Every money whose
owner could thin it
has been thinned

This page puts the words first on purpose. Bitcoin is full of words that sound like ordinary English but are not. A wallet holds no money. Mining digs nothing out of the ground. And a whale is a person. Learn these, and the rest of the page reads at normal speed.

Listen & Repeat reads each word, then waits for you to say it back. Speaking drill turns on your microphone and checks that the words came out clearly. On each card, say it reads the word aloud. Tell me more opens a longer explanation and a sentence to try.

Almost every argument in The Bitcoin Standard belongs to a fight that is a hundred years older than bitcoin. Two schools of economics disagree about what money is for. Nearly everything else follows from that. Get them clear now, before section 07 asks you to judge the book.

The whole fight in one picture

If you remember nothing else from this section, remember this. It uses the two things section 06 will compare: an apple and a bar of gold.

Keynesians are comfortable with money that slowly loses value. To them, money is like an apple. It is for using, not for keeping. If it rots a little each year, that is acceptable — and some would say useful, because it pushes you to spend or invest instead of sitting on it. So when the economy stalls, make more of it. Critics call this "printing money".

Austrians want money to behave like gold. You work, you save, and what you put away is still worth something in thirty years. You can build a pile over a lifetime. To them, money that rots is a broken measuring stick. Making more of it is quiet theft from everyone already holding it.

One side thinks money should move. The other thinks money should keep.

This has happened before, and Rome is the case the book leans on

The oldest version of this argument has nothing to do with central banks. It is about a coin. It is also the story The Bitcoin Standard opens with.

Ammous likes what happened next. Around 309 AD, Constantine made a new gold coin called the solidus. Nobody debased it. It kept its weight and its purity for about seven hundred years. The western empire that stopped using it fell long before. The eastern half kept the good coin — and kept going until 1453.

The dollar used to be a claim on gold. Then it stopped being one.

For most of modern history, a banknote was a receipt. You could hand it in and get metal. That is not a comparison. The promise was printed on the note, and you could enforce it in court.

Ammous calls 1971 the turning point of the modern world. It really was the moment the last outside limit on making money disappeared. What you think of everything after that is the argument.

What "printing money" actually means

Almost nobody prints anything. When people say a government is "printing money", they usually mean one of two very different things. The difference matters.

The first is borrowing. A government spends more than it collects in tax. It sells bonds to cover the gap. That is debt, not new money. Someone's existing savings are being lent out.

The second is what a central bank does. It buys those bonds with money it makes by typing a number into its own accounts. No printing press. No paper. This is quantitative easing, and it is the closest thing to what people picture. The money supply grows because the central bank decided it should.

And then there is the debt

One thing since 1971 is easy to measure. Governments have borrowed more than at any peacetime moment in history.

Both schools agree the money supply is a decision. They disagree about whether anyone should be trusted to make it.

Bitcoin was built into that disagreement. Its fixed supply is not just a convenient technical detail. It is one side of this argument, written into software, by someone who had clearly picked a side. Was that side right? The rest of this page keeps coming back to that question. It has no technical answer.

Take away the price charts and bitcoin is an accounting trick. It is one list. Every payment that has ever happened, in order. The trick is this: thousands of unrelated computers hold the same copy, and they agree on what it says. None of them is in charge.

That sounds small. It is not. Until 2009, nobody had solved one particular problem. Digital things copy perfectly. If money is a file, I can send you the file and keep a copy too. Then I have spent it twice. Every earlier attempt at digital money fixed this the same way. It chose somebody to keep the master list — a bank, a company, a server. That works. But it means there is a somebody. And a somebody can be pressured, hacked, bought, or told by a government whose money to freeze.

Bitcoin's answer had two parts. Make the list public. And make adding to it deliberately expensive. Anyone can add the next page. But you must burn real electricity to earn the right. The winner gets new coins for the trouble. To rewrite an old page, you must redo that work for every page since — while the rest of the network keeps racing ahead. So the past is not protected by a rule or a promise. It is protected by arithmetic and an electricity bill.

