fix: correct hallucinated PREMIS terms and Schema.org namespace mismatch

PREMIS ontology fixes (8 schema files): - Replace invalid premis:hasRepresentation with dcterms:hasFormat - Replace invalid premis:hasAccessRestriction with odrl:hasPolicy - Replace invalid premis:hasPreservationPolicy with dcterms:conformsTo - Replace invalid premis:hasAccessPolicy with dcterms:accessRights - Replace invalid premis:hasStoragePolicy with dcterms:conformsTo - Replace invalid premis:ProcessingStatus with skos:Concept - Add proper close_mappings to valid PREMIS classes (premis:Representation, etc.) - Document hallucinated terms in Rule 51 (AGENTS.md) for future prevention Schema.org namespace fixes (3 frontend files): - Update OntologyTermPopup.tsx: add normalizeSchemaOrgUri() function - Update ontology-loader.ts: change schema prefix to https://schema.org/ - Update linkml-schema-service.ts: change schema prefix to https://schema.org/ - The schemaorg.owl file uses https:// but code was using http:// These changes ensure ontology term lookups work correctly for Schema.org terms and that LinkML schema files only reference valid ontology predicates.
2026-01-13 14:16:33 +01:00 · 2026-01-13 14:16:33 +01:00 · f2b10fca19
commit f2b10fca19
parent caa2690ba4
21 changed files with 94 additions and 52 deletions
--- a/frontend/public/schemas/20251121/linkml/instances/enums/archive_processing_status.yaml
+++ b/frontend/public/schemas/20251121/linkml/instances/enums/archive_processing_status.yaml
@ -218,13 +218,13 @@ instances:
      scopeNote: >-
        May include conservation treatment (paper repair, deacidification),
        rehousing into archival containers, digitization for preservation/access,
-        or environmental stabilization. PREMIS premis:PreservationEvent.
+        or environmental stabilization. PREMIS premis:Event.
      example:
        - Paper documents undergoing deacidification
        - Photographs being rehoused in archival sleeves
        - Fragile documents being digitized for access
    ontology:
-      class_uri: premis:PreservationEvent
+      class_uri: premis:Event
    typical_duration: "Days to years for major conservation"
    treatment_types:
      - Paper repair
@ -416,7 +416,7 @@ usage:
    Color-code by stage for visual processing pipeline.
  rdf_generation: >-
    Map to rico:Activity for RDF serialization.
-    Use premis:PreservationEvent for IN_PRESERVATION stage.
+    Use premis:Event for IN_PRESERVATION stage.
  workflow_automation: >-
    Enable status transitions in workflow management systems.
    Track timestamps for processing metrics.
--- a/frontend/public/schemas/20251121/linkml/instances/enums/auxiliary_place_type.yaml
+++ b/frontend/public/schemas/20251121/linkml/instances/enums/auxiliary_place_type.yaml
@ -87,7 +87,7 @@ instances:
      note: "vault (architecture)"
    ontology_mappings:
      schema_org: schema:Place
-      premis: premis:storageLocation
+      premis: premis:StorageLocation
    environmental_requirements:
      temperature: "Controlled (varies by material)"
      humidity: "45-55% RH"
--- a/frontend/public/schemas/20251121/linkml/manifest.json
+++ b/frontend/public/schemas/20251121/linkml/manifest.json
@ -1,5 +1,5 @@
 {
-  "generated": "2026-01-13T12:50:30.701Z",
+  "generated": "2026-01-13T13:09:16.459Z",
  "schemaRoot": "/schemas/20251121/linkml",
  "totalFiles": 2886,
  "categoryCounts": {
--- a/frontend/public/schemas/20251121/linkml/modules/classes/AccessPolicy.yaml
+++ b/frontend/public/schemas/20251121/linkml/modules/classes/AccessPolicy.yaml
@ -40,13 +40,13 @@ imports:
 - ../slots/has_appointment_required_flag
 classes:
  AccessPolicy:
-    class_uri: premis:RightsDeclaration
+    class_uri: premis:RightsStatus
    description: "Access policy defining conditions under which heritage collections can be accessed.\n\n**PURPOSE**:\n\n\
      AccessPolicy captures the access conditions governing a Collection:\n- WHO can access (public, researchers, staff only)\n\
      - HOW access is granted (open, by appointment, with credentials)\n- WHEN access is available (opening hours, embargo\
-      \ periods)\n- WHAT restrictions apply (fragile materials, privacy, cultural sensitivity)\n\n1. **PREMIS**:\n   - `premis:RightsDeclaration`\
-      \ - \"An assertion of one or more rights or\n     permissions pertaining to an object and/or its content.\"\n   - Links\
