The first time I bought virtual money, in October 2017, bitcoins, the cryptocurrency everyone by now has heard of, were trading at $5,919.20. A month later, as I started writing this, a single coin sold for $2,000 more. “Coin” is a metaphor. A cryptocurrency such as bitcoin is purely digital: it is a piece of code—a string of numbers and letters—that uses encryption techniques and a decentralized computer network to process transactions and generate new units. Its value derives entirely from people’s perception of what it is worth. The same might be said of paper money, now divorced from gold and silver, or of gold and silver for that matter. Money is a human invention. It has value because we say it does. (...)
The central obstacle to a fully automated monetary system run exclusively by computers is validation: how to ensure that the transactions on the network are legitimate. The bitcoin software devised by Nakamoto employs a number of features to deal with this. The first is basic encryption. A bitcoin is nothing more than a record of value—you have seven bitcoin, I have five bitcoin, and so on—encoded and stored on the bitcoin system as an address. To release that bitcoin to buy something or to cash it out, its owner must use a private encryption key, known only to him or her, which is associated with that account. Matching the private key with the address is done automatically by the decentralized network of computers. If they don’t match up, or if the owner of the private key is attempting to spend his or her bitcoin more than once, the computers reject the transaction.
The “miners” who verify and collect these transactions into a block—“miners” being a term for those who run the computers on the network—are also required by the bitcoin software to perform an additional validating function before the block can be added to the bitcoin ledger. Called “proof of work,” it is essentially a computational lottery in which all the mining computers vie to guess an algorithmically generated number between zero and 4,294,967,296 with the correct number of zeros preceding it. Finding the target number takes trillions of guesses and a tremendous amount of computing power.
The idea behind “proof of work,” according to Daniel Krawisz, of the Satoshi Nakamoto Institute, is that it is “an added complication, like a ritual, so as to make blocks more difficult to generate…. [It] is…a means for a group of self-interested people, none of whom is subordinate to any other, to establish a consensus against a considerable incentive to resist it.” Because it takes so much computing power to find this number, miners are motivated to ensure that the transactions they are processing are valid and nonconflicting. But they are motivated to participate in the first place because the software generates a reward: the miner who finds the “proof of work” number first is paid in (an algorithmically determined number of) bitcoins. Though that is how new bitcoins are created, or “mined,” and added to the system, as the Tapscotts point out, mining is
Still, it’s not exactly free money. Marco Streng, the cofounder of Genesis Mining, estimates that it costs his company around $400 in electricity alone to mine each bitcoin. That’s because bitcoin mining is not only computationally intensive, it is energy-intensive. By one estimate, the power consumption of bitcoin mining now exceeds that of Ireland and is growing so exponentially that it will surpass that of the entire United States by July 2019. A year ago, the CEO of BitFury, Valery Vavilov, reckoned that energy accounted for between 90 and 95 percent of his company’s bitcoin-mining costs. According to David Gerard—whose new book, Attack of the Fifty Foot Blockchain, is a sober riposte to all the upbeat forecasts about cryptocurrency like the Tapscotts’—“By the end of 2016,” a single mining facility in China was using “over half the estimated power used by all of Google’s data centres worldwide at the time.”
One way bitcoin miners offset these costs is by collecting the very thing digital money, traded peer-to-peer, was supposed to make obsolete: transaction fees. By one estimate, these fees have risen 1,289 percent since March 2015. On any given day, the fees will be in the millions of dollars and now cost upward of twenty dollars per transaction. While transaction fees are not mandatory, they are a way for users to attempt to jump the queue in a system rife with bottlenecks, since those who offer miners a fee to have their transactions included in a block have a better chance of that happening. With so many transactions lined up, waiting to be processed, miners have discretion over which will make it to the head of the line; the higher the fee, the more likely it is to be chosen. As the explanatory website Unlock Blockchain puts it: “when miners mine a block, they become temporary dictators of that block. If you want your transactions to go through, you will have to pay a toll to the miner in charge…. The higher the transaction fees, the faster the miners will put [the transactions] up in their block.” As a consequence, transactions can be held up for hours or days or dropped altogether.
