This article is general educational information from Metadiscover. It is not investment, legal, tax, or technical advice, and it does not recommend any product, network, or digital asset.
A blockchain (ブロックチェーン) is a shared digital record book that many computers keep in sync, and that adds new entries in linked batches called blocks. Once an entry has been added and confirmed, changing it later is extremely difficult, because every later block depends on it. That single idea supports digital currencies, tokenised records, and a growing list of other applications.
This article explains what a blockchain is and how it works, step by step, without assuming any technical background. It was written for readers in Japan and around the world who want a clear picture before reading anything more detailed. Where a number or a rule is quoted, its source is named, and a glossary near the end gives the Japanese term for each key word.
What is blockchain?
The United States National Institute of Standards and Technology (NIST) describes blockchains as tamper-evident and tamper-resistant digital ledgers that are implemented in a distributed way, without a central repository and usually without a central authority such as a bank, company, or government. Every word in that description matters, so it helps to take it apart.
- Ledger. A ledger is simply a list of records, most often records of something moving from one account to another. A blockchain is a ledger first and everything else second.
- Distributed. Instead of one master copy held by one organisation, thousands of computers each hold a full copy and compare notes constantly. No single copy is the official one; the copy that follows the rules and that most of the network agrees on is the truth.
- Tamper-evident and tamper-resistant. Records are grouped into blocks, and each block is cryptographically tied to the one before it. Any attempt to alter an old record leaves an obvious trace and would require an enormous amount of work to hide.
- Usually without a central authority. In public networks, the rules are enforced by software that everyone runs, not by an administrator who can quietly edit the database.
A useful mental picture is a town notebook that every household keeps a copy of. Each page is a block. Each new page begins with a short code that summarises the previous page. If anyone tries to rewrite an old page, the code at the top of the next page no longer matches, and everybody else's copy exposes the change.
Compare that with an ordinary database, such as the one a bank uses. The bank holds the master copy, its staff can correct or reverse entries, and customers trust the bank to do that honestly. A blockchain replaces that single trusted party with a set of rules and a large group of participants who check one another. Neither model is better in every situation; they solve different problems.
What is stored in a block?
A block is a batch of records plus a small amount of housekeeping data. On most public networks the records are transactions: a note that a certain amount of a digital asset, or a certain instruction, moves from one address to another. Blocks do not normally store documents, photographs, or personal files; they store compact records and, on some networks, small programs.
Alongside the transactions, a block header typically contains:
- the hash of the previous block, which is what links the blocks into a chain;
- a timestamp;
- a summary hash of all the transactions inside the block;
- on proof-of-work networks, a number called a nonce that is used in the puzzle described later in this article.
Networks add blocks at different rhythms. Bitcoin's documentation states that new blocks are added approximately every ten minutes on average, and the network adjusts its puzzle difficulty every 2,016 blocks to keep that pace. Ethereum's developer documentation describes fixed twelve-second slots, with one validator selected to propose a block in each slot. The rhythm is a design choice, not a law of nature, and it shapes how quickly a record is considered settled.
What is a hash and why does it matter?
A hash is a fixed-length fingerprint of any piece of data, produced by a mathematical function. The same input always produces the same fingerprint. Change even a single character, and the fingerprint changes completely and unpredictably. Given only the fingerprint, there is no practical way to work backwards to the original data.
A small example makes the point. The word "hello" and the word "Hello" differ by one capital letter, yet their hashes look nothing alike. NIST lists the properties that make hash functions suitable for blockchains: it should be practically impossible to find an input that produces a chosen fingerprint, to find a second input that produces the same fingerprint as a given input, or to find any two inputs that share a fingerprint.
Blockchains use hashes in two places. First, the transactions in a block are summarised into one hash, so anyone can check that the contents have not been altered. Second, each block header stores the hash of the previous block header. This second use is what makes the chain a chain. If someone edited a transaction in an old block, that block's hash would change, the next block's stored reference would no longer match, and the mismatch would ripple through every block that followed. To hide the edit, the attacker would have to rebuild every later block and then convince the rest of the network to accept the rebuilt version instead of the one it already holds.
How does blockchain work, step by step?
Putting the pieces together, here is what happens when a record is added to a public blockchain. The details differ between networks, but the sequence is broadly the same.
- A user creates a transaction. Using wallet software, the user states what should move where and signs the transaction with a private key. The signature proves that the owner of the sending address authorised it, without revealing the key itself.
- The transaction is broadcast. The wallet sends the signed transaction to a few computers on the network, which pass it on to others until most of the network has seen it.
- Nodes check the rules. Each computer that receives the transaction verifies it independently: the signature must be valid, the sending address must hold what it is trying to send, and the format must follow the protocol. Invalid transactions are simply dropped.
