Skip to content

Wiki

Falsification and next step

What would count as progress and what would count as failure on every open item — and the intended scope of each restricted finite result. Live queue opens at v13.24.

Method · Method · Open · 2026-10-01 · compiled 2026-10-10 · 12 min read · 32 claims · 7 X sources

Original on X · Package: package17_bundle · JSON · Cite

falsification · open · method

On this page

Causal stack

  1. 00 Astra / global consistency
  2. 01 Genesis Pin
  3. 02 Retained pruning-order history
  4. 03 Admissible slice
  5. 04 L1 (closed)
  6. 05 L2 (closed)
  7. 06 L3 hyperedge memory T_H + D_L3
  8. 07 L4 H₄ rank lift
  9. 08 Discrete atlas / scalar accessibility
  10. 09 GEM-bridge / gravity

The program is already careful about open claims. This page makes that operational. It does not add physics. It restates, for each open item, what would count as progress and what would count as failure — and, for restricted finite results, the intended scope so a later post cannot quietly generalize them.

  • Pretty predicate ≠ pass. v15.99 can look true and still be INVALID. The meter outranks the predicate.
  • Completeness ≠ law. v16.36 certified the canonical 18-pair set. Gate D stays UNRESOLVED. A complete record is not a source law.
  • Barrier ≠ metric. v16.35 barriers are combinatorial. They are not distances, energies, or time.
  • Bounded unit ≠ universal law. v16.37 and v16.38 found no nonunit pair inside n≤6. v16.39 SC is sufficient, not a characterization.
  • Family theorem ≠ one graph, and ≠ every tree. v16.45 covers arbitrary leaf degrees in its stated domain. It does not cover chains or arbitrary retained trees.
  • Height 1 ≠ small ambiguity, and ≠ one move. A packed star m=k=q has m! exact components at barrier 1. The repair can be a long sequence whose maximum excursion is 1.
  • Uniform height ≠ a no-go. It rules out a source that depends only on barrier height. It does not prove the retained structure cannot contribute. The chain induction explains the bound inside chains. It does not select P.
  • v16.50 is an existence theorem for one suitable sequence on full binary trees. The bound is the total excursion, not each coordinate. The 1920 cases are not a census. Full binary does not mean balanced.
  • Do not call the barrier merely a uniform tax. Palette width can grow, by max(a,b) or a+b, while the peak departure stays 1.
  • The three-child question was run as v16.51, one ternary root, not a nonunit witness. v16.52 repeats ternary composition. v16.53 stops at arity 4. v16.54 is floor two only. v16.55 is a bounded guard, not a universal root repair.
  • A11.X60 is a colored-carrier checkpoint. It is not universal arbitrary-root minimum repair. A three-cover bound on minimum paths in one control is not an obstruction to every temporary incidence.
  • A11.X61 is a sufficient batch interface. It is not three-cover necessity and not a universal existence certificate. X62, X63, and X64 have no closeout.
  • A12.1–A12.5 is not numbered v16 certification and not physical validation. The stronger graph-descriptor impossibility is withdrawn as unproved, not declared false. The A12.5 no-go does not exclude every locality notion.
  • Post-A12 scoped proofs at 6a99418 named auxiliary cycle breaking as the next line. The 10 Oct tip does not publish a closeout under that name and does not keep it as the frontier. A temporary auxiliary incidence is not a physical primitive. No A12.6.
  • Spare capacity is not reachable reserve. A non-isolated local-progress obstruction is not a general impossibility. Static feasibility is not connectivity.
  • Universal core region means the declared tau 3 to 4 hidden-completion test. It is not general connectivity and not observer access.
  • Scoped C3 repair is not broad C3 and not guaranteed progress. Count reconstructions do not derive field access. Safe core writes do not initialize the hidden record.
  • An archived Gemini acceptance is not peer review, not numbered v16 certification, and not a STATUS bump.

Geometry line failure modes (1 Oct 2026)

Failure: teaching v15.99 INVALID as a confirmation. Failure: teaching v16.13F as Einstein or as derived quantum mechanics. The obstruction composition stays underdetermined. Failure: teaching v16.35 barriers as a metric, an energy, or time. Failure: teaching v16.36 Gate D UNRESOLVED as LOCAL_UNIT, or the closed 18-pair audit as a predictive barrier law. The old signature uses an already computed barrier. An independent construction can still explain the bound. Failure: teaching PR #81 as a merge to main. Failure: reading barrier height 1 as “the profile almost determines the component,” or as one move. Failure: reading a uniform unit bound as proof that the retained structure cannot contribute to a source. Failure: dismissing v16.48 because an older README says PRE-CLOSURE. The integration-branch closure is chains only. Failure: teaching v16.49 as arbitrary trees, or scoring unit as failure and nonunit as success. Failure: teaching v16.50 as every path, as a short path, as a balanced-only theorem, or as a mere uniform tax. Failure: teaching the 1920 directed cases as a census. Failure: teaching a three-child overlap example as a nonunit witness, or a failed normalization as a certified barrier. Failure: teaching A11.X60 as universal arbitrary-root repair. Failure: teaching A11.X61 as three-cover necessity, or X62–X64 as closed. Failure: teaching the A12 closeout as numbered v16 certification, as physical validation, or as proof that a response graph cannot describe a state. That stronger sentence was withdrawn as unproved, not declared false. Failure: teaching auxiliary cycle breaking as finished, or a temporary auxiliary incidence as a physical primitive. None of v15.75–v16.55, and not the overlapping-guard branch, is merged to main. Supported four unchanged.

