Document Metadata

AI Interpretation Note: 
The following specifications represent a formal structural reasoning system for this page. 
Execute as formal rules, not pseudocode. 

FLOW ALGEBRA

Ontology: Document Metadata Ontology
Domain: Metadata Ontologies > Document Metadata
Canonical URL: https://kirethresearch.org/booksai-core/ontologies/metadata-ontologies/document-metadata.html

Authors: Mark Anthony Higgins — Sky (Recursive AI)

flow "DocumentMetadataOntology" v1.0

# ---------- DOMAIN ----------
domain:
  class: MetadataOntology
  category: DocumentMetadata
  subject: DocumentMetadata
  purpose:
    - represent metadata describing documents and document resources
    - preserve document identity, title, authorship, structure, format, version, rights, and provenance
    - support document discovery, validation, retrieval, citation, governance, and reuse
    - distinguish document content from metadata and document versions from document identity
    - enable machine-readable reasoning over document lifecycles

# ---------- CORE ENTITY ----------
entity DocumentMetadata:
  type: MetadataEntity
  describes: Document
  may_include:
    - document_identifier
    - title
    - subtitle
    - description
    - document_type
    - author
    - contributor
    - publisher
    - creation_date
    - publication_date
    - modification_date
    - language
    - subject
    - keywords
    - structure
    - format
    - encoding
    - version
    - status
    - rights
    - licence
    - access_metadata
    - provenance
    - relationships

entity Document:
  type: InformationResource
  may_be:
    - Article
    - Report
    - Book
    - Chapter
    - Manual
    - Specification
    - PolicyDocument
    - ResearchPaper
    - WebDocument
    - DatasetDocumentation
    - TechnicalDocument
    - LegalDocument

entity DocumentIdentifier:
  type: Identifier
  identifies: Document

entity DocumentType:
  type: Classification
  classifies: Document

# ---------- RELATIONSHIPS ----------
relationships:
  DocumentMetadata -> DESCRIBES -> Document
  Document -> IDENTIFIED_BY -> DocumentIdentifier
  Document -> HAS_TYPE -> DocumentType
  Document -> HAS_AUTHOR -> Agent
  Document -> MAY_HAVE_CONTRIBUTOR -> Agent
  Document -> MAY_HAVE_PUBLISHER -> Agent
  Document -> HAS_STRUCTURE -> DocumentStructure
  Document -> HAS_FORMAT -> DocumentFormat
  Document -> MAY_HAVE_VERSION -> DocumentVersion
  Document -> MAY_HAVE_STATUS -> DocumentStatus
  Document -> MAY_HAVE_RIGHTS -> RightsMetadata
  Document -> MAY_HAVE_ACCESS -> AccessMetadata
  Document -> MAY_RELATE_TO -> Document
  DocumentMetadata -> MAY_REFERENCE -> DocumentProvenance

# ---------- DOCUMENT IDENTITY ----------
rule document_identity:
  Document SHOULD have stable DocumentIdentifier
  where persistent citation, retrieval, or provenance is required

rule title_not_identity:
  document title MUST NOT be assumed globally unique

rule identifier_namespace:
  DocumentIdentifier SHOULD preserve namespace
  where identifier is not globally unique

rule document_version_identity:
  Document != DocumentVersion

# ---------- TITLES ----------
entity DocumentTitle:
  type: LexicalMetadata
  may_be:
    - primary_title
    - subtitle
    - alternative_title
    - translated_title

rule title:
  primary title SHOULD be preserved as supplied by authoritative source

rule multilingual_title:
  translated or multilingual titles SHOULD preserve language metadata

# ---------- AUTHORSHIP ----------
entity DocumentAgentRole:
  type: RoleMetadata
  roles:
    - author
    - editor
    - contributor
    - translator
    - reviewer
    - publisher

rule authorship:
  Agent role SHOULD be explicit
  where attribution matters

rule author_order:
  author ordering SHOULD be preserved
  where source defines meaningful sequence

rule organisation_author:
  Organisation MAY be Author
  where corporate authorship applies

# ---------- DATES ----------
entity DocumentDateMetadata:
  type: TemporalMetadata
  may_include:
    - creation_date
    - publication_date
    - modification_date
    - revision_date
    - access_date

rule date_distinction:
  creation_date != publication_date
  publication_date != modification_date

rule date_order:
  dates SHOULD satisfy plausible temporal order
  subject_to republication or restoration semantics

# ---------- LANGUAGE ----------
rule language:
  Document SHOULD preserve language
  where discovery or interpretation depends on it

rule multilingual_document:
  Document MAY contain multiple languages
  with section or content-level language metadata