Bitcoin’s high transaction fees and slow transaction times were two of the reasons I chose to buy ether. But there was another reason as well: while bitcoin was invented to bypass traditional currency by tendering a new kind of money, ether, another cryptocurrency that can be bought, sold, and used to purchase goods and services, was created to raise capital to fund a project called the Ethereum network. The principals behind it are building out what is being trumpeted as the next iteration of the Internet, Web 3.0, also known as “the blockchain.”
A blockchain is, essentially, a way of moving information between parties over the Internet and storing that information and its transaction history on a disparate network of computers. Bitcoin, for example, operates on a blockchain: as transactions are aggregated into blocks, each block is assigned a unique cryptographic signature called a “hash.” Once the validating cryptographic puzzle for the latest block has been solved by a mining computer, three things happen: the result is timestamped, the new block is linked irrevocably to the blocks before and after it by its unique hash, and the block and its hash are posted to all the other computers that were attempting to solve the puzzle. This decentralized network of computers is the repository of the immutable ledger of bitcoin transactions. As the Tapscotts observe, “If you wanted to steal a bitcoin, you’d have to rewrite the coin’s entire history on the blockchain in broad daylight.”
While bitcoin operates on a blockchain, it is not the blockchain. The insight of Vitalik Buterin, the young polymath who created Ethereum, was that in addition to exchanging digital money, the blockchain could be used to facilitate transactions of other kinds of digitized data, such as property registrations, birth certificates, medical records, and bills of lading. Because the blockchain is decentralized and its ledger immutable, those transactions would be protected from hacking; and because the blockchain is a peer-to-peer system that lets people and businesses interact directly with each other, it is inherently more efficient and also cheaper than systems that are burdened with middlemen such as lawyers and regulators.
A company that aims to reduce drug counterfeiting is using the blockchain to follow pharmaceuticals from provenance to purchase. Another outfit is doing something similar with high-end sneakers. Yet another start-up, this one called Paragon, is currently raising money to create a blockchain that “registers everything that has happened to a cannabis product, from seed to sale, letting consumers, retailers and the government know where everything came from.” “We are treating cannabis as a normal crop,” Paragon’s founder and CEO Jessica VerSteeg, a former Miss Iowa, told a reporter for the website Benzinga. “So, the same way that you would want to know where the corn on your table came from, or the apple that you had at lunch came from, you want to know where the weed you’re consuming came from.”
While a blockchain is not a full-on solution to fraud or hacking, its decentralized infrastructure ensures that there are no “honeypots” of data available for criminals to exploit. Still, touting a bitcoin-derived technology as the answer to cybercrime may seem a stretch in light of the high-profile—and lucrative—thefts of cryptocurrency over the past few years. Gerard notes that “as of March 2015, a full third of all Bitcoin exchanges”—where people stored their bitcoin—“up to then had been hacked, and nearly half had closed.” There was, most famously, the 2014 pilferage of Mt. Gox, a Japanese-based digital coin exchange, in which 850,000 bitcoins worth $460,000,000 disappeared. Two years later another exchange, Bitfinex, was hacked and around $60 million in bitcoin was taken; the company’s solution was to spread the loss to all its customers, including those whose accounts had not been drained. Then there was the theft via malware of $40 million by a man in Pennsylvania earlier this year. He confessed, but the other thieves slipped away, leaving victims with no way to retrieve their funds.
Unlike money kept in a bank, cryptocurrencies are uninsured and unregulated. That is one of the consequences of a monetary system that exists—intentionally—beyond government control or oversight. It may be small consolation to those who were affected by these thefts that neither the bitcoin network nor the Ethereum network itself has been breached, which perhaps proves the immunity of the blockchain to hacking.
