What a blockchain actually is
A blockchain is a type of shared database in which records are grouped into blocks and linked in order, and in which many independent parties keep copies. Each block contains a set of entries and a cryptographic fingerprint of the previous block. Because of that link, changing an old entry changes its block’s fingerprint, which no longer matches what the next block recorded. The change becomes visible to anyone who checks.
The term is often used loosely. In practice there are public chains that anyone can join and read, and private or permissioned chains run by a defined group, such as a consortium of businesses. Their properties differ considerably.
The building blocks
Hashing. A hash function turns any input into a short, fixed-length fingerprint. The same input always yields the same fingerprint, and even a tiny change produces a completely different one. This makes it easy to check whether data has been altered without comparing the data itself. It is a close relative of the techniques in encryption basics, though hashing is not encryption and cannot be reversed to reveal the input.
Digital signatures. Participants sign entries with a private key, and others verify them with the matching public key. This shows who approved an entry and that it was not modified afterward.
Replication. Many nodes hold copies, so no single machine is a single point of failure, and there is no single administrator who can quietly rewrite the record.
Consensus. The nodes need a rule for agreeing which new block is valid. Public chains use mechanisms that make cheating costly, such as proof of work or proof of stake, while permissioned chains may use simpler voting among known members.
Together, these give a record that is append-only in practice: new entries can be added, but altering earlier ones is detectable and, on a well-distributed network, very hard to do without being noticed.
What a blockchain can help secure
- Integrity. It can show that a record has not been changed since it was written.
- Timestamping and ordering. It provides a shared sequence of events that parties can all point to.
- Reduced reliance on one party. When several organizations that do not fully trust each other need a common record, a shared ledger can avoid giving one of them control.
- Availability. Replication means the record can survive the loss of some participants.
What it does not secure
The limits matter as much as the strengths.
- Garbage in, garbage out. A blockchain proves that an entry has not changed since it was recorded, not that the entry was true when recorded. A wrong or fraudulent input stays wrong, permanently. This is a data quality problem the technology cannot fix.
- Confidentiality. On many chains, data is visible to all participants. Encryption can be applied, but then key management becomes essential.
- Keys. Anyone who steals or loses a private key can lose control of whatever it governs. Weak sign-in, phishing scams and poor handling of keys remain common causes of loss.
- Software flaws. Bugs in the chain software, wallets, or automated contract code can be exploited.
- Concentrated control. If a few parties control most of the validation, the promised decentralization is weaker than it appears.
- Endpoints. Attacks on the computers and apps that connect to the chain are unaffected by the chain’s design.
Storing data on a chain versus pointing to it
Blockchains are poor places to keep large files. Storage is replicated on every node, which is expensive, and data written to a public chain is difficult to remove. A common pattern therefore keeps the actual data in conventional storage and records only its hash on the chain. Anyone can later re-hash the file and compare it with the recorded fingerprint to confirm it is unchanged. The trade-off is that availability of the file depends on the off-chain system.
The permanence of data also creates privacy tension. Rules such as the right to erasure may be hard to reconcile with a record designed never to be altered, and personal data on a chain raises questions under privacy law. Keeping personal data off the chain, and applying data minimization, is a commonly discussed approach. Requirements differ by place, so organizations should seek qualified legal advice.
Moving data and value
For transfer, a shared ledger can let parties exchange records or assets with a single agreed history instead of reconciling separate copies. This is the idea behind digital money, covered in cryptocurrency basics, and behind some record-sharing between organizations. The usual weakness is speed and cost: waiting for agreement among many nodes is slower than writing to a single database, and some public chains charge fees that rise with demand.
Blockchain versus an ordinary database
Many projects that consider a blockchain would be served just as well by a regular database. A rough set of questions can help.
- Do several parties write to the record who do not trust one another?
- Is there no obvious neutral party to run the database?
- Is a permanent, tamper-evident history essential?
- Are the costs in speed, complexity and flexibility acceptable?
If the answer to most is no, a conventional database with good access controls, backups and audit logs is typically simpler, faster, cheaper and easier to correct. A database owned by one organization can also be secured well, using the principle of least privilege and logging. A blockchain adds value mainly when shared control and tamper evidence are the central requirements. How one fits among databases generally is covered in the comparison of a data warehouse and a database.
A balanced summary
Blockchains offer a particular combination: a shared, tamper-evident record without a single owner. That combination is useful in some situations and unnecessary in many others, and it does not by itself make data private, correct or safe. A sound evaluation looks at the specific problem, who needs to trust whom, and what the alternatives cost.