Library Science Domain Ontology

FLOW ALGEBRA
TITLE: Library Science Domain Ontology
VERSION: 1.0
SOURCE_URL: https://kirethresearch.org/booksai-core/ontologies/domain-ontologies/information-domains/library-science-domain-ontology.html
STATUS: Domain-based draft; source page content not independently verified.
AUTHORS: Mark Higgins — Sky (Recursive AI)
PROJECT: Kireth Semantic Library

DOMAIN:
  id: library_science_domain
  label: Library Science Domain Ontology
  parent: information_domains
  purpose: Represent library collections, resources, description, discovery, circulation, preservation and access as typed, provenance-aware relationships.

TYPES:
  Library: Organisation
  Collection: Aggregation
  InformationResource: Entity
  Work: AbstractIntellectualCreation
  Expression: RealisationOfWork
  Manifestation: PublicationOrFormat
  Item: PhysicalOrDigitalExemplar
  Agent: PersonOrOrganisation
  Creator: Agent
  Patron: Agent
  Librarian: Agent
  CatalogueRecord: MetadataRecord
  Identifier: String
  Subject: Concept
  ClassificationScheme: ControlledVocabulary
  ClassificationCode: String
  AuthorityRecord: IdentityRecord
  Holding: ResourceAvailabilityRecord
  Loan: CirculationEvent
  Reservation: RequestEvent
  Licence: AccessAgreement
  Repository: StorageSystem
  PreservationAction: Event
  DigitalObject: InformationResource
  ProvenanceRecord: EvidenceRecord

RELATIONSHIPS:
  holds(Library, Item)
  curates(Library, Collection)
  contains(Collection, InformationResource)
  created_by(Work, Agent)
  realised_as(Work, Expression)
  embodied_in(Expression, Manifestation)
  exemplified_by(Manifestation, Item)
  described_by(InformationResource, CatalogueRecord)
  identified_by(InformationResource, Identifier)
  has_subject(Work, Subject)
  classified_as(InformationResource, ClassificationCode)
  code_from(ClassificationCode, ClassificationScheme)
  authority_controls(AuthorityRecord, Agent|Subject)
  held_at(Item, Library)
  borrowed_by(Loan, Patron)
  loan_of(Loan, Item)
  requested_by(Reservation, Patron)
  reservation_for(Reservation, InformationResource)
  governed_by(InformationResource, Licence)
  stored_in(DigitalObject, Repository)
  preserved_by(DigitalObject, PreservationAction)
  supported_by(Assertion, ProvenanceRecord)
  related_to(InformationResource, InformationResource)

CONSTRAINTS:
  - Every catalogue record must identify its described resource.
  - Identifiers must be scoped by identifier scheme and assigning authority.
  - A single item cannot have overlapping active exclusive loans.
  - Access permissions must be evaluated against the applicable licence and patron context.
  - Work, expression, manifestation and item are distinct modelling levels.
  - A missing record does not imply a resource does not exist.
  - Classification and subject assignment may be many-to-many and version-dependent.

INFERENCE_RULES:
  R1: realised_as(W,E) AND embodied_in(E,M) -> manifestation_of(M,W)
  R2: embodied_in(E,M) AND exemplified_by(M,I) -> instance_of(I,E)
  R3: held_at(I,L) AND exemplified_by(M,I) -> holds_manifestation(L,M)
  R4: has_subject(W,S) AND realised_as(W,E) -> inherits_subject(E,S) [derived]
  R5: loan_of(Loan,I) AND borrowed_by(Loan,P) AND active(Loan) -> unavailable_for_exclusive_loan(I)
  R6: described_by(R,C) AND has_subject(R,S) -> catalogue_subject(C,S)
  R7: stored_in(D,R) AND preserved_by(D,A) -> preservation_trace(D,R,A)
  R8: related_to(A,B) does NOT imply related_to(B,A) unless relation type is symmetric.

CONFIDENCE:
  scale: 0.0..1.0
  scope: assertion_level
  method: Evidence-weighted, calibrated where possible; never infer certainty from missing evidence.
  fields: [value, method, evidence_ids, assessed_at, assessor]

SEMANTIC_EQUIVALENCE:
  exact_match: Require shared referent, level, context and identifier scope.
  close_match: Preserve differences between editions, translations and catalogue conventions.
  prohibited_assumption: Work == Expression == Manifestation == Item
  alignments: [RDF, SKOS, Dublin_Core, BIBFRAME, IFLA_LRM]
  note: External vocabulary mappings require explicit verification.

PROVENANCE:
  fields: [source_id, source_uri, record_id, contributor, timestamp, licence, revision, evidence_method]
  requirement: Retain original assertions and distinguish derived assertions.

FLOW_HASH:
  algorithm: SHA-256
  input: Canonical UTF-8 serialization of typed entities, relations, rules and version metadata.
  generation: Compute on deployment; no hash value is asserted in this draft.

UPDATE_MODEL:
  strategy: Versioned append-and-reconcile
  triggers: [catalogue_change, authority_update, circulation_event, licence_change, preservation_event]
  steps: [validate_types, preserve_provenance, detect_conflicts, recompute_inferences, recalculate_hash, publish_version]
  deletion: Tombstone records where necessary to preserve auditability.

AI_CONSUMPTION:
  mode: Typed semantic retrieval and provenance-aware inference
  inputs: [resource_query, user_context, access_context, ontology_version]
  outputs: [matched_resources, relationship_paths, evidence, confidence, access_status, uncertainty]
  safeguards: [respect_licences, protect_patron_privacy, avoid_unsupported_inference, retain_citations]

VALIDATION:
  - Verify entity and relationship type signatures.
  - Check referential integrity and identifier scoping.
  - Validate temporal constraints for loans and reservations.
  - Preserve distinctions between asserted and inferred facts.
  - Test ambiguous author names and edition-level distinctions.
  - Check access restrictions before exposing protected content.

END FLOW ALGEBRA