Finite Realism is an operational framework for epistemology, metaphysics, and physics that makes finite resource constraints explicit. Ontological commitment is licensed only within a declared window W, and only when a structure passes a joint earning criterion: it must (i) compress sequential predictions beyond a predeclared baseline (prequential/MDL or PAC-Bayes gains) and (ii) remain stable under admissible interventions, as measured by a window integral probability metric DW built from implementable tests and guarded by data-processing closures. Objectivity is not assumed but derived: it is the communication fixed point at which capacity-limited agents converge when exchanging bounded messages honestly within W. The formalism develops a window semantics using presheaves over measurement contexts and DPI-guarded IPMs, with all optimizations carried out on finite partitions induced by the admissible test class. A standing contract fixes three metrics — the licensing metric DW, a hybrid compute/report norm dₕyb, and a product total-variation norm dPi used for LP projections — together with explicit transfer bounds between them. An output-level budget inequality aggregates decoherence residuals, detector mismatch, intervention sensitivity, ND-projection error, and sampling error into a single auditable certification bound. Two case studies anchor the framework in physics. In general relativity, a finite-sample latency diagnostic tests whether a noisy beacon network earns a conformal structure: the classic theorem that causal order determines the conformal class serves as a consistency target, with Hoeffding-based concentration controlling mis-order probability. In quantum theory, pointer outcomes are certified as ε-facts — budgeted claims that a measurement result is operationally definite at resolution ε — using explicit Lipschitz witnesses for robust contextuality (worked out for CHSH) and visibility-based decoherence surrogates with declared scope conditions. Together, these components replace informal appeals to explanatory virtue with quantitative compression-under-control, and make the scope and limits of every ontological claim auditable within its declared window.
Building similarity graph...
Analyzing shared references across papers
Loading...
Craig Holmander (Sat,) studied this question.
www.synapsesocial.com/papers/6992652ceb1f82dc367a10e0 — DOI: https://doi.org/10.5281/zenodo.18639311
Craig Holmander
Building similarity graph...
Analyzing shared references across papers
Loading...