Blockchain, Crypto, and Hysteria
This article is for those who have passively consumed press about Bitcoin, Blockchain, ICOs, and the like over the past couple years, trying to track the changing topology of such a vast, complex landscape. So rather than focus on a particular portion of the rise and fall of cryptocurrencies and distributed ledger technologies, this article intends to build a map around the events, technology, and possibilities that continue to shape this winding narrative.
It was late fall, 2017, at the height of cryptocurrency mania. Bitcoin was a seemingly unstoppable rocket to the moon to an eventual peak of $18,000 US per coin. People sold all of their possessions, including family homes, in a frenzy to buy a few bits associated with an address on a global ledger.
I had flown into Las Vegas for a conference when I first realized this boiler was primed to explode. The biggest online wallet for storing cryptocurrencies, Coinbase, was on pace to become the first legitimate crypto unicorn based solely on buying and selling an unproven digital currency that was both highly insecure (the successful hacks were too numerous to track) and a terrible currency (a volatile currency is hardly useful). Upon learning that I was in the technology business, my Uber driver from the airport asked if he should buy Bitcoin or Litecoin. By November 2017, those coins were pretty pedestrian choices. When I arrived at my destination, although the conference was ostensibly about cloud technologies, the only topic of conversation were so-called altcoins, namely, Ripple (XRP), which was trading at 25¢ per coin. Five weeks later it would trade at $3.80 each, an over 1500% ROI in the time it takes to get a tax refund.
It was clearly a bubble, but how high would it go before the dip, and how far would it crash when it did? For those of us involved in trading crypto since the early 2010s, peaks and long winter valleys were the norm but it always rallied stronger in the next cycle. We were an insular group, namely trading amongst ourselves. The most famous early transfer of crypto funds was 10,000 Bitcoin (BTC) for two pizzas, valuing a single BTC at under half a cent each. Eight years later, that would be a $100 million pizza.
Cryptocurrency became a staple from CNBC to Fox Business, and institutional investors made claims that crypto should be a part of healthy portfolios. Crypto whales (high value individuals) were regularly featured on Wall Street Journal, and the leaders in the field were being hosted at global economic forums. Times, they were a changin’. At least, for a few months. There was a depression era joke that said if your shoeshine boy started talking about a stock, it was time to sell, and my Uber driver was a clear indicator that the burst was nigh.
We were being taken seriously, and it felt good. When the bubble burst in 2018, many were financially ruined, dreams were dashed. Benoit Coeure, executive board member from the European Central Bank (ECB), called Bitcoin, “a combination of a bubble, a Ponzi scheme, and an environmental disaster.” The party looked to be over before the headlining act took the stage. So much for revolutionizing the financial industry and supplanting fiat currencies.
A cottage industry grew around supporting crypto, but what to do now? A glimmer of hope remained for those dreaming of selling pickaxes (a gold rush metaphor for those who got rich, not by mining gold, but by selling pickaxes to all those eager miners dreaming of avarice). If the future of cryptocurrency was unclear, what about the underlying technology itself, blockchain? The infrastructure was there, the knowledge and expertise. To borrow a phrase from the startup world, the crypto players were making a pivot from currencies to blockchain.
To understand the current and future state of the blockchain world, let’s first cover what blockchain is and what it might be used for. Then we’ll dig into some technical details on how blockchain provides its unique value proposition, followed by an audit of some of the work left to do. At the end, we’ll revisit the question of whether blockchain is the technology you should be investigating, or should we be looking elsewhere for answers.
Blockchain aka Distributed Ledger Technology
Imagine a huge spreadsheet or ledger of transactions (financial or otherwise), a list of every transfer of funds or data from one party to another, for all time. If I send a payment from my account (delineated by a unique address) to a restaurant’s account to cover the cost of a cheesesteak sandwich, that transaction is permanently encoded in the global ledger. Blockchain is the technology that manages this ledger, and the currency built to operate on it is a cryptocurrency. Blockchain technology was initially invented for the creation and transfer of a non-fiat currency called Bitcoin, which spawned an entire ecosystem of thousands of competing cryptocurrencies of varying values.
What makes blockchain interesting for currencies, is that it creates a stable system in which the double-spending of funds is impossible, without the need of a bank or central authority. Over time this technology generalized beyond currency exchange, and we started calling them distributed ledger technology (or DLT). DLT technology can be thought of as a giant write-only database where data is spread across every computer in the network, rather than in a central location like a traditional database. This design has a peculiar tradeoff which is the source of its complexity and strength.
DLTs enforce the fundamental rule of irreversibility, where all new records added are immutable, unlike a regular database where you can fix data by updating it. This means, if I write data to the blockchain, it’s permanent. The only way to fix a mistake is when the affected party intentionally reverses it.
