[amaybaum]

Owner-adopted OI foundation — observer-relative finiteness (2026-10-30)

Foundational research requirement: for each specified physically bounded embedded observer/interface (O), the resolved observation (\pi_O:S\to V_O) has a finite set of distinguishable outcomes (|V_O|<\infty). There is no general requirement that the total microscopic substratum (S) be finite. Finite, countable and continuous carriers remain candidate substrates subject to separately stated dynamics, measurable observables and normalized states/priors. See #1007 owner decision.

Status and guard: This is a deliberately adopted foundational formulation, not a theorem that a general infinite carrier reproduces quantum mechanics, Yang–Mills, GR or exact recurrence. Existing finite-(S) proofs and finite-alphabet models remain valid only with their original stated finite-carrier premises. Finite distinguishability at one bounded interface does not imply a bound on unbounded accumulated histories, synchronized visible return, finite total entropy or observational gauge equivalence at all times. The latter requires matching specified accessible records/operational laws.

Specific consequence here: This PR provides explicit exact countercontrols: finite instantaneous observation does not bound cumulative records, imply uniform recurrence, or establish all-time equivalence between microscopic carriers.

Scientific migration: #1015 exact-source minimum-diff ledger, #1014 boundary/horizon entropy, #872 continuum/operational limits. Keep research-only and draft; no changed theorem grades, manuscript text, main, F/E, workflow dispatch or merge by virtue of this decision. Existing historical checkpoints below are preserved.

---

Current exact-head research checkpoint — November 4, 2026

• Head SHA: bae611e4d96ac69a172d7635c10b6981edaf867a, draft/unmerged. Base: d78055d9, which merges main 1609dc44. Earlier repair: 6216a0a8 restored the LaTeX escapes and removed a control byte.
• Independent review: ACCEPT at bae611e4, with no blocking findings. It resolves findings 1–7 and 9–13 of the review of d78055d9. Two recommendations are open:
  • R1: enumerate horizons n = q and n = q+1 directly;
  • R2: the main-side audit row, below.
• Author-side replay of the probe (this is the author's own run, not the reviewer's): python3 -I -B verification/programmes/substratum/continuous-substates/boundary_controls_probe.py exits 0. Its output matches the expected block recorded in the note, ending PASS TOTAL: 17758 checks. The total is counted from the checks that ran, not a literal.
• Corrected scope:
  • The finite cyclic register's visible law agrees exactly with the infinite stream's for every n ≤ q. It first differs at n = q+1, where b_{q+1}⊕b_q = b_1⊕b_0 in every realization.
  • Agreement for n ≤ q and the first difference at q+1 are derived on paper. The probe enumerates n = q−1 and the full return T^(2q) = I.
• Outstanding on main, outside this PR: verification/audits/manuscript/observer-relative-finiteness-20261030/DEPENDENCY-AUDIT.md:320–322 says the surrogate diverges at 2q; the first difference is at q+1. This needs a separate correction.
• Governance: research-only. Two verification files. No manuscripts, F/E, dispatch, ready or merge.

Earlier checkpoint — October 30, 2026 (superseded by the checkpoint above)

• Head SHA: b23ab626eaa239b3567e17d606f76a417fa5a58d, draft/unmerged; base main f6989a27815c4d715223eebf2859422c1231e3f8.
• Exact local replay: standard-library script with rooted-matrix assertions passed 9 finite-prefix q=n+1 tests for n=1...9, the 7 full finite-order controls q=2...8 (T^(2q)=I), and the toggle Γ₁/Γ₂ check. These are author-side executions, not independent review.
• Automatic workflow: run 41079412005 was queued at checkpoint. No exact-head whole-CI, Lean or independent reviewer acceptance is claimed.
• New crucial scope: finite hidden cycle agrees with unbounded hidden-stream statistics for any preselected n<q, but finite T^(2q)=I is distinguishable at time 2q from the infinite stream, where P(b_{2q}=b₀)=1/2. Finite observable resolution does not imply all-time equivalence.
• Governance: research-only, two new verification files, no manuscripts, main, F/E, ready/merge, or manually dispatched workflows.

