Standing and claims
LunarTrace reports every result with an explicit standing. A standing describes how far a result has been executed, reproduced or admitted. It does not describe whether the result is correct in the wider scientific sense, and no standing in this project implies publication, endorsement or a DOI.
Standing levels
Section titled “Standing levels”| Standing | Meaning | Where it is used |
|---|---|---|
METHOD_EXECUTED | The qualified slope_quality/v1 method ran on the stated inputs and mask. Nothing is claimed about admission. | Every workbench ROI result |
NOT_ADMITTED | The result has not been admitted as scientific evidence. | Paired with METHOD_EXECUTED on every workbench result |
NUMERIC_REPRODUCED | The numeric outputs of the frozen Site01 study were reproduced from its frozen inputs. | The frozen Site01 study |
SYNTHETIC | A fixture generated for tests or demonstration; not data. | Seeded Rows and Evidence stories only |
CANDIDATE | Work in progress; the surface or document is not yet a settled standard. | The R3 workbench documentation |
The workbench status for a user ROI is therefore fixed at METHOD_EXECUTED · NOT_ADMITTED. A
user ROI is evaluated over a user-drawn mask (cell-centre-inside rule), which differs from the
lattice closed rule used for the source ROIs. For this reason, a user ROI is never reported as
NUMERIC_REPRODUCED.
Claim boundaries
Section titled “Claim boundaries”The following claims are made by the workbench and its documentation:
- The method was executed in Python over the stated frozen inputs, identified by SHA-256 digests.
- Included-cell counts, inclusion fractions, quantiles and histograms are the outputs of that execution for the stated ROI and τ values.
- Figures and CSV exports are regenerated from the same Python output.
- Display rounding is labelled; exact values are available in tooltips, exports and the Python preset.
The following are not claimed:
- Scientific admission of any ROI result, including the user ROIs.
- Validation, accuracy, publication, DOI or endorsement of any result.
- Error bars or uncertainty. τ is a slope-RMS data-quality cutoff only.
- Lunar area. The valid-cell fraction is a fraction of grid cells, not an area.
- Any result beyond the PGDA Site01 5 m inputs listed in the Source instrument.
- Any claim about rendering appearance in a real browser beyond what has been observed. Cesium rendering and live Perspective highlighting are not yet confirmed in a browser capture.
Interpretation rules
Section titled “Interpretation rules”- Labels and tooltips must not present a τ as an uncertainty.
- A missing value is shown as missing (
—), never as zero. - A cell absent from the sample is shown as “not in the sample”, not as a computed value.
- A Lab notebook or downloaded Python script is a research aid. Its output is not promoted to a Study, a capsule of admitted evidence, or a published result.
- A saved run capsule records
durability: local file only; not openBIS, not a Study, not admitted.
ARA-42
Section titled “ARA-42”ARA-42 requires that unsupported claims are not made. In practice this means:
- No claim of scientific admission, DOI, endorsement or validation appears in the workbench interface, its exports or its documentation.
- No public repository link is made. The repository is private. The licence (BSD-3-Clause for code, CC BY 4.0 for prose) and authorship are decided; public release and a DOI are still pending.
- Capabilities that are not verified are described as not established, and listed under Residuals.
Admission records
Section titled “Admission records”research/ADMISSION.md records admission of the historical MoonTrace R1 Probe aggregate
evidence (FINDINGS.md, RESULT_LOCK.md, GOLDEN_CONTRACT.json); the original Probe runs and
observation or pair rows remain unavailable. That record concerns the historical Probe basis only.
It does not admit the Site01 slope_quality/v1 result, which stays NUMERIC_REPRODUCED and
NOT_ADMITTED, nor any workbench ROI. Consult the admission file for the authoritative wording.