Next named door — v13.24 (Open)

Post-QRSL branch reassessment / metric-affine continuum evidence. Which Pillar-3 claims survive without an ontology-native coarse-to-fine section. Do not add a refinement axiom to unlock RATS. RATS remains sealed / unexecuted. Pillar 3 / Einstein remains OPEN. Not a gravity derivation. Live board: /wiki/september-2026-ladder. Evidence: GitHub STATUS.md (no X URL).

This is the ledger, made checkable

Supported harnesses have numeric targets on the proofs page. Open items have progress/failure tests here. Restricted results name the family, the graph size, or the sector they actually cover.

Open items

Wiki freeze treated as frontier

Open

This upgrade. Wiki may lag X; wiki may not lag STATUS.md by more than one gate.

  • Progress — live-gates table matches STATUS.md the same day a gate merges.
  • Failure — Teaching v15.02 as HEAD while v15.03 exists. Teaching a(r) as a frozen theorem before STATUS lists it.

Fine data used to pick the fine world

Restricted

v13.23 circularity. QRSL irreducible.

  • Progress — Selector using only pre-selection coarse data. None found.
  • Failure — Using CMI / fine provenance to choose the completion it already inhabits.

Homogeneous solver treated as |Σ|

Restricted

v13.27 projective ray.

  • Progress — Independent geometric calibration, not the homogeneous current solver.
  • Failure — Calling the min-norm current |Σ| or a derived coupling.

Free multiplier called derived coupling

Restricted

v13.28 / author response. RGCL REQUIRES_NEW_AXIOM.

  • Progress — DERIVED, OBSTRUCTED, or REQUIRES_NEW_AXIOM. Nothing else.
  • Failure — Inserting a dimensionful G to match Einstein.

PCA / random isometry / “5=5” called J

Restricted

v14.04, v15.02. 120 inequivalent bijections.

  • Progress — Frozen functor or labeled NEW ASSUMPTION.
  • Failure — Choosing the mapping that looks most gravity-like.

Common parent treated as a source

Restricted

v15.01. Parent/support representation is not a provenance source.

  • Progress — An archived parent source class/tangent.
  • Failure — Calling C^125 a source because it is the parent Hilbert space.

λ or frame index called time

Restricted

No prune law. λ_source is a source parameter.

  • Progress — An independently motivated pruning / time law.
  • Failure — Reading the Path-A frame index as physical time.

Einstein residual used as selector

Restricted

Claim-boundary ban. Physics-facing tests must not select κ.

  • Progress — A correspondence that recovers Einstein without aiming at Einstein.
  • Failure — Tuning source/κ to shrink an Einstein residual.

Next named door: representation principle J

Open

GitHub-only 12 Sep. Latest completed v14.04. provenance → [P] STOPPED PENDING REPRESENTATION PRINCIPLE. v13.28 RGCL REQUIRES_NEW_AXIOM. Do not invent an intertwiner or coupling law to continue. RATS remains sealed. Pillar 3 / Einstein remains OPEN. Not a gravity derivation.

  • Progress — A frozen natural intertwiner already in the archive, an independently motivated identification of the provenance and support spaces, or an explicit NEW ASSUMPTION labeled as such.
  • Failure — Inventing an embedding, coupling law, or refinement axiom; filing v14.04, v14.03, or the v13.27 demo as Einstein; calling Pillar 3 closed.

WetLabEngine protein campaigns

Restricted

In-silico coarse-grained campaigns (p53, MECP2) plus posted Villin HP35 and frozen v9 on 1UAO/1L2Y. Engine is a tool.

  • Progress — Independent rerun of findings.json verdicts from raw CSVs; a pooled frozen v9 confirmation; Villin R_t on a protein that is not a controlled sample.
  • Failure — Calling p53 or MECP2 a therapy, promoting v9 to native folding, or filing Villin AUC as AlphaFold refuted. Dropping the MECP2 Run 5 DEVIATION.

Package 17 GEM-bridge / GR limit

Open

Derivation program. Friedmann / BBN / CMB listed as open seams.

