The Nakamoto Legacy

Having noticed recently that every other email in my spam folder at the moment is concerned with some crypto-currency or other, I thought it might be timely to revisit a 2015 article looking at the origins and context of Bitcoin and the Blockchain…

In 2008 a technical paper quietly appeared featuring a theoretical technology which explicitly solved a specific problem, the ‘double payment problem’, while implicitly solving a number of others.? The paper proposed an elegant solution, and it spawned a real world technology which was to make a big splash, even if (at the time of writing this article) it’s impact was very mixed and it’s long term future was far from certain. The ostensible author of the paper, Satoshi Nakamoto, didn’t actually exist and (again at the time of writing) the actual author has still not been definitively identified.

The paper was entitled Bitcoin: a Peer-to-Peer Electronic Cash System.?

Crypto-currency and Byzantine Generals

The origins of Bitcoin lie with an even more obscure paper published in 1982 in ACM Transactions on Programming Languages and Systems, written by Leslie Lamport, Robert Shostak and Marshall Pease. Interestingly, I think, it was part-sponsored by NASA and the US Ballistic Missile Defence Systems Command.? It was called The Byzantine Generals Problem.??

Imagine a bunch of Byzantine generals camped with their troops around an enemy city.? They communicate by messenger, and have to agree a consensus plan of battle.? The catch is that one or more of the generals may be a traitor.? The problem is to figure out a method that allows the loyal generals to come correctly to an agreement.

It turns out that it’s only possible to do this (at least with oral messages) if at least two-thirds of the generals are loyal.? The proof is far from straightforward, and on the way takes in Lieutenants and Albanians, so we’ll give that a miss.

Er, so what?

The Byzantine Generals were actually a kind of mathematical allegory for computer components.? The authors were addressing the issue of a failed component simply sending incorrect information to other parts of the system, and their solution became known as ‘Byzantine Fault Tolerance’.

Fast forward 26 years, and by now the subject matter isn’t components so much as individual computers in a network, the Internet, and the problem isn’t faulty equipment, it’s active subversion by hackers and their ilk.? Arguably a closer allegory of the original problem in fact.

Virtual Coins and the Double Payment Problem

Let’s suppose someone sends you a virtual coin, a Bitcoin.? What is it?? What have they actually sent you?

They’ve sent you information.? In this case, that information comprises a ‘hash’ of the last transaction (which was their own incarnation of the coin) and your ‘public key’, signed digitally (i.e. encoded) using their own ‘private key’.? A hash, in this context, is a kind of encoding system which takes input data and turns it into a number with a fixed length.? Changing the input data even slightly will change the hash.

When you receive the ‘coin’ you can use the sender’s public key to verify it.? What you can’t do, though, is prove whether or not the sender has already spent the coin somewhere else: hence, the double payment problem.? Back in the Olden Days you’d have to solve this via a Trusted Third Party (probably a bank), who’d decide which transactions were valid and which ones weren’t.??

The question then arises as to whether a solution can be found that doesn’t need such a central authority.? Because, hey, it’s 2008, and banks, well, come on man…

?Enter the Blockchain

The solution proposed by Satoshi Nakamoto, whoever (or whatever)? that is, was a very clever piece of virtual technology dubbed the ‘Blockchain’.??

The basic idea is that when someone sends you a coin, they send everyone else the same transaction – they broadcast it to everyone in the network.? In order for that transaction to be valid, it then has to find its way into the Blockchain.

A number of nodes in this network have copies of the Blockchain and beaver away updating it.? They gather together recently broadcast transactions into a single block, along with two other things: the hash of the previous block, and a useless number called a ‘nonce’.? No, really.??

