We present version 6.0 of the covariant dissipative programme as a proposal for unified physics to be scrutinized, tested, and, if necessary, rejected by the scientific community. The guiding principle remains Einsteinian: physics is what survives finite-resolution observation. Observation is a channel, a channel loses information, and the finite-resolution shadow is therefore governed by contraction, convexity, partial minimization, defect monotonicity, commutants, benchmark entropy, and dissipative closure. The aim of the present version is to make that vision theorem-level exact at the remaining categorical, infrared, ultraviolet, numerical, and editorial bottlenecks. The main formal upgrades are the following. Operational universality is expressed through the realized image together with its essential image, while the canonical reduced architecture is proved terminal on the correct pseudo-fiber of retained-data realizations. Whitney descent is carried out in a fixed tame o-minimal category. The Gaussian large-deviation theorem is support-aware in the semidefinite case. The Schur theorem separates coercivity from the joint convexity needed for positivity. The entropy module now starts from the surface-dependent first replica coefficient GW(μ;Σ) and only then specializes to an area-universal scalar GW(μ); the corresponding infrared package is written both in its fully surface-dependent completion and in its area-universal specialization, with pure Einstein branches defined as simultaneously entropy-local, operator-clean, and area-universal. The active carrier is saturated by a finite benchmark-separating family built into the finite observable-type principle. The gauge module is formulated with primitive rank‑t abelian sublattices, a finite semisimple target list at fixed active dimension, and a representation-resolved chiral anomaly system. Version 5.8 pushed the theorem-level closure to its natural pre-numerical frontier. First, the scalar area row is replaced by a fully surface-covariant intrinsic reduced map with its own absolute global uniqueness theorem, and the surface-dependent theorem is now equipped with an explicit practical degree-one homotopy certificate on canonically oriented quotient sublevels. Second, the free-oriented quotient hypothesis is supplemented not only by an orbifold-weighted global selection theorem with an explicit isotropy cap, but also by an exact saturation criterion for the weighted degree: when the orbifold degree equals 1/Γ∗ and every local sign is positive, one surviving orbit is forced. Third, the branchwise hypotheses are reorganized into an explicit discharge packet of finite constants — singular-value gap, Gauss–Newton alignment gap, singular-front margin, boundary homotopy margin, spectral-gap margin, and frame Gram gap — and this packet is now complemented by a finite branch-instantiation packet that records the active frame, encountered Wedderburn types, observability minor, and area-universality defect on the concrete physical branch. Fourth, the abelian gauge search is arithmetized by Smith-normal-form control, height/denominator bounds, target-pattern arithmetic cutoffs, and kernel-lattice uniqueness criteria, so that the Standard-Model-like hypercharge problem is reduced to a finite exact integer search whenever the physically relevant arithmetic cutoff is fixed. Fifth, the first even T8/T10 closure problem is reformulated as a finite matched-FRG ledger reduction, and the reduced quasi-stationary T8 variables are proved stable under constant reparametrization of the retained λ8 coordinate. Sixth, the covariance-aware benchmark metric is promoted to a theoremic evidence package with explicit stability margins under physically reasonable covariance refinements. In that precise sense v5.9 isolates the exact theoremic/numerical frontier: once a concrete branch is fixed, what remains is finite branchwise computation rather than further open-ended theoremic invention. The new step of v5.9 is to formalize that residual branchwise package completely: the matched-FRG scheme lock is upgraded to explicit extraction formulae for the effective T8 and T10 coefficient blocks, the forward κλ module is defined directly from the full running potential rather than from a frozen ansatz, the ultraviolet fixed-point search is organized into a finite atlas of persistent branch identities on canonical sublevels, the three-generation lift is reduced to a finite extension of the active-carrier and anomaly modules, and the observed cosmological constant is written as a final matched infrared flow functional rather than as an additional open theoremic principle. Version 5.9 normalized the residual branchwise package itself. The new step of v6.0 is to close the last pre-numerical conceptual gap explicitly: the manuscript now states a theorem of theoremic adequacy, isolates the exact first-principles T8 gate that decides current-branch physical closure, and shows how a unique physical orbit becomes a unique canonical representative point after a transverse normalization slice is fixed. In that precise sense v6.0 answers the remaining pre-numerical question sharply: the general theoremic architecture is now complete up to explicit finite branchwise computation, and the correct next step is the disciplined numerical campaign beginning with the first-principles T8 ledger. The decisive uniqueness repair remains equally explicit. A Jacobian gap alone does not identify the critical points of Φ=12∥F∥2 with the zeros of F for a rectangular reduced map. The manuscript therefore uses the correct criterion: critical-zero equivalence is assumed or derived from a Gauss–Newton alignment estimate. Together with canonical barrier exhaustion, singular-front exclusion, positivity of the local degree at genuine zeros, and genuine quotient degree one on canonically oriented quotient sublevels, this reduces sharp global uniqueness on an area-universal free-oriented branch to a finite verification ladder of geometric certificates. The numerical section is correspondingly disciplined: it records both the audited T6 ridge and the current T8 closeout campaign, states the present metric as a diagonal diagnostic benchmark specialization of a covariance-aware scheme, distinguishes the official T6 atlas from the denser and stress-replay domains, and shows that the first even T8 extension already admits a globally mild-verified pocket together with a semi-analytic rational closure family lying inside that pocket, while direct validation of the extended T6 domain and a first-principles derivation of the effective T8 coefficients remain open. In that precise and honest sense the paper proposes an Einstein-style route toward the selection of one physics: the conceptual structure is closed conditionally, and what remains is explicit finite computation on concrete ultraviolet candidates with active, graded, anomaly-compatible matter data together with branchwise verification of the finite uniqueness certificates. The manuscript is therefore offered as a proposed unified-theory framework to be vetted by the scientific community on exactly those explicit mathematical and phenomenological grounds.
Andrea Caffagni (Wed,) studied this question.