Let’s say I place an order on Amazon for a pair of shoes and a shirt, then change my mind and decide I don’t want the shirt. The standard database that Amazon uses lets me update my order to remove the shirt from my cart, leaving only the shoes. DLT works differently. If Amazon decided to track all purchases on a blockchain, then I have to live with my mistake. If I don’t want the shirt, I need to create a second transaction later, and ship the shirt back. While this a horrible attribute for shopping carts, it’s exactly the kind of behavior you want for financial transactions. And, it turns out, many other kinds of data.
A Trusted Third Party
How many institutions exist for the purpose of trust? When I write a check to my dog groomer, they must trust I have the cash to cover the amount, but we are not the only players in the equation. I’m leveraging the trust of my bank to pay the recipient from my account. If there’s any question for the need to trust the bank, try the direct approach on your next purchase by writing your grocery store a personal IOU for some apples. They’re unlikely to happily accept. It’s the trusted third-party bank that ensures my debt is paid. Central banks are a lender of last resort, playing a generally similar role between banking institutions. Fiat currencies are based on the trust of the backing government to guarantee the underlying value.
When you buy a house, an escrow agent is a trusted third party for holding funds that are meant to transfer between you and the seller until all terms are met. A notary acts as a third party that verifies you signed your closing documents. And when you visit your loan officer’s website, security is guaranteed by a trusted third party called a certificate authority, who vouches for the encryption of the data in a page you visited. In each of these cases, blockchain could be the trusted third party, acting as a middleman between you and who or what agent you’re dealing with. Rather than trusting a person, agency, or institution, you can instead trust a shared distributed technology network. It’s with this understanding that we can investigate some of the use-cases where distributed ledger technologies like blockchain can fit, from enterprise industries to consumer products.
Sustainability
Sustainability is a hot topic in DLT circles these days, because there are so many players in a sustainable system who must trust each other while also competing. Sustainability requires traceability and transparency of goods in a way that can be trusted by industry, auditors and the public. When the provenance of leather is tracked in every pair of sneakers, it’s pretty easy to prove whether a particular pair was farmed from a location that contributes to rainforest depletion. If every step in the manufacture of a high-end handbag is tracked from the factory, it’s possible to track if the workers are paid fair wages. IXO is one of many vendors in this space, tracking environmental impact of personal and institutional decisions, such as carbon emissions and credits. We’ll cover how it can be possible to ensure this sort of physical-to-digital tracking later.
Supply Chain
In the summer of 2018 the technology company IBM and global logistics provider Maersk announced a deal to co-create a new supply chain software company called TradeLens. This is normally the kind of announcement that few would bat an eyelash at, were it not for the crucial ingredient of IBM’s DLT called Hyperledger Fabric. The goal of TradeLens is to digitize the massive amount of paper that exists in the field of international logistics, from bills of lading to customs forms. The true innovation here is not yet another digital system with dreams of global domination (we already have SAP for that), but the network topology of Fabric. Imagine that your job was to convince governments to digitize their legal documents. If you go to Russia and ask them to put their customs forms on standard computer service, those servers have to be physically located somewhere. Russia is likely to demand that their data lives on servers that reside within their own borders. In fact, it’s safe to assume every country desires a similar option. Furthermore, each country would need guaranteed that transaction records were secure, unalterable, and auditable. Finally, the network should be safe from nation-state hack attempts, without a single point of failure. Luckily, we have such a technology in blockchain. In a very real way, DLT’s are one of the best technologies available to help governments trust in international data exchange, which is a huge step forward to digitalizing governments’ addiction to paper.
Healthcare
My first job out of college was working as a programmer at a large healthcare software company in the middle of Nowhere, USA. What immediately struck me was how inefficient the healthcare industry was, and hesitant to integrate health records between individual hospitals or networks. While there were plenty of integration standards (HL7 was all the rage 20 years ago), there was no national database of healthcare records in the US. It’s possible there never will be, unless patients start demanding it. With the growth of healthcare options, it’s increasingly important to individual patients to start keeping their own records, which again, requires a third party to trust. Or, alternatively, a DLT which contains a trusted history of all health records, where the patient can choose which records to share and with whom, independent of any particular health provider or even government. Medicalchain is an organization attempting such a feat. An interesting twist is that it allows users to even monetize their private health records for research purposes. Now that’s owning your own data.
In the US, 34% of healthcare costs are administrative. Anyone who has ever filled a prescription or scheduled a medical procedure knows the wait between the decision to proceed and the actual medical act. Much of this wait is due to insurance administration, from correctly coding the procedure to be covered to ensuring the right data exists for a prior authorization. A shared but common trusted system for transmitting this data instantly could reasonably bend healthcare costs by double digits. Moreover, the transparency between wildly different costs between identical procedures taking place in different hospitals might actually allow patients to force price competition and cost alignment into an opaque system.
