Under the Hood
The shared notebook is brilliant — until someone picks up an eraser. Here’s the cryptographic trick that makes the ledger’s history effectively permanent.
At the end of Part 1, we had a public ledger shared across thousands of computers. Anyone can read it. That’s the feature. But now a question: if anyone can see it, can anyone edit it? And if not — why not?
The answer sits in two ideas that work together: hashing (digital fingerprints) and chaining (linking every page to the one before it). Once you see how these interact, the word ‘immutable’ stops being jargon and starts being obviously, mechanically true.
What a hash actually is
A hash function is a mathematical blender. Feed it any input — a word, a number, an entire encyclopedia — and it produces a fixed-length string of characters called a hash. The output looks random. It always has the same length regardless of what you put in. And crucially, the same input always produces exactly the same output.
THE ANALOGY
Think of it as a wax seal on a letter — but one that automatically shatters the moment anyone touches the contents. Change even a single full stop inside the letter and the seal reforms into a completely different pattern. Anyone holding the letter can check whether the seal matches the contents, instantly, without reading the letter itself.
In blockchain terms: the ‘letter’ is a block of transactions. The ‘seal’ is its hash. And the shattering is cryptographically guaranteed.
Here’s a real example of how sensitive hashes are. The word ‘hello’ produces a completely different hash than ‘Hello’. One capital letter. Totally different fingerprint. Change a single penny in a transaction and the hash of the entire block transforms. This property is called the avalanche effect, and it’s the foundation of blockchain security.
The chain: why every block is linked to the one before it
Now here’s the second idea that makes everything click. Each block on a blockchain doesn’t just contain its transactions — it also contains the hash of the previous block.
Picture a receipt roll at a till. Each receipt is printed with a code derived from the receipt before it. Tear out receipt number 47 and every number after it is suddenly inconsistent. Anyone can check. The corruption is visible immediately across the entire roll.
KEY CONCEPT
This is the chain in blockchain. Every block fingerprints its parent. Change a transaction two years ago and you change that block’s hash, which breaks the link into the next block, which cascades through every block that follows it — all the way to today. You’d need to recompute every single subsequent block faster than the entire honest network is building new ones. At Bitcoin’s scale, that’s practically impossible.
Who gets to write to the ledger — and why they can be trusted
So far we’ve explained why the past is hard to change. But there’s still a question: who decides what gets written in the first place? This is the job of consensus — the mechanism by which thousands of independent computers agree on which transactions are valid and in what order.
Two dominant designs exist, and they solve the problem with completely different economics.
Proof of Work (used by Bitcoin): Computers called miners compete to solve a brute-force mathematical puzzle. There’s no shortcut — you just have to try billions of combinations until you find one that works. The winner adds the next block and earns newly created coins. The catch: solving the puzzle requires enormous amounts of real electricity and hardware. To attack the network and rewrite history, you’d need to out-compute the entire honest mining industry simultaneously. The energy cost is the security cost.
Proof of Stake (used by Ethereum since 2022): Instead of burning electricity, validators lock up their own coins as collateral. If they try to cheat, the network destroys — or ‘slashes’ — their stake. Honest participation earns rewards. Security is purchased with capital at risk rather than energy, cutting electricity consumption by over 99% compared to Proof of Work.
WORTH NOTING
Neither mechanism is perfect. Proof of Work is energy-intensive by design. Proof of Stake gives more influence to those with more coins, which has its own centralisation risks. Both are genuine trade-offs, not marketing choices.
Why the people maintaining the ledger bother at all
This is the question most explainers skip, but it’s essential: miners and validators are strangers with no obligation to run this infrastructure. Why do they?
Because they’re paid in the network’s own currency. Bitcoin miners earn newly created BTC. Ethereum validators earn ETH staking rewards. The network’s coin is both the incentive to participate and the thing the participation secures. This is genuinely elegant: the currency needs the network secure to have value; the network stays secure because the currency has value. It’s a self-reinforcing loop — and it’s why the economics of crypto cannot be separated from the technology.
You now understand how the ledger stays honest. Hashes make every block tamper-evident. The chain makes tampering cascade visibly through history. Consensus makes adding fake blocks extraordinarily expensive. And economic incentives keep honest participants maintaining it.
But here’s the question that opens Part 3: everything we’ve described so far only records one type of event — ‘Alice sent Bob X coins.’ What if the ledger could follow instructions instead? What if it could hold the logic of an entire contract, and execute it automatically, without a lawyer, a judge, or any human in the loop?
NEXT IN THE SERIES
“Recording transactions is step one. But what if the ledger could run code — execute a contract, release funds, transfer ownership — all without a single human authorising it? That’s when things get genuinely strange.”
→ Part 3: Beyond Currency
