Mining & consensus

Mining is often described as solving complex maths problems. It is not. Miners guess numbers, hash the block header with each guess, and keep going until a hash lands below the target. The work is deliberately dumb and deliberately expensive, and that expense is the security.

Difficulty is the self correcting part. Every 2,016 blocks the network retargets so blocks keep arriving roughly every ten minutes, however much hashrate has joined or left. Nobody adjusts it and nobody can override it. The block subsidy halves every 210,000 blocks, and that schedule is what enforces the supply cap.

Consensus is the other half of the section, and it is where the terms get slippery. A block that gets orphaned is not the same as one that goes stale. A chain reorganisation is routine at one block and alarming at six. Selfish mining sits here too, along with the distinction between consensus and consensus rules that most explanations skip. The fork vocabulary lives under protocol and upgrades, and the 51 percent attack under privacy and security, because both are about what someone does to the network rather than how it runs.

Pools, ASICs, hashrate and the energy question are here too, because an honest answer about how much electricity bitcoin uses starts with knowing what a terahash is and who is paying for it.