Updated: Apr 8, 2020
A Medium article by David Kelnar
The following is a Medium article published by David Kelnar.
The blockchain winter is thawing. We offer an accessible introduction to an important emerging technology.
In 2018, expectations for blockchain technology, and the value of blockchain currencies, fell to earth. The blockchain ‘winter’ has been bitter. There are signs of green shoots. Corporate blockchain adoption has reached a ‘turning point’ (Deloitte). In 2019, 83% of large companies see compelling use cases for blockchain and 40% plan to invest at least $5m in blockchain initiatives (Deloitte) in the year ahead. Global demand for blockchain engineers grew 517% in 2019, four times greater than for any others (Hired). The Governor of the Bank of England wrote that blockchains could “drive efficiency and resilience”, “reduce risks” and “improve financial inclusion” in the financial system (Mark Carney). The number of cryptocurrency wallets has grown 42% in 12 months (Blockchain.com). But blockchain is complex and debate is characterised by extremes: exuberance or cynicism; idealism or defeatism; greed or fear. Below, we provide an accessible and balanced introduction.
We explain how blockchains — shared databases — let us place trust in peer-to-peer networks instead of central authorities, enabling decentralisation. We describe how blockchains work, and their advantages and drawbacks. Why is the technology suitable for some use cases but not others? We explore blockchain’s potential to unlock ‘Web 3.0’ — a decentralised future — and how, together with ‘tokens’, blockchain could offer new business models, greater and more equitable innovation, and a shift in value in the software stack. ‘Amara’s law’ posits that we over-estimate the impact of new technologies in the short term while under-estimating their long-term impact. Today, expectations have re-set while activity is growing. In the years ahead, Amara’s Law may ring true. Blockchain moves trust to the network
“Blockchain moves trust from central parties to network participants.” Most consumer experiences (sending money, buying goods or sharing activities) and corporate processes (lending money, managing a supply chain or scoring credit) rely on databases that belong to trusted, central authorities (Barclays, Amazon, Facebook; Lloyds, WalMart, Experian). Central parties provide services, maintain the record of what’s true (the ‘system of record’) and offer customer engagement and recourse. If you forget your banking password, your bank will reset it. Blockchains are shared records — databases stored, maintained and updated by multiple parties, with no central authority. Blockchains move trust from central parties to network participants, enabling information and governance to be shared. By doing so, Blockchains enable decentralisation .
Public blockchains trade efficiency for decentralisation Blockchains can be ‘public’ or ‘private’. Typically:
Public blockchains enable anyone to participate. Anonymous users can access and store a copy of the ledger, and submit and validate transactions. Public blockchains — such as the bitcoin currency — are highly distributed, with thousands or millions of users.
Private, also known as ‘permissioned’, blockchains feature administrators who restrict who can participate and what they can do. Often used by corporate consortiums, such as groups of banks, private blockchains allow only known parties to access and store data, and submit and validate transactions.
We explain blockchain’s advantages and drawbacks below — but many trade-offs relate to a blockchain’s degree of decentralisation. Public blockchains, which are open and decentralised, enable new, global services free from central control. But without trusted participants, they struggle to operate with speed and efficiency. Private blockchains operate rapidly and at scale because trusted participants administer them. But they are not decentralised, which necessitates trusting administrators and limiting security — dynamics which many people seek to avoid.
Cryptography and consensus mechanisms enable decentralisation “Blockchains use cryptography and consensus mechanisms to maintain a distributed ledger.” If no single party administers a blockchain — a distributed ledger — how can it function? How can we ‘move trust to the network’? Blockchains use two mechanisms to ensure legitimate use and maintain a single version of the truth:
Cryptography (a branch of mathematics) enables only legitimate transactions to be processed.
‘Consensus mechanism’ schemes determine how a shared ledger is updated, so everyone shares a single view of what’s true.
We’ll explain the detail of cryptography and consensus mechanisms in a future post. In brief: Cryptography draws on established mathematical techniques used for secure online communication. Related pairs of mathematical ‘keys’ have useful properties. By exploiting these properties, blockchain networks can validate account ownership and verify that transactions were authorised by their originators. Consensus mechanisms are processes for ledger coordination. Achieving consensus around a public blockchain — a ledger that anyone can access and update — is difficult. The mechanism must include incentives to protect against malicious behaviour, such as altering the ledger to enable an attacker to spend money twice. Typically, consensus mechanisms select a single network participant to amalgamate valid transactions and update the ledger. Selecting the leader must combine randomness with a financial incentive to promote good behaviour. Different consensus mechanisms use different approaches:
‘Proof-of-work’ schemes, including that used by the Bitcoin blockchain, require parties to expend resources — computer processor cycles, and the cost of powering them — if they wish to update the ledger. Proof-of-work is intended to make disruptive behaviour computationally impractical and expensive.