---

Purpose

Bounded independent research for #1014, coordinated by #1007, complementary to #1009. Research-only draft; not a GR, OI-to-QM, area-law, gauge, or publication claim.

Exact mathematical results

For one finite binary observer interface and a compact uncountable two-sided Bernoulli hidden carrier with invariant Haar probability, compare three invertible microscopic maps on the same carrier and prior:

• Decoupled shift: visible bit static; exactly invisible hidden extension.
• Shift-plus-XOR: each new public bit reveals one fresh hidden bit; all length-(n+1) visible histories are independent fair bits, so their cumulative entropy is n+1 bits although instantaneous boundary entropy remains <=1 bit. The visible process is Markov despite actual hidden coupling; no synchronized visible return.
• Fixed-hidden toggle: rooted Gamma_1 is uniform, Gamma_2=I, giving a P-indivisibility witness and conditional history readback.

These are scope countercontrols: neither continuous hidden cardinality nor finite visible resolution alone fixes memory, recurrence, or cumulative accessible histories. Instantaneous visible entropy is at most 1 bit in all three models. A bound on cumulative histories needs a further duration, capacity, reset or information-flow condition.

Changed files and reproduction

1. verification/programmes/substratum/continuous-substates/BOUNDARY-CONTROLS.md — complete assumption list, algebraic infinite-product proof, scope guards and the expected probe output.
2. verification/programmes/substratum/continuous-substates/boundary_controls_probe.py — standard-library exact enumeration that iterates the implemented maps for horizons 1–9 (q=n+1) and the full returns q=2–8, with inverse, involution and exact rank checks; failures print FAIL and exit 1.

Requirements for review

Independently inspect inverse/Haar invariance, full-history uniformity and entropy calculation, rooted matrices, finite-prefix parity, the difference between cumulative observation records and fixed-time horizon entropy, and the non-native status of the chosen interface/coupling. Please report any counterexample or missing physical premise. Do not promote mathematics to a GR prediction.

Governance: draft, no manuscript, Lean, generated publication, main branch or existing historical receipt modifications. No manual workflow dispatch, F/E, readiness change, merge or theorem promotion. Independent review mandatory.

[amaybaum]

Owner direction cross-reference: finite observer distinguishability, unrestricted microscopic carrier is recorded in #1007 and in this PR description. Draft #1017 now provides minimum source/theorem migration criteria. The #1016 binary interface + hidden-stream result illustrates why finite-resolution does not bound total history entropy or establish all-time gauge equivalence. Preserve the finite-horizon versus exact all-time comparison and source/measure hypotheses; no GR upgrade.

[amaybaum]

Repair and reconciliation (owner-approved, #1015 c.6624266088).

• 6216a0a8: restores the LaTeX escapes in BOUNDARY-CONTROLS.md lines 108–115: \(, \), \oplus, \bigoplus, \Pr and \ldots. It also removes the literal backspace byte (0x08) at line 108. No other text changed.
• d78055d9: merges current main (1609dc44) into the branch as a merge commit, not a rebase. The diff against main is still only the two research files (+256).

Next steps

• Fresh PR CI on d78055d9.
• An independent review of this exact head. It re-runs the PR's own probe, rechecks the algebra by hand, and covers the repaired lines.

Corrections noted for later, not made yet: the PR body's "independent local version" means an author-side run, and the status sentence at :134–136 goes stale once the PR lands. Both are review items.

Integration as scoped mathematical evidence still needs the owner's exact-SHA approval after review and CI.

---

Generated by Claude Code

[amaybaum]

First independent review at d78055d9: ACCEPT WITH CHANGES, as scoped mathematical evidence. There are 2 blocking findings, 6 should-fix and 5 nits. Both blocking findings concern the text only.