  • Progress — χ derived as a fixed point; a GR-limit statement with named residual; a Hubble confrontation that is not a proxy sketch.
  • Failure — Calling the bridge closed, posting a fitted χ as the law, or treating Challenge 14 as the derivation.

Astra continuum identification

Open

Finite identities F, t*, d(t) on a controlled model. This family, this week.

  • Progress — A continuum object that recovers the frozen identities without a fitted response law.
  • Failure — Appending a new gravity page, or reading F / t* / d(t) as Einstein or Origin Law.

v9.167 compatibility lab

Restricted

Matched four-weight family on three five-level systems. Calculated states, not a laboratory observation. Not v9.156.

  • Progress — Independent rerun that hits 315 vs 311 and the 48.03% vs 46.55% pair inside the posted tolerance.
  • Failure — Generalizing to a hidden-information theorem, or calling the 27-second film gravitational motion.

v12.41 constraint algebra identification

Open

Qubit KKS identity {C_A,C_B}=C_{−i[A,B]} with Δ_closure=0 on 121 relational transforms. Gravity’s constraints not identified. λ is not time.

  • Progress — A named identification of gravity’s constraints with this algebra that is not a fitted generator law.
  • Failure — Filing v12.41 as derived gravity, ADM, or Einstein equations.

QRSL origin / canonical refinement selector

Restricted

v13.23 GitHub-only. QRSL irreducible relative to the current frozen ontology. Genesis Pin, W1/W2/W3, positivity, recoverability order, quotient naturality do not select a canonical fine world.

  • Progress — A new primitive that selects a unique fine world from coarse data without circularly using the fine data.
  • Failure — Filing QRSL as derived gravity, or treating Genesis Pin as a refinement selector.

HCPR frozen-ledger non-factorization

Restricted

v13.10 GitHub-only. HCPR is not a function of the current frozen ledger. Same one- and two-body data, different rigidity.

  • Progress — A source-conditioned higher-order completion law that factorizes HCPR, or a new ledger datum that does.
  • Failure — Treating same-ledger as same-reality, or filing HCPR as a gravity theorem.

Finite-state QMAR / CMI jet locality

Restricted

v13.11–v13.18 GitHub-only. Finite-state QMAR and O(√CMI) / O(‖P‖ √CMI) recoverability-to-geometry bounds on faithful polar-gapped strata.

  • Progress — A named identification of QMAR with Einstein/ADM that is not a fitted correspondence.
  • Failure — Filing QMAR or jet locality as Einstein equations, ADM, or a derived stress-energy law.

DESI-facing prediction ledger

Open

5 Mar 2025 directional evolving-DE prior is scored as timing, not a posterior. 30 Apr w(z) / A_L / cluster-core remain tests. #nodarkmatter is program voice.

  • Progress — A named w(z) (or equivalent) that beats ΛCDM on DESI DR3 BAO in z=0.5–1.0 at the posted ~2–3σ, with Package 17 failure modes still in force.
  • Failure — Filing DESI DR2 2.8–4.2σ as a UQCF-GEM confirmation, as a no-dark-matter result, or scoring the 21 Mar restatement as a prediction. Promoting the 30 Apr sketch to Supported.

Hubble pipeline

Not claimed

13 Apr H0DN proxy and 27 Mar D_f sim are sketches. Package 17 still lists the confrontation as open.

  • Progress — A named pipeline against H0DN ~73.5 vs ΛCDM ~67.2 that survives the Package 17 failure modes.
  • Failure — Filing the ~9% coherence-gradient recalibration, or the 5 May cosmology sentence, as a resolution.

Lyapunov of V[Ω, C, μ]

Open

Named 17 May after V555. The freeze is the perturbation → restored-futures invariant, not the Lyapunov.

  • Progress — A descent proof under the posted controlled attacks, with the same claim boundary as V555.
  • Failure — Treating the 17 May infographic as a closed theorem, or promoting heat-kernel / ADM / V760 with it.

It-from-bit projection to metric physics

Open

10–11 Jun freeze is bind + order. Wheeler’s slogan is prior art. V1698 is discrete atlas residual only.

  • Progress — A derived projection operator from retained recombination to macroscopic metric physics.
  • Failure — Treating bind + order as a Wheeler theorem, restoring V1688 as holonomy, or calling the mouse GR.

Source generator Q_i

Open

23 Jul gate. In-thread review: not from admissibility + BKM reciprocity alone.

  • Progress — A construction that aligns every incident edge’s leading source line from global provenance.
  • Failure — Declaring Q_i generated because reciprocity held on a commuting toy.

Heat-kernel / ADM week

Open

7–9 May. dx²/long operator frozen; heldout spatial-R 25.15% → 16.55%. Next object δρ = A·q.

  • Progress — An ADM action with the posted residual channel, still inverted (do not discretize Einstein–Hilbert).
  • Failure — Calling V498–V555 or V760 a gravity derivation, or restoring tensor/ADM after the V824 prune.