Proof-of-stake schemes grant validation rights in proportion to the share of value in the network that participants own. Participants with large stakes have more to lose in the event of network disruption. Aligning influence with ownership is an attempt to promote good behaviour.
There are other approaches to achieving consensus, including innovative schemes that offer leaderless consensus.
Blockchains have advantages and drawbacks “Peer-to-peer exchange of information and value can remove intermediaries, offering more structurally efficient economies.” Decentralising authority — moving trust to the network — has advantages and drawbacks. Blockchains offer many benefits…
Immutability: a tamper-proof record For technical reasons, once information is entered into a blockchain it’s almost impossible to rewrite. Centralised databases, on the other hand, may be altered by malicious insiders or outsiders who circumvent security procedures. Many ideas and business models are impractical or impossible because reliable data are inaccessible. A decentralised system, in which information is immutable by design, unlocks new opportunities in sectors ranging from insurance to healthcare.
Disintermediation: cut out the middle-man Peer-to-peer exchange of information and value lets us remove intermediaries, enabling more structurally efficient economies. Removing financial service intermediaries who receive a fee — for money transfer, foreign exchange transactions, transaction clearing and escrow services — can reduce friction and costs.
Automation: an ‘intelligent network’ Accessible, reliable digital information shared using a common protocol enables greater automation. ‘Smart contracts’ — self-enforceable contracts programmed in software — can sit on top of blockchains and undertake actions automatically and autonomously when specified conditions are met. ‘Escrow-on-a-blockchain’, for example, could use a smart contract to release parties’ funds when a defined transfer of value has occurred — without involving an escrow provider. In extremis, blockchains enable Distributed Autonomous Organisations (DAOs) -– entire organisations written in code and executing autonomously on a blockchain.
Resilience: no single point of failure Because decentralised databases are shared by many market participants, they aren’t susceptible to central points of failure. Even our most reliable centralised networks, such as Amazon Web Services (AWS), suffer downtime from time to time.
Censorship Resistance: freedom from state control Because public blockchains can be accessed by anyone with an internet connection, they can be used to exchange information and value in ways that nation states cannot control. While this could facilitate the exchange of immoral or illegal goods, it can profoundly benefit citizens in repressive, unstable, or appropriative regimes.
“In the West, people say they trust Google, Facebook or their banks. But the rest of the world doesn’t trust organizations and corporations that much — I mean Africa, India, Eastern Europe, or Russia. It’s not about the places where people are really rich. Blockchain’s opportunities are the highest in the countries that haven’t reached that level yet.” (Vitalik Buterin) …while having several drawbacks
Scalability: limited throughput, high latency The throughput of most of today’s public blockchains are limited. While centralised systems, such as Visa, can process thousands of transactions per second, Bitcoin and Ethereum manage just seven and 20 transactions per second respectively. Public blockchains also struggle with delays (‘latency’). For distributed ledgers to become a new, foundational layer for the internet — to enable large-scale applications that will draw on the anticipated 75 billion devices that will be connected to the internet by 2025 (IHS Market) — public blockchains must offer high throughput and low latency. Today, many parties are working on solutions to blockchain’s scalability challenge. Experiments include: bigger block sizes (enabling more transactions per block); ‘sharding’ (splitting block verification and running the process in parallel); and using less data from the root chain to streamline verification. Corresponding solutions are emerging. The Lightning Network operates on top of blockchain-based currencies, like Bitcoin, and enables fast transactions between participating nodes. Some solutions expand scalability at the expense of decentralisation. An application-specific balancing act between scalability and decentralisation may be required.
Waste: extensive energy consumption Most public blockchains, such as the Bitcoin network, are resource-intensive. Many networks’ consensus mechanisms are premised on market participants demonstrating the expenditure of computing resources and electricity (‘Proof-of-Work’). Together, participants in many public blockchains consume vast amounts of energy, which is expensive and environmentally damaging. Currently, the Bitcoin network consumes more energy annually than Austria. Blockchain developers are experimenting with less consumptive consensus mechanisms, such as Proof-of-Stake.
