ANCIENT_OCCULTATION_VALIDATION_PROGRAM - TheDaniel166/moira GitHub Wiki
Ancient Occultation Validation Program
This note defines the separate validation track for ancient occultations.
Why It Is Separate
Ancient occultations do not belong in the same oracle family as modern and future occultation path geometry.
Modern and future validation can rely on:
- IOTA graze and limit path publications
- machine-readable predicted path tables
- direct path-line and boundary comparison
Ancient validation cannot.
Its highest authority is instead:
- scholarly historical-astronomy record corpora
- published reductions of those records
- site-identification and chronology scholarship
So the validation mode changes from path parity to historical event reconstruction plausibility.
Program Boundary
This program owns:
- admissibility rules for ancient occultation records
- date/time reduction doctrine
- site-identification doctrine
- star/body identity doctrine for historical naming systems
- uncertainty classification for each record
- reconstructed local-event validation against reduced historical sources
This program does not own:
- modern/future path geometry parity
- IOTA prediction-path validation
- Swiss
whereparity
Proposed Phases
Phase 0 — Source Survey
Identify which corpora are actually admissible for Moira validation.
Expected source classes:
- ancient Chinese occultation/appulse catalogs
- peer-reviewed historical-astronomy reductions
- site-specific observational studies with explicit chronology notes
Deliverable:
- curated source list with provenance and admissibility notes
Phase 1 — Admissibility Doctrine
Define what counts as a valid ancient validation record.
Each record must be explicit about:
- event kind
- observing site
- target identity
- dating basis
- uncertainty class
- source citation
Deliverable:
HistoricalOccultationRecorddoctrine and admission rules
Phase 2 — Reduction Doctrine
Define how records are normalized into computational inputs.
Must cover:
- calendar conversion
- night-boundary interpretation
- local time phrasing
- site geolocation
- star-name mapping
- uncertainty propagation
Deliverable:
- reduction standard for ancient occultation records
Phase 3 — Pilot Corpus
Build a small gold-standard pilot set.
Recommended size:
- 3 to 5 well-studied bright-star lunar occultations
Deliverable:
- machine-readable pilot corpus with reduction notes
Phase 4 — Validation Harness
Validate Moira by reconstructing the local event at the historical site and testing whether the reconstructed event is plausible within the stated uncertainty envelope.
Expected assertion classes:
- occultation visible vs not visible
- disappearance / reappearance plausibility
- altitude above horizon
- near-graze plausibility where appropriate
- contact-window plausibility when the source warrants it
Current Status
Status: deferred
Reason:
- Moira has the computational substrate
- Moira does not yet have the curated historical corpus and reduction doctrine
Current Active Track
The active occultation validation program remains:
modern_future_occultation_path_validation
Primary authority:
- IOTA graze/limit publications
Secondary authority:
- Swiss
where
Ancient validation should not be represented as if it were already covered by that modern/future track.