyone could rewrite the ledger and give themselves a billion bitcoins, the cryptocurrency would never work. Blockchain was a solution to the forgery problem. By mathematically linking each new entry on the ledger to the previous one, blockchain created a theoretically unalterable chain of data through time—making it impossible to rewrite previous entries in bitcoin’s ledger.

But blockchain didn’t solve all of bitcoin’s problems. Without banks and governments, who was to say which new entries were valid, and who would enforce those decisions? If bitcoin users were going to be anonymous, how could you trust them to follow the rules? To overcome these challenges, bitcoin adopted “proof of work,” the harrowingly complex procedure by which new blocks are linked into bitcoin’s blockchain by miners. Bitcoin users have to obey its proof-of-work rules thanks to the principal of algorithmic governance.

Algorithmic governance sounds very high tech, but actually it’s wonderfully simple. An algorithm is a set of rules—in this case, the rules of the bitcoin network. If you follow those rules, you can interact with the network and exchange bitcoins. If you break the rules, the network won’t recognize your transactions and you’ll be automatically cut out. So the rules are self-enforcing. Bitcoin can get by without banks or governments because algorithms are its highest law.

Making algorithms your supreme law does, however, create a giant hassle whenever they have to be changed. Such changes are inevitable, because we can’t predict the future. Bitcoin, for instance, was ill-prepared for its own popularity. It’s now struggling to process an increasing volume of transactions. Attempts to solve this problem have caused a series of disruptive splits in the network, called hard forks.

A hard fork can happen anytime someone proposes new rules for the network that are incompatible with the old ones. Under algorithmic governance, every user votes with their feet as to which rules to follow. Unless almost everyone makes the same decision, the network will split into two. These splits are all but inevitable and contribute to instability in systems like bitcoin.

In November 2017, a long-planned hard fork in bitcoin was called off at the last minute because there wasn’t a consensus about what should be changed. In reaction, bitcoin lost and then regained about 25 percent of its value in a week. To put that volatility in perspective, the value of the British pound only fluctuated 12 percent in the week following the epoch-making Brexit vote.

To a corporate world that felt Brexit like a once-in-a-century earthquake, bitcoin’s madcap price variations are absolutely unacceptable—especially because they highlight the difficulty of managing the cryptocurrency’s network. Yet, corporations have noticed that blockchain is quite useful for creating secure and efficient data systems. So the challenge for business-minded developers has been whether they can get the security and efficiency of blockchain without the chaos and uncertainty of Bitcoin.

It turns out that almost everything a corporation might not like about bitcoin comes from algorithmic governance. And algorithmic governance is only necessary if you want your network to be anonymous, radically decentralized, and open to anyone who wants to use it. Because most corporations don’t want those features anyway, many businesses interested in blockchain are leaving algorithmic governance behind.

What’s emerged is a whole range of permissioned blockchains, in which there are some rules around who can join. Hyperledger, an open-source software project founded in 2015 that specializes in permissioned blockchains, now has dozens of members, including the Chinese search engine Baidu, the French aircraft giant Airbus, and the Bank of England—all of which are interested in permissioned blockchain technology.

In a typical permissioned system, all the participants will agree in advance about how the network should be run and sign a legal agreement setting out the rules. For example, imagine a consortium of banks using blockchain to clear and settle transactions (an idea recently piloted by the Bank of Canada). In this system, blockchain algorithms are used to store and exchange data in a secure, decentralized manner, but the banks remain in control and manage the whole network according to their contract. New members would have to sign the contract in order to join.

That contract—sometimes called a network constitution—would contain the procedure for changing the system’s algorithms and also govern aspects of the network that no algorithm can control. When it comes to privacy, for instance, a computer program can strictly control who has access to what data, but it can’t control what someone does with the information once its inside their head. To stop an employee from spilling clients’ secrets at the bar, you still need a nondisclosure agreement.

Some purists don’t think these systems should be called blockchains at all, because they trade algorithmic governance for usability. You’ll sometimes hear them described as distributed ledgers or consortium blockchains. There’s no consistent terminology. And the fight over terms is really a fight about principles. To a purist, a blockchain must be a system, like bitcoin, that decentralizes both data and power. These alternate blockchain designs decentralize only the data, not the power.

From a bank’s perspective, this has many advantages. All the participants are known, which is required by regulators. Algorithms can be changed as needed, according to the contract, which adds stability. And whereas all bitcoin transactions are visible to anyone, permissioned designs can give different participants different levels of access—keeping clients’ data private and business secrets secret. A permissioned system can also avoid bitcoin’s onerous and environmentally ruinous proof-of-work system for adding new entries. Hyperledger is experimenting with several alternatives, and members can choose the algorithm that works best for their purpose, whether it’s banking, supply chain management, or keeping medical records. “Any industry that involves transactions will be interested in blockchain, which is all of them,” said Brian Behlendorf, Hyperledger’s executive director.

Seeing this huge potential market, a number of major tech companies have started to offer blockchain services too, and they have also focused on permissioned designs. Executives at IBM and Microsoft, two of the leading corporate blockchain players, told me that their clients overwhelmingly prefer systems where all the members are known and have agreed to follow the rules. According to Matthew Kerner, a general manager in charge of blockchain at Microsoft, clients also demand speed and usability. Trying to design around the idea that any member might be untrustworthy makes systems too slow and cumbersome. “Compared to their enterprise systems of today,” said Kerner, that type of blockchain “just doesn’t stack up.”

This isn’t to say that these corporations are right to dismiss bitcoin-style blockchains. The corporate track record on handling disruptive technologies is hardly spotless. In the realm of cryptocurrency startups and initial coin offerings, permissionless blockchains are still all the rage—attracting billions from investors.

What’s crucial is that would-be investors understand that not all blockchains are created equal. Those making sweeping predictions about the future should take note of this too. Blockchain boosters tout the idea that cryptocurrencies will soon displace conventional monetary systems, eliminating banks, and that a tidal wave of decentralization will then speed through every industry, washing away the entrenched middlemen in its path. Yet bitcoin has been mired in wrangling over capacity problems while permissioned blockchains have gained backing from the very same middlemen that the cryptocurrency was supposed to eliminate. Those institutions seem to have found a way to co-opt many of the most useful technical features of bitcoin without the cryptocurrency’s revolutionary implications. The bitcoin blockchain model is up against some terrifically powerful vested interests, which have no intention of putting themselves out of business.

This article is part of Future Tense, a collaboration among Arizona State University, New America, and Slate. Future Tense explores the ways emerging technologies affect society, policy, and culture. To read more, follow us on Twitter and sign up for our weekly newsletter.