-      \ to Collection via premis:hasRightsDeclaration\n\n2. **Dublin Core**:\n   - `dcterms:accessRights` - \"Information\
+      \ periods)\n- WHAT restrictions apply (fragile materials, privacy, cultural sensitivity)\n\n1. **PREMIS**:\n   - `premis:RightsStatus`\
+      \ - \"Information about the rights status of an object.\"\n   - Links\
+      \ to Collection via premis:rightsStatus\n\n2. **Dublin Core**:\n   - `dcterms:accessRights` - \"Information\
      \ about who may access the resource\n     or an indication of its security status.\"\n\n3. **RiC-O**:\n   - `rico:hasOrHadAllMembersWithAccessConditions`\
      \ - Links RecordSet to access\n     conditions applying to all members\n\n4. **RightsStatements.org**:\n   - Standardized\
      \ rights statements for cultural heritage\n   - E.g., \"In Copyright\", \"No Copyright\", \"Unknown Copyright\"\n\n\
@ -75,7 +75,7 @@ classes:
      \ not current access\n- Access restricted until triggering conditions (time, event)\n- \"Gray literature\" or un-catalogued\
      \ backlogs awaiting processing\n"
    exact_mappings:
-    - premis:RightsDeclaration
+    - premis:RightsStatus
    - dcterms:accessRights
    close_mappings:
    - rico:Rule
@ -236,7 +236,7 @@ classes:
    - Temporal validity enables policy versioning and embargo expiration
    - DimArchive (dark archive) uses AccessPolicy to express preservation-only access
    see_also:
-    - http://www.loc.gov/premis/rdf/v3/RightsDeclaration
+    - http://www.loc.gov/premis/rdf/v3/RightsStatus
    - https://rightsstatements.org/
    - https://localcontexts.org/
    - https://www.ica.org/standards/RiC/ontology#Rule
--- a/frontend/public/schemas/20251121/linkml/modules/classes/CustodianCollection.yaml
+++ b/frontend/public/schemas/20251121/linkml/modules/classes/CustodianCollection.yaml
@ -73,7 +73,7 @@ classes:
    - rico:RecordSet
    - bf:Collection
    related_mappings:
-    - premis:hasRepresentation
+    - premis:relationship
    - dcterms:hasPart
    slots:
    - has_or_had_access_right
--- a/frontend/public/schemas/20251121/linkml/modules/classes/DarkArchive.yaml
+++ b/frontend/public/schemas/20251121/linkml/modules/classes/DarkArchive.yaml
@ -58,10 +58,10 @@ classes:
      \ granted\n- Declassification decision\n- Original system failure (disaster recovery)\n\n**MULTILINGUAL LABELS**:\n\
      - Dark Archive (de) [uses English term]\n\n**RELATED TYPES**:\n- LightArchive (Q112815447) - broadly accessible\n- DimArchive\
      \ (Q112796779) - limited access\n- ClosedSpace - physical restricted access areas\n\n**ONTOLOGICAL ALIGNMENT**:\n- **SKOS**:\
-      \ skos:Concept (type classification)\n- **PREMIS**: premis:RightsDeclaration for access restrictions\n- **RiC-O**: rico:RecordSet\
+      \ skos:Concept (type classification)\n- **PREMIS**: premis:RightsStatus for access restrictions\n- **RiC-O**: rico:RecordSet\
      \ with access restrictions\n- **Wikidata**: Q112796578\n\n**PREMIS INTEGRATION**:\n\nDark archives typically use PREMIS\
-      \ for preservation metadata:\n- `premis:rightsGranted` with no access rights\n- `premis:rightsEndDate` for embargo expiration\n\
-      - `premis:linkingAgentIdentifier` for custodian\n"
+      \ for preservation metadata:\n- `premis:RightsStatus` to document access restrictions\n- `premis:endDate` for embargo expiration dates\n\
+      - `premis:Agent` to identify the responsible custodian\n"
    slot_usage:
      wikidata_entity:
        equals_string: Q112796578
@ -92,7 +92,7 @@ classes:
    exact_mappings:
    - wd:Q112796578
    close_mappings:
-    - premis:RightsDeclaration
+    - premis:RightsStatus
    - rico:RecordSet
    - skos:Concept
    broad_mappings:
--- a/frontend/public/schemas/20251121/linkml/modules/classes/DimArchives.yaml
+++ b/frontend/public/schemas/20251121/linkml/modules/classes/DimArchives.yaml
@ -98,7 +98,7 @@ classes:
    exact_mappings:
    - wd:Q112796779
    close_mappings:
-    - premis:RightsDeclaration
+    - premis:RightsStatus
    - rico:RecordSet
    - skos:Concept
    broad_mappings:
--- a/frontend/public/schemas/20251121/linkml/modules/classes/METSAPI.yaml
+++ b/frontend/public/schemas/20251121/linkml/modules/classes/METSAPI.yaml
@ -166,7 +166,7 @@ classes:
          '
        range: boolean
      technical_metadata_standard:
-        slot_uri: premis:hasObjectCharacteristics
+        slot_uri: premis:characteristic
        description: 'Standard used for technical metadata.


