What a blockchain actually is

The word describes a data structure, not a product

The term “blockchain” is used loosely, so it helps to start with what it names precisely. A blockchain is a way of organising records so that a group of computers that do not trust one another can still agree on a single, shared history. Ethereum’s own documentation puts it plainly: “A blockchain is a public database that is updated and shared across many computers in a network.” Everything else follows from that one sentence.

Two properties do the heavy lifting. First, the database is shared: many machines each keep a full copy rather than trusting one central operator. Second, the database is append-only: new records are added, but existing records are not meant to be edited or removed. A blockchain is the combination of those two ideas, held together by cryptography.

Why the ledger is copied everywhere

In a conventional system, one organisation runs the authoritative database. If you want to know your bank balance, you ask the bank, and you trust its answer. A blockchain removes that single point of authority. Instead, the record is replicated across independent participants, called nodes. Each node stores the ledger and checks new data against the same rules.

This is what people mean when they call a blockchain “decentralised”. No single node is in charge, and no single node has to be trusted, because every node can independently verify the whole history for itself. The trade-off is that all of those copies must be kept in step. As Ethereum’s documentation notes, “every computer in the network must agree upon each new block and the chain as a whole.” Getting thousands of separate machines to agree, without a referee, is the central problem a blockchain solves.

Blocks: transactions grouped and stamped

Records are not added one at a time. They are collected into batches called blocks. A block is essentially a list of transactions that happened in a short window of time, plus some metadata about the block itself. Batching is an efficiency measure: it lets the network confirm many transactions together and agree on one snapshot of history at a time, rather than negotiating over every individual entry.

Each block also carries a cryptographic fingerprint, called a hash. A hash function takes an input of any size and produces a fixed-length string that is unique to that input in practice. Change a single character of the input and the hash changes completely and unpredictably. This gives every block a compact identifier that depends on its exact contents.

The chain: why blocks are linked

Here is the detail that turns a pile of blocks into a blockchain. Every block records the hash of the block immediately before it. Ethereum’s documentation describes blocks as “batches of transactions with a hash of the previous block in the chain.” Each block therefore points backwards to its parent, and that parent points to its own parent, all the way to the very first block.

Because a block’s hash depends on its full contents, and because that hash is written into the next block, the blocks are bound together in a specific order. You cannot quietly change an old transaction. If you did, that block’s hash would change, which would no longer match the reference stored in the following block, which would then need changing too, and so on. In Ethereum’s words, “one change in any block in history would invalidate all the following blocks as all subsequent hashes would change.”

This is the source of the “tamper-evident” property. The design does not physically prevent someone from editing their own copy of an old block. What it guarantees is that any such edit is immediately obvious to everyone else, because the chain no longer links up. Honest nodes reject the altered copy.

Consensus: deciding what gets added next

Linking blocks explains why the past is hard to rewrite. It does not, by itself, explain who gets to write the next block, or how the network settles disagreements when two valid-looking blocks appear at once. That is the job of a consensus mechanism.

Different networks use different consensus rules. Bitcoin uses proof of work, in which participants expend computing effort to earn the right to add a block; the Bitcoin developer documentation explains that this design means “the cost to modify a particular block increases with every new block added to the block chain.” Ethereum today uses proof of stake, where, as its documentation describes, validators lock up funds and are “randomly selected to propose blocks that other validators check and add to the blockchain.” The mechanics differ, but the goal is the same: to make adding honest blocks cheap and rewriting history expensive, so that independent nodes converge on one shared chain without a central authority.

Reading the ledger versus changing it

A useful mental split is between reading and writing. Reading a public blockchain is open to anyone: the full history is replicated across nodes and can be inspected freely, which is why block explorers exist. Writing is deliberately constrained: a new entry must be a valid, properly authorised transaction, it must be packaged into a block by whoever the consensus rules select, and that block must be accepted by the rest of the network. The asymmetry, easy to read and verify, hard to forge, is the whole point.

What a blockchain is not

Precision cuts both ways, so it is worth naming some common misreadings. A blockchain is not automatically “unhackable”; the ledger’s integrity says nothing about the security of the software, wallets, or exchanges built around it, which is where most real-world losses occur. It is not a guarantee that recorded information is true; the chain faithfully preserves whatever was submitted, including mistakes and lies. It is not inherently anonymous; on public chains, activity is pseudonymous and permanently visible. And it is not, on its own, a currency, an investment, or a company. A blockchain is a record-keeping structure. Coins, tokens, contracts, and applications are things people build on top of it.

Held to that definition, the technology becomes far easier to reason about. Every later article in this series, on keys, transactions, fees, contracts, and explorers, is really a description of how people read from and write to this one shared, append-only, cryptographically linked ledger.

How a new participant catches up

One question sharpens the whole picture: if there is no central server, how does a computer joining the network for the first time obtain the correct history? It downloads blocks from its peers and verifies them itself, rather than taking anyone’s word. Starting from the agreed first block, it checks that each block’s recorded parent hash matches the block before it, that every transaction is valid under the rules, and that the consensus conditions for each block are met. Only a history that passes every one of these checks is accepted. This is why the network does not need a trusted coordinator: correctness is something each participant can confirm from the raw data, and a forged or altered history simply fails verification.

It also clarifies what “trustless” means, a word often used carelessly. It does not mean no trust is involved anywhere; it means you do not have to trust a particular company or person to be honest, because you can verify the shared record against fixed, public rules. Trust shifts from an institution to a set of mechanisms anyone can inspect. That shift, from trusting a party to verifying a process, is the quiet idea underneath all the noise about blockchains.

Sources

• Ethereum.org — Introduction to Ethereum
• Ethereum.org — Blocks
• Ethereum.org — Consensus mechanisms
• Bitcoin Developer Guide — Block Chain
• Nakamoto, S. — Bitcoin: A Peer-to-Peer Electronic Cash System (whitepaper)