Identity and Access Management
Like most Americans, I have a social security card, tattered, stained, and non-laminated. Because the USA, like many countries, lacks a national ID system, this little shredded paper contains an insecure, incremented (and when I die, recycled) SSID. This represents a standard identifier for a multitude of purposes, from attaining a credit card to getting a new job. That SSID is associated with thousands of databases, which each contain different, and potentially conflicting data about my identity, from credit scores to criminal history. A central, secure, distributed ledger of major life events that I could track, verify, and only share with others when I provide digital permission would be a major improvement to a smudged and tenebrous paper document. This kind of Digital Identity is a human authenticity guarantee to others that I am who I say I am. Oftentimes, that proof provides certain benefits (a social security check) or access (entry to a private club).
Beyond people, there are objects that need to be proven authentic, and rights bestowed upon individuals or systems that meet the criteria. Digital Rights Management via blockchain is a burgeoning area of research. Consider a star athlete or musician, who wants paid a nominal fee whenever their image is seen, or a song is played. Digital Ledger Technologies can track when a digital file is used, and charge for the right to access it. The contract is built entirely around use. Unless you pay, it can be impossible to decrypt or access the file. Moreover, you can pay in real-time for partial use. Say you listen to one minute of a three minutes song. A smart contract can extract micro payments against the file stream, and you only pay one third of the total price.
Voting
Voting machines are terrible. Every election season, we’re inundated with stories about the insecurity of voting machines, how easy it is to alter votes, and how impossible it is to verify that our votes were counted as we believed they should be. What’s the next best solution? Paper ballots. We can do better. An irreversible, trackable, anonymous (you get your own secret ID per election) but publicly readable (you can count the votes yourself if you like) distributed ledger of votes. I’m not even sure what to add here. This is something we should do, globally, immediately, for the integrity of democracy from here and into the future.
Carfax for X
I love cars. I grew up in Auburn, Indiana, south of Detroit, home of the Auburn line of cars, self-proclaimed classic car capital of the world and home to the Auburn-Cord-Duesenberg car auction every year. Buying a car is a necessity and rite of passage in my home town, but you don’t want to get taken for a sucker. “Show me the Carfax,” the commercial goes, a ledger of every relevant automotive event from wrecks to maintenance history. Having that history is valuable for the secondary market, helping prospective buyers make an informed decision. But there are plenty of other secondary markets for gently used items. Houses, smartphones, footwear (sneakerheads love their old Jordans), jewelry, or anything where a positive maintenance or authenticity verification makes the product more valuable. The trick here is getting as many players in the space to enter into the same network, by way of consortiums. This is a much harder problem to overcome than the specific DLT technology, but thanks to the network effect, some small set of blockchains will likely come out on top throughout a broad variety of industries.
Sharing Economy
Airbnb is one of the most valuable hotel chains in the world, despite not owning a single hotel. A similar comparison mostly holds for other brokers, like Uber and Lyft. One of the more exciting use cases of blockchain is the ability to rent out equipment that you own to others, without the need for a middleman taking a cut. ArcadeCity is an open competitor to Uber, a ride sharing app backed by blockchain, where anyone can post their intention to drive users around, who pay via token directly on the platform. No middleman other than the blockchain. OpenBazaar has slightly broader ambitions, ostensibly taking on Amazon or eBay or Etsy as an open marketplace where anyone can post or purchase goods. Where Amazon takes a 30% cut or charges a fee to open a store, OpenBazaar is a completely open platform, and the only fees are the small amount of funds required to run the code necessary to post or purchase goods and services. An interesting feature of OpenBazaar is that it provides an escrow feature, where any disputes lock funds until a mediator can intervene, proving that sometimes you just need a human in the loop.
Use-cases for trust abound, and we could fill an entire book on potential opportunities to leverage DLTs. Thus-far, you’ve had to trust me that this technology can be trusted. Now is time to dive deeper into exactly why and how we can be so sure that this trust is earned by drilling into the DLT that started it all, the Bitcoin blockchain and its Proof of Work mechanism.
A Gentle Breakdown of Proof of Work
There are many technologies one can understand with a scarce understanding of the technical details. Blockchain is not one of them. Its very soul is technical, and understanding its value is an exercise is subtly. Even if the implementation is complex, the concept is simple enough.
Fundamentally, Proof of Work (or PoW) makes it far too expensive, in terms of physical computing power and electricity, to alter history. Proof of Work runs on a cross-section of cryptography, statistics, and the physical reality of what humans can reasonably compute.