The innovation was not the coin. It was a way for strangers who do not trust each other to agree on the order things happened in.

Now put that next to section 02. Every money in that section failed the same way. Somebody could make more of it, and in the end they did. Rome could thin the denarius because Rome made it. A central bank can grow the money supply because it is the central bank. So why can nobody grow bitcoin's supply? Not because a rule says no — rules have never stopped anyone. It is because nobody is in a position to do it. The accounting trick and the money argument are the same trick, seen from two ends.

Everything else on this page follows from that one move. That includes the parts people dislike. The electricity is not waste added to the design. It is the design. A few people holding most of the coins is what happens to an open network after seventeen years. And is a thing with no income, no issuer and a fixed supply really money? Or a commodity? Or a bubble that has not burst yet? That is the argument the rest of the page sets out.

Three ideas, in order. None of them is new on its own. The combination is.

First, the hash. A hash function takes any amount of data and gives back a fingerprint of fixed length. Bitcoin uses SHA-256, which always returns 64 characters. Two things make it useful. Change one character of the input, and the whole output changes — in a way nobody can predict. And you cannot run it backwards. Given a fingerprint, the only way to find the data is to guess.

Second, the chain. Each block of payments includes the hash of the block before it. So block 3's fingerprint depends on block 2's. Block 2's depends on block 1's. Change one letter in an old block and its hash changes. Then the next block's "previous hash" no longer matches. And so on, all the way to today. One edit breaks everything after it. That is exactly why we call it a chain.

Third, proof of work. A computer could fix a broken chain in a fraction of a second. So there is one more rule. A block is only accepted if its hash starts with a run of zeros. There is no clever way to make one. You add a meaningless number called a nonce. You hash the block. If it does not start with enough zeros, you change the nonce and try again. Billions of times a second, all over the world. The result is a puzzle that is brutally expensive to solve and instant to check. That gap is the whole point: hard to write, cheap to check.

What mining actually is

Mining sounds clever. It is not. It is guessing.

A miner takes the block and adds a meaningless number to it, called a nonce. It hashes the whole thing and looks at the answer. Does it start with enough zeros? Almost always, no. So the miner adds one to the nonce and tries again.

There is no shortcut. You cannot work out the right nonce. You cannot start from the answer you want and run backwards to find it. The only method is to try, and keep trying. That is the whole job. A miner is simply a machine that guesses very fast, and owning two machines gives you two chances instead of one.

When you press mine above, your browser does exactly that. It starts at zero and counts upward until a hash comes back with four leading zeros. It usually takes around sixty-five thousand tries, and your laptop does that in well under a second.

Why you could not do this to the real bitcoin

You have just rewritten history in four blocks. Doing it to the real network is much harder, and for three separate reasons.

The chain keeps growing while you work. You are not only redoing the old blocks. Everybody else is adding new ones the entire time. So you must redo all the old work, and catch up, and then get ahead. Nodes follow whichever chain has the most work behind it.

So you need more machines than everyone else put together. Not more than the biggest miner — more than all of them combined, for as long as the attack takes. That is the 51% attack. It is not impossible. It is expensive, and it gets more expensive every time somebody plugs in a machine.

On 31 October 2008, someone posted a nine-page paper to a cryptography mailing list. They used the name Satoshi Nakamoto. Lehman Brothers had collapsed five weeks before. The paper was called Bitcoin: A Peer-to-Peer Electronic Cash System. Its first sentence names the enemy. Business on the internet, it says, "relies almost exclusively on financial institutions serving as trusted third parties."

Nobody knows who wrote it, and this page will not solve that. People have named many candidates over the years. Each one has either denied it or failed the simple test. That test is easy: sign a message with the keys that mined the first blocks. Nobody has.

What we do know is where the ideas came from. That part is unusually well recorded, because it all happened in public, on mailing lists you can still read today.