Lack of recourse: what if there’s a problem? Decentralisation and automation have a downside: to whom do you turn when you have a problem? If you lose the private key to your cryptocurrency wallet, or have a dispute with a counterparty, who can help? People find value in companies’ services, not just their products. Further, when a smart contract or distributed autonomous organisation produces an unexpected or unwelcome result, because a particular scenario was unimagined or an individual has exploited a flaw — should the outcome be reversed? Parties are offering solutions to some of the above challenges, such as custody services for blockchain keys. But many re-introduce centralisation — a trusted authority — and therefore risk.
Security: can blockchains be compromised? The decentralised nature of blockchains present a potential security vulnerability — the ‘51% attack’. If a majority of nodes in a network (in practice, sometimes more or less) are controlled by a single entity, the system becomes centralised and can be manipulated. Successful attackers could reverse transactions completed during their period of control, enabling them to double-spend coins, and prevent new transactions from being confirmed. Attacks on smaller blockchains, which use limited computational power, have occurred. In May 2018, the Bitcoin Gold blockchain suffered a successful 51% attack, in which perpetrators stole more than $18m of the cryptocurrency by double-spending. Successful attacks on major blockchains, however, are less likely. Perpetrators would require greater computing power than the combined capability of millions of global miners, and a vast budget for the electricity to power them. The Bitcoin blockchain has never been compromised. In blockchains that use alternative, proof-of-stake consensus mechanisms, effective governance and community coordination can protect networks even after a 51% attack. Honest nodes can ‘fork’ the network to render the attackers’ resources worthless. In practice, however, this can be difficult to achieve.
Governance: who sets the rules? Governance is the process by which groups of people organise to make decisions. In blockchain ecosystems, governance impacts how transactions are processed and which changes to the codebase are made. Most governance frameworks evolved within centralised systems and cannot readily be applied to decentralised ones. Decisions can be made ‘on-chain’, with governance rules hard-coded into the blockchain protocol and proposals decided by stakeholder voting. On-chain governance offers decentralised decision-making, permanent code changes, transparency and rapid consensus given established voting periods. But infrastructure for on-chain governance is incomplete and voter apathy can enable manipulation. ‘Off-chain’ governance, by contrast, is social. Away from the network’s code base, publicly or privately stakeholders try to persuade each other to accept or decline proposals. While offering benefits, off-chain governance can tend towards centralisation. Governance is, perhaps, the most fundamental challenge in blockchain — because it establishes the incentives for people to address all others — and must continue to evolve.
Poor user experience: few user-friendly experiences Today, except for buying and selling cryptocurrencies, only a small number of people engage with blockchain technology because there are few user-friendly experiences. Developing blockchain solutions is also challenging. When engaging with blockchain jargon abounds, addresses are meaningless streams of letters and numbers, and programming requires special tools such as the Solidity language for smart contracts. Emerging technologies require compelling user experiences to drive adoption. Blockchain solutions that offer attractive user experiences, such as Coinbase, can thrive. To expand beyond innovators and early adopters, blockchain-based solutions must offer compelling experiences.
Legal uncertainty: unresolved legal questions Many legal questions about blockchain ecosystems are unresolved. How enforceable are smart contracts? Which blockchain tokens (an innovation we describe below) should be classified as securities? In April 2019, the Securities and Exchange Commission (SEC) issued guidance, but further clarity is required and many questions are unresolved. Uncertainty, coupled with a ‘regulate by enforcement’ strategy, is challenging the blockchain ecosystem while causing difficulty for regulators.
Given the disadvantages and drawbacks of decentralisation, it is vital important to select appropriate use cases for blockchain deployment and investment. In many situations, decentralisation has more drawbacks than benefits. In some others, it offers unique possibilities. “With blockchain, it’s vital to select appropriate use cases for deployment and investment.”
Blockchain is suitable for select, existing use cases… Given the advantages and drawbacks of decentralisation, Blockchain may be valuable when:
a process involves the transfer of information or value; and
many parties seek to access and update a database; and
existing processes require overhead to deliver shared access to secure and trusted data.
