Blockchain Revolution

Problem to Be Solved: On the Internet, people haven’t been able to transact or do business directly for the simple reason that money isn’t like other information goods and intellectual property per se. You can send the same selfie to all your friends, but you ought not give your friend a dollar that you’ve already given to someone else.

The money must leave your account and go into your friend’s. It can’t exist in both places, let alone multiple places. And so there’s a risk of your spending a unit of digital currency in two places and having one of them bounce like a bad check. That’s

called the double-spend problem. That’s good for fraudsters who want to spend their money twice. It’s bad for the recipient of the bounced amount and bad for your reputation online. Traditionally, when making online payments, we solve the double- spend problem by clearing every transaction through the central databases of one or many third parties, such as a money transfer service (like Western Union), a commercial bank (Citicorp), a government body (Commonwealth Bank of Australia), a credit card company (Visa), or an online payment platform (PayPal). Settlement can take days or even weeks in some parts of the world.

Breakthrough: Satoshi leveraged an existing distributed peer-to-peer network and a bit of clever cryptography to create a consensus mechanism that could solve the double-spend problem as well as, if not better than, a trusted third party. On the bitcoin blockchain, the network time-stamps the first transaction where the owner spends a particular coin and rejects subsequent spends of the coin, thus eliminating a double spend. Network participants who run fully operating bitcoin nodes—called miners—gather up recent transactions, settle them in the form of a block of data, and repeat the process every ten minutes. Each block must refer to the preceding block to be valid. The protocols also include a method for reclaiming disk space so that all nodes can efficiently store the full blockchain. Finally, the blockchain is public.

Anyone can see transactions taking place. No one can hide a transaction, and that makes bitcoin more traceable than cash.

Satoshi sought not only to disintermediate the central banking powers but also to eliminate the ambiguity and conflicting interpretations of what happened. Let the code speak for itself. Let the network reach consensus algorithmically on what happened and record it cryptographically on the blockchain. The mechanism for reaching consensus is critical. “Consensus is a social process,” blogged Vitalik Buterin, pioneer of the Ethereum blockchain. “Human beings are fairly good at engaging in

consensus . . . without any help from algorithms.” He explained that, once a system scales beyond an individual’s ability to do the math, people turn to software agents. In peer-to-peer networks, the consensus algorithm divvies up the right to update the status of the network, that is, to vote on the truth. The algorithm doles out this right to a group of peers who constitute an economic set, a set that has skin in the game, so to speak. According to Buterin, what’s important about this economic set is that its members are securely distributed: no single member or cartel should be able to overtake a majority, even if they had the means and incentive to do so.⁷

To achieve consensus, the bitcoin network uses what’s called a proof of work

(PoW) mechanism. This may sound complicated but the idea is a simple one. Because we can’t rely on the identity of the miners to select who creates the next block, we instead create a puzzle that is hard to solve (i.e., it takes a lot of work), but easy to verify (i.e., everyone else can check the answer very quickly). Participants agree that

whoever solves the problem first gets to create the next block. Miners have to expend resources (computing hardware and electricity) to solve the puzzle by finding the right hash, a kind of unique fingerprint for a text or a data file. For each block they find, miners receive bitcoin as a reward. The puzzle is mathematically set up to make it impossible to find a shortcut to solve it. That’s why, when the rest of the network sees the answer, everyone trusts that a lot of work went into producing it. Also, this puzzle solving is continuous “to the tune of 500,000 trillion hashes per second,” according to Dino Mark Angaritis. Miners are “looking for a hash that meets the target. It is statistically bound to occur every ten minutes. It’s a Poisson process, so that sometimes it takes one minute and sometimes one hour, but on average, it’s ten minutes.” Angaritis explained how it works: “Miners gather all the pending transactions that they find on the network and run the data through a cryptographic digest function called the secure hash algorithm (SHA-256), which outputs a 32-byte hash value. If the hash value is below a certain target (set by the network and adjusted every 2,016 blocks), then the miner has found the answer to the puzzle and has ‘solved’ the block. Unfortunately for the miner, finding the right hash value is very difficult. If the hash value is wrong, the miner adjusts the input data slightly and tries again. Each attempt results in an entirely different hash value. Miners have to try many times to find the right answer. As of November 2015, the number of hash attempts is on average 350 million trillion. That’s a lot of work!”⁸

You may hear about other consensus mechanisms. The first version of the Ethereum blockchain—Frontier—also uses proof of work, but the developers of Ethereum 1.1 expect to replace it with a proof of stake mechanism. Proof of stake requires miners to invest in and hang on to some store of value (i.e., the native token of the blockchain such as Peercoin, NXT, etc.). They needn’t spend energy to vote. Other blockchains, such as Ripple and Stellar, rely on social networks for consensus and may recommend that new participants (i.e., new nodes) generate a unique node list of at least one hundred nodes they can trust in voting on the state of affairs. This type of proof is biased: newcomers need social intelligence and reputation to participate. Proof of activity is another mechanism; it combines proof of work and proof of stake, where a random number of miners must sign off on the block using a cryptokey before the block becomes official.⁹ Proof of capacity requires miners to allot a sizable volume of their hard drive to mining. A similar concept, proof of storage, requires miners to allocate and share disk space in a distributed cloud.

Storage does matter. Data on blockchains are different from data on the Internet in one important way. On the Internet, most of the information is malleable and fleeting, and the exact date and time of its publication isn’t critical to past or future information. On the blockchain, bitcoin movement across the network is permanently stamped, from the moment of its coinage. For a bitcoin to be valid, it must reference

its own history as well as the history of the blockchain. Therefore, the blockchain must be preserved in its entirety.

So important are the processes of mining—assembling a block of transactions, spending some resource, solving the problem, reaching consensus, maintaining a copy of the full ledger—that some have called the bitcoin blockchain a public utility like the Internet, a utility that requires public support. Paul Brody of Ernst & Young thinks that all our appliances should donate their processing power to the upkeep of a blockchain: “Your lawnmower or dishwasher is going to come with a CPU that is probably a thousand times more powerful than it actually needs, and so why not have it mine? Not for the purpose of making you money, but to maintain your share of the blockchain,”¹⁰ he said. Regardless of the consensus mechanism, the blockchain ensures integrity through clever code rather than through human beings who choose to do the right thing.