- A publishing node assembles a block. Certain participants, called miners on proof-of-work networks and validators on proof-of-stake networks, collect valid transactions into a candidate block and add the header data described above.
- The network agrees on the next block. The consensus rules decide whose candidate block is accepted. On proof-of-work networks that means solving a computational puzzle first; on proof-of-stake networks it means being selected to propose and then receiving enough attestations from other validators.
- Every copy is updated. The accepted block is appended to the chain, and every node updates its copy of the ledger. As further blocks are added on top, the record becomes progressively harder to alter. People often describe these later blocks as confirmations.
Two ideas in that sequence deserve their own explanation: addresses and keys, and consensus.
Addresses and keys
An address is a little like an account number that anyone may see and send to. A private key is the secret that proves ownership of that address and authorises anything leaving it. The two are mathematically linked, but the private key cannot be worked out from the address. Whoever holds the private key controls the address, which is why key management matters so much for anyone who uses a blockchain directly.
What do nodes do?
A node is any computer that runs the network's software and takes part in keeping the ledger. Not all nodes do the same job.
- Full nodes store a complete copy of the blockchain and check every block and transaction against the rules. They do not have to trust anyone else's copy, because they verify everything themselves.
- Publishing nodes, meaning miners or validators, do everything a full node does and also propose new blocks.
- Lightweight clients, including many wallet apps, do not hold the full ledger. They ask full nodes for the information they need and rely on those nodes to answer honestly.
In a public network, anyone with suitable hardware and an internet connection may run a node. That openness is what keeps the ledger distributed; the more independent copies exist, the harder it is for any one party to rewrite history or to shut the network down.
How do networks agree? Proof of work and proof of stake
Consensus (コンセンサス) is the set of rules that lets thousands of computers, which do not know or trust one another, agree on a single next block. The two most widely used approaches are proof of work and proof of stake.
Proof of work
In proof of work (プルーフ・オブ・ワーク), publishing nodes compete to solve a computational puzzle. NIST describes the essential property: solving the puzzle is difficult, but checking that a solution is correct is easy. In practice, miners repeatedly change the nonce in the block header and hash the header until the result falls below a target set by the network. The first miner to find a valid solution publishes the block. The network raises or lowers the target so that blocks keep arriving at the intended pace; Bitcoin's documentation describes this adjustment happening every 2,016 blocks.
The security of proof of work rests on cost. Rewriting the chain would require redoing the puzzles for every block after the edited one, faster than the honest network adds new blocks, which in practice means controlling more than half of the network's total computing power. That cost is also its main drawback: the electricity used by large proof-of-work networks is substantial. Independent estimates, such as the index maintained by the Cambridge Centre for Alternative Finance, put Bitcoin's annual electricity use in 2026 at a level comparable to that of a medium-sized country.
Proof of stake
In proof of stake (プルーフ・オブ・ステーク), there is no puzzle. Validators lock up, or stake, a quantity of the network's own asset as a form of security deposit. Ethereum's documentation describes proof of stake as a way to prove that validators have put something of value into the network that can be destroyed if they act dishonestly. On Ethereum, running a validator requires a deposit of 32 ETH, the network's own asset ether, into its deposit contract; this is a protocol parameter, not a suggestion about what anyone should own.
The network selects one validator to propose a block in each twelve-second slot, other validators attest that the block is valid, and a block becomes final once checkpoints attract votes representing at least two-thirds of all staked value. A validator that proposes conflicting blocks or otherwise breaks the rules can have part of its stake destroyed, a penalty called slashing. Ethereum moved from proof of work to proof of stake in September 2022, in an upgrade widely known as the Merge, and ethereum.org, citing an estimate by the Crypto Carbon Ratings Institute, reports that the change reduced the network's annual electricity consumption by more than 99 per cent.
A side-by-side view
| Question | Proof of work | Proof of stake |
|---|---|---|
| Who adds blocks? | Miners who solve a computational puzzle | Validators selected from those who have staked value |
| What makes cheating expensive? | Hardware and electricity needed to out-compute the honest network | Staked value that can be destroyed through slashing |
| Energy use | High, by design | Low |
| Example network | Bitcoin | Ethereum, since September 2022 |
Other consensus designs exist, particularly on permissioned networks where participants are known, but proof of work and proof of stake cover the public networks most people have heard of.
Can a blockchain be changed?
Blockchains are often called immutable. NIST is careful to say that this is not strictly true: they are tamper-evident and tamper-resistant. The difference matters.
- Recent blocks are less settled than old ones. Two publishing nodes occasionally produce competing blocks at the same time. The network resolves this by following the branch with the most accumulated work on proof-of-work networks, or the most validator attestations on proof-of-stake networks, so a block at the very tip of the chain can be replaced. This is why a recipient usually waits for several confirmations before treating a transfer as final.