V1699 Pillar 3 GR/ADM correspondence

Open

Pillar 1 closed. Pillar 2 is a finite variational candidate. Pillar 3 is the continuum identification.

  • Progress — A correspondence that maps the discrete action onto ADM without assuming the Einstein equations.
  • Failure — Reading A[q] = ½ ‖R_R + Lq − J_R + B_R‖² as a physical gravitational action.

Unique pruning law through A_G

Open

A_G is a poset of admissible retained supports, not yet a unique history. Posted 20 Jul 2026.

  • Progress — A law that selects a unique pruning sequence through A_G from provenance, not a fitted schedule.
  • Failure — Treating the poset itself as a unique cosmology or a GR history.

Dorau–Much protocol

Open

Can pre-geometric recoverability already know modular relative entropy? Protocol, not a derivation. 19 Aug 2026.

  • Progress — A stated correspondence that maps a recoverability object onto modular relative entropy and survives a null that preserves geometry-only closure.
  • Failure — Calling UCRP/ECSM derived, or treating the 19 Aug alignment as a gravity proof.

Objective-reduction timescale

Open

4 Sep 2026 question: what recoverability object is the gravitational self-energy E in t = ħ/E? Gravity not claimed.

  • Progress — A named recoverability object E and a lab-facing protocol whose measurement would constrain the priority of the timeless layer.
  • Failure — Filing Penrose OR as derived, or treating the question as a result.

Quadratic RG cosmology scaffold

Open

Apr 2026 solved core σ*≈0.7333, γ≈0.2667. DESI frozen operator null. Remaining gap is microscopic amplitude / ρ=B.

  • Progress — A microscopic amplitude that keeps the interior fixed point and is not null on DESI.
  • Failure — Calling Package 17 closed, or treating σ*/γ as a Hubble resolution.

Applied-lab unification

Open

Protein, TSP, critical path, and C3++ share a posted remember-what-survived pattern. 10 May asked what equation unifies them.

  • Progress — A single operator with named residual that transfers unchanged across the posted protein, TSP, and critical-path harnesses.
  • Failure — Treating recurrence rhetoric, or χ≈0.2667 on 2 May, as a closed unification law.

Editorial tests compiled from package READMEs and the freeze posts. Not an independent certification. Gravity stays a candidate.

Restricted results: do not generalize

  • Astra F, t*, d(t), g_★, D_N — this finite family, this week. Continuum identification still required. g_★ and D_N are X-only.
  • v12.41 — qubit KKS identity, 121 steps, Δ_closure = 0. Not ADM, not Einstein constraints, λ is not time.
  • v13 QRSL / HCPR / QMAR — GitHub-only 12 Sep Restricted no-gos. Not Einstein. RATS remains sealed. Queue at v13.24.
  • v9.167 lab — matched four-weight family, calculated states, 315 vs 311. Not v9.156, not a laboratory observation.
  • V1698 — five-chart ledger, residual ~3.27×10⁻¹⁶. Discrete atlas. Not continuum GR.
  • V824 — scalar / conformal sector only.
  • Challenge 14 — frozen SPARC 171 scaffold. WALLABY 12D is a separate unseen-gas transfer.
  • V760 atrium — posted simulation only. Not physical GR or physical time.
  • 17 May recoverability framework — the mapping perturbation → restored futures. Lyapunov of V[Ω,C,μ] is named, not closed.
  • It-from-bit 10–11 Jun — bind + order. Projection to metric physics remains Open.

What a successful run does not show

A green residual on a named harness is not a spacetime theorem. The proofs page repeats the box on each run-this-first notebook. If a later post quietly drops the box, the badge on this page is the one that still governs.

Open

Package 17 GEM-bridge derivation / GR limit remains open. Challenge 14 and WALLABY are restricted empirical packets, not the derivation.

Source on X

Open

Lyapunov descent of V[Ω, C, μ] is named, not closed. The 17 May freeze is the restored-futures invariant.

Source on X

Open

Projection from retained recombination to macroscopic metric physics is to be derived, not assumed. Bind + order is the 10–11 Jun freeze.

Source on X

X sources

Original posts this article was compiled from. The notebook copy is editorial; the status URL is the source.

Previous Supported fourNext Predictions — DESI and no-dark-matter

Causal stack

  1. 00 Astra / global consistency
  2. 01 Genesis Pin
  3. 02 Retained pruning-order history
  4. 03 Admissible slice
  5. 04 L1 (closed)
  6. 05 L2 (closed)
  7. 06 L3 hyperedge memory T_H + D_L3
  8. 07 L4 H₄ rank lift
  9. 08 Discrete atlas / scalar accessibility
  10. 09 GEM-bridge / gravity

On this page

X sources

Related articles

GitHub packages

Field notes