{
  "schema_version": "1.2.0",
  "dataset": "Neruda Archives public catalogue",
  "record_format": "One archival item per JSONL line or CSV row.",
  "null_policy": "Unknown scalar values are null in JSONL and empty in CSV.",
  "array_policy": "JSONL preserves arrays; CSV joins repeated values with ' | '.",
  "provenance": "Generated from unauthenticated public Omeka S API values and validated published Historical context JSON sidecars.",
  "formats": {
    "catalogue.jsonl": "Complete research and AI export, including approved public text, machine-review status and ordered media metadata.",
    "catalogue.csv": "Tabular form of the complete public catalogue for research reuse.",
    "catalogue-index.json": "Compact browser-interface feed for fast search, filtering and collection assignment; it deliberately does not replace the complete exports.",
    "catalogue-record.schema.json": "JSON Schema Draft 2020-12 contract for each object in catalogue.jsonl.",
    "catalogue-index.schema.json": "JSON Schema Draft 2020-12 contract for the complete compact index document.",
    "catalogue-manifest.json": "Dataset identity, distribution URLs, media types, byte sizes and SHA-256 hashes.",
    "checksums.sha256": "Conventional SHA-256 list for verifying downloaded public distributions."
  },
  "fields": {
    "identifier": "Stable Neruda Archives identifier; unique and required.",
    "title": "Public scholarly title of the archival item.",
    "description": "Concise public description or summary.",
    "date": "Public date transcription or normalized ISO date when known.",
    "sort_year": "Four-digit year derived from the beginning of date; null when unknown.",
    "sort_date": "Compact-index-only ISO date used for chronological catalogue ordering when the collection profile explicitly enables chronological display; null when a usable date is unavailable.",
    "date_certainty": "Public statement of certainty when explicitly recorded.",
    "languages": "Human-readable public language names.",
    "language_codes": "Two- or three-letter language codes supplied in the public record.",
    "creators": "Public sender or creator values; original wording is preserved.",
    "recipients": "Public recipient values; original wording is preserved.",
    "place_sent": "Reviewed normalized place from which the document was sent, taken only from the approved nra:placeSent value. Null means that no reviewed source value is available; it is never inferred from the order of dcterms:spatial values.",
    "spatial_coverage": "All public dcterms:spatial values. This field can mix origin, destination, and broader regions and is not yet a controlled origin-place facet.",
    "subjects": "Public subject and discovery terms.",
    "document_types": "Public detailed document type values. A controlled broad category is planned.",
    "primary_collection": "Exactly one authoritative collection derived from the item's active Omeka Item Set.",
    "image_count": "Number of public Omeka media references attached to the item.",
    "media": "Ordered public access-image records with display order, media identifier, role, description, alternative text, format, rights, API URL, public access-copy URL, filename, byte size, SHA-256 checksum, pixel dimensions and safe public thumbnails.",
    "media_json": "CSV-only JSON serialization of the ordered public media array.",
    "thumbnail_url": "Compact-index URL of the first safe public access thumbnail; never an original file.",
    "transcription_segments": "Ordered public transcription segments.",
    "translation_segments": "Ordered public English translation segments.",
    "research_statuses": "Approved public statements about the present state of contextual or historical review. Open questions remain open and must not be inferred as resolved.",
    "transcription_statuses": "Approved public statements describing how the transcription was produced or reviewed.",
    "translation_statuses": "Approved public statements describing how the English translation was produced or reviewed.",
    "citation": "Item-level public citation when supplied.",
    "rights": "Item-level public rights statements; authoritative for reuse.",
    "rights_holder": "Public rights-holder values.",
    "access_rights": "Public access-rights values.",
    "stable_url": "Canonical public NRA-L item URL.",
    "api_url": "Unauthenticated public Omeka S API record.",
    "iiif_manifest_url": "IIIF Presentation 3 manifest preserving page order.",
    "omeka_modified": "Public Omeka record modification date in ISO format.",
    "historical_context": "Null when no selective Historical context is published; otherwise the authoritative public JSON sidecar URL, title, version and review date.",
    "historical_context_title": "CSV-only title of the published Historical context.",
    "historical_context_url": "CSV-only authoritative Historical context JSON URL.",
    "historical_context_version": "CSV-only Historical context version.",
    "historical_context_reviewed_on": "CSV-only Historical context review date.",
    "modified": "Effective public page modification date: the newer of the Omeka record date and the published Historical context review date."
  },
  "data_contract": {
    "json_schema_dialect": "https://json-schema.org/draft/2020-12/schema",
    "record_schema_url": "https://archive.neruda-archives.org/catalogue-record.schema.json",
    "index_schema_url": "https://archive.neruda-archives.org/catalogue-index.schema.json",
    "manifest_url": "https://archive.neruda-archives.org/catalogue-manifest.json",
    "checksum_url": "https://archive.neruda-archives.org/checksums.sha256",
    "checksum_algorithm": "SHA-256",
    "distribution_model_reference": "https://www.w3.org/TR/vocab-dcat-3/"
  }
}