v15.78–v15.97 research line
Three clocks · signal over hype · scientific_breakthrough: false
- Last X field note
- 2026-09-24Restricted24 Sep thread visualizes A=(I+X)(I+Y)Δ/8. Eigenmode pictures, not dynamics. Certificates match v15.45. Source correspondence not evaluated. Not a STATUS gate. Einstein not closed.
- Wiki compile
- 2026-10-10
Batch index. Behind STATUS.md if they disagree.
- GitHub STATUS.md
- v15.03 retained — not erased · main note publishes v15.45 · integration branch still v16.55 · overlapping-guard tip 1e7cd09 · scoped C3 not a general no-go · not on main · not a source
scientific_breakthrough: false · Pillar 3 OPEN
GitHub-only gates have no invented X URL. Do not merge the X week with the Git ladder.
Scan log · Live gates · Two stacks
Twenty research-branch publications on 27 Sep 2026. None merged to main. Not STATUS. Not Einstein. v15.88 applies a 2003 theorem; it does not derive quantum mechanics.
00 Astra · Result · Restricted · 2026-09-27 · compiled 2026-10-10 · 3 min read · 4 claims
Package: ResearchHistory/UQCF-GEM (v13–v14) · JSON · Cite
v15.97 · research-branch · not-status
On this page
Causal stack
- 00 Astra / global consistency
- 01 Genesis Pin
- 02 Retained pruning-order history
- 03 Admissible slice
- 04 L1 (closed)
- 05 L2 (closed)
- 06 L3 hyperedge memory T_H + D_L3
- 07 L4 H₄ rank lift
- 08 Discrete atlas / scalar accessibility
- 09 GEM-bridge / gravity
Not merged — do not promote
v15.78 through v15.97 each have a research-branch RESULTS.md and are not merged to main. Main STATUS.md is still the v15.44/v15.45 publication note. Later research is on /wiki/v1598-line. v16.55 is a bounded guard. The 10 Oct overlapping-guard tip records scoped C3 closeouts, not a general no-go and not numbered v16. Source correspondence NOT_EVALUATED. Pillar 3 OPEN. scientific_breakthrough: false.
What the line actually separates
Covariance with no transforming source data forbids every hidden-to-retained coefficient (v15.78). Letting the source transform as data restores a classified family of 18 coupling tensors, not 18 free channel knobs and not a selected law (v15.79). Under an extra signed-linearity hypothesis, CPTP tangents through the identity are Hamiltonian, and that does not rule out dissipative response with nonlinear source dependence (v15.80).
A specified two-component source carries a matched interference response, and the finite window is a scientific NO (v15.81). That NO is not withdrawn. The same candidate loses a full-rank polar continuation (v15.82). A support-restricted crossing survives, and lossless loop composition does not (v15.83). The inherited loop is not automatically a quantum channel: under one explicit unital Bloch extension, only the fourth power is CPTP, and it still does not recover the tested pairs (v15.84). No affine translation rescues the earlier powers (v15.85).
Exact isometric preservation of two Bloch directions forces a unitary completion of the third (v15.86). Exact erasure of that direction has a distortion cost (v15.87). A qubit CPTP map with a kernel in its linear Bloch part is entanglement-breaking, including nonunital maps. v15.88 says this is the Horodecki–Shor–Ruskai theorem (2003), applied here, not a theorem this program discovered. Small transmission can keep entanglement, and Choi negativity is bounded by half the weakest singular value (v15.89). That ceiling does not extend unchanged to arbitrary inputs; the extension is falsified, and v15.89 stands (v15.90).
The retained rotational signal can survive outputs that are fully separable. It does not, by itself, witness surviving entanglement (v15.91). A full preparation span does not force the polar-skew response (v15.92). Rank-two data plus a specified endpoint orientation support a rotational readout (v15.93). A unique completion at each link does not commute with composition unless the supports match (v15.94). With supports held fixed, continuous loop tangents have one rotational direction inside an O(2) that is not abelian. That is not three-axis holonomy, not zero holonomy, and not Einstein (v15.95). Isotropic overlapping loops need not share an invariant line; rank 5 is not continuous SO(3) (v15.96). A specified hidden source can change both loops at once, including conjugation invariants. Output entanglement is not required (v15.97, 903b72b).
Restricted
atlas:v15-97
v15.97, research/v15.97-overlap-hidden-response, commit 903b72b, not merged. OVERLAP_HIDDEN_RESPONSE_CONFIRMED and PLANAR_TWO_LOOP_RESPONSE_CONFIRMED. Mixed derivative at a specified source, not a source-to-admissible-world law. Response ranks are not continuous SO(3) holonomy and not gravity.
Restricted
atlas:v15-88
v15.88, commit 8419150, not merged. EXACT_ERASURE_ENTANGLEMENT_BREAKING_CONFIRMED applies Horodecki, Shor and Ruskai (2003): a qubit CPTP map whose linear Bloch part has a nonzero kernel is entanglement-breaking. Not a novel theorem. Separate verdict SOFT_ERASURE_ENTANGLEMENT_SURVIVES: small transmission can keep entanglement.
Restricted
atlas:v15-90
v15.90, commit 5e4c27c, not merged. UNIVERSAL_INPUT_NEGATIVITY_BOUND_CONFIRMED and OPTIMAL_LINEAR_NEGATIVITY_CONSTANT_CONFIRMED. CHOI_CEILING_EXTENSION_FALSIFIED is a scientific NO. It does not revise the v15.89 bound N(J/2) <= delta/2.