The central obstacle to a fully automated monetary system run exclusively by computers is validation: how to ensure that the transactions on the network are legitimate. The bitcoin software devised by Nakamoto employs a number of features to deal with this. The first is basic encryption. A bitcoin is nothing more than a record of value—you have seven bitcoin, I have five bitcoin, and so on—encoded and stored on the bitcoin system as an address. To release that bitcoin to buy something or to cash it out, its owner must use a private encryption key, known only to him or her, which is associated with that account. Matching the private key with the address is done automatically by the decentralized network of computers. If they don’t match up, or if the owner of the private key is attempting to spend his or her bitcoin more than once, the computers reject the transaction.
The “miners” who verify and collect these transactions into a block—“miners” being a term for those who run the computers on the network—are also required by the bitcoin software to perform an additional validating function before the block can be added to the bitcoin ledger. Called “proof of work,” it is essentially a computational lottery in which all the mining computers vie to guess an algorithmically generated number between zero and 4,294,967,296 with the correct number of zeros preceding it. Finding the target number takes trillions of guesses and a tremendous amount of computing power.The idea behind “proof of work,” according to Daniel Krawisz, of the Satoshi Nakamoto Institute, is that it is “an added complication, like a ritual, so as to make blocks more difficult to generate…. [It] is…a means for a group of self-interested people, none of whom is subordinate to any other, to establish a consensus against a considerable incentive to resist it.” Because it takes so much computing power to find this number, miners are motivated to ensure that the transactions they are processing are valid and nonconflicting. But they are motivated to participate in the first place because the software generates a reward: the miner who finds the “proof of work” number first is paid in (an algorithmically determined number of) bitcoins. Though that is how new bitcoins are created, or “mined,” and added to the system, as the Tapscotts point out, mining is
an awkward analogy because it conjures images of experts whose talent might confer some competitive advantage…. It doesn’t. Each miner is running the software like a utility function in the background, and the software is doing all the computations…. There’s no skill involved.When the bitcoin network began operating in 2009, people could run the validation program on their personal computers and earn bitcoins if their computer solved the puzzle first. As demand for bitcoin increased, and more people were vying to find the random, algorithmic proof of work validation number, speed became essential. Mining began to require sophisticated graphics cards and, when those proved too slow, special, superfast computers built specifically to validate transactions and mine bitcoins. Individual miners have dropped out for the most part, and industrial operators have moved in. These days, mining is so computer-intensive that it takes place in huge processing centers in countries with low energy costs, like China and Iceland. One of these, in the town of Ordos, in Inner Mongolia, has a staff of fifty who oversee 25,000 computers in eight buildings that run day and night. A company called BitFury, which operates mining facilities in Iceland and the Republic of Georgia and also manufactures and sells specialized, industrial processing rigs, is estimated to have mined at least half a million bitcoins so far. At today’s price, that’s worth around $7.5 billion.
Still, it’s not exactly free money. Marco Streng, the cofounder of Genesis Mining, estimates that it costs his company around $400 in electricity alone to mine each bitcoin. That’s because bitcoin mining is not only computationally intensive, it is energy-intensive. By one estimate, the power consumption of bitcoin mining now exceeds that of Ireland and is growing so exponentially that it will surpass that of the entire United States by July 2019. A year ago, the CEO of BitFury, Valery Vavilov, reckoned that energy accounted for between 90 and 95 percent of his company’s bitcoin-mining costs. According to David Gerard—whose new book, Attack of the Fifty Foot Blockchain, is a sober riposte to all the upbeat forecasts about cryptocurrency like the Tapscotts’—“By the end of 2016,” a single mining facility in China was using “over half the estimated power used by all of Google’s data centres worldwide at the time.”
One way bitcoin miners offset these costs is by collecting the very thing digital money, traded peer-to-peer, was supposed to make obsolete: transaction fees. By one estimate, these fees have risen 1,289 percent since March 2015. On any given day, the fees will be in the millions of dollars and now cost upward of twenty dollars per transaction. While transaction fees are not mandatory, they are a way for users to attempt to jump the queue in a system rife with bottlenecks, since those who offer miners a fee to have their transactions included in a block have a better chance of that happening. With so many transactions lined up, waiting to be processed, miners have discretion over which will make it to the head of the line; the higher the fee, the more likely it is to be chosen. As the explanatory website Unlock Blockchain puts it: “when miners mine a block, they become temporary dictators of that block. If you want your transactions to go through, you will have to pay a toll to the miner in charge…. The higher the transaction fees, the faster the miners will put [the transactions] up in their block.” As a consequence, transactions can be held up for hours or days or dropped altogether.
