Blockchain Explained: How Blocks, Nodes and Consensus Work

Blocks and the chain

A blockchain is a list of records grouped into batches called blocks. Each block holds a set of transactions, a timestamp and a special code called a hash.

A hash works like a fingerprint for data. Feed any information into a hash function and you get a short code of fixed length. Change even one character of the input and the fingerprint changes completely. Each block includes the hash of the block before it, and that is what links the blocks into a chain.

This link is the clever part. If someone altered an old transaction, that block's fingerprint would change, breaking the link to the next block, and the next, all the way to the latest one. Tampering becomes obvious straight away.

Nodes: many copies of the same record

A blockchain is not stored in one place. It is copied across many computers called nodes, often run by independent people and organisations in different countries. Each node keeps its own copy of the record and checks new transactions and blocks against the network's rules.

Because there are so many copies, there is no single server to shut down or quietly edit. If one node goes offline or tries to cheat, the others simply ignore it. On public blockchains such as Bitcoin and Ethereum, anyone with suitable hardware can usually run a node.

Consensus: how the network agrees

With thousands of computers involved, the network needs a fair way to decide which new block comes next and which version of the record is the real one. This process is called consensus. The two most common methods are:

• Proof of work. Miners use computing power to race to solve a puzzle. The winner adds the next block and earns a reward. Bitcoin uses this method. It has a long track record but uses a lot of electricity.
• Proof of stake. Validators lock up some of the network's coins as a deposit. The protocol picks validators to propose and confirm blocks, and dishonest behaviour can cost them part of that deposit. Ethereum has used this method since 2022. It uses far less energy, though critics argue it can favour those who already hold many coins.

Both methods aim for the same result: make cheating costly and reward those who follow the rules.

Why history is hard to change

To rewrite a past transaction on a large proof of work chain, an attacker would need to redo the work for that block and every block after it, faster than all honest miners combined. In practice that means controlling most of the network's computing power, which on a large network would be enormously expensive. This scenario is known as a 51% attack.

On proof of stake chains, an attacker would need to control a large share of all staked coins, and the rules allow dishonest validators to lose part of their deposit. On both systems, the older a transaction is, the harder it becomes to reverse. That is why people talk about waiting for confirmations.

Smaller blockchains with few miners or validators are easier targets, and several have suffered successful attacks of this kind. Size and decentralisation matter.

Limits and trade-offs

• Speed and cost. Every node checks every transaction, which makes public blockchains slower and often more expensive per transaction than an ordinary payment database.
• The trilemma. Designers often talk about balancing security, decentralisation and scalability. Improving one usually means giving up a little of another.
• Garbage in, garbage out. A blockchain can show that a record has not been changed, but it cannot prove the record was true when it was entered.
• Permanent mistakes. The same feature that blocks tampering also blocks corrections. A payment to the wrong address generally stays wrong.
• Not everything needs one. For many business tasks, a normal database run by a trusted company is simpler and cheaper.

When a project claims it needs a blockchain, a useful question is: who would we otherwise have to trust, and is removing that trust worth the extra cost and complexity?