Resource-Sensitive Computation

Resource-sensitive computation is the paradigm of programming and system design in which the consumption, transformation, and disposal of resources are first-class concerns of the formal model, not afterthoughts of implementation. In a resource-sensitive framework, every computation step is understood as consuming inputs and producing outputs, and the formalism must track whether resources are used once, many times, or not at all.

This paradigm is grounded in linear logic and affine logic, where the structural rules of proofs encode resource discipline. It extends to process calculi, where channels are resources that must be acquired and released; to type systems, where ownership and borrowing track memory; and to quantum computing, where the no-cloning theorem enforces resource sensitivity at the level of physical law.

The broader field of resource analysis seeks automated methods for inferring and verifying resource bounds, moving resource tracking from the programmer's burden to the compiler's responsibility.

Resource-sensitive computation is not a niche concern for embedded systems programmers. It is the correct way to think about computation in a physical universe. The assumption that memory is infinite, energy is free, and time is unbounded is not abstraction — it is denial. Every program runs on a physical substrate, and resource-sensitive computation is the refusal to pretend otherwise.