Blockchain: What It Is and How It Works

How a blockchain works

A blockchain combines four ingredients: cryptographic hashes, blocks, a peer-to-peer network and a consensus rule.

1. Blocks and hashes. Transactions are collected into a block. Each block header contains a hash of the previous block, a short fingerprint produced by a function such as SHA-256. Change a single character in an old block and its hash changes, which breaks the link to every block after it. That is what makes the ledger tamper-evident.

2. A network of nodes. Thousands of independent computers, called nodes, each keep a copy of the chain and check every new transaction and block against the same rules: valid signatures, no double-spending, correct format. Nodes that receive an invalid block simply reject it.

3. Consensus. Because anyone can propose a block, the network needs a rule to decide which block comes next. This is the consensus mechanism:

• In proof of work, used by Bitcoin, miners spend computing power to find a valid block; the chain with the most accumulated work wins.
• In proof of stake, used by Ethereum since 2022, validators lock tokens as collateral and are selected to propose and attest blocks.

4. Finality. Once a block has enough blocks built on top of it (or, on some chains, once a finality vote passes), reversing it would require redoing that work or losing a large stake. Records become, in practice, permanent.

A simple example: when Alice sends coins to Bob, her wallet signs the transaction with her private key. Nodes check the signature and her balance, a miner or validator includes it in a block, and from then on anyone can see the transfer on a block explorer.

History

The idea of chaining records with hashes is older than cryptocurrency. In 1991, researchers Stuart Haber and W. Scott Stornetta described a system for time-stamping digital documents so that their dates could not be backdated, by linking each record to the previous one cryptographically.

The missing piece was a way for strangers to agree on one ledger without a central operator. Satoshi Nakamoto's Bitcoin whitepaper (31 October 2008) solved it by combining hash-linked blocks with proof-of-work mining and economic incentives. The Bitcoin network started on 3 January 2009 and became the first widely used public blockchain.

Ethereum (2015) added general-purpose smart contracts, turning blockchains into platforms for applications. Later networks experimented with faster consensus, sharding, and modular designs that split execution from data availability, while layer 2 networks began to run on top of existing chains. In 2018 the US National Institute of Standards and Technology published Blockchain Technology Overview (NISTIR 8202), a widely cited reference on the technology's components.

Types of blockchain

Public chains are further divided into layer 1 networks, which settle their own transactions, and layer 2 or sidechain systems that rely on another chain for security or settlement.

What blockchains are used for

• Digital money: cryptocurrencies such as Bitcoin, and stablecoins pegged to fiat currencies.
• Decentralized finance (DeFi): lending, trading and derivatives run by smart contracts.
• Digital ownership: NFTs, domain names and in-game items.
• Tokenized real-world assets: funds, bonds and tokenized stocks recorded on-chain.
• Record-keeping: supply-chain tracking, timestamps and audit trails.

A blockchain is not the best tool for every database. It trades speed and storage efficiency for shared control and resistance to tampering; when a single trusted party already runs the system, a conventional database is often simpler.

An everyday example: when Alice sends ETH to Bob, the transaction is broadcast to Ethereum nodes. Validators check Alice's signature and balance, include the transaction in a new block, and link that block to the previous one by its hash. Within one block, about 12 seconds on Ethereum, Bob's balance shows the deposit, and anyone can look the transaction up on a public block explorer such as Etherscan, a level of transparency that bank ledgers do not offer.

An enterprise example and its lesson: A.P. Moller - Maersk and IBM announced TradeLens, a blockchain-based platform for sharing container-shipping documents and events, in 2018. It worked technically, but on 29 November 2022 the two companies said they would shut it down because it had not achieved the full global industry collaboration needed and had not reached commercial viability. Enterprise blockchains depend on the participants agreeing on a shared standard as much as on the technology.

Strengths and limits

Strengths: no single point of control, a public and verifiable history, and transactions that settle around the clock without an intermediary.

Limits: throughput is lower than centralized systems, fees rise when demand for block space is high (see gas fee), mistakes such as sending to a wrong address are usually irreversible, and data written on a public chain is visible to everyone. The "blockchain trilemma" describes the difficulty of maximizing decentralization, security and scalability at the same time.

Forks: because the rules are software, a chain can split. A soft fork tightens the rules in a backward-compatible way; a hard fork changes them so that old and new software disagree, which can leave two competing chains, as when Bitcoin Cash split from Bitcoin on 1 August 2017 or when Ethereum Classic continued after Ethereum's 2016 DAO fork. See fork.