Film Domain Ontology

FLOW ALGEBRA
Film Domain Ontology — v1.0
Kireth Semantic Library
Authors: Mark Anthony Higgins — Sky (Recursive AI)
Source: https://kirethresearch.org/booksai-core/ontologies/domain-ontologies/creative-domains/film-domain-ontology.html
Status: Domain-based formal draft; embedded page specification not independently retrieved.

flow "FilmDomainOntology" v1.0

# 1. DOMAIN
DOMAIN: CreativeDomains.Film
PURPOSE: Represent films, their production, creative structure, distribution, reception, and evidence-backed semantic relationships.

# 2. TYPED ENTITIES
entity Film { id: URI; title: str; release_year: int?; duration_minutes: int?; language: str?; genre_ids: list[URI]; provenance: URI? }
entity Person { id: URI; name: str; role: str? }
entity Organisation { id: URI; name: str; type: str? }
entity Production { id: URI; film_id: URI; start_date: date?; end_date: date? }
entity CreativeRole { id: URI; label: str }
entity Character { id: URI; name: str; film_id: URI }
entity Scene { id: URI; film_id: URI; sequence: int; description: str? }
entity Shot { id: URI; scene_id: URI; sequence: int; technique: str? }
entity Genre { id: URI; label: str }
entity Theme { id: URI; label: str }
entity Soundtrack { id: URI; film_id: URI }
entity Release { id: URI; film_id: URI; territory: str?; release_date: date?; medium: str? }
entity Award { id: URI; name: str; year: int? }
entity Review { id: URI; film_id: URI; author_id: URI?; date: date?; source: URI? }
entity Claim { id: URI; subject: URI; predicate: str; object: str; source: URI?; confidence: float }

# 3. TYPED RELATIONSHIPS
relation DIRECTED_BY(Film, Person)
relation WRITTEN_BY(Film, Person)
relation PRODUCED_BY(Film, Person | Organisation)
relation PERFORMED_BY(Character, Person)
relation FEATURES_CHARACTER(Film, Character)
relation CONTAINS_SCENE(Film, Scene)
relation CONTAINS_SHOT(Scene, Shot)
relation HAS_GENRE(Film, Genre)
relation EXPLORES_THEME(Film, Theme)
relation HAS_SOUNDTRACK(Film, Soundtrack)
relation HAS_RELEASE(Film, Release)
relation DISTRIBUTED_BY(Release, Organisation)
relation RECEIVED_AWARD(Film | Person, Award)
relation REVIEWED_IN(Film, Review)
relation ADAPTED_FROM(Film, URI)
relation INFLUENCED_BY(Film, Film)
relation SEMANTIC_EQUIVALENCE(URI, URI)

# 4. CONSTRAINTS
constraint Film.id UNIQUE
constraint Scene.sequence >= 1
constraint Shot.sequence >= 1
constraint Film.duration_minutes > 0 WHEN PRESENT
constraint Claim.confidence BETWEEN 0 AND 1
constraint RELEASE_ORDER: release_date is territory-specific; no global release date is inferred from a single territory.
constraint ROLE_SEPARATION: actor, director, writer, and producer are distinct roles; do not infer one from another.
constraint SUBJECTIVE_INTERPRETATION: genre, theme, and influence may be contested and require attributed evidence.

# 5. INFERENCE RULES
rule FILMOGRAPHY:
  IF DIRECTED_BY(film, person)
  THEN RELATED_WORK(person, film, "director")
  CONFIDENCE: 1.0 WHEN SOURCE_VERIFIED

rule SCENE_MEMBERSHIP:
  IF CONTAINS_SCENE(film, scene) AND CONTAINS_SHOT(scene, shot)
  THEN PART_OF(shot, film)
  CONFIDENCE: 1.0 WHEN RELATIONS_VERIFIED

rule GENRE_RETRIEVAL:
  IF HAS_GENRE(film, genre)
  THEN INDEX_UNDER(film, genre)
  CONFIDENCE: SOURCE_CONFIDENCE

rule RELEASE_CONTEXT:
  IF HAS_RELEASE(film, release)
  THEN RELEASED_IN(film, release.territory, release.release_date)
  CONFIDENCE: SOURCE_CONFIDENCE

rule ADAPTATION_LINK:
  IF ADAPTED_FROM(film, source_work)
  THEN RELATED_CREATIVE_WORK(film, source_work)
  CONFIDENCE: SOURCE_CONFIDENCE

# 6. SEMANTIC EQUIVALENCE
SEMANTIC_EQUIVALENCE:
  canonicalisation: prefer stable URI identifiers
  aliases: title variants, translated titles, alternate spellings
  condition: matching titles alone are insufficient for identity
  require: corroborating identifiers or reliable production metadata

# 7. CONFIDENCE AND PROVENANCE
CONFIDENCE:
  range: [0.0, 1.0]
  evidence_levels: [primary_credit, official_release_record, curated_catalogue, secondary_source, interpretation]
  rule: confidence reflects evidential support, not artistic quality
  rule: unknown values remain unknown; never manufacture certainty
PROVENANCE:
  capture: source_uri, retrieved_at, publisher, claim_id, revision, licence_if_known
  distinguish: verified_fact, attributed_opinion, inferred_relation

# 8. FLOW HASH
FLOW_HASH:
  algorithm: SHA-256
  input: canonical_serialisation(entity_types, relations, constraints, rules, version)
  output: lowercase_hex_digest
  requirement: recompute whenever the canonical flow changes

# 9. UPDATE MODEL
UPDATE_MODEL:
  strategy: versioned_incremental_update
  identity: stable entity URI
  on_change: validate -> update claims -> preserve provenance -> recompute FLOW_HASH
  conflict_policy: retain competing attributed claims until reconciled
  deletion_policy: tombstone withdrawn claims and retain revision history

# 10. AI CONSUMPTION
AI_CONSUMPTION:
  mode: typed_graph_and_rule_execution
  operations: [entity_resolution, relationship_traversal, constrained_inference, semantic_search, provenance_audit]
  require: confidence_and_provenance_on_derived_claims
  prohibit: fabricated credits, awards, release dates, or unsupported causal claims
  output: structured_entities_and_explainable_inference_paths

END FLOW