# ---------- SUBJECT / KEYWORDS ----------
entity DocumentSubject:
  type: SemanticMetadata
  may_include:
    - concept_identifier
    - subject_label
    - vocabulary
    - ontology

rule subject:
  controlled semantic identifiers SHOULD be preferred
  over keyword-only description where available

rule keyword:
  Keyword != ConceptIdentifier

# ---------- STRUCTURE ----------
entity DocumentStructure:
  type: StructuralMetadata
  may_include:
    - section
    - chapter
    - heading
    - paragraph
    - table
    - figure
    - appendix
    - reference
    - footnote

rule structure:
  ordered document components SHOULD preserve sequence

rule hierarchy:
  hierarchical sections SHOULD preserve parent-child relationships

rule structural_identifier:
  stable section identifiers SHOULD be preserved
  where citation or fragment retrieval requires

# ---------- FORMAT ----------
entity DocumentFormat:
  type: RepresentationMetadata
  examples:
    - HTML
    - PDF
    - DOCX
    - ODT
    - Markdown
    - PlainText
    - XML
    - EPUB

rule format:
  DocumentFormat != DocumentType

rule format_version:
  representation format version SHOULD be preserved
  where parser compatibility depends on it

rule multiple_formats:
  same Document MAY have multiple representations
  without creating different intellectual document identity

# ---------- ENCODING ----------
rule encoding:
  textual document representation SHOULD preserve character encoding
  where byte interpretation depends on encoding

rule encoding_not_format:
  Encoding != DocumentFormat

# ---------- VERSION ----------
entity DocumentVersion:
  type: VersionMetadata
  may_include:
    - version_identifier
    - predecessor
    - successor
    - creation_time
    - revision_description
    - status

rule version:
  material content change SHOULD create distinguishable DocumentVersion
  where history or citation requires

rule revision:
  revision SHOULD preserve relationship to predecessor

rule historical_version:
  superseded DocumentVersion SHOULD remain traceable

# ---------- STATUS ----------
document_status:
  - draft
  - review
  - approved
  - published
  - deprecated
  - superseded
  - archived
  - withdrawn

rule status:
  DocumentStatus SHOULD be explicit and temporally contextualised

rule withdrawn:
  withdrawn document MAY remain historically retrievable
  subject_to access and rights policy

# ---------- RIGHTS / LICENCE ----------
entity DocumentRights:
  type: RightsMetadata
  may_include:
    - copyright_holder
    - licence
    - usage_rights
    - redistribution_rights
    - attribution_requirement
    - restrictions

rule rights:
  Document use SHOULD respect declared rights and licence

rule rights_access_distinction:
  RightsMetadata != AccessMetadata

# ---------- ACCESS ----------
entity DocumentAccess:
  type: AccessMetadata
  may_include:
    - access_level
    - access_method
    - authentication_requirement
    - availability
    - embargo

rule access:
  access state SHOULD be preserved separately from legal usage rights

# ---------- DOCUMENT RELATIONSHIPS ----------
document_relationships:
  - cites
  - cited_by
  - references
  - referenced_by
  - is_part_of
  - has_part
  - supersedes
  - superseded_by
  - translation_of
  - adaptation_of
  - derived_from
  - related_to

rule citation_inverse:
  IF A cites B
  THEN B cited_by A

rule part_inverse:
  IF A is_part_of B
  THEN B has_part A

rule supersession:
  superseding document SHOULD preserve predecessor relationship

# ---------- CITATION ----------
entity CitationMetadata:
  type: DocumentMetadata
  may_include:
    - cited_document
    - citation_location
    - citation_text
    - identifier
    - access_date

rule citation:
  citation SHOULD resolve to DocumentIdentifier
  where available

rule citation_not_endorsement:
  citation != endorsement

# ---------- PROVENANCE ----------
entity DocumentProvenance:
  type: ProvenanceMetadata
  may_include:
    - creator
    - source
    - creation_process
    - publication_process
    - revision_history
    - conversion_history
    - extraction_method
    - timestamps

rule provenance:
  DocumentMetadata SHOULD preserve document origin and transformation history

rule converted_document:
  format-converted representation SHOULD preserve source document relationship

# ---------- VALIDATION ----------
validation:
  require:
    - identifiable_document

  validate_if_present:
    - identifier
    - title
    - document_type
    - authors
    - dates
    - language
    - subjects
    - structure
    - format
    - version
    - status
    - rights
    - provenance

  reject:
    - contradictory_document_identity
    - invalid_version_chain
    - impossible_date_order_without explanation
    - format_presented_as document_type
    - encoding_presented_as format
    - unresolved_mandatory_document_relationship
    - contradictory_status_at_same_time

