The title hints at what the white paper intends to explain. Bitcoin explains real-world electronic cash that can be exchanged using the internet, just like we do with fiat money in the real world.
The white paper tells the problem it is trying to solve in our current digital monetary system. You spend a dollar bill at a store and get something in return. You cannot spend that same dollar bill again to purchase something else. But with the digital version of the dollar bill, you can take a copy of the same dollar bill and spend at two or more places simultaneously, contributing to the ‘double-spend problem.’ In digital money, this problem is solved by intermediary businesses and banks by keeping track of each transaction and adjusting a digital ledger book accordingly. The Bitcoin solution removes the intermediaries' and mentions using hashing to seal transactions to preserve authenticity. Hashing is converting a digital data set into a unique shorter string used to identify original data. Then to make the digital ledger immutable, the paper suggests chaining the hashed data into a data structure by spending computational work such that to change the ledger will be a rework of the whole computational work.
The paper introduces the limitations of a trust-based monetary system where you need to involve some trusted institution to enable reversible transactions. The new system proposed mentions replacing trust with mathematical cryptographic proofs and a peer-to-peer global computer system that functions by itself using an incentive mechanism that comes with the network protocol.
The new digital money is a chain of digitally signed transactions in a universal ledger. During the transfer of digital money from one person to another, the current owner signs the previous transaction hash with their private key* and adds the public key* of the payee. All transactions are pseudonymously announced in the ledger for the payee to verify the chain of past transactions. The system has a mechanism for the entire network to reach a consensus on a single history of all immutable transactions.
*The public key is public and open to anyone in the system. However, the private key is private, only stored on the user’s device, and is used to decrypt data. Bob wants to send Alice an encrypted email. To do this, Bob takes Alice’s public key and encrypts his message to her. Then, when Alice receives the message, she takes the private key known only to her to decrypt the message from Bob.
Timestamping transaction block is a mechanism to prevent duplicate transactions. A timestamp server is essentially a node in the Bitcoin network that takes the hash of a block of transactions and publically broadcasts the hash on the network. A timestamp is added to the hash for the point of time the PoW is identified and the block hash broadcasted. Also, each timestamp includes the previous block timestamp in its hash. This hash serves as proof that a set of transactions existed at some point in time, and network participants can verify the order in which transactions blocks are broadcasted.
Proof-of-Work is at the heart of the Bitcoin blockchain, which regulates the competing nodes to add new blocks to the ledger. Proof-of-Work is essentially computing power spend by miner nodes. A new block is accepted when a miner comes with a winning PoW first, and the miner earns Bitcoin as a reward. Finding the winning PoW is very difficult, and each miner spends expensive specialized computations to achieve this. The miner’s goal is to create a hash matching a current target set by the Bitcoin protocol. They must create a hash with enough zeros at the front for which the probability is very low. But miners across the world execute trillions of such computations in a second to reach the goal first. Once a winning PoW is reached, the Bitcoin protocol will set the next target. The goal of PoW is to prevent counterfeit Bitcoin and the double-spending problem. In essence, PoW is the bank implementation that regulates bitcoin transactions.
This is how a Bitcoin network works at a high level. New transactions are broadcast to all nodes. Each node collects all simultaneously occurred transactions into a block for convenience. The nodes compete for PoW, and when success usually takes 10 minutes, broadcasts the block to all nodes. One of the reasons why the Bitcoin network is mentioned as unscalable and not good for high-volume transactions is that most nodes will have to verify all transactions in a block and reach consensus by adding the new block into the blockchain. Nodes always consider the longest chain as the correct one and keep extending it.
Bitcoin is an all-or-nothing system. In a sense, if the system exists as designed, it is good for all of humanity, and if someone tries to attack the system, it is bad or a loss for everyone, including the attacker. The incentive mechanism and the coin economics make Bitcoin special compared to past peer-to-peer systems like BitTorrent. The node operators in Bitcoin are incentivized to maintain the network and follow the protocol rules. Every time a PoW generates a new block, the miner of that block gets bitcoin till the protocol limit of Bitcoin supply lasts. Transaction fees are also collected from the transaction's sender to prevent spam transactions and give miner fees.
The Bitcoin blockchain network keeps growing infinite as more and more transactions are added to the network. This, in turn, keeps adding the storage space of participating nodes. To reclaim disk space, block data are arranged in a special data structure called Merkle tree, and data from old transactions are discarded, except the root data is kept in the block’s hash.
SPV is a mechanism that allows users to quickly verify if a transaction is successful in the Bitcoin network without downloading the whole blockchain data. The user only needs only three things to perform SPV. The list of block headers, the transaction details, and the Merkle proof of the transaction can be obtained by a user by querying the bitcoin protocol. Bitcoin lightweight wallet clients make use of SPV to verify transactions really fast.
Bitcoin is not an account-based ledger but a transaction-based ledger. In an account-based ledger, the flow of currency between accounts is verified. A transaction-based account has inputs & outputs and is chained to the previous transaction or source using hash pointers. At the most, a transaction can have any number of inputs and a max of two outputs, one is the actual payment, and the other is the return change if any. Since Bitcoin is transaction-based, the value of a coin can be split and transacted. For example, if you buy 1 bitcoin from a decentralized exchange, you might be getting .2 from one seller .7 from another seller, and .1 from another, all combined inputs to receive 1 Bitcoin as output. This is done for efficiency and anonymity.
The Bitcoin network is pseudonymous. This means even though all transactions are published on the public ledger, they are not linked to a real-world identity. What is used as identities are public keys which look like a gibberish string of characters. Users can keep changing their public keys so that no one can really track specific transactions of a public key address and try to link some real-world entity.
The Bitcoin network becomes more secure as it grows. The scenario of an attacker trying to generate an alternate chain is impossible because that will require winning multiple PoW in a row. Also, the 51% attack usually mentioned with the Bitcoin network will require more than half of the miners to conspire to compromise the network. This is also impossible as the cost required to achieve far outplays the benefits.
In summary, the proposed system for digital transactions replaces trust with a peer-to-peer system using proof-of-work incentivized to work for itself with digital signatures that control ownership.
Finally, Satoshi Nakamoto — a bunch of human brains, in my opinion, concludes by saying:
“The network is robust in its unstructured simplicity”