When a certain number of transactions (like, send two Bitcoin to my mom’s account) are made on the network, computer nodes called miners attempt complex math puzzles to bundle up the most recent transactions into a new block. If you think of transactions as rows on a spreadsheet, a block is basically a certain number of rows in a single file. The miners race to collect transactions and “sign” the next spreadsheet. The winning computer in this contest gets to permanently lock out all other nodes from making changes, so when a miner wins the contest, the remaining nodes move on to creating the next spreadsheet in the chain.
The sequence of all blocks create an ordered series of transactions across the planet called the blockchain. It is, quite literally, a digital chain of blocks.
Winning the Blockchain Lottery
When a miner collects a list of transactions to build a block, they encode all of this and other data as a hash. A hash is just a big number, common in cryptography, that can take any amount of data, and encode a single fixed value. For example, the word “breakdance” hashes (in the SHA256 algorithm) to B0A9E90D61F43A0F5166593523CD6B661AA27AE7D871ADB3A1CF4B4C67E3C5CA. If you were to hand this number to someone, they could never convert that number back to the word “breakdance”. It’s a one-way check. However, if you were given the word “breakdance” you could always generate that long hash number, exactly the same, every time. When slight changes are made to the block (via an incremented digit called a nonce), and the correct hash is generated according to certain rules (for example, the number must start with 18 zeroes), the miner wins.
Let’s consider a game where you throw 50 pennies into the air, and they each have to land on heads to win. The odds of any throw working out your favor are 1 in 2⁵⁰, or about one in one quadrtillion (one followed by 15 zeroes). This is about the odds of a miner winning the chance to create the next Bitcoin block on the first try. If it sounds suspiciously like a lottery, it very much is. Although the chance of winning the race to make the next block has very low probability in a single attempt, multiply this by millions of computers making billions of attempts, and a new block is statistically likely to be created somewhere on the internet within ten minutes. When a computer finally generates the winning hash, it informs the rest of the network, where it’s pretty easy for other computers to verify the winner.
The integrity of the blockchain is proven due to the sheer amount of work necessary to generate a new valid block that hashes correctly, hence, Proof of Work. It takes a tremendous amount of computing power, and thus electricity, in order to stumble upon the right block hash. That’s the brilliance of the system, it is far more complex to generate the winning number than it is to validate that the number is correct, so faking history is nigh-impossible, but verifying history is trivial.
From Block to Blockchain
Now that we’ve created a block, how does that make a chain? For simplicity, we had skipped over a bit of detail about the block. Blocks are not exclusively a spreadsheet of transactions. They also contain other data about the block itself and its place in the network, called the block header. This metadata is as important as the transaction ledger data.
The block header contains mining values (like a timestamp, number of transactions, and a nonce), a rollup value of the transactions and most importantly, it contains the hash of the previous block. The current block points at the hash of the last block, which itself points at the hash of the prior block, and so on. It’s like a backward conga line, where everyone places their hands on the hips of the person behind them. Every block points at the previous block, making a chain all the way back to the very first block ever created called the genesis block.
So that’s it! You’ve now built a chain of signed blocks that are reasonably impervious to attack by humans on earth with modern technology.
In Bitcoin, the genesis block was created January 3, 2009 by the original mysterious author, who wrote under the pseudonym Satoshi Nakamoto. As proof that the block was created on or after January 3, 2009, it contains the following note: `The Times 03/Jan/2009 Chancellor on brink of second bailout for banks`. For a currency meant to disrupt the financial status quo, it was quite a comment on the fractional reserve banking structure.
To recap thus far, in severe jargon: miners gather transactions into a block, and iteratively increment a nonce in an attempt to generate a winning cryptographic hash while ensuring the newest block header contains the hash of the previous block. This linked list of blocks creates a write-only ordered ledger of all transaction records representing the transfer of cryptocurrency from one address to another. The entire list of ordered blocks starting from the genesis block is called the blockchain, which is a kind of distributed ledger technology or DLT. Don’t worry if you blanked on some specific terms, it’s the concept that matters.
Who are these Miners?
We still haven’t explained why miners take part in this work. The short answer is money, because the reward for mining a block is paid in Bitcoin. Mining is not only how the transaction ledger maintains integrity, it is also how new currency is introduced to the network. When a miner wins the block creation lottery, the miner is rewarded by having new bitcoins created and “deposited” to the miner’s own address. The reward for generating a single winning block can be worth tens of thousands of dollars each, with a new lottery every ten minutes. This is a billion-dollar industry.