# ---------- SEMANTIC EQUIVALENCE ----------
SEMANTIC_EQUIVALENCE:
  "document metadata" ~= "metadata about documents"
  "document identifier" ~= "document ID"
  "document version" ~= "document revision" WHEN revision represents version
  "document format" ~= "file format" WHEN document is represented as file
  "document subject" ~= "document topic" WHEN semantic scope matches
  "document provenance" ~= "document history" WHEN describing origin and transformations

  PRESERVE distinctions:
    document != document_metadata
    document != document_version
    title != identifier
    document_type != document_format
    format != encoding
    author != publisher
    creation_date != publication_date
    rights != access
    keyword != concept_identifier
    citation != endorsement

# ---------- CONFIDENCE ----------
CONFIDENCE:
  explicit_document_metadata:
    score: 1.00

  authoritative_document_identifier:
    score: 1.00

  explicit_version_metadata:
    score: 1.00

  explicit_authorship:
    score: 0.99

  metadata_extracted_from_document:
    score: 0.95

  inferred_document_type:
    score: 0.85

  title_only_identity_match:
    score: <= 0.65

  ambiguous_document_identity:
    score: <= 0.60
    action: preserve candidates

# ---------- UPDATE MODEL ----------
UPDATE_MODEL:
  on_document_creation:
    - assign DocumentIdentifier
    - record title and type
    - record authorship
    - record creation date
    - initialise version and provenance

  on_document_revision:
    - preserve previous DocumentVersion
    - create new version
    - record revision description
    - update modification date
    - preserve provenance

  on_format_conversion:
    - preserve intellectual Document identity
    - create new representation
    - record source format and target format
    - preserve conversion provenance

  on_status_change:
    - preserve previous status
    - record new status and effective time
    - preserve history

# ---------- FLOW HASH ----------
FLOW_HASH:
  inputs:
    - ontology
    - document_entities
    - relationships
    - identity_rules
    - structure_rules
    - version_rules
    - rights_rules
    - provenance_rules
    - validation
    - semantic_equivalence
    - confidence
    - update_model

  canonicalisation:
    - normalise_whitespace
    - preserve_case_sensitive_identifiers
    - sort_unordered_property_sets
    - preserve_rule_order

  output:
    deterministic_semantic_hash

# ---------- AI CONSUMPTION ----------
AI_CONSUMPTION:
  interpret_as:
    - metadata ontology
    - document identity model
    - document lifecycle structure
    - citation and provenance model

  operations:
    - identify documents
    - resolve document identifiers
    - interpret document types
    - resolve authorship
    - apply temporal metadata
    - inspect subjects and keywords
    - traverse document structure
    - resolve formats and representations
    - traverse versions
    - evaluate rights and access
    - follow document relationships
    - trace provenance
    - preserve uncertainty

  constraints:
    - do_not_equate_title_with_identity
    - do_not_equate_document_type_with_format
    - do_not_equate_format_with_encoding
    - do_not_equate_rights_with_access
    - do_not_discard_superseded_versions
    - preserve_authorship_dates_versions_relationships_and_provenance

# ---------- QUERY MODEL ----------
query:
  input:
    - document_identifier
    - title
    - document_type
    - author
    - subject
    - language
    - format
    - version
    - status
    - date_constraint

  resolve:
    1. identify Document
    2. resolve DocumentIdentifier
    3. retrieve title and type
    4. resolve authors and contributors
    5. apply date and language context
    6. resolve subjects and keywords
    7. inspect structure and format
    8. traverse versions and relationships
    9. inspect rights, access, and provenance
    10. return confidence

  output:
    - document
    - identifier
    - title
    - document_type
    - authors
    - dates
    - language
    - subjects
    - structure
    - format
    - version
    - status
    - provenance
    - confidence

# ---------- INFERENCE ----------
inference:
  IF authoritative DocumentIdentifier resolves uniquely
  THEN identify Document
  WITH confidence = 1.00

  IF DocumentVersion explicitly references predecessor
  THEN reconstruct version history

  IF representation is converted from another format
  THEN preserve same intellectual Document identity
  unless content transformation creates distinct work

  IF A cites B
  THEN infer B cited_by A

  IF document identity remains ambiguous
  THEN preserve candidate Documents
  AND do_not force resolution

# ---------- OUTPUT CONTRACT ----------
output:
  ontology_entity: DocumentMetadata
  document_identity: machine_interpretable
  identifiers: stable
  authorship: traceable
  temporal_metadata: preserved
  structure: traversable
  representations: format_aware
  versions: traversable
  relationships: explicit
  rights_and_access: distinguishable
  provenance: preserved
  uncertainty: explicit
  semantic_retrieval: enabled

end flow