Principles & Vision
On this page
A modular synthesizer for mathematics — where analytic structure, computation, and perception merge into one instrument.
1. Overview
Oakfield Operator Calculus (OOC) is the theoretical and architectural foundation of the Oakfield simulation engine. It describes how fields evolve under operators and how those operators compose, integrate, and interact. The runtime implements numerical, analytic, stochastic, and fractional operations, with experimental neural inference and hybrid operators. Symbolic solving, training workflows, and quantum extensions are outside the current runtime’s capabilities.
The calculus guides the code. Every header, kernel, and integrator in the Oakfield runtime expresses one layer of that design. Each operator is a module; each field is a signal; and the runtime is a modular analytic synthesizer.
2. Core Principles
1️⃣ Composability
In the calculus, operators are morphisms between field spaces. In the runtime, you compose them by binding their input and output fields and choosing an integrator. A diffusion operator and a time integrator, for example, provide a numerical realization of the heat equation:
Composition is the common workflow. You patch operators like synthesizer modules, with explicit parameters, field representations, and numerical settings described in the Lua API reference.
2️⃣ Analytic Continuity
The mathematical model describes functions over continuous domains, with derivatives and spectral representations. The runtime realizes those models with numerical field buffers, analytic kernels, and spatial or spectral operations.
The field API allocates real or complex arrays and records physical or spectral representation metadata. Resolution, timestep, boundary conditions, and the chosen method determine how closely a numerical experiment follows its continuous model.
3️⃣ Dual Representation (Symbolic + Numeric)
Operators have mathematical definitions and numerical kernel implementations. The registry exposes descriptions, parameter symbols, and mathematical views alongside executable contracts.
Today’s mathematical views help explain a numerical operator. The runtime does not provide general symbolic simplification, factorization, exact solutions, or automatic differentiation.
4️⃣ Physical Intuition
Oakfield is designed for people who think in flows, waves, fields, and forces. Every operator has a physical interpretation. ∇² is diffusion. ∂ₜ is time evolution. ∇× is rotation.
Visualizers, field statistics, and sound-observation operators make a simulation’s behaviour available for inspection.
5️⃣ Locality and Diffeomorphism
Local operators act on neighbourhoods; spectral transforms and other whole-field operations have broader dependencies. Scheduling and supported acceleration paths follow those dependencies and each operator’s execution contract.
The calculus describes fields on manifolds and transformations of their coordinates. Runtime experiments use discretized fields with the dimensional, spacing, representation, and boundary rules specified by each operator.
6️⃣ Observability and Feedback
Contexts expose field statistics, step metrics, profiling, logging, and visualization queries. These observations help connect a field’s mathematical description to its numerical behaviour.
Observability is a design principle: you should be able to inspect what the simulation is doing and how its numerical choices affect the result.
3. Architectural Ethos
Oakfield is structured in layers, each corresponding to a level of mathematical abstraction:
| Layer | Purpose | Manifestation |
|---|---|---|
| Mathematical | Define operators and their relationships | Operator calculus, functors, mathematical views |
| Computational | Implement operators numerically | Kernels, integrators |
| Runtime | Schedule and execute operations | Scheduler, memory manager |
| Perceptual | Inspect results as signals | Visualizers, statistics, sound observation |
Each layer is independent but tightly coupled. The mathematical layer doesn’t “know” about GPU kernels. The runtime doesn’t “know” about audio synthesis. Yet they work together seamlessly because the abstraction is correct.
4. Vision
The operator ecosystem includes stochastic forcing, fractional memory and subordination, and experimental neural inference and hybrid operators.
The goal is not to build a tool. The goal is to build a medium — a continuous, expressive, hands-on way of exploring mathematical structure.
Oakfield is to mathematics what the modular synthesizer is to sound — a hands-on, continuous, expressive medium for exploring structure.