Let’s pause to consider the brilliance of this system. Because of the difficulty in mining a new block, it’s effectively impossible for anyone to modify records encoded in the network. Since each block contains the hash of the previous block, the only way to rewrite history is to rebuild every block for all time, which isn’t physically possible on the Bitcoin blockchain given the world’s computing power by cost. Furthermore, because there’s a potential financial reward for taking part in verifying transactions, there’s a global network of computers fighting for the chance to protect all transactions. Because those miners are rewarded in Bitcoin, they have a financial incentive to ensure the integrity of the system as a whole. If any particular group got greedy and decided to mine so much that they effectively owned a majority of the network, it could destabilize the value of Bitcoin since users would no longer trust it. The miners would then have a majority share of a worthless asset. In other words, everyone has an interest in taking part in the network and protecting its integrity. This is a network designed, not merely on cryptography (mining) and statistical likelihoods (block creation will eventually happen somewhere), but also rooted in game theory and ultimately human psychology. The more stake an individual has in the network, the higher their incentive to ensure its integrity.
With this basic structure in place, some folks realized that the Blockchain ecosystem could be leveraged for more than storing data through pre-defined transactions. It could be generalized further into a global network of general-purpose computing resources. This is the turning point in our story where everything changed.
Initial Coin Offerings
The blockchain use-cases for most early cryptocurrencies are similar and emerge from a sort of libertarian appeal to avoid central control and fiat manipulation at all costs. The blockchain is public, ensuring it provides a radical transparency into all transaction history, to ensure no alteration has occurred. The network must also be trustless, which means that all computers on the network are equal, and anyone can take part in trading on, or operating (mining for) the network. There can be no Animal Farm scenario, where some computers are more equal than others. No miner is privileged above any other, there is no central authority or control.
But humans are still involved, so philosophical conflicts happen. These disagreements between experts were often resolved by making changes, or “forks” in the code that the miners choose to run, effectively creating new currencies. This has happened many times, most famously, a disagreement between groups of programmers in the Bitcoin protocol forked the blockchain into two competing currencies: Bitcoin (BTC) and Bitcoin Cash (BCH). But even when a fork in the codebase happens, there’s no loss or rewrite of history, it’s just that two roads diverged in a wood and different people chose different paths going forward.
It’s against this backdrop in 2013 that a nineteen-year-old programmer and Bitcoin enthusiast named Vitalik Buterin had an idea. Rather than trying to make the core protocol of Bitcoin everything to everyone, instead create a blockchain network that is capable of running custom software. Furthermore, why not let the miners earn money executing useful code instead of merely racing to see who can generate the biggest random number fastest? Sure, there would still be mining, currency transfers, and block creation, but there would also be a readily available computation platform where anyone could pay to execute distributed applications (a.k.a. dApps). Unlike the Bitcoin blockchain network which only trade Bitcoins, anyone could code their own tokens bound by whatever rules they want.
In 2015 the network he and others launched was named Ethereum (née Frontier) on the back of a crowdfunding campaign that raised $18 million USD. This network has its own cryptocurrency named Ether (ETH), which can be traded, similar to Bitcoin, but can also be used as a reward to miners who were willing to execute code. You can think of the Ethereum network as a giant, global virtual computer made up of millions of physical computers, ready to execute any code you pay them to run. It’s with this view in mind that the Ethereum execution network is called the Ethereum Virtual Machine, or EVM. Anyone who knows how to program an EVM scripting language, Solidity, can write code that others can run on the network. To ensure idempotency, the source code is stored on the blockchain, just like a financial transaction might be.
These dApps are public and execution is irreversible once someone has chosen to run the code. For those with a broad imagination, they are akin to legal contracts that are self-executing, earning them the name smart contracts. Since smart contacts are just computer code, there are few limits on what behaviors you can code into them. You can set up a smart contract where any user can rent a car, and while you drive the car the contract will continuously transfer funds to the car’s owner until you take out the key. You can create a smart contract that attaches to a beer tap and charges per pour. You can create a smart contract that generates images of cartoon cats and sells those images on a marketplace of cartoon cat breeders (seriously, this is a real thing called CryptoKitties).
The DAO
In early summer of 2016, some of the core creators of Ethereum had a prescient idea: if you could codify a sophisticated enough smart contract, could you design an organization that was run largely autonomously? Any interface between the organization and individuals would be dictated by this contract, effectively rendering the org little more than a set of rules, a pool of currency, and a governing body whose governors were elected by shareholders and limited to actions explicitly in the contract. Anyone could be a shareholder merely by putting money into the system, through execution of a public smart contract. Any payment of funds for work required approval by the governors, and such officers could be removed and replaced at any time by a shareholder vote. The organization was a distributed autonomous organization (or DAO).
This large public offering had some analogs to the kind of initial public offering (IPO) of stock that you would expect to buy through a stock exchange like NASDAQ. Since your ownership in the company took the form of tokens or coins in the Ethereum network, rather than stock certificates, this kind of public crowdfunding of a blockchain based business became known as an initial coin offering (or ICO).
