Based on the Linked Art documentation, here’s a synthesis of the key patterns used throughout the model:
\nCore Foundational Patterns
\nLinked Art uses intermediate entities representing activities to connect other entities, rather than direct relationships. For instance, instead of directly linking an artwork to an artist, there’s a creation activity carried out by the artist that can have additional properties like time, place, techniques, and other objects . This creates a “What, Who, Where, When, Why and How” framework.
\nThe model extensively uses the classified_as property to type entities using controlled vocabularies (particularly the Getty AAT). This allows flexible categorization without creating specific relationships for every case. Concepts can have both broader terms and classifications, with broader terms encompassing the concept directly while classifications categorize it .
\nEvery entity has essential properties for usability:
\n\t•\tUnique identifier (id)
\n\t•\tType classification (type)
\n\t•\tHuman-readable label (_label)
\n\t•\tNames and identifiers (identified_by)
\nWhen data doesn’t support full specificity or is best expressed in human-readable form, the pattern uses the LinguisticObject class to record descriptive text that can be referenced by any other entity .
\nThe minimal timespan model uses begin_of_the_begin and end_of_the_end properties with fully qualified ISO 8601 timestamps, plus human-readable Names for display .
\nPrimary information uses simple patterns while more complex data expands upon it without interference - for example, basic Production uses carried_out_by for the Artist, while previous attributions use AttributeAssignment to expand upon this .
\nThese patterns work together to create a flexible, event-based system for describing cultural heritage that balances machine-readability with practical usability.
\nBased on the Linked Art model and real-world implementations, here’s how to best leverage it for websites and web apps with optimal UI/UX:
\nTechnical Implementation Strategy
\n\t•\tUse the Linked Art API endpoints as your data layer
\n\t•\tImplement JSON-LD consumption on the frontend (it’s JSON, so JavaScript-friendly)
\n\t•\tCache frequently accessed relationships to reduce API calls
\n\t•\tUse the search API with HAL links for discoverability
\nThe exhibition browser exemplar demonstrates creating sustainable presentations with minimal resources - ideal for small institutions or projects with limited maintenance capacity . Start simple and layer complexity.
\n\t•\tTransform Linked Art’s event-based model into user-friendly views
\n\t•\tPre-compute common queries (artist’s works, object provenance chains)
\n\t•\tUse the classification system to drive filtering and navigation
\nUI/UX Best Practices
\nDisplay Patterns for Core Entities:
\nObjects/Artworks:
\n\t•\tLead with high-quality images and essential metadata (title, artist, date)
\n\t•\tUse expandable sections for production details, materials, dimensions
\n\t•\tShow provenance as a visual timeline rather than dense text
\n\t•\tLink related works through “Explore Similar” sections
\nArtist/Agent Pages:
\n\t•\tBiography with timeline of key activities (productions, exhibitions)
\n\t•\tGrid of associated works
\n\t•\tMap showing places of activity/residence
\n\t•\tNetwork visualization of relationships with other artists
\nExhibitions:
\nMuseum apps can use linked data to provide richer contextualized information - visitors scan QR codes to access biographical information, artwork history, and connections to related works . Create immersive exhibition experiences both in-person and online.
\nSearch & Discovery:
\n\t•\tFaceted search using classifications (medium, time period, style)
\n\t•\tAuto-suggest powered by identifiers and alternate names
\n\t•\t“Related to this” sections powered by the activity connections
\n\t•\tVisual browsing through image grids before detail views
\nSpecific UX Recommendations
\nThe Art Tracks project defined four complexity levels for provenance, treating simpler levels as subsets so records can be displayed at appropriate granularity based on available data . Apply this concept broadly:
\n\t•\tShow basic info by default
\n\t•\tOffer “See full details” for researchers
\n\t•\tProvide both human-readable and technical views
\nThe event-based model creates rich connections - turn these into intuitive navigation:
\n\t•\t“Who else exhibited here?”
\n\t•\t“What else was made in this technique?”
\n\t•\t“Where else did this object travel?”
\n\t•\tTimeline views for chronological data
\n\t•\tMap views for geographic data
\n\t•\tNetwork graphs for relationships
\n\t•\tTraditional list/grid for browsing
\nWhen detailed data isn’t available, Linked Art uses descriptive text. In UI:
\n\t•\tDisplay these prominently with attribution
\n\t•\tUse them as fallbacks when structured data is missing
\n\t•\tAllow toggling between “what we know” and “what’s documented”
\nPerformance Considerations
\n\t•\tPagination: Use the API’s built-in pagination for large result sets
\n\t•\tLazy Loading: Load detailed relationships on-demand
\n\t•\tImage Optimization: Leverage IIIF for responsive images if available
\n\t•\tCaching Strategy: Cache entity pages but invalidate when relationships change
\nAccessibility & Internationalization
\n\t•\tThe _label property is perfect for screen readers
\n\t•\tMultiple Name objects support multilingual content
\n\t•\tClassifications link to Getty AAT which has translations
\n\t•\tProvide text alternatives for all visualizations
\nWould you like me to create a prototype interface demonstrating any of these patterns?
","updatedAt":"2018-10-20T01:46:40.000Z","checksum":"d45bbbb02d70a6955ef5388bbb083e3b7a3acd90063a5112e6c80ae6ecd65a14","checksumPrefix":"d45bbbb02d70","anchorCount":0,"lineCount":85,"rawUrl":"/api/docs/content?path=linked-art%2FLinkedArtPatterns.md","htmlUrl":"/docs?doc=linked-art%2FLinkedArtPatterns.md","apiUrl":"/api/docs/content?path=linked-art%2FLinkedArtPatterns.md"}