Mathematics: correct, checked by hand.

• D, R and Q are bijective and preserve Haar measure, and R's inverse is right.
• R gives b_t = b₀⊕x₀⊕…⊕x_{t−1}, with an i.i.d. fair record and Γ_t = J/2.
• Q has Γ₁ = J/2 and Γ₂ = I, so it is P-indivisible.
• The repaired line 108 is correct, and T_R^{2q} = I.
• The result is sharper than stated: the finite and infinite laws agree for all n ≤ q and first differ at n = q+1.

Repair: confined to lines 108–115, with no control bytes left.

Probe: it exits 0, printing 9 shift-XOR cases, 7 long-horizon cases and the toggle.

Blocking

• :68–77. The "Crucial GR distinction" paragraph makes GR claims, which goes outside the no-GR scope. It should be restated as instantaneous versus cumulative visible entropy.
• :134–136. The status line goes stale once the PR lands (§A.27, §A.30).

Should-fix

• Probe: D and Q counters. They write out the closed forms instead of iterating the maps.
• Probe: PASS TOTAL. It is a hard-coded string that leaves out the long-horizon cases.
• Probe: bare assert. Every check uses one.
• :122–128. The GR/SM framing points beyond scope.
• Agreement range. It should be stated as n ≤ q, with first divergence at q+1.
• PR body. It calls an author-side run "independent".

Noted for main, outside this diff: DEPENDENCY-AUDIT.md:320–322 says the surrogate "diverges at 2q". That wording implies agreement through 2q−1, which is wrong. It needs its own correction.

Next: fix the findings (text and probe robustness), then a delta review, then CI, then the owner's exact-SHA decision.

---

Generated by Claude Code

[amaybaum]

Final acceptance record, head bae611e4d96ac69a172d7635c10b6981edaf867a.

• Object. Two research verification files:
  • verification/programmes/substratum/continuous-substates/BOUNDARY-CONTROLS.md
  • boundary_controls_probe.py
  • Lineage: 6216a0a8 (repair) → d78055d9 (merges main 1609dc44) → bae611e4 (review fixes). No manuscript, Lean, workflow or generated-artifact change.
• Independent review. ACCEPT at bae611e4, no blocking findings. Findings 1–7 and 9–13 of the review of d78055d9 are resolved; finding 8 (this PR's description) was updated at that head.
• CI at bae611e4: 25 passed, 1 skipped, 0 failed.
• Accepted as scoped mathematical evidence:
  • Exact derivation on paper for the infinite product system.
  • Exhaustive enumeration over cyclic registers q = 2,…,10, plus full-return checks for q = 2,…,8. The probe exits 0, and its output matches the expected block recorded in the note.
  • The finite cyclic register's visible law agrees with the infinite stream's for n ≤ q and first differs at n = q+1. That range is derived on paper; the probe enumerates n = q−1 and the full return.
  • Research-only; no manuscript claim depends on it.
• Not part of this acceptance:
  • Direct enumeration at horizons n = q and n = q+1 (reviewer recommendation R1) is an optional follow-up.
  • The DEPENDENCY-AUDIT.md:320–322 correction ("diverges at 2q" → agreement through n ≤ q, first difference at q+1) comes as a separate small follow-up.
• Owner approval: normal merge at exactly bae611e4, given in chat on 2026-11-04. The head stays unchanged.

---

Generated by Claude Code

[amaybaum]

Merge record.

• Merge commit: normal merge b2e053f3e5625bc25896ce00e0b1557472b59a73 of the approved head bae611e4.
• Parents: main 1609dc44 and bae611e4.
• Tree: 897dff09, identical to bae611e4's tree. The merge adds only the two research files.
• Main CI on b2e053f3: green, 25 passed and 1 skipped.
• Follow-up: the audit-sentence correction is PR audit: #1016 surrogate horizon is n ≤ q, first difference at q+1 #1086.

---

Generated by Claude Code