The DAO ICO was an exciting time and heralded a new form of both business and funding model. In a world where Worldcom and Enron executives enriched themselves at the expense of shareholders, such total transparency was a breathtaking experiment. The DAO raised 11.5 million ETH from over 11,000 individual investors (including yours truly), valued at the time at $150 million US.
What the DAO really was, ultimately, was an exercise in perfect bureaucracy. No one was above the law, but there was also no real authority to appeal to when something went wrong. In a regular contract, honest disputes can be resolved by courts, but what recourse against a conceptual flaw in a contract that is publicly executable and irreversible? The fundamental error of the DAO centered around a bug, not in technology or software, but in logic. A user found a flaw in the DAO contract, and drained the account of nearly $50 million US before the community was able stop the bleeding. Many loathe to call this a hack or a theft, and that’s because this person’s actions where not so clear cut. He or she did not break any rules of the system as designed, but merely outsmarted the thousands of people who had put funds into the DAO account. What the user did was entirely within the bounds of the contract as designed.
Following a rapid community debate and vote on what to do next, the core Ethereum developers opted for the nuclear option, to fork the blockchain and return funds to the investors. This fork created two currencies, becoming the mainline Ethereum (ETH) for the majority, then Ethereum Classic (ETC) for those who believed that maintaining the integrity of the blockchain is more important than $50M. Ironically, a blockchain designed to avoid forks was ultimately forked.
The Rise of the ICO Phoenix
The failure of the DAO would have been a black eye for any emerging technology, but history has shown that this loss of funds deterred no one. The takeaway that the community learned wasn’t one of tempering irrational exuberance (18 months later ETH would be trading at a 13000% increase), but rather that you could raise large sums of money without any product at all. Just an idea and a smart contract raised $150M on an unclear experiment, imagine how much you could raise with a bigger idea! And thus, the ICO boom was born, a phoenix from the ashes of the DAO, fueling a two-year race to raise ever larger amounts of funds with very little effort.
In 2018, ICO mania drive an overall market of untested startups to raise over $13 billion US. The most sophisticated single project named EOS raised over $4 billion US to build an alternative to the Ethereum network. The jury is still out on whether they’ll be successful, but it was very attractive to those who felt as if they had missed out on the early days of ETH and BTC. FOMO, fear of mission out, was a powerful driver of many ICOs. Initial coin offerings were able to blur the line between holding a stake in a company (a security), holding tokens that can be transferred for goods and services, and an asset like Bitcoin that you can use as a general currency.
So, while the DAO launched the mania for companies run on smart contracts, why go through all of the work of building a company, when you can just launch an alternative currency and hype it on Ethereum? This new set of coins were collectively called altcoins. The set of coins that all ran on the EVM were so similar, they even created a standard protocol called ERC20. This ICO mania was inexorably linked with the hunger for more cryptocurrencies to invest in, which fueled the creation of more altcoins. For a brief time in 2017, anyone with a little programming ability and gift of hype could be a millionaire by merely copying an existing altcoin smart contract and create a slick website to sell the new currency. Over the span of months, most ICOs failed, either through incompetence or outright fraud. This placed a chilling effect on market exuberance and the value of blithely generating new coins.
With the promise of instant millions a fading dream, and the value of cryptocurrencies plummeting, is the market dead? Is there nothing left of this grand experiment? It’s not quite time to despair, because as we’ve already explored, there are other uses for a write-only distributed ledger technology. But there are some technical problems we need to work through first before DLT is ready for a broader adoption.
An Imperfect Storm (Curing the Ills of DLT)
We ran through a few use-cases for DLTs and there are countless more. We checked out how a type of DLT works, and how that expanded into dApps. But we glossed over many weaknesses of the DLT technology as we’ve described it, and how the greater blockchain industry is working to resolve them, with varying degrees of success.
Trusting the Physical World
The power of blockchain relies on the fact that any record placed in the blockchain cannot be altered, mutated, or reversed. But we have to keep in mind the old computer information mantra, “garbage in, garbage out” (or GIGO). Just because blockchain ensures data is irreversible, how do you trust that any data put into the system can itself be trusted? If someone creates a blockchain that tracks my heart-rate in order to get a discount on life insurance for good results, it’s only useful if my vitals are read by a device that itself can be trusted to be correct, whose results cannot be tampered with, and can securely submit those results to the blockchain. If the underlying blockchain requires me to manually submit my vitals on a monthly basis, I have every incentive to lie about bad results to get the discount. It has to be secure all the way down. When blockchains are populated by external values, the trusted systems that convert data from the real world to the blockchain are called oracles. Oracles are only useful insomuch as they can be trusted, so tend to enforce that trust with encryption that can only be decrypted by the smart contract, and secure hardware that ensures the values aren’t tampered with in some other way. Anything less that total roundtrip security can open the blockchain up to garbage data, which make it difficult to rely on. GIGO. One of the more recent companies looking to solve this problem through secure hardware is Chainlink. It’s an interesting option for tying IoT data to a blockchain.