The Financial Services sector is experimenting most with blockchain. In Financial Services, multiple parties seek to transfer value and it is burdensome to manage access to the secure, trusted data required. Transferring money, settling and clearing trades, providing escrow services and undertaking Know-Your-Client (KYC) checks all require intermediaries. If, instead, parties can interact with a trusted, shared ledger then: money can be transferred without banks; trades cleared without clearing houses; and transactions completed without escrow providers (via smart contracts). Increasingly, other sectors are engaging with blockchain. Financial Services’ share of blockchain projects has fallen from 75% in 2017 to 30% in 2019 (Fig. 3, Gartner) as manufacturers, healthcare organisations, media companies and retailers increasingly experiment with blockchain technology. Fig. 3: Increasingly, sectors beyond Financial Services are experimenting with blockchain. (Source: Gartner)…but has the potential to unlock ‘Web 3.0’ “Decentralisation could weaves the ‘missing layers’ into the fabric of the web.” While many blockchain projects seek to improve existing processes, long-term value creation could originate from blockchain-driven business model innovation. Blockchain offers profound new possibilities by enabling the decentralised web – ‘Internet 3.0’. Today’s internet has a big limitation: it doesn’t have a built-in mechanism to capture, or transfer, historic or current conditions — ‘state’. State represents information, which enables the capture and transfer of value. The internet lacks native state because of the limited set of ‘protocols’ on which it’s built. Protocols are rules agreed between parties that enable cooperation. In the 1990s (‘Internet 1.0’), protocols were established to enable the world wide web and email. The TCP/IP protocol enabled data exchange. HTTP powered communication between servers and web browsers. POP/IMAP enabled email. But many protocols were missing. For example, there are no protocols for identity (your name and information) or payment built into the web. In the absence of plentiful protocols, in the 2000s companies stepped up to capture data and value (‘Internet 2.0’). Facebook is useful because it’s a store of people’s identity — their information, interests, relationships and activities. Facebook captured data, which it monetised by displaying advertising. In a blockchain (‘Internet 3.0’) world, data is stored securely on shared databases. With common access to permissioned data, companies cannot create value through data lock-in. If users control their identity data, and permission it to authorised parties when required, they can easily swap between applications that use it. In an Internet 3.0 world, users could swap social networks as easily as we swap between internet browsers, or email clients, today. Blockchains, therefore, enable ‘protocolisation’. They offer open standards for the capture and transfer of information where protocols are absent. A decentralised future weaves the ‘missing layers’ into the fabric of the web. Sharing value across open networks — breaking data silos — could render old business models obsolete while enabling new ones, break monopolies, catalyse new products and services, and re-imagine stakeholder participation. We describe these possibilities below. Creating the decentralised internet requires great effort. Base layers in the software infrastructure ‘stack’ – from computation to storage and communication – must be developed. These, in turn, enable higher-order services. The market for decentralised applications (‘DApps’, Fig. 4), which replace centralised services with decentralised offerings in which users control their data, is in its infancy. Most of the 2,150 DApps live or in development today (stateofthedapps.com) are experimental. Few sustain more than a few hundred users. For many, functionality capability and user experience inhibit adoption. As base layer infrastructure matures, however, DApps should deliver greater utility and offer an on-ramp to ‘Internet 3.0’. Fig. 4: DApps, which replace centralised with decentralised applications, offer an on-ramp onto ‘Internet 3.0’ (Source: MMC Ventures, derived from Matteo Gianpietro Zago (https://bit.ly/2poFEqV))The ‘token economy’ offers new possibilities “Tokens support the creation, operation and adoption of blockchain networks — and offer new business models and opportunities.” The technical innovation of decentralised databases coincides with a business model evolution: tokenisation. ‘Tokens’ are digital resources that exist on blockchain networks. Coded to be limited in supply, they can be owned by individuals or entities. Tokens support the creation, operation and adoption of blockchain networks — and offer new business models and opportunities. There are three types of token:
Utility tokens offer holders access to a service.
Security tokens represent an ownership share in an asset.
Cryptocurrencies are digital currencies used to make or receive payments.
Tokens can power the creation of blockchains by enabling funding. In an ‘Initial Coin Offering’ (ICO), participants provide capital in return for tokens in a new network. Participants’ money funds network development. If the network prospers, the tokens rise in value and participants profit. The principle has potential and over 70% of the capital raised through ICOs in 2017 went to credible projects (Statis Group). But 80% of 2017 ICOs, by volume, were fraudulant, exploiting a flood of speculative money that flowed into the blockchain sector following cryptocurrency price appreciation in 2017/18 (Statis Group). The ICO bubble burst in 2018 — a welcome development within the community. Over time, a ‘plateau of reality’ is emerging. Some tokens support the operation of blockchains, by facilitating distributed consensus. On the Bitcoin blockchain, miners receive valuable tokens in return for validating transactions and creating the next entries on the ledger.
