b-money and Bit Gold: Bitcoin's Direct Predecessors

In November 1998, Wei Dai published a short paper on the cypherpunks mailing list describing b-money, a "scheme for a group of untraceable digital pseudonyms to pay each other with money and to enforce contracts amongst themselves without outside help."

b-money proposed two protocols. In the first, every participant maintains their own copy of a ledger tracking how much money each pseudonym holds. Money is created by broadcasting the solution to a previously unsolved computational problem, with the cost of computation determining the value. In the second, a smaller set of servers maintains the ledger, with users monitoring for honesty.

The second protocol bears striking similarity to Bitcoin's mining system, though Dai acknowledged it was incomplete and didn't solve the problem of reaching consensus among servers.

Nick Szabo, a computer scientist and legal scholar, independently developed Bit Gold around the same period. Szabo was deeply influenced by the properties of physical gold: scarce, costly to produce, and valuable precisely because of that cost.

Bit Gold worked as follows: a user creates a proof-of-work string (similar to HashCash), timestamps it with a distributed timestamp service, and adds it to a distributed property title registry. Each new proof-of-work string includes the hash of the previous one, creating a chain. The cost of computation gives the tokens intrinsic scarcity, and the chain provides ordering and tamper resistance.

Szabo recognized that proof-of-work outputs from different eras would have different computational costs, making them non-fungible. He proposed a market mechanism to standardize units, but this remained an unsolved problem.

Both b-money and Bit Gold articulated the vision of decentralized digital money with remarkable clarity, but neither solved all the technical challenges. The hardest problems were:

Double-spending prevention without a central authority: How do you stop someone from spending the same token twice when there's no bank to check? b-money required all participants to agree on account balances, but provided no mechanism for resolving disagreements.

Sybil resistance: In a decentralized system, what prevents an attacker from creating thousands of fake identities to overwhelm the consensus process? Neither proposal had a complete answer.

Difficulty adjustment: If computational power increases over time, tokens become cheaper to produce. Bit Gold's non-fungibility problem stemmed from this. Bitcoin would solve it elegantly with automatic difficulty recalibration.

Satoshi Nakamoto's Bitcoin whitepaper cites both b-money (explicitly) and incorporates ideas from Bit Gold (structurally). The connection is clear:

b-money's distributed ledger becomes Bitcoin's blockchain. Bit Gold's chain of proof-of-work outputs becomes Bitcoin's block chain. Both proposals' use of computational puzzles to create money becomes Bitcoin's mining.

What Satoshi added was the longest-chain rule: nodes always follow the chain with the most accumulated proof-of-work. This elegant mechanism solved consensus, double-spending, and Sybil resistance simultaneously. Combined with difficulty adjustment and a fixed supply schedule, it completed the vision that Dai and Szabo had outlined years earlier. Bitcoin didn't emerge from nothing — it was the culmination of a decade of iterative design.