The Problem with Large Files
There’s a lot of talk about blockchain for trading large files, from healthcare records to digital contracts. But think back to how blockchain actually works. It’s much more akin to a spreadsheet of transactions than a large file store like iCloud or Dropbox. If someone creates an application that stores legal documents somewhere on a standard cloud service, and signing that document creates a record in blockchain, what stops me from modifying that document directly on the cloud file store? Or just outright deleting the file? Sure, the signature is there, but what exactly did I sign? Just like our problem with trusting the physical world, we need to trust that large files are distributed, always accessible, and that changes are irreversible. There are a few blockchain-based solutions for this issue. The first was IPFS, or Interplanetary File System. What makes IPFS interesting is that it starts with a blockchain-like write-only data structure (Merkle DAG) to ensure that every change to a file is kept throughout history and combines that with a similar concept behind the famous peer-to-peer file-sharing protocol BitTorrent. Servers store pieces of various files and are rewarded the most for holding onto the rarest blocks, ensuring that every piece of every file in the network is held somewhere by some computer. Reconstructing those pieces is as simple as asking the network where those blocks are and connecting them. Storj is another large file system, along with a handful of others. What matters most is that these file stores have features similar to blockchain: distributed and irreversible. As long as such large file stores exist, blockchain can leverage pointing to the file and version being transacted against.
Blockchains are Slow
One of the oldest concerns about the Bitcoin blockchain is how slow it was to resolve a transaction. The Bitcoin network famously supported nine transactions a second, and each of those transactions take ten minutes or more to resolve. Many fixes have happened over the years, with various solutions. There are side-chains, which are basically stand-alone blockchains with custom rules that can eventually converge into the main blockchain. There is the Lightning network, that is a type of temporary private channel, dedicated to supporting many rapid transactions between a limited number of traders, whose final result are broadcast to the main chain. The Ripple network can validate a transaction in 4 seconds. Zilliqa has a blockchain which can validate 15,000 transactions a second. The credit card company Visa can support 24,000. An emerging blockchain called Ternio is working on an implementation that can cover 1 million transactions per second. Other laboratory tests have even higher rates, one claiming even 7 million per second. What all of this says is that the famous slowness of early implementations are being resolved while still remaining distributed irreversible ledgers.
Environmental Impact
This one is a big deal. While the benefits of a blockchain architecture are myriad for some of the use-cases we’ve outlined, and many more, it’s hard to make claims that DLTs based on Proof of Work are a good thing for the environment. As of 2018, executing a single transaction took the same amount of energy to power about 25 households for a day in the US. Based on the given rate of Bitcoin transactions, that is equivalent to the energy consumption for the entire country of Austria. Essentially, the lottery system we looked at is a way to make it very hard for someone to generate a block, theoretically giving everyone a fair shake at getting to make the next one based statistically on how much electricity they’re willing to burn. But there are other ways of creating new block, generally called consensus protocols.
The top alternative contender is called Proof of Stake (or PoS), which rewards those who are willing to put the most stake (a.k.a., cash) into the election system. The theory being, those with the most to lose are the most likely to ensure the integrity of the system. If you’re a bad actor, you’ll be punished financially. What’s considered “bad” depends on the system, but commonly bad is voting against the common consensus. In Casper, the PoS system for Ethereum, the set of conditions that punish bad actors is called Slasher. Then, there is Proof of Authority (of PoA), which gives a sort of political bent to PoS, where stake and being a good actor earns you authority, giving you a right to vote. Similar to PoS, acting badly means punishment loss of stake as well as being demoted as a blessed Authority. Another method of leader election championed by Intel is called Proof of Elapsed Time (or PoET), where solving secure puzzles in a given timeframe help choose. Solving those puzzles requires specialized trusted execution environments that Intel happens to manufacture in a chip form called SGX. There are other proofs, Proof of Activity, Proof of Burn, Proof of Replication, and so on. While it’s still too early to tell which, if any, will emerge the victor, there is plenty of work in the space to cure blockchain of its addiction to electricity.