Tokens also drive blockchain adoption. Great value is accrued and delivered by today’s centralised applications — so why should investors, entrepreneurs, companies and users invest in decentralised systems? Stakeholders in a blockchain — including investors and network participants — frequently receive tokens in return for their support. Earlier participants tend to receive more. This creates powerful network effects. Generous token allocations incentivise early adopters. Their success attracts other participants. Over time, select protocols develop critical mass and create a magnet for future activity. Recently, people have sought greater flexibility in the incentive structures they could create – that is, more flexible tokens. Because tokens are expressed in software, developers can experiment with token characteristics to shape the economic systems they desire. In traditional economies, central banks influence outcomes by shaping monetary policy, including the supply of money and inflation rates. But their tools are limited, change is slow and impact is unclear. In token economies, options are myriad and impact is rapid. In token economies, developers can test the impact of ‘economic policy’ at the speed of software.
Blockchain could have profound implications Blockchain could have profound implications — including new services and business models, greater and more equitable innovation, and a shift in value in the software stack:
New possibilities: When combined, decentralised databases and tokens offer new value propositions and business models. Cryptocurrencies enable peer-to-peer payment and might become global stores of value independent from governments. Accessible data could catalyse the open source movement — if Facebook users could share their ‘likes’, to what uses could they be put? Decentralised applications could deliver more autonomous, secure services without relying on the whim of today’s platform providers. Smart contracts, which trigger token transfers when specified conditions are met, could replace select intermediaries such as escrow providers. New participation models, using tokens, could reward service users for their activity in ways previously impossible or impractical. More broadly: in the last four decades we’ve created digital representations of many objects, assets and events. But combining shared data (blockchains) with programmability (via tokens) presents myriad new possibilities.
Greater, more equitable competition: Many ‘internet 2.0’ companies create value by amassing people’s data and becoming its only repository. This creates strong network effects’— the more people who use a service (say, Instagram), the more useful it becomes to any individual. Network effects create a ‘catch 22’ that makes it difficult for new companies to gain market share, even if they offer better features. If data resides on a shared database, however, data network effects fall away. Companies can’t defend market share through data lock-in and high switching costs. More innovative solutions — ‘disruptors’ — can gain share more quickly and easily. More open, equitable competition will have secondary consequences. Given low switching costs, might blockchain applications retain value for shorter periods of time?
A shift in value capture: Traditionally, applications and platforms — not protocols — capture the value created on the internet. Google captures value through its email client, via advertising, but the creators of the POP/IMAP protocol that enables email exchange receive nothing. Shared databases might reverse this dynamic, so protocols capture more value than applications — the ‘fat protocol’ thesis. Why?
Application providers lose a means of value creation — ‘data lock-in’. If databases are shared, application providers all have access to the same data. They cannot create value my amassing individuals’ data (say, their ‘likes’) and becoming the only source of its availability.
Protocols gain a mechanism for value capture. Increasing activity on a blockchain protocol drives demand for the tokens that enable its operation— so tokens enable value to flow from applications to the protocols that power them.
Amara’s Law may ring true “People expect new technologies to have a linear impact over time, but new technologies follow ‘S-curves’ of influence.” ‘Amara’s Law’ posits that people tend to over-estimate the impact of new technologies in the short term, and under-estimate their impact in the long term. This is because people expect new technologies to have a linear impact over time, but new technologies follow ‘S-curves’ of influence. Expectations for new technologies can be high from the outset, but the immaturity of new technologies limits their use. The gap between expectations and early reality causes bubbles to form and burst (Fig. 5). Fig. 5: New technologies form ‘S-curves’ of adoption. The gap between expectations and early reality causes bubbles to form and burst. (Source: MMC Ventures / Amara).In 2018, lofty expectations collided with blockchain’s nascent capabilities — forming a ‘blockchain winter.’ Limitations in blockchain’s capability, scalability, governance and user experience meant there were few use cases where the advantages of decentralisation outweighed its benefits. Over time, however: scalability and user experiences are likely to improve; layers of blockchain infrastructure are emerging; companies are identifying more appropriate use cases; and the ICO bubble has burst and normalised. Blockchain’s greatest obstacle is likely to be the change it necessitates. Replacing entrenched architectures, investments and behaviours will prove an immense challenge. Artificial intelligence withstood seven ‘AI winters’ before it became mainstream.
Tim Berners-Lee developed the protocol for the web in 1989. 10 year later, in 1999, its potential was glimpsed — but technological, commercial and economic challenges brought expectations back to earth with a crash. 15 years after that, Tim’s vision for globalised information, e-commerce, and communication was realised. The Bitcoin white paper was published in 2008. 10 years later its potential was glimpsed — but technological, commercial and economic limitations brought expectations down to earth with a crash. Today, expectations have re-set while activity is growing. Investors, developers, companies and central banks are advancing blockchain. In the years ahead, Amara’s Law may ring true.