The problem a blockchain tries to solve
Most records we rely on, such as a bank balance, a land registry or a loyalty-points account, are kept by one organization. You trust that organization to write entries honestly and not to alter them later. That arrangement works well in many cases, but it has a single point of control and a single point of failure.
A blockchain is a way of keeping a record where no single party has to be trusted to hold the pen. Many computers each keep a copy of the same ledger, and a set of rules lets them agree on what gets added next. The design is not magic, and it is not the right answer to every problem. It is a specific trade: you give up some speed and efficiency in return for a record that is hard to change quietly.
What a block is
A ledger is a list of entries. In a blockchain, entries are collected into groups called blocks. A block typically holds three things:
- A batch of records, often transactions such as “account A sends an amount to account B”.
- A timestamp or similar ordering information.
- A reference to the block that came immediately before it.
That third item is what turns a pile of blocks into a chain. The reference is not a page number. It is a fingerprint of the previous block, produced by a hash function.
Hashing: a fingerprint for data
A hash function takes any input, from a single word to an entire file, and produces a fixed-length output that looks random. A widely used example, SHA-256, always produces a 256-bit result, regardless of whether the input is one letter or a whole book. Under the hood it is all binary code being mixed through a fixed series of operations.
Good hash functions have a few properties that matter here:
- Deterministic. The same input always gives the same output.
- Sensitive. Changing even one character of the input produces a completely different output, with no visible pattern.
- One-way. Given an output, there is no practical way to work backwards to the input other than guessing.
- Collision resistant. It is believed to be impractical to find two different inputs that give the same output.
The word “believed” is worth noticing. These properties rest on mathematics and on the fact that nobody has found a practical shortcut, not on a proof that none exists.
How hashes chain blocks together
Each block stores the hash of the previous block. Suppose someone wants to secretly change a transaction in an old block. Altering that block changes its hash. But the next block recorded the old hash, so it no longer matches. To hide the change, the attacker would have to recompute the next block’s hash, which then breaks the block after that, and so on to the end of the chain.
This is what people mean when they call a blockchain tamper-evident. Edits are not impossible. They are visible, because anyone holding a copy can check that each block’s stored reference matches the block it points to.
Shared copies and agreeing on the next block
If one computer held the chain, an attacker would only need to rewrite it there. So a blockchain network keeps many copies, held by participants often called nodes. The harder question is how the nodes agree on which block comes next when they are all working independently and some may be dishonest. The agreement method is called a consensus mechanism. Two families are common.
Proof of work. Participants compete to find a block whose hash meets a difficult-to-hit condition, for example a hash that starts with many zeros. The only way to find one is to try enormous numbers of guesses, which uses a lot of computing power from a graphics card or specialized chip, or a general central processing unit in the early days. The first to succeed announces the block, and others can verify it with a single quick hash. Rewriting history would require redoing that work for every later block faster than the honest network adds new ones, which is why long chains are considered costly to attack. Bitcoin uses this approach and is designed to add a block roughly every ten minutes on average.
Proof of stake. Instead of burning computing power, participants lock up some of the network’s own units as a deposit and are selected to propose and vouch for blocks. Dishonest behavior can cost them their deposit. Ethereum switched from proof of work to proof of stake in 2022, largely to cut its energy use.
Both methods are attempts to make cheating more expensive than honesty. Neither gives an absolute guarantee. A group controlling a majority of the mining power or staked value could, in principle, manipulate a network, and smaller networks are generally more exposed to this than large ones.
Public, private and permissioned chains
Not every blockchain is open to everyone. Public chains let anyone read the ledger and take part. Private or permissioned chains restrict who can write or validate, often because a group of companies wants a shared record without exposing it to the world. Restricting participation makes agreement faster and simpler, but it also reintroduces the trust in a small group that public chains were designed to avoid. Whether that is still better than an ordinary shared database is a fair question, and the honest answer depends on the case.
What a blockchain does and does not guarantee
A blockchain is good at one thing: showing that a record has not been changed since it was written, without relying on one keeper. It does not guarantee that the information was true when it was entered. If a false statement is written to the chain, the chain will faithfully preserve the false statement.
Other limits are worth knowing:
- Privacy is not automatic. Many public chains are pseudonymous rather than anonymous. Entries are visible to everyone, and patterns can sometimes be linked back to real people, which is a permanent kind of digital footprint.
- Mistakes are hard to undo. The same stubbornness that resists tampering also resists corrections.
- Speed and cost. Having many computers repeat the same checks is slower and uses more resources than one database would.
- The edges are still vulnerable. The chain may be sound while the surrounding software, exchanges and user accounts are not. Stolen passwords, phishing and data breaches at companies that hold people’s keys are common routes to loss.
Where the idea gets used
The best-known use is recording ownership and transfers of digital coins, covered in more detail elsewhere on this site. Ledgers are also being tested or used for tracking goods through supply chains, recording credentials, and moving money between institutions, with mixed results and plenty of projects that did not go beyond pilots. A useful rule of thumb is to ask who the participants are, whether they trust one another, and whether a plain shared database would do the job. Where one trusted party already exists and everyone accepts it, a blockchain often adds cost without adding much.
Keeping the idea in proportion
Strip away the vocabulary and a blockchain is a list of records, a fingerprint linking each group of records to the last, many copies of the list, and a set of rules for agreeing on additions. Those ingredients are individually simple, and the interesting part is how they combine. Knowing that makes it easier to read claims about the technology critically, including claims that it will fix everything and claims that it is useless.