Public and Trustless
While public tractability of all transactions is a good feature for promoting trust in cryptocurrencies, it’s not the best quality for a DLT system designed to store personal healthcare records. And while the trustless nature of blockchain is the killer feature for Bitcoin, an inter-governmental platform shouldn’t allow just anyone to join their customs data network. The first decade of DLT was all about adding features to a public, trustless blockchain. Now academic and industry eyes are focused on building private, trusted blockchains. While the technical details of how they differ can vary, the use cases tend to be much more aligned with the need for large organizations like nations and industrial players to share data with each other in closed marketplaces. Corda and Fabric are two leaders in this space. Many private blockchains look to leverage the mindshare of the existing developer community by making available the same smart contract interface of Ethereum’s Solidity code, but repurposed for a closed infrastructure.
While it’s probably correct to think that these blockchain implementations are antagonistic to Satoshi’s promise of radical transparency and decentralization via blockchain, it’s also possible that these kinds of quasi-private consortiums are going to have the biggest and longest impact long-term. While the jury is still out on the world’s hunger for a hard digital currency with no government backing, the future is bright with possibilities for DLTs that support increased transparency between large organizations and states.
The End of Vertical Integration
In the early days of 2019, cryptocurrency values are dropping to an unknowable floor. The majority of ICOs are failures, and fraud in the blockchain space looms rampant. Not to mention all of the environmental impact of Proof of Work, the remaining technical unsolved issues that plague DLTs, and the fact that the space is still so nascent and misunderstood that it seems hard to grasp specific use-cases that DLTs can solve better than existing distributed databases. You’d be forgiven for harboring a sense of futility around the whole exercise.
It’s important to separate the turgid, get-rich-quick hype of cryptos fueled by ICOs from the technical guarantees of DLTs. There are clear use-cases for distributed irreversible ledgers ranging from personal conveniences to governmental reform. One of the exciting sea changes is how DLTs can evolve enterprise companies of the future into smaller, leaner, and distributed networks with well-defined contracts as code.
The nearest term value of blockchain rests in leveraging the technology to build consortiums of companies with a common alignment, where there’s relatively little risk for joining, for example, distributed authenticity tracking to fight black market counterfeits or sustainable circular economies in cotton production. This first wave of consortiums are a first pass at decentralization, allowing companies to relax centralized control in a low-risk way. The benefits of giving up certain rights of governance in order to access the economies of scale of an orderly cooperative will be too good to pass up.
As organizations become more comfortable joining networks of other organizations, backed by a technology infrastructure that make such consolidations effortless, this may even expand into increasingly core operations that were once vertically integrated in favor of open marketplaces. For the buyers, exposing the details of their operations will allow providers to better service the enterprise, while membership for the sellers evolves from competitive advantage to necessity.
Consider the case of transportation. Why would a grocer choose to have a contract with a single large-scale shipping company, when they can instead place bids on a marketplace populated by a consortium of transportation agents at a lower cost, higher quality, with full transparency. Attached to each agent has a set of services that they can provide, and a history of outcomes, where the buyer merely places a transportation bid against the consortium’s smart contract with certain defined requirements and accept the lowest qualified bid. With such an open marketplace, why would a driver choose to stay with Uber when they can work directly with the buyer? Uber, for their part, would eventually need to join this consortium or at least create one of their own, perhaps with other major players. It’s hard to push back forever against the relentless force of strong atomized networks.
While the DAO (distributed autonomous organization) experiment didn’t immediately work for those of us who invested early, the concept can’t be dismissed out of hand. Organizations will be increasingly distributed, and that such distribution will be held together by autonomous operations. Robot Process Automation (RPA) is happening globally today and giving way to an AI based version called Cognitive Automation (CA). As more operational processes are automated internally, in increasingly intelligent ways, eventually that automation will tread into operational tasks that interface with external agencies. The increasing automation will require a system through which companies can engage with rules that enforce good behavior. There will always be cheats, and AI has shown to be one of the greatest cheaters in history, so the system itself needs to keep an irreversible ledger of all historical operations and enforce certain rules, for which smart contracts would be a reasonable fit.
Finally, we have to admit, it’s possible that blockchain as a technology is not the right fit for many of the use-cases we’ve discussed. What’s hard to argue is that we do and will increasingly require systems that can act as trusted third parties, by way of allowing data to be shared between individuals and organizations that may not necessarily have honest intentions. In this case, the system itself must enforce its own rules, and such a system requires the trust of its users. That trust can be a collective that everyone agrees has the right to act on everyone’s behalf (i.e., a governing body) or a technology system that by its very nature disallows all but very specific behaviors (i.e., smart contracts, dApps). While it’s very possible what we currently call blockchain may be antiquated in a few years, what’s likely to continue persisting is the need for some kind of smart distributed ledger technology. It’s worth pursuing now. Since the future of corporations trend to more networks and relationships between entities, at the very least, blockchain can act as training wheels for this inevitable shift.