It is worth saying plainly what kind of story this is. The page has been building it since section 02. For two thousand years, money was quietly thinned by whoever controlled it. Somebody looked at that pattern and drew a hard conclusion. The problem was not who controlled the money. The problem was that anyone controlled it. So he built the alternative alone, under a false name. Then he gave it away, kept nothing, and disappeared before it made him famous.

As rebel stories go, that is close to perfect. And that is exactly why you should be careful with it. A story this good makes people stop checking. So the panel below lists what he was really fighting, in his own words. For each one, it asks a simple question: did it work?

The headline in the first block

So the why is not really in doubt, whatever you think of it. The whitepaper is a technical document. The first block is a statement. And his forum posts in 2009 and 2010 are blunt: "The root problem with conventional currency is all the trust that's required to make it work. The central bank must be trusted not to debase the currency, but the history of fiat currencies is full of breaches of that trust."

Was he right? That is section 07's argument. That this was his reason is simply on the record.

And then he left

Satoshi wrote code and forum posts through 2010. Then he handed the code over to Gavin Andresen. In April 2011 he sent one last email. He had "moved on to other things," he said. Nothing signed with those keys has appeared since.

Start with the simplest possible version. You pick apples in October. You want to eat in March. The apples are a store of value — you have moved this autumn's labour into next spring — and they are a terrible one, because by March most of them are brown.

Every kind of money humans have used tries to fix that one failure. The goal is not to be useful — apples are useful, and that is not the problem. The goal is to survive the journey. There are three journeys. The economist Saifedean Ammous, whose book is section 07, calls this salability: selling across time, across space and across scale. In plain words: can it survive being kept? Can it survive being moved? And can you break it into the size of the thing you want to buy?

The list below compares five things across nine properties. The scores are judgements, not measurements, and the note says so. But the shape of each one is not really in dispute. The interesting part is that every column has a hole in it.

What the holes tell you

Nothing scores well on everything, and the failures are not random. Gold's weakness is checking it. A bar with tungsten inside looks exactly like a real one. So gold ends up locked in vaults, and people trade paper claims instead of metal. That quietly brings back the counterparty gold was supposed to remove. Cash wins on liquidity: you can spend it instantly, anywhere, at full value. It is excellent at every property but one — and that one is the property that matters over a lifetime. Real estate has something none of the others have: it pays you rent while you hold it. It pays for that by never moving, and by taking months to sell.

Bitcoin's one truly new score is verifiability. Anyone can check the whole supply, and check any coin, for free, on a laptop. There is nothing like that for gold. It pays for this with the three worst scores on the board. It has only seventeen years of history. It has had three drawdowns of about seventy-five per cent or more — three falls that deep, from a peak. And its security is weak. An asset that halves in a year did not store value that year, whatever it does over ten.

The security row is the one to argue about, and this page has changed its mind about it once already. Look at what it is really asking: will you still have this in ten years?

Against theft and your own mistakes, bitcoin is the worst thing here. Someone steals your key. An exchange fails. You type one wrong address. The coins are gone. There is no fraud department, no insurance, and no court that can move them back. Mt. Gox lost about 850,000 coins. FTX held customer balances that were never really there. Nobody got their money back. Meanwhile, stealing a house is genuinely hard.

But against everything else in that question, it is the strongest thing on the list. Nobody can freeze it. Nobody can print more of it and make yours worth less. No bank holding it can fail and take it with them. Money in an account is none of those things: it can be frozen in an afternoon, it loses a little value every year by design, and in Cyprus in 2013 depositors simply lost part of their balance to save the banks.

So bitcoin scores 7 here and cash scores 6 — but that is a judgement about which danger is bigger, not a measurement. A reader in a stable country with a good bank may reasonably flip those two around. That is exactly what the boxes below are for.

One number hides something, and the hidden part is the interesting part. Split security into two halves, and bitcoin and real estate swap places completely.

