How Blockchains Work – Without the Jargon

Think of a blockchain as a shared ledger of which many computers each hold a copy. When someone wants to make a transfer, it is broadcast across the network, checked against the rules, and grouped with other transactions into a block. The block receives a cryptographic fingerprint that also incorporates the fingerprint of the previous block – that is the chaining.

Why the chaining matters

If someone alters an old transaction, the fingerprint of that block changes, and so does the fingerprint of every block after it. The discrepancy is immediately visible to anyone holding a copy of the ledger. A blockchain does not make data impossible to change – it makes changes obvious and, in practice, very costly to carry out.

Who gets to write to the ledger?

This is where the consensus mechanism comes in. In proof of work, whoever performs the computational work may propose the next block; in proof of stake, participants are chosen based on how much of the network's asset they have locked up as collateral. Both models try to solve the same problem: enabling many unknown participants to agree on a shared history without a central administrator.

• Node – a computer that stores the ledger and enforces the rules.
• Block – a package of transactions with a timestamp and fingerprint.
• Consensus – the rules for which version of history is valid.
• Confirmation – the number of blocks added after your transaction.

The limits of the technology

A blockchain guarantees that the ledger is internally consistent. It does not guarantee that the information written to it is true, that an application built on top of the chain is secure, or that whoever receives your funds is trustworthy. Transfers are also generally irreversible: there is no customer service desk that can reverse a transaction sent to the wrong address.

Irreversibility is the technology's defining feature – and what makes fraud so costly.