This December I wrote a widely-circulated article on the inapplicability of blockchain to any actual problem. People objected mostly not to the technology argument, but rather hoped that decentralization could produce integrity.

Let’s start with this: Venmo is a free account to alteration dollars, and bitcoin transfers are not free. Yet after I wrote an commodity last December saying bitcoin had no use, someone responded that Venmo and Paypal are raking in consumers’ money and people should switch to bitcoin.

What a surreal adverse amid blockchain’s non-usefulness/non-adoption and the confidence of its believers! It’s so absolutely axiomatic that this person didn’t become a bitcoin enthusiast because they were attractive for a convenient, free way to alteration money from one person to addition and apparent bitcoin. In fact, I would assert that there is no single person in existence who had a botheration they wanted to solve, apparent that an accessible blockchain band-aid was the best way to solve it, and accordingly became a blockchain enthusiast.

The number of retailers accepting cryptocurrency as a form of acquittal is declining, and its better accumulated boosters like IBM, NASDAQ, Fidelity, Swift, and Walmart have gone long on press but short on actual rollout. Even the most arresting blockchain company, Ripple, doesn’t use blockchain in its product. You read that right: the company Ripple decided the best way to move money across all-embracing borders was to not use Ripples.

A blockchain is a authentic technology, not a metaphor

Why all the activity for article so abortive in practice?

People have made a number of doubtful claims about the future of blockchain—like that you should use it for AI in place of the type of behavior-tracking that google and facebook do, for example. This is based on a confounding of what a blockchain is.

A blockchain isn’t an aerial thing out there in the cosmos that you can “put” things into, it’s a specific data structure: a linear transaction log, about replicated by computers whose owners (called miners) are adored for logging new transactions.


There are two things that are cool about this authentic data structure. One is that a change in any block invalidates every block after it, which means that you can’t tamper with actual transactions. The second is that you only get adored if you’re alive on the same chain as anybody else, so each actor has an allurement to go with the consensus.

The end result is a shared absolute actual record. And, what’s more, because accord is formed by each person acting in their own interest, adding a false transaction or alive from a altered history just means you’re not accepting paid and anybody else is. Following the rules is mathematically enforced—no government or police force need come in and tell you the transaction you’ve logged is false (or extort bribes or bully the participants). It’s a able idea.

So in summary, here’s what blockchain-the-technology is: 

Let’s create a very long arrangement of small files — each one absolute a hash of the antecedent file, some new data, and the answer to a difficult math botheration — and divide up some money every hour among anyone accommodating to accredit and store those files for us on their computers.

Now, here’s what blockchain-the-metaphor is: “What if anybody keeps their annal in a tamper-proof athenaeum not owned by anyone?”

An analogy of the difference: In 2006, Walmart launched a system to track its bananas and mangoes from field to store. In 2009 they alone it because of logistical problems accepting anybody to enter the data, and in 2017 they re-launched it (to much fanfare) on blockchain.

If addition comes to you with “the mango-pickers don’t like doing data entry,” “I know: let’s create a very long arrangement of small files, each one absolute a hash of the antecedent file” is a nonsense answer, but “What if anybody keeps their annal in a tamper-proof athenaeum not owned by anyone?” at least addresses the right question!

Blockchain-based abidingness falls apart in practice

People treat blockchain as a “futuristic candor wand”—wave a blockchain at the problem, and aback your data will be valid. For almost annihilation people want to be valid, blockchain has been proposed as a solution.

It’s true that analytical with data stored on a blockchain is hard, but it’s false that blockchain is a good way to create data that has integrity.

To accept why this is the case, let’s work from the applied to the theoretical. For example, let’s accede a widely-proposed use case for blockchain: buying an e-book with a “smart” contract. The goal of the blockchain is, you don’t trust an e-book vendor and they don’t trust you (because you’re just two individuals on the internet), but, because it’s on blockchain, you’ll be able to trust the transaction.

In the acceptable system, once you pay you’re hoping you’ll accept the book, but once the vendor has your money they don’t have any allurement to deliver. You’re relying on Visa or Amazon or the government to make things fair—what a recipe for being a chump!

In contrast, on a blockchain system, by active the transaction as a record in a tamper-proof athenaeum not owned by anyone, the alteration of money and agenda artefact is automatic, atomic, and direct, with no agent needed to adjudge the transaction, behest terms, and take a fat cut on the way. Isn’t that better for everybody?

Hm. Perhaps you are very accomplished at autograph software. When the biographer proposes the smart contract, you take an hour or two to make sure that the arrangement will abjure only an amount of money equal to the agreed-upon price, and that the book — rather than some other file, or annihilation at all — will absolutely arrive.


Auditing software is hard! The most-heavily scrutinized smart arrangement in history had a small bug that nobody noticed — that is, until addition did notice it, and used it to steal fifty actor dollars. If cryptocurrency enthusiasts putting calm a $150m advance fund can’t appropriately audit the software, how assured are you in your e-book audit?

Perhaps you would rather write your own counteroffer software contract, in case this e-book author has hidden a recursion bug in their adaptation to drain your ethereum wallet of all your life savings?

It’s a complicated way to buy a book! It’s not trustless, you’re dupe in the software (and your adeptness to defend yourself in a software-driven world), instead of dupe other people.