Against theft and mistakes, a house is almost unbeatable and bitcoin is the worst thing here. Against seizure, it flips exactly. You cannot hide a house. You cannot move it. And the government holds the record of who owns it. That makes it the easiest thing on this list to tax, freeze or take. A phrase in your head is the hardest. Gold sits in the middle, and has suffered both. People steal it. And the United States made owning it illegal from 1933, for forty years.

So which way should the row point? It depends entirely on what you are afraid of, and that is not a technical question at all. Fear burglars, hackers and your own carelessness, and bitcoin is the worst choice here. Fear the government where you live, or fear your savings quietly shrinking, and it is the best one. That second fear is exactly what it was built around — read the genesis block in section 05 again.

One thing on that chart is genuinely new, and it is not any single score. Look at where each asset lives. Cash, bank deposits, shares and bonds all sit inside the banking system. They are entries in somebody's ledger. That somebody can freeze them, lose them, charge you for them, or be ordered to hand them over. Gold and property sit outside the system. But you cannot carry useful amounts of gold across a border, and a house cannot move at all. For the whole of history, "outside the banking system" and "able to move" could not both be true.

Bitcoin is the first thing that is both. There is no account, no custodian, no counterparty and no form to fill in. It settles anywhere with an internet connection, in about an hour. Whatever you think about the price, that combination did not exist before. It is worth stopping to notice how strange this is. One anonymous person built it. And the answer to "who do I ask for permission?" is nobody.

This is also why the security score is 7 and not 10. Each of those missing things has a second face. No custodian also means nobody to call. No counterparty also means no insurer. No permission also means no way to get your money back. The revolutionary property and the dangerous property are not two features. They are one feature. Which name you give it depends on what happens to you.

"Store of value" is not something a thing simply is. It is a claim about a distance and a length of time. And it is false for every asset, at some distance and some length.

Which is why the honest answer is a dull one. Over a week, a bank deposit beats everything. Over thirty years, the bank deposit is the one that quietly lost the most. Over five thousand years, only gold has a record at all — and that record is the one thing bitcoin cannot have yet, and will not have in your lifetime.

Wiley published The Bitcoin Standard in 2018. It is easily the most influential book about bitcoin. It is worth reading — and worth reading carefully. About two-thirds of it is the history of money. The last third is a political argument. The book does not always tell you which one you are reading.

The argument, as fairly as it can be put

His main idea is hardness. He measures it with the stock-to-flow ratio. You take the pile that already exists and divide it by the amount made in a year. If that number is high, even a very busy year of production hardly changes the total. So producers cannot flood the market and destroy everyone's savings.

His claim about history is this. Governments do not choose money by order. Money wins a competition. And societies that chose easy money were ruined by it. The examples are real, and they are the best part of the book.

The people of Yap used huge limestone discs. Cutting them on another island was slow and dangerous. Then a shipwrecked Irish-American trader arrived with modern tools and explosives. He made the stones cheaply, and they stopped being money.

West African communities kept their wealth in glass beads. Making those beads locally was slow and expensive. Then Europeans arrived with factory glass. They bought a continent's savings with trinkets.

In both cases the money did not fail because it was primitive. It failed because somebody found a cheap way to make more.

Where it gets contested

Three separate objections, which are often mashed together and should not be.

What survives all of this. Easy money does destroy savings. The people holding it were robbed slowly instead of suddenly. That is well evidenced and hard to argue with. Currencies really have lost most of their value over a century. The Yap stones and the glass beads really happened. You can accept every one of those facts and still say no to the conclusion. A fixed-supply digital asset may not be the answer. Many careful readers end up exactly there.

This is the clearest thing on the page that nobody planned. Satoshi wrote "one CPU, one vote". He expected people to mine at home. Within about four years that was over. Mining moved to ASICs: chips that can do one calculation and nothing else. It moved into big sheds next to cheap power. Nobody broke a rule. The money simply pointed that way, and nothing in the design pointed anywhere else.

