SCIENCE · DERIVATION
First Principles Derivation
Three independent branches of physics. One convergent constraint. No assumptions beyond the mathematics.
A first-principles derivation begins with no assumptions beyond the axioms of mathematics and the observed structure of the system. It does not start with a desired outcome and work backward. It starts with what is true about the system — its symmetries, its dynamics, its scale structure — and derives what must follow. The Omni-Compass governing law is derived this way. The result is not an engineering choice. It is a mathematical consequence.
Step 1 — Identify the Symmetry (Noether)
The first step is to identify the continuous symmetry of the system. Compute infrastructure exhibits translational symmetry in workload: the same workload, on the same hardware, at different times, produces the same result. This is not an assumption. It is an observable property of deterministic computation. Emmy Noether's theorem — proved in 1915 and described by Einstein as the most significant creative mathematical genius thus far produced — states that every continuous symmetry of a physical system corresponds to a conserved quantity. The translational symmetry of compute workloads therefore guarantees the existence of a conserved quantity. That conserved quantity is the governing law's invariant: the Work-per-Bill ratio at the system's optimal operating point.
Step 2 — Construct the Lyapunov Function
The second step is to construct a Lyapunov function that measures deviation from the conserved quantity. Aleksandr Lyapunov proved in 1892 that a dynamical system is asymptotically stable if there exists a scalar function V(x) that is positive definite, zero at the equilibrium, and strictly decreasing along every trajectory of the system. The governing engine's Lyapunov function is the deviation from the optimal Work-per-Bill ratio: V(x) = ||WpB(x) − WpB*||². Every control action the engine takes is constrained to reduce V(x). The system converges to the optimal operating point. It cannot drift. It cannot oscillate. The mathematics forbids it.
Step 3 — Prove Scale Invariance (Renormalization Group)
The third step is to prove that the governing law holds at every scale. Kenneth Wilson's renormalization group theory — for which he received the 1982 Nobel Prize in Physics — describes how a system's effective laws transform under changes of scale. When the laws are invariant under scale transformation, the system exhibits the same behavior at every scale. The governing law's constraint is scale-invariant: the same closed-form expression that governs a single Kubernetes pod governs a multi-region cloud deployment. No per-scale modification is required. No per-environment tuning is needed. The law applies universally.
Step 4 — Close the Circle
The three proofs close the circle. Noether proves a conserved quantity exists. Lyapunov proves the system converges to it. Renormalization group proves the law holds at every scale. The Five Operations of the Four-Stack Unified Law implement the constraint: Observe (measure the complete system state), Classify (identify the current operating regime), Constrain (evaluate the admissibility envelope), Actuate (commit only admissible transitions), Verify (confirm convergence and update the Lyapunov function). The circle is closed. Every operation feeds the next. The system cannot exit the governed state without violating the constraint — and the constraint is enforced before any action is committed.
The Formal Statement
Let S be a distributed compute system with state space X, transition function T: X → X, and workload W: X → ℝ⁺. Define the Work-per-Bill ratio WpB(x) = W(x) / C(x) where C(x) is the resource cost of state x. Let WpB* = sup{WpB(x) : x ∈ X} be the optimal ratio. Define the Lyapunov function V(x) = ||WpB(x) − WpB*||². The Omni-Compass governing law is the constraint: T(x) is admissible if and only if V(T(x)) ≤ V(x). Under this constraint, the system is asymptotically stable with equilibrium at WpB*. The constraint is scale-invariant by the renormalization group argument applied to the translational symmetry of W. The conserved quantity WpB* exists by Noether's theorem applied to the same symmetry. The derivation is complete.
What the Derivation Proves and What It Does Not
The derivation proves: (1) a conserved quantity exists for any compute system exhibiting translational workload symmetry; (2) the system converges to that conserved quantity under the governing constraint; (3) the constraint is scale-invariant. The derivation does not prove: (1) that every possible inefficiency is eliminated — the constraint excludes inadmissible transitions, not all suboptimal ones; (2) that the observation of system state is perfect — the governor operates on measured state, which is subject to measurement error; (3) that the constraint applies to systems that do not exhibit translational workload symmetry. The proof is honest. The claims are bounded. The receipts are public.
Formal Notation Reference
SDistributed compute systemXState space of ST: X → XTransition functionW(x)Work completed in state xC(x)Resource cost of state xWpB(x)Work-per-Bill ratio = W(x)/C(x)WpB*Optimal Work-per-Bill ratioV(x)Lyapunov function = ||WpB(x) − WpB*||²C(x,T(x))Admissibility constraint∇VGradient of Lyapunov functionCitations
Noether, E. (1915). Invariante Variationsprobleme. Nachrichten von der Gesellschaft der Wissenschaften zu Göttingen.
Lyapunov, A.M. (1892). The General Problem of the Stability of Motion. Kharkov Mathematical Society.
Wilson, K.G. (1971). Renormalization Group and Critical Phenomena. Physical Review B.
Einstein, A. (1918). Letter to Felix Klein on Noether's theorem. Göttingen Academy.
The complete benchmark harness implementing this governing law is publicly available on GitHub for open evaluation and independent replication by any engineer, institution, or evaluator worldwide. No license is required to verify. A license is required to monetize or commercialize.
GitHub