What is blockchain technology, without the price chart

What is blockchain technology, in the paper that introduced Bitcoin, is a way to put transactions in an order that participants can check, without a bank sitting in the middle. This page stays with that paper. It is not investment advice, not a prediction, and not a guide to hiding money.

The source is Satoshi Nakamoto, Bitcoin: A Peer-to-Peer Electronic Cash System. The abstract says a peer-to-peer version of electronic cash would let payments go from one party to another without a financial institution. Digital signatures are part of that, but a trusted third party would still be needed to stop double-spending unless something else replaces it. The paper’s something else is a peer-to-peer network that timestamps transactions by hashing them into an ongoing chain of hash-based proof-of-work.

What is blockchain technology in that design

Section 3 describes a timestamp server. It takes a hash of a block of items and publishes that hash. The timestamp shows the data existed in time to be included. Each timestamp includes the previous timestamp in its hash, so the timestamps form a chain, and each new one reinforces the ones before it. That chain of hashed blocks is the structure people later called a blockchain. The paper’s name for the job is a public history of the order of transactions.

Section 2 says an electronic coin is a chain of digital signatures. An owner transfers it by signing a hash of the previous transaction and the next owner’s public key. A payee can check the signatures. The payee cannot, from signatures alone, know the coin was not already spent. A mint could check that, but then every payment depends on the mint. The paper’s answer is to announce transactions publicly and agree on one history of the order they were received. The earliest transaction is the one that counts.

Proof-of-work, as the paper defines it

Section 4 says a distributed timestamp server needs proof-of-work, similar to Hashcash, instead of a newspaper post. The work is finding a value that, when hashed, begins with a required run of zero bits. Checking it is a single hash. Changing the block means redoing the work, and changing an older block means redoing the blocks after it. The paper says the longest chain is the one with the most of that work, and that honest nodes with a majority of CPU power will grow the longest chain. That rule is the point of this page, not a brand of database.

The paper says the difficulty adjusts with a moving average so blocks do not arrive too fast when hardware gets quicker. It does not give you a number to trade on. The calculations section shows how an attacker’s chance of catching up falls as more blocks are added, under the paper’s assumptions. Those tables stay in the paper. This page will not copy them into a promise about your payment.

How a block gets accepted

Section 5 lists the steps. New transactions are broadcast. Each node collects them into a block and works on a proof-of-work. When a node finds one, it broadcasts the block. Other nodes accept it only if the transactions are valid and not already spent. They show acceptance by building the next block on that hash. If two blocks appear at once, nodes keep the first they saw and save the other branch until one chain is longer. Dropped messages are tolerated. A node that misses a block can ask for it when the next one arrives.

Section 6 says the first transaction in a block can create a new coin for the block’s creator, which is how new coins enter circulation without a central issuer. The paper compares that to miners spending resources. It also says a transaction can pay a fee when the outputs are worth less than the inputs, and that the incentive could later be fees alone. It argues a node with a lot of CPU power may earn more by following the rules than by undoing payments. That is an argument in the paper, not a forecast about a market.

What you can check, and what you should not add

Section 8 says a user can check a payment without running a full node by keeping block headers of the longest chain and a Merkle branch that ties the transaction to a block. The paper says this is reliable while honest nodes control the network, and weaker if an attacker overpowers the network. Businesses that take many payments may still want their own node. It does not mean trusting a screenshot. It means you can point at a place in the chain.

Section 10 says publishing every transaction removes the privacy of a private bank ledger. Privacy in the paper’s design is keeping public keys from being tied to names, and using a new key pair for each transaction so they are harder to link. Multi-input transactions can still show that inputs shared an owner. The paper does not describe a way to break that link after the fact, and this page will not either.

Do not treat the paper as a manual for attacking a network, mixing funds, or avoiding a law. The calculations section is about how hard it is for an attacker to replace a recent payment, not a recipe. Do not treat the paper as a reason to buy anything. It is a 2008 design for electronic cash. Later systems changed pieces of it. When a page says blockchain and means a private database with an administrator, that is a different design. Say which paper you mean.

Tomorrow, read the abstract and sections 3, 4, and 5. Write four sentences: what a timestamp chain is, why signatures alone do not stop double-spending, what proof-of-work adds, and how nodes accept a block. If a sentence has a price, a yearly storage figure, or an attack probability, delete it and link the paper. What is blockchain technology is done when those four sentences match the text.

Keep the author name and the URL in the notes. A diagram of three boxes labeled block, hash, and previous hash is enough for a reader who has not opened the PDF. If the diagram needs a market claim, it is no longer the paper.

When someone asks whether a payment is final, answer with the paper’s condition: more blocks after it make a rewrite costlier if honest nodes hold a majority of the work. You do not know that majority from a blog. You know the rule. Point at section 4 and stop.

Section 7 says old spent transactions can be dropped to save space once they are buried, if transactions are hashed in a Merkle tree and only the root sits in the block hash. The paper explains why the header can stay small. The storage arithmetic in that section belongs to the year it was written. Do not repeat it as a current hardware claim. The idea to keep is the tree, not the disk size.

Section 9 says value can be split and combined. A transaction can have more than one input and usually one output for the payment and one for change. You do not need a separate transaction for every small unit. That is bookkeeping in the paper, not a product feature list.

The conclusion says the system records a public history with proof-of-work so that changing it is impractical if honest nodes control a majority of CPU power. Nodes can leave and rejoin, and they accept the chain as proof of what happened while they were gone. They vote by extending valid blocks and refusing invalid ones. What is blockchain technology, in this conclusion, is that agreement mechanism. Anything sold as a blockchain that still needs one company to approve each write is outside the paper. Say so.