The mechanism people get wrong

So what is all that electricity actually for? Section 02 is the answer. It is not paying for payments — those are almost free. It is buying one thing: nobody can quietly make more coins, and nobody can change the past. Is that worth a medium-sized country's electricity? That is a real question. Both answers are serious. If it seems obvious to you either way, look again at what sits on each side.

The carbon is much harder to measure than the electricity. It depends on where the machines are. That has changed a lot. China held most of the mining until it banned it in mid-2021. The industry moved to the United States, Kazakhstan and Russia. Was that better or worse? It is not clear. It lost a lot of Chinese hydropower. It gained Kazakh coal and American gas. Published estimates differ by three times, depending on which power mix the writer assumes. So anyone who gives you one confident number has picked one.

There is a real argument on the other side, and it deserves a fair hearing. A mine can be built anywhere, and it can be switched off in seconds. Very few big users of power can do both. So you can put a mine at a gas field where the gas is being burnt off and wasted. You can put one on a grid with more wind than demand at three in the morning. And when the grid needs the power back, the mine stops at once. Texas pays miners to do exactly this. Is that a real benefit to the grid, or just a good excuse? People argue about it honestly. Remember too that the industry's own green figures come from the industry itself.

There is also the hardware. An ASIC can do nothing else, so when a better one arrives the old one is rubbish. One well-known study put this at tens of thousands of tonnes of e-waste a year. The industry disagrees with how that was counted. The argument is about how long the machines really last.

Why a fast computer is not a fast miner

Try the AI workstation on the list above. It is a genuinely powerful machine, the kind used to train models, and it costs about four thousand dollars. At this job it is roughly twice as fast as a gaming graphics card. That is all it buys you.

Nothing is wrong with the workstation. AI and mining ask a computer for opposite things, and seeing why explains the whole shape of this section.

An AI model is arithmetic on a huge scale. Millions of numbers multiplied together, over and over, in big blocks. That work splits perfectly: thousands of small cores each take a piece and none waits for the others. The model is also enormous, often many gigabytes, so the machine needs a lot of fast memory to hold it. Many parallel cores and wide memory are the expensive parts of that chip. For AI they are exactly the parts that matter.

Mining needs neither. A hash is a short chain of simple steps on a few bytes: shift some bits, add, mix, repeat sixty-four times. Every step needs the answer from the step before, so one hash cannot be split across cores at all. And the whole job fits in about 256 bytes. Not gigabytes. Bytes.

So on the workstation, every expensive part sits idle while it mines. The memory goes unused. The decimal maths goes unused. The tensor units — the thing that makes it an AI machine in the first place — are not touched at all. You have bought a warehouse to store one envelope.

An ASIC is the opposite trade. Almost every transistor in it is a SHA-256 circuit, copied over and over. It cannot run a model. It cannot run anything else at all.

That is the whole answer. A general computer can do anything, and paying for that flexibility is what makes it lose here by a factor of a hundred thousand. The chip that wins gave up the ability to do anything else. It is why home mining ended around 2013, and why it is not coming back.

How you would actually do it

The software part is much simpler than people expect. It is five steps.

1. Get a wallet address. This is where the money would go. You do not need the coins first, and you do not need permission.

2. Choose solo or a pool. Solo means you keep a whole block when you find one, and the table above shows how long that wait is. A pool joins your machine to thousands of others. You are paid a small share every day instead of a huge amount almost never. Nearly everyone uses a pool.

3. Point the machine at the pool. You type in one address — the pool's server — and your own wallet address as the user name. That is the whole setup.

4. Watch the heat. A real miner turns almost all its electricity into heat and noise. A modern one is about as loud as a vacuum cleaner and needs 3,500 watts, which is more than most kitchen sockets will give you.

5. Get paid. The pool sends coins when your share passes its minimum.

The useful thing this shows is where the line falls. Below a certain electricity price, a modern machine makes money. Above it, the same machine loses money every day it is switched on. Nothing about the machine changed. That is why mining moved to Texas, Kazakhstan and Alberta, and it is the whole of section 08 in one number.

