The essence of Bitcoin: computing system or organizational system? The key to understanding Bitcoin lies not in "computing power", but in the "boundary of consensus". Structurally, Bitcoin can be clearly divided into two orthogonal dimensions: Consensus and Policy. Firstly, the sole objective of the consensus layer is to determine the global order of transactions in a decentralized system. It gradually converges the entire network to a universally recognized history through the "longest chain principle" and PoW mechanism. The key here is not the calculation itself, but whether the result is accepted by all nodes. Therefore, consensus is not "calculated", but "recognized". Once the transaction order is determined, the double spending problem naturally disappears. In other words, the consensus of Bitcoin is essentially constructing an irreversible time series. Secondly, the strategy layer belongs to the local behavior of nodes, such as transaction filtering, fee sorting, packaging preferences, etc. These logics can be freely modified without the need for consistency across the entire network, and will not affect whether the system forks or not. Therefore, strategy represents the "local rationality" of nodes, which belongs to a typical computational problem. This division brings an important breakthrough: computation and consensus are completely decoupled. The role of computing power is only to compete for "accounting rights", rather than to determine "what is true". What truly determines history is the selection structure of the entire network, not the single point computing power. If further abstracted, Bitcoin is actually a three-layer structure: Computing layer: Perform transaction verification and state transition with complete certainty Organizational layer: Implementing distributed coordination through consensus mechanism, with probabilistic convergence characteristics Time layer: forming an irreversible historical sequence through block accumulation Therefore, the essence of Bitcoin is not a 'computing machine', but a decentralized organizational machine. It is not responsible for solving problems, but for solving a more fundamental proposition: How to reach a consensus on which outcome is effective in the world without a center. That's also why it can be said: Computing power determines who writes history, but it does not determine what history is.