Arithmetic evidence. Explicit limits.
STARMAP is the in-browser mapping engine for S.T.A.R. Labs. Explore arithmetic projections using 12 handwritten illustrative entries and 148 deterministic synthetic fixtures. No fixture here is verified source data, and browser demonstrations do not execute registered experiments.
EXP-MAP-A01: bounded arithmetic result
The preserved investigator-A primary run reports local rank/elevation coherence against the locked NULL-MAP-001 arithmetic rank-permutation null. It rejects that null at α = 0.01 for the single preregistered endpoint.
- Observed T
- -114.40183841751903
- Corrected permutation p
- 0.001
- Null exceedances
- 0 / 999
- Analysis rows
- 38,042 / 64,687 source rows
T = −mean(|elevationᵢ − elevationⱼ|) on the fixed k = 10 arithmetic neighbor graph; larger T means greater local coherence. Exceedances count T_null ≥ T_observed. The locked correction is p = (1 + 0) / (999 + 1) = 0.001, the smallest attainable p-value with 999 permutations, not zero. The cohort selects curve-number-1 isogeny representatives.
This arithmetic null does not test a survey-selection-matched cosmological null or distinguish ACSC from other arithmetic structure. No cosmological correspondence or physical ACSC support is established. Independent rerun and replication remain outstanding; archive/hash inspection is package verification. The 160 browser fixtures and canvases are illustrations, separate from this recorded result.
Current registry gates: experiment execution eligible: true; dataset execution eligible: true; controlled-support eligible: false; physical-support eligible: false. Execution eligibility is recorded for the experiment and dataset. A source-locked runner preflight is still required for each controlled run. Earlier package descriptions of disabled execution describe their historical state.
Preserved primary transaction: 20261016T041335361140Z-e60fd05117fc · source revision 1b0f2d7853608bfbab79cf2dc04b002f1f193eb2.
Historical 2_STARMAP.md illustration: (Δ, N, r, Reg) → (θ, φ, ρ, λ). Separate from EXP-MAP-A01.
What the program asks
Explore the research questions and canonical claim registry. Independent data, preregistration, and null controls are required before support can be considered.
Explore topology
Persistent H0, kNN graph cycles, and permutation examples on illustrative fixtures. These browser runs are unregistered demonstrations.
Historical Φ illustration
θ(Δ), φ(N), ρ(r), λ(Reg). Switch ACSC against MCJ, PTD, FT, and the historical rank-elevation family.
Scale-dependent Hubble
Adjust an illustrative period-weighting model. Its chosen normalization approaches a reference anchor; this is not an observational fit.
Charge vs scalar
Q_RTCH[Σ₂] = ∫ Φ*ω is global. 𝒥_Q = ¼ F_Q² is the local invariant that may enter A. They must not be conflated.
Established vs conjectural
Pullbacks and conformal matter coupling are differential geometry. ACSC arithmetic-as-thermodynamics remains a conjecture until independently validated.
Null hypothesis
Appendix B of the variational paper isolates a single extra parameter. The extended model is supported only insofar as β_Q improves prediction while surviving independent nulls.
Historical ACSC illustration
The 5-tuple from 2_STARMAP.md. Rank has no direct axis unless Rank lift is on, but generated regulator, period, and height values depend on rank. This is not an independent rank test or the EXP-MAP-A01 protocol.
Controlled notebooks
Open in Labs- 00Registry overview
- 01Preprocessing
- 03Mapping benchmark
- 05Rank topology
- 06Null models
- 07SFR prediction