The ledger is public. So you can count exactly how much sits in each address. Almost no other market lets you do that. But you cannot count the people. That gap between the two is where most of the loose talk starts.

The concentration you can actually name

Four blocks of supply are large, identifiable and worth more than any address chart:

The coins that are gone. Between three and four million bitcoin are lost forever. Keys were thrown away. Hard drives went to the dump. Some owners died. Those coins still show in the supply, but they will never move again. So the real supply is well below twenty-one million.

Satoshi's one million or so. Nobody has touched them since 2010, as section 05 describes. If they ever moved, the price would react violently. Everyone quietly assumes they never will.

The funds. American regulators approved spot bitcoin ETFs in January 2024. They bought coins far faster than anyone expected. Together they now hold well over a million bitcoin. This changed the market. A large part of it now sits in regulated custody — held for you by a company under supervision. Ordinary pension savers own it. And the buying and selling is published every day.

The companies. One company alone holds several hundred thousand bitcoin. It bought most of them with borrowed money and by selling new shares. Its holding is so large that its money decisions move the market by themselves.

Those last two numbers change every few months, so they are left loose on purpose. Both are public. The funds report every day and the company announces what it buys. Look them up. Do not trust a number written here in 2026.

Manipulation: what has been shown, and what is folklore

For years the industry's own numbers were simply false. A 2019 study given to American regulators found that about ninety-five per cent of reported trading volume was fake. This is called wash trading: exchanges trade with themselves to look busy. This has improved a lot, thanks to regulated exchanges and the ETFs. Few people notice that change.

The market itself makes this easy, and the reasons are simple. It never closes and nothing can pause it. The order books are thinner than the reported volume suggests — the liquidity is thin. Ordinary traders can borrow fifty or a hundred times their own money, which is called leverage. So one big sell order can force many other traders to sell too. Each forced sale is a liquidation. The price then falls much further than that first order deserved. You do not need a plot for this. You need a big order and a thin book.

There is an awkward mirror here, and the page should not walk past it. Section 02 complained about normal money. The complaint was this: a few people nobody elected decide things that change everyone's savings. Now look at sections 08 and 09. A few large holders and a few mining pools do something similar. Nobody elected them either. Bitcoin removed the committee. It did not remove concentration. Open systems concentrate — that is what seventeen years does to them. This does not disprove the main claim. Nobody can print more coins, and that was the real promise. But it is far from a network with no powerful players in it.

Separate the two things one last time. Nearly every bad argument about bitcoin comes from gluing them together.

The engineering question is closed. A network with no owner has processed payments since January 2009 without stopping. Nobody has ever rewritten it. It kept running while most of the companies around it collapsed. It solved the problem it set out to solve. You tested a small version of it yourself in section 04.

The money question is wide open, and it is not technical at all. Here it is. Can a thing hold value across a human lifetime, when it has no issuer, no income, no legal status, and only seventeen years behind it? The code cannot answer that. Only behaviour can. Will people keep holding it through ten years in which nothing interesting happens? That is the one test it has never faced.

The idea this page opened with survives, but in a smaller form than people usually claim. This much is true and well evidenced. When someone could make more of the money, they did — in Rome, on Yap, in West Africa, and in every currency your grandparents saved in. It is also true that the last outside limit came off in 1971, and that debt and prices have climbed since.

But one thing does not follow from that. It does not prove that a fixed-supply digital asset is the answer. The book asks you to accept that without quite arguing for it. Getting the illness right is not the same as getting the cure right. This page has tried to keep those two apart on every claim it makes.

Gold's whole case is that people did not abandon it for five thousand years. Nobody can make that argument any faster.

This page has tried to do one thing above all: show you the joins. Which claims can you check on your own laptop? Which come from a study you can go and read? And which are somebody's opinion dressed as a fact? The coloured labels on every box are there for exactly that. If you take one habit away from this page, take that one. It works on the next thing you read about bitcoin, and the one after that. Most of them will not show you the joins.