- A majority attack is expensive rather than impossible. A party controlling more than half of a network's mining power or staked value could, for a time, reorganise recent blocks. NIST notes that such an attack is not technically difficult, merely expensive, and for large networks the expense is enormous.
- Coordinated changes have happened. Developers, node operators, and users all exercise some control over a network. NIST cites an occasion in 2013 when a new version of the Bitcoin client introduced a flaw that split the chain in two, and the developers chose to revert to the earlier version, discarding one of the chains. Such events are rare and highly visible, which is the point of a tamper-evident system.
- Permissioned networks can be changed by their owners. When a company or consortium runs a blockchain, the operator can legitimately alter rules or replace blocks, subject to whatever governance the participants agreed.
In everyday terms: records on a large public blockchain are extraordinarily hard to alter, but the guarantee comes from cost, transparency, and the number of independent participants, not from magic.
Public and private blockchains
NIST divides blockchains into permissionless and permissioned networks, which people often call public and private.
| Feature | Public (permissionless) | Private or consortium (permissioned) |
|---|---|---|
| Who can read the ledger? | Anyone | Approved participants, sometimes the public |
| Who can submit transactions? | Anyone with a wallet | Approved participants |
| Who can run publishing nodes? | Anyone who meets the protocol's requirements | Selected organisations |
| Who governs the rules? | Open-source developers, node operators, and users, through rough consensus | The owning company or consortium |
| Typical purpose | Open digital assets and applications | Record sharing between organisations that already know one another |
Permissioned networks trade some openness for finer control and higher throughput. They still benefit from the linked-block structure and from having several organisations verify the same records, but they rely on the participants being identifiable and accountable rather than on economic puzzles.
Is blockchain anonymous?
No. Public blockchains are pseudonymous. Addresses are strings of characters rather than names, but every transaction on a public network is visible to anyone, permanently. Bitcoin's own documentation states plainly that Bitcoin is not anonymous and that all transactions are public, traceable, and permanently stored.
An address becomes linked to a person whenever that person reveals it: when using a regulated service that verifies identity, when publishing the address, or when patterns of activity are analysed. NIST also notes that a blockchain's key infrastructure does not, by itself, support identity; proving who is behind an address requires something outside the chain.
What are smart contracts?
A smart contract (スマートコントラクト) is a program stored on a blockchain that runs when certain conditions are met. Ethereum's documentation describes it as a reusable snippet of code that a developer publishes into the network's shared state. Once published, the program executes exactly as written, on every node, whenever someone sends it a transaction.
Despite the name, a smart contract is not a legal contract. It is software, and it has the limitations of software. Bugs in a published contract can be difficult or impossible to fix, because changing the code usually means deploying a new contract. A smart contract also cannot see the outside world on its own; it only knows what is on the chain. To act on real-world facts such as a delivery date or a temperature reading, it must rely on a data feed, often called an oracle, and the reliability of the whole arrangement then depends on the reliability of that feed. Specialists call this the oracle problem.
Where is blockchain used?
The best-known use of public blockchains is moving crypto-assets (暗号資産（仮想通貨）) between people without an intermediary holding the funds. Beyond that, the technology is used or being trialled in several areas. The following are examples of where the linked-ledger idea is applied, not endorsements of any project.
- Records of ownership. Tokens on a blockchain can represent ownership of a digital collectible, a ticket, or a claim on something held elsewhere, with the transfer history visible to everyone.
- Supply chain records. Several organisations that handle the same goods can write to one shared ledger, so that each step in a product's journey is recorded once and checked by all.
- Programmable agreements. Smart contracts allow applications, from simple escrow arrangements to complex automated systems, to run according to published rules.
- Digital credentials. Pilot projects use blockchains to anchor certificates and identity attestations so that a third party can verify they have not been altered.
- Settlement between institutions. Groups of financial institutions and some central banks have run trials of permissioned ledgers for settling transfers between themselves.
In each case, the question worth asking is whether the application really needs several parties to share a tamper-evident record without trusting a single administrator. When it does not, an ordinary database is usually simpler and cheaper.
What are the limits of blockchain?
NIST opens its discussion of misconceptions by observing a tendency to overhype and overuse most new technology, and blockchain is no exception. A fair picture includes the limits.
- Energy. Proof-of-work networks deliberately consume large amounts of electricity. Proof of stake avoids this, at the price of a different security model.
- Speed and cost. Because every node processes every transaction, public blockchains handle far fewer transactions per second than centralised payment systems, and fees rise when the network is busy.
- Key management. There is no password reset. If a private key or its backup is lost, whatever it controlled is lost with it, and if it is stolen, the thief can act as the owner.
- No built-in identity. The chain knows addresses, not people. Anything that needs to know who someone is must add that layer separately.
- Software risk. Wallets, nodes, and smart contracts are software and can contain bugs. The trust that a blockchain removes from institutions is partly moved onto code and its authors.