Bitcoin’s high transaction fees and slow transaction times were two of the reasons I chose to buy ether. But there was another reason as well: while bitcoin was invented to bypass traditional currency by tendering a new kind of money, ether, another cryptocurrency that can be bought, sold, and used to purchase goods and services, was created to raise capital to fund a project called the Ethereum network. The principals behind it are building out what is being trumpeted as the next iteration of the Internet, Web 3.0, also known as “the blockchain.”
A blockchain is, essentially, a way of moving information between parties over the Internet and storing that information and its transaction history on a disparate network of computers. Bitcoin, for example, operates on a blockchain: as transactions are aggregated into blocks, each block is assigned a unique cryptographic signature called a “hash.” Once the validating cryptographic puzzle for the latest block has been solved by a mining computer, three things happen: the result is timestamped, the new block is linked irrevocably to the blocks before and after it by its unique hash, and the block and its hash are posted to all the other computers that were attempting to solve the puzzle. This decentralized network of computers is the repository of the immutable ledger of bitcoin transactions. As the Tapscotts observe, “If you wanted to steal a bitcoin, you’d have to rewrite the coin’s entire history on the blockchain in broad daylight.”
While bitcoin operates on a blockchain, it is not the blockchain. The insight of Vitalik Buterin, the young polymath who created Ethereum, was that in addition to exchanging digital money, the blockchain could be used to facilitate transactions of other kinds of digitized data, such as property registrations, birth certificates, medical records, and bills of lading. Because the blockchain is decentralized and its ledger immutable, those transactions would be protected from hacking; and because the blockchain is a peer-to-peer system that lets people and businesses interact directly with each other, it is inherently more efficient and also cheaper than systems that are burdened with middlemen such as lawyers and regulators.
A company that aims to reduce drug counterfeiting is using the blockchain to follow pharmaceuticals from provenance to purchase. Another outfit is doing something similar with high-end sneakers. Yet another start-up, this one called Paragon, is currently raising money to create a blockchain that “registers everything that has happened to a cannabis product, from seed to sale, letting consumers, retailers and the government know where everything came from.” “We are treating cannabis as a normal crop,” Paragon’s founder and CEO Jessica VerSteeg, a former Miss Iowa, told a reporter for the website Benzinga. “So, the same way that you would want to know where the corn on your table came from, or the apple that you had at lunch came from, you want to know where the weed you’re consuming came from.”
While a blockchain is not a full-on solution to fraud or hacking, its decentralized infrastructure ensures that there are no “honeypots” of data available for criminals to exploit. Still, touting a bitcoin-derived technology as the answer to cybercrime may seem a stretch in light of the high-profile—and lucrative—thefts of cryptocurrency over the past few years. Gerard notes that “as of March 2015, a full third of all Bitcoin exchanges”—where people stored their bitcoin—“up to then had been hacked, and nearly half had closed.” There was, most famously, the 2014 pilferage of Mt. Gox, a Japanese-based digital coin exchange, in which 850,000 bitcoins worth $460,000,000 disappeared. Two years later another exchange, Bitfinex, was hacked and around $60 million in bitcoin was taken; the company’s solution was to spread the loss to all its customers, including those whose accounts had not been drained. Then there was the theft via malware of $40 million by a man in Pennsylvania earlier this year. He confessed, but the other thieves slipped away, leaving victims with no way to retrieve their funds.
Unlike money kept in a bank, cryptocurrencies are uninsured and unregulated. That is one of the consequences of a monetary system that exists—intentionally—beyond government control or oversight. It may be small consolation to those who were affected by these thefts that neither the bitcoin network nor the Ethereum network itself has been breached, which perhaps proves the immunity of the blockchain to hacking.
by Sue Halpern, NY Review of Books | Read more:
Image: Yoshikazu Tsuno/AFP/Getty Images
