Proof-of-Work (PoW)
Proof-of-Work is how a network agrees who writes the next block: the right goes to whoever first finds the solution to a computational puzzle. The search costs electricity, and that cost is what stops the history being rewritten.
How it works
A miner searches for a number that makes the block's hash fall below a target. It cannot be reasoned out, only tried, billions of attempts a second. Verification, though, is a single operation: other nodes confirm the answer instantly.
That asymmetry between an expensive search and a cheap check is the whole mechanism. The network trusts nobody and votes on nothing: it simply accepts whichever chain has more work behind it.
Difficulty readjusts every 2,016 blocks, roughly a fortnight, holding the gap between blocks near ten minutes no matter how much hashpower has joined.
Why attacking it is expensive
Rewriting history means controlling more than half the network's hashpower and holding it.
| Hardware required | > 50% of the network |
| Energy draw | comparable to a mid-size country |
| What it achieves | a double-spend |
An attacker cannot conjure coins or touch anyone else's addresses, only reverse their own recent transfers. The cost runs to a large company's budget, and the network they attacked devalues along with the prize.
Strengths and weaknesses
- Proven by timeBitcoin has run on it since 2009 without a single successful attack on consensus.
- It needs no trust in participantsAnyone with hardware can join, and nobody decides who is admitted.
- Energy-hungryThat is the price of the security, not a side effect: cheap defence would mean a cheap attack.
- Throughput is limitedTen minutes a block is a deliberate trade-off. Speed is sought in layer-2 solutions rather than in the base network.