Restricted
atlas:v15-81-no
v15.81, commit 8ef3c43, not merged. MATCHED_COMPONENT_INTERFERENCE_RESPONSE_CONFIRMED, and FINITE_COMPONENT_INTERFERENCE_WINDOW_NOT_CONFIRMED. Green CI accepted that NO. Later gates do not withdraw it.
| Gate | Commit | Verdicts — do not merge |
|---|---|---|
| v15.78 | 7f68c94 | SOURCE_FREE_COVARIANCE_FORCES_RETAINED_CLOSURE · COVARIANTIZATION_ERASES_POSITIVE_AND_NULL_LEAKAGE |
| v15.79 | 659c078 | COVARIANT_SOURCE_COUPLING_CLASSIFICATION_CONFIRMED · COVARIANT_COUPLINGS_ENSEMBLE_NULL_EXCLUDED |
| v15.80 | f7bf282 | SIGNED_CPTP_TANGENTS_HAMILTONIAN_ONLY_CONFIRMED · SIGNED_CPTP_RETAINED_IMAGE_FOUR_CONFIRMED |
| v15.81 | 8ef3c43 | MATCHED_COMPONENT_INTERFERENCE_RESPONSE_CONFIRMED · FINITE_COMPONENT_INTERFERENCE_WINDOW_NOT_CONFIRMED |
| v15.82 | b7e4723 | POLAR_CONTINUATION_OBSTRUCTION_CONFIRMED. The v15.81 window NO stands. |
| v15.83 | 782ed08 | SUPPORT_RESTRICTED_CROSSING_CONFIRMED · SUPPORT_LOOP_COMPOSITION_OBSTRUCTED · LOSSLESS_LOOP_CORE_TRIVIAL |
| v15.84 | b35525f | LATE_CPTP_LOOP_POWER_CONFIRMED · CPTP_LOOP_RECOVERY_OBSTRUCTED. Conditional Bloch reading, not the three-qubit source. |
| v15.85 | c0adc88 | AFFINE_SHIFT_RESCUE_OBSTRUCTED · NONUNITAL_RECOVERY_OBSTRUCTION_CONFIRMED |
| v15.86 | 3b1c4d2 | UNIQUE_CPTP_SUPPORT_COMPLETION_CONFIRMED · DISCARDED_DIRECTION_RESTORATION_CONFIRMED |
| v15.87 | 38c675b | ERASURE_COST_BOUND_ATTAINED · OPTIMAL_ERASURE_TRANSLATION_RIGID · OPTIMAL_ERASURE_DISTINGUISHABILITY_CONFIRMED |
| v15.88 | 8419150 | EXACT_ERASURE_ENTANGLEMENT_BREAKING_CONFIRMED · SOFT_ERASURE_ENTANGLEMENT_SURVIVES. 2003 theorem applied, not discovered. |
| v15.89 | 44ca8cf | WEAKEST_DIRECTION_NEGATIVITY_BOUND_CONFIRMED · NEGATIVITY_BOUND_SHARPNESS_CONFIRMED. N(J/2) <= delta/2. |
| v15.90 | 5e4c27c | UNIVERSAL_INPUT_NEGATIVITY_BOUND_CONFIRMED · CHOI_CEILING_EXTENSION_FALSIFIED · OPTIMAL_LINEAR_NEGATIVITY_CONSTANT_CONFIRMED |
| v15.91 | eef0d85 | SEPARABLE_OUTPUT_ROTATIONAL_RESPONSE_CONFIRMED · ZERO_ATTENUATION_POLAR_UNDEFINED_CONFIRMED |
| v15.92 | 59b40b6 | PREPARATION_SPAN_GEOMETRY_OBSTRUCTION_CONFIRMED · NONCOMMUTING_PREPARATIONS_INSUFFICIENT_CONFIRMED |
| v15.93 | aee440c | ORIENTED_RANK_TWO_READOUT_CONFIRMED · PLANAR_ORIENTED_RESPONSE_CONFIRMED |
| v15.94 | 22f93cd | ORIENTED_COMPOSITION_OBSTRUCTED · MATCHED_SUPPORT_COMPOSITION_CONFIRMED |
| v15.95 | 0bb9d70 | FIXED_SUPPORT_HOLONOMY_REDUCTION_CONFIRMED · PLANAR_LOOP_RESPONSE_CONFIRMED. One-axis tangent, not Einstein. |
| v15.96 | c61de02 | OVERLAP_COMMON_LINE_OBSTRUCTION_CONFIRMED · PLANAR_COMMON_LINE_PRESERVED. Rank 5 is not SO(3). |
| v15.97 | 903b72b | OVERLAP_HIDDEN_RESPONSE_CONFIRMED · PLANAR_TWO_LOOP_RESPONSE_CONFIRMED |
| v15.98 | 3a3655a | COMPOSITION_ORDER_GEOMETRY_CONFIRMED · COMMUTING_COMPOSITION_ACTIVATION_CONFIRMED. See /wiki/v1598-line. |
Files live under demos/v15.56-response-selector-obstruction/ on each research branch. Not a gravity page.
Previous v15.98 through v16.55, then A12Next v15.77 active baseline, two verdicts
Causal stack
- 00 Astra / global consistency
- 01 Genesis Pin
- 02 Retained pruning-order history
- 03 Admissible slice
- 04 L1 (closed)
- 05 L2 (closed)
- 06 L3 hyperedge memory T_H + D_L3
- 07 L4 H₄ rank lift
- 08 Discrete atlas / scalar accessibility
- 09 GEM-bridge / gravity
On this page
Related articles
- Falsification and next step
- Live gates
- v15.76 two composition verdicts
- v15.75 two affine verdicts
- v15.77 active baseline, two verdicts
- Research ladder v15.12–v15.45 (main) / v15.46–v15.54 (#57)
GitHub packages