- Governance disputes. When participants disagree about the rules, a network can split into two chains, an event called a fork. Deciding how a public network should evolve is a social process, not a technical one.
- Trust is reduced, not eliminated. NIST points out that users still trust the cryptography, the software, the developers, and the honesty of most participants. "Trustless" is a shorthand, not a literal description.
A note for readers in Japan
Japanese law has used the term crypto-assets, 暗号資産, since 2020, replacing the earlier 仮想通貨, which literally means virtual currency and remains common in everyday speech. Japan's Financial Services Agency publishes general information about the terminology and about the registration system for businesses that exchange crypto-assets. This article is about how the technology works and does not cover any regulatory or financial question; readers who need that information should consult the relevant official bodies directly.
Glossary in English and Japanese
| English term | Japanese term | Meaning |
|---|---|---|
| Blockchain | ブロックチェーン | A shared, distributed ledger that adds records in linked blocks. |
| Distributed ledger | 分散型台帳 | A record book kept as identical copies on many computers. |
| Block | ブロック | A batch of transactions plus a header that links to the previous block. |
| Hash | ハッシュ | A fixed-length fingerprint of data produced by a mathematical function. |
| Node | ノード | A computer that runs the network software and helps keep the ledger. |
| Consensus | コンセンサス | The rules by which the network agrees on the next block. |
| Proof of work | プルーフ・オブ・ワーク | Consensus based on solving a computational puzzle. |
| Proof of stake | プルーフ・オブ・ステーク | Consensus based on locking up value that can be destroyed for misbehaviour. |
| Smart contract | スマートコントラクト | A program stored on a blockchain that runs when its conditions are met. |
| Crypto-asset | 暗号資産（仮想通貨） | A digital asset recorded and transferred on a blockchain. |
| Ether | イーサ | The asset used by the Ethereum network, including for validator deposits. |
| Private key | 秘密鍵 | The secret that proves ownership of an address and authorises transfers. |
| Address | アドレス | A public identifier that can receive transactions. |
| Wallet | ウォレット | Software or a device that stores keys and creates transactions. |
| Fork | フォーク | A split in a network when participants follow different rules. |
Frequently asked questions
What is blockchain in simple terms?
A blockchain is a record book shared by many computers. New entries are added in batches called blocks, and each block is locked to the one before it with a digital fingerprint. Because everyone holds a copy and every block depends on the previous one, changing an old entry is extremely difficult and immediately visible.
Is blockchain the same as Bitcoin?
No. Bitcoin is one network that uses a blockchain, and it was the first widely used example. Many other networks use the same underlying idea with different rules, and some blockchains have nothing to do with currency at all.
Who controls a blockchain?
On a public network, no single party does. Developers write the software, node operators choose which version to run, publishing nodes add blocks, and users decide which network to use. NIST notes that saying "no one controls a blockchain" is not strictly true; control is spread across these groups rather than absent. Permissioned networks are controlled by the organisation or consortium that runs them.
Can data on a blockchain be deleted?
Not in the ordinary sense. Once a block is buried under many later blocks on a large public network, removing or editing it would require rebuilding the chain and persuading the network to accept the new version. This is one reason public blockchains are unsuitable for personal or sensitive data.
Do all blockchains use mining?
No. Mining is specific to proof-of-work networks such as Bitcoin. Proof-of-stake networks, including Ethereum since September 2022, use validators who stake value instead of solving puzzles, and permissioned networks often use other agreement methods entirely.
Is a blockchain just a database?
A blockchain is a kind of database, but a specialised one. Ordinary databases are fast, flexible, and controlled by an administrator who can edit records. A blockchain is slower and harder to change by design, in exchange for letting many parties who do not fully trust one another share one tamper-evident record.
How long does a blockchain transaction take?
It depends on the network and on how many confirmations the recipient wants. Bitcoin adds a block roughly every ten minutes on average, and Ethereum has a twelve-second slot in which one block may be proposed, but a recipient may wait for several further blocks before treating a transfer as settled.
Sources
- National Institute of Standards and Technology (NIST), Blockchain Technology Overview, NIST Internal Report 8202, 2018.
- ethereum.org developer documentation: Intro to Ethereum, Blocks, Proof-of-stake, Proof-of-work, and Energy consumption, 2026 editions.
- bitcoin.org, Developer Guide: Block Chain, and Frequently Asked Questions, 2026.
- Cambridge Centre for Alternative Finance, Cambridge Bitcoin Electricity Consumption Index, 2026.
- Financial Services Agency of Japan, information page on crypto-assets and registered exchange service providers, 2026.
What to read next
To see how a wallet puts these ideas into practice, our article MetaMask(メタマスク) Security Guide for Beginners explains how a self-custody wallet manages keys and how to recognise the common scams that target newcomers.