They then throw raw computing power at creating a hash with a particular number of leading zeroes.? This is really hard, because the hash number is very long (so there are gazillions of possible outcomes), and it turns out that the difficulty goes up exponentially depending on how many zeroes are required.? Calculate the hash, check the number of leading zeroes and if you don’t get a match, increment the nonce and do it again.? If you do get a match, you add this new block to the chain, the hash of which will be included in the next block.? This process is called ‘proof of work’, and it usually takes about ten minutes.? The guys who do it are dubbed ‘miners’, and are rewarded with new Bitcoins and/or transaction fees (the first transaction in the new block is a new coin owned by the creator of that block).? If the proof of work gets too easy (as measured by the number of valid new blocks per hour being generated), the number of leading zeroes is increased to make it hard again.

This means of course that it’s possible in principle to have different versions of the Blockchain kicking about at the same time.? No matter: the rule is that the longest chain is always the right one.? Nodes working on another chain will drop it once a longer one is received, and switch to working on the longer one.? The consensus Blockchain is therefore the definitive ledger of Bitcoin transactions, and replaces the trusted third party.??

If you’re a treacherous Byzantine general in this network, you’ve got your work cut out.? In order to commit a fraud you have to rewrite history, as recorded by the Blockchain ledger.? But any change you make to an earlier block will change the way that it is hashed in subsequent blocks.? So you’d have to redo all that proof of work.? Unless you dominate the computing power of the network, this is effectively impossible.? You’d be better off becoming an honest miner and making money that way.?

There’s a bit more to it than this, but it’s already migraine-inducing stuff, so let’s move on.

The Trouble with Bitcoin

The technology behind Bitcoin is, without a doubt, very clever indeed.? The economics on the other hand is another matter.? With no central bank to regulate the supply of money in the system, Bitcoin is subject to wild gyrations in conversion rate to other currencies.? It has an underlying deflationary mechanism (remember those miners and their new bitcoins), which will cease when the number of coins becomes fixed at a limit of 21 million, at which point if demand is rising, inflation will rule the day.? And that’s leaving aside the climate impact of proof-of-work. Whether it (or another Bitcoin variant) is ever likely to become a mainstream success I have no idea, but it’s not a no-brainer that it will.

Innovations and developments in recent years have sought to mitigate some of these downsides. Some crypto-currencies have sought to use alternatives to the colossally wasteful proof-of-work mechanism. And some central banks are actively exploring their own ‘stablecoin’ digital currencies.

Blockchain Redux

Although ‘Nakamoto’ wrote specifically about using Blockchain to implement a crypto-currency, the concept isn’t dependent on this: in principle the blocks could contain any structured information.? The mechanism “enables trust between mutually distrusting parties” (Joi Ito).? It doesn’t have to be trust about a transfer of currency, so much as a transfer of value.

And so whether or not Bitcoin itself has a bright future, the Blockchain is independently a source of interest. Because in principle the technology can be applied to any information where consensus and trust are vital amongst parties who don’t necessarily trust each other, and especially where that trust is currently provided vicariously by some kind of third party – registrars, trustees, custodians and so on.

Two other factors have the potential to take this idea into the realms of the extraordinary – the potential for ‘hybrid’ Bitcoins, and the Internet of Things.??

What I call hybrid bitcoins is a simple recognition that other information can be transferred with a coin.? The Internet of Things on the other hand is a converging technology – the ability of all manner of devices to communicate independently via the Internet.

Imagine a car, or a house, for example, whose ownership is registered on a distributed Blockchain rather than at a centralised government database, and which is in contact with it’s own block.? A change in ownership is registered on the Blockchain and the vehicle or property is immediately aware of this, and can act on the new information.

A raft of applications have been envisaged, including ‘smart’ contracts and escrow arrangements; digital assets – decentralised asset registries, including home ownership; smart assets, like our car above; audit trails; chains of share ownership – shareholder registers, stock exchanges; music royalties; global airline payments; and voting.

There’s been more than a degree of Blockchain related hype in recent years, but actual uptake of applications like this has so far been limited. It may well be that the energy demanded by proof-of-work has made climate conscious executives queasy. All the same, it might be instructive to think about the parts of our business lives that depend on some third party to establish trust, and how that might fundamentally change if the Blockchain becomes the AirBnB of the custodial professions.