--- a/frontend/public/schemas/20251121/linkml/modules/enums/ArchiveProcessingStatusEnum.yaml
+++ b/frontend/public/schemas/20251121/linkml/modules/enums/ArchiveProcessingStatusEnum.yaml
@ -121,8 +121,8 @@ enums:
          
          **Duration**: Days to years for major conservation
          
-          **PREMIS**: premis:PreservationEvent
-        meaning: premis:PreservationEvent
+          **PREMIS**: premis:Event (preservation activity)
+        meaning: premis:Event
        
      PROCESSED_PENDING_TRANSFER:
        description: |
@ -203,11 +203,11 @@ enums:
      - "Tracks operational archive lifecycle BEFORE integration into CustodianCollection"
      - "Processing backlogs commonly span decades in archival institutions"
      - "RiC-O rico:Activity used for processing activities"
-      - "PREMIS premis:PreservationEvent for preservation activities"
+      - "PREMIS premis:Event for preservation activities"
      - "Status changes should be tracked with timestamps for metrics"
    
    see_also:
      - "https://www.ica.org/standards/RiC/ontology#Activity"
-      - "http://www.loc.gov/premis/rdf/v3/PreservationEvent"
+      - "http://www.loc.gov/premis/rdf/v3/Event"
      - "https://nde.nl/ontology/hc/class/CustodianArchive"
      - "https://nde.nl/ontology/hc/class/CustodianCollection"
--- a/frontend/public/schemas/20251121/linkml/modules/slots/digital_surrogate.yaml
+++ b/frontend/public/schemas/20251121/linkml/modules/slots/digital_surrogate.yaml
@ -4,22 +4,22 @@ title: Digital Surrogates Slot
 prefixes:
  linkml: https://w3id.org/linkml/
  hc: https://nde.nl/ontology/hc/
+  dcterms: http://purl.org/dc/terms/
+  premis: http://www.loc.gov/premis/rdf/v3/
 imports:
 - linkml:types
 default_prefix: hc
 slots:
  digital_surrogate:
-    slot_uri: premis:hasRepresentation
+    slot_uri: dcterms:hasFormat
    range: string
    multivalued: true
    description: 'Digital representations/surrogates of collection materials.


-      PREMIS: hasRepresentation links intellectual entities to digital representations.
+      Dublin Core: hasFormat links resources to alternative format representations.

-      A premis:Representation is a digital object that instantiates or renders an
-
-      IntellectualEntity (the conceptual work).
+      PREMIS: premis:Representation class models digital instantiations of intellectual entities.


      Values can be:
@ -40,3 +40,12 @@ slots:
      - "https://archive.org/details/manuscript-collection-x"

      '
+    exact_mappings:
+    - dcterms:hasFormat
+    close_mappings:
+    - premis:Representation
+    annotations:
+      custodian_types: '["*"]'
+      custodian_types_rationale: Applicable to all heritage custodian types.
+      specificity_score: 0.6
+      specificity_rationale: Digital surrogate links are moderately specific to digitized collections.
--- a/frontend/public/schemas/20251121/linkml/modules/slots/digitization_status.yaml
+++ b/frontend/public/schemas/20251121/linkml/modules/slots/digitization_status.yaml
@ -1,15 +1,23 @@
 id: https://nde.nl/ontology/hc/slot/digitization_status
 name: digitization_status_slot
 prefixes:
+  linkml: https://w3id.org/linkml/
+  hc: https://nde.nl/ontology/hc/
  premis: http://www.loc.gov/premis/rdf/v3/
+  adms: http://www.w3.org/ns/adms#
+imports:
+- linkml:types
+default_prefix: hc
 slots:
  digitization_status:
-    slot_uri: premis:hasRelatedStatementInformation
+    slot_uri: adms:status
    range: string
    description: 'Current status of collection digitization efforts.


-      PREMIS: hasRelatedStatementInformation for preservation metadata.
+      ADMS: status provides standardized vocabulary for asset lifecycle states.
+
+      PREMIS: premis:note can capture additional preservation metadata.


