What Is a Blockchain, Actually?

Blockchain has spent a decade as a buzzword, which is a shame, because underneath the noise sits one genuinely interesting idea: a ledger that nobody owns and everybody can check. This post explains what that means mechanically, what it is actually good for, and what it costs, without the evangelism in either direction.

The mechanics

A blockchain is a database with two unusual properties: many computers hold identical copies, and entries can be added but never edited. Transactions are grouped into blocks, and each block cryptographically fingerprints the one before it (that is the chain). Altering an old entry would mean re-forging every block after it on thousands of independent machines at once. In practice, the past is locked.

The clever part is how strangers who do not trust each other agree on which new block comes next. A consensus mechanism makes lying expensive. In Proof of Work (Bitcoin) it demands costly computation; in Proof of Stake (Ethereum, since 2022) it demands a deposit that gets destroyed on cheating. Either way, the result is the property everything else builds on: the network agrees on one history without a bank, notary or company in the middle.

Everything on the usual feature list falls out of this. Decentralization: no single point of control. Immutability: the locked past. Transparency: public chains are readable by anyone. Security: cryptography plus economics, not a password on someone's server.

What it is actually for

The honest test for a blockchain use case: do the participants distrust each other, and is there no referee they would all accept? Where the answer is yes, it earns its cost.

• Money without banks. The original use, still the biggest. Bitcoin moves value peer to peer with no institution able to block or reverse it, which matters most where banks are inaccessible, unstable or politically weaponized. The billion-plus unbanked people with phones are the standing argument.
• Finance without intermediaries. DeFi rebuilds exchanges, lending and market-making as open code. Settlement takes minutes rather than days, and anyone can audit it. It also faithfully rebuilt several of finance's failure modes, at higher speed.
• Records nobody can quietly edit. Provenance for goods, credentials, land titles in places where the land registry is the thing you cannot trust. Real, but narrower than the hype. A blockchain guarantees the record is untampered, not that whoever typed it in told the truth. Garbage in, immutable garbage out.

Where a trusted referee already exists and works, as in most corporate databases, a blockchain is just the slow, expensive way to run a normal database. Most "enterprise blockchain" projects of the 2010s died of exactly this.

The honest costs

Throughput. Every node checking everything means base chains are slow. Bitcoin handles a few transactions per second and Ethereum about 15, versus tens of thousands for a card network. The workarounds are layered systems (Lightning on Bitcoin, rollups on Ethereum) or newer chains like Solana that trade heavier hardware requirements for speed. It works, at the cost of complexity.

Energy. Mostly a Proof-of-Work criticism. Bitcoin mining consumes on the order of a small country's electricity, by design. Proof of Stake cut Ethereum's usage by about 99.95%, and most newer chains started there. Bitcoin is not switching, so the argument stays alive.

Usability. Still the biggest real barrier. Seed phrases with no password reset, gas fees, irreversible mistakes, and a scam density that punishes every lapse. Self-custody currently demands a level of operational care most people reasonably refuse. Wallet UX improves yearly; the gap to a banking app remains wide.

Regulation and fragmentation. Jurisdictions have gone every direction from ban to embrace, and businesses dislike that uncertainty more than they would dislike strict but clear rules. Meanwhile chains multiply and interoperate badly, and cross-chain bridges are the industry's most reliably hacked component.

Where it stands

Fifteen years in, the honest summary: blockchains found undeniable product-market fit as uncensorable money rails, stablecoins became the sleeper hit that actually gets used for payments and remittances at scale, DeFi works but remains a professionals' arena, and most non-financial applications are still looking for their reason to exist. That is neither the revolution promised in the whitepapers nor the pure casino claimed by critics. It is a young settlement technology that does one thing no prior system could, still figuring out what else that unlocks for it. For a trader, it is also the base layer of the entire asset class we trade every day, which makes the mechanics worth an evening of your time whatever your side in the ideology war.