      Values:
@ -36,5 +44,12 @@ slots:
      - "IN_PROGRESS - expected completion 2026"

      '
-    broad_mappings:
+    exact_mappings:
    - adms:status
+    close_mappings:
+    - premis:note
+    annotations:
+      custodian_types: '["*"]'
+      custodian_types_rationale: Applicable to all heritage custodian types.
+      specificity_score: 0.6
+      specificity_rationale: Digitization status is moderately specific to digital preservation contexts.
--- a/frontend/public/schemas/20251121/linkml/modules/slots/has_or_had_access_condition.yaml
+++ b/frontend/public/schemas/20251121/linkml/modules/slots/has_or_had_access_condition.yaml
@ -24,7 +24,7 @@ slots:
      - rico:hasOrHadAllMembersWithContentType
      - schema:conditionsOfAccess
    related_mappings:
-      - premis:hasRightsStatement
+      - premis:rightsStatus
    broad_mappings:
      - dcterms:rights
    annotations:
--- a/frontend/public/schemas/20251121/linkml/modules/slots/has_or_had_access_control.yaml
+++ b/frontend/public/schemas/20251121/linkml/modules/slots/has_or_had_access_control.yaml
@ -35,12 +35,12 @@ slots:
    range: string
    slot_uri: hc:hasOrHadAccessControl
    close_mappings:
-    - premis:hasRightsStatement
+    - premis:rightsStatus
    - dcterms:accessRights
    related_mappings:
    - schema:conditionsOfAccess
    comments:
-    - premis:hasRightsStatement links to rights statement objects; this slot describes the control mechanisms themselves.
+    - premis:rightsStatus links to rights status objects; this slot describes the control mechanisms themselves.
      Related but not semantically identical.
    annotations:
      custodian_types: '["*"]'
--- a/frontend/public/schemas/20251121/linkml/modules/slots/has_or_had_access_level.yaml
+++ b/frontend/public/schemas/20251121/linkml/modules/slots/has_or_had_access_level.yaml
@ -53,7 +53,7 @@ slots:
    close_mappings:
    - schema:conditionsOfAccess
    related_mappings:
-    - premis:hasRightsStatement
+    - premis:rightsStatus
    annotations:
      custodian_types: '["*"]'
      custodian_types_rationale: Applicable to all heritage custodian types.
--- a/frontend/public/schemas/20251121/linkml/modules/slots/has_or_had_access_policy.yaml
+++ b/frontend/public/schemas/20251121/linkml/modules/slots/has_or_had_access_policy.yaml
@ -54,7 +54,7 @@ slots:
    close_mappings:
    - schema:conditionsOfAccess
    related_mappings:
-    - premis:hasRightsStatement
+    - premis:rightsStatus
    - schema:publishingPrinciples
    comments:
    - schema:conditionsOfAccess is semantically closer (conditions for access) than schema:publishingPrinciples (editorial
--- a/frontend/public/schemas/20251121/linkml/modules/slots/has_or_had_access_policy_reference.yaml
+++ b/frontend/public/schemas/20251121/linkml/modules/slots/has_or_had_access_policy_reference.yaml
@ -18,13 +18,13 @@ slots:
      Links to AccessPolicy class defining access conditions.


-      PREMIS: hasRightsDeclaration for access rights.
+      PREMIS: rightsStatus for access rights.

      '
    range: uri
-    slot_uri: premis:hasRightsDeclaration
+    slot_uri: premis:rightsStatus
    exact_mappings:
-    - premis:hasRightsDeclaration
+    - premis:rightsStatus
    close_mappings:
    - dcterms:references
    - schema:citation
--- a/frontend/public/schemas/20251121/linkml/modules/slots/has_or_had_access_restriction.yaml
+++ b/frontend/public/schemas/20251121/linkml/modules/slots/has_or_had_access_restriction.yaml
@ -49,7 +49,7 @@ slots:
    slot_uri: hc:hasOrHadAccessRestriction
    close_mappings:
    - dcterms:accessRights
-    - premis:hasRightsStatement
+    - premis:rightsStatus
    related_mappings:
    - rico:hasOrHadAllMembersWithContentType
    comments:
--- a/frontend/public/schemas/20251121/linkml/modules/slots/has_or_had_access_right.yaml
+++ b/frontend/public/schemas/20251121/linkml/modules/slots/has_or_had_access_right.yaml
@ -34,7 +34,7 @@ slots:
    close_mappings:
    - schema:conditionsOfAccess
    related_mappings:
-    - premis:hasRightsStatement
+    - premis:rightsStatus
    - edm:rights
    broad_mappings:
    - dcterms:rights
--- a/frontend/src/components/ontology/OntologyTermPopup.tsx
+++ b/frontend/src/components/ontology/OntologyTermPopup.tsx
@ -41,8 +41,8 @@ const STANDARD_PREFIXES: Record<string, string> = {
  'dcterms': 'http://purl.org/dc/terms/',
  'dct': 'http://purl.org/dc/terms/',
  
-  // Schema.org (support both http and https)
-  'schema': 'http://schema.org/',
+  // Schema.org - use https to match schemaorg.owl which uses https://schema.org/
+  'schema': 'https://schema.org/',
  
  // Domain ontologies
  'rico': 'https://www.ica.org/standards/RiC/ontology#',
@ -155,6 +155,17 @@ function expandCurie(curie: string): string {
  return curie;
 }

+/**
+ * Normalize Schema.org URIs to use https:// consistently
+ * The schemaorg.owl file uses https://schema.org/ but some references may use http://
+ */
+function normalizeSchemaOrgUri(uri: string): string {
+  if (uri.startsWith('http://schema.org/')) {
+    return uri.replace('http://schema.org/', 'https://schema.org/');
+  }
+  return uri;
+}
+
 /**
 * Get the prefix from a CURIE
 */
@ -217,8 +228,9 @@ async function loadTermInfo(curie: string): Promise<TermInfo | null> {
    return fetchWikidataEntity(curie);
  }
  
-  // Expand the CURIE to full URI
-  const fullUri = expandCurie(curie);
+  // Expand the CURIE to full URI and normalize Schema.org URIs
+  let fullUri = expandCurie(curie);
+  fullUri = normalizeSchemaOrgUri(fullUri);
  
  // Find the ontology file that contains this URI
  const ontologyFile = getOntologyFileByUri(fullUri);
@ -239,8 +251,10 @@ async function loadTermInfo(curie: string): Promise<TermInfo | null> {
    // Load and parse the ontology
    const ontology = await loadOntology(ontologyFile.path);
    
-    // Search for the term in classes
-    const classMatch = ontology.classes.find(c => c.uri === fullUri);
+    // Search for the term in classes (normalize both sides for Schema.org)
+    const classMatch = ontology.classes.find(c => 
+      normalizeSchemaOrgUri(c.uri) === fullUri
+    );
    if (classMatch) {
      return {
        uri: fullUri,
@ -255,8 +269,10 @@ async function loadTermInfo(curie: string): Promise<TermInfo | null> {
      };
    }
    
-    // Search for the term in properties
-    const propMatch = ontology.properties.find(p => p.uri === fullUri);
+    // Search for the term in properties (normalize both sides for Schema.org)
+    const propMatch = ontology.properties.find(p => 
+      normalizeSchemaOrgUri(p.uri) === fullUri
+    );
    if (propMatch) {
      return {
        uri: fullUri,
@ -274,8 +290,10 @@ async function loadTermInfo(curie: string): Promise<TermInfo | null> {
      };
    }
    
-    // Search for the term in individuals
-    const indMatch = ontology.individuals.find(i => i.uri === fullUri);
+    // Search for the term in individuals (normalize both sides for Schema.org)
+    const indMatch = ontology.individuals.find(i => 
+      normalizeSchemaOrgUri(i.uri) === fullUri
+    );
    if (indMatch) {
      return {
        uri: fullUri,
--- a/frontend/src/lib/linkml/linkml-schema-service.ts
+++ b/frontend/src/lib/linkml/linkml-schema-service.ts
@ -281,7 +281,7 @@ const COMMON_PREFIXES: Record<string, string> = {
  org: 'http://www.w3.org/ns/org#',
  prov: 'http://www.w3.org/ns/prov#',
  pico: 'https://personsincontext.org/model#',
-  schema: 'http://schema.org/',
+  schema: 'https://schema.org/',
  foaf: 'http://xmlns.com/foaf/0.1/',
  rico: 'https://www.ica.org/standards/RiC/ontology#',
  tooi: 'https://identifier.overheid.nl/tooi/def/ont/',
--- a/frontend/src/lib/ontology/ontology-loader.ts
+++ b/frontend/src/lib/ontology/ontology-loader.ts
@ -177,7 +177,7 @@ const NAMESPACES: Record<string, string> = {
  dcterms: 'http://purl.org/dc/terms/',
  skos: 'http://www.w3.org/2004/02/skos/core#',
  foaf: 'http://xmlns.com/foaf/0.1/',
-  schema: 'http://schema.org/',
+  schema: 'https://schema.org/',
  prov: 'http://www.w3.org/ns/prov#',
  org: 'http://www.w3.org/ns/org#',
 };