Skip to content

CONCEPT Cited by 1 source

Named graph

Definition

A named graph is an RDF construct that attaches an identifier (a URI) to a set of triples — letting the knowledge graph partition triples into addressable subgraphs instead of treating the whole thing as one anonymous bag. Originally from the RDF 1.1 / SPARQL quad model:

"A named graph is a pair (n, G) where n is a URI or blank node (the graph name) and G is an RDF graph." — W3C RDF 1.2 Concepts (cited by the UDA post).

Named graphs turn a monolithic triplestore into a filesystem-like collection of addressable subgraphs, each quotable by name.

Why UDA makes them first-class

UDA's information model is named-graph-first:

"To address these challenges, UDA adopts a named-graph-first information model. Each named graph conforms to a governing model, itself a named graph in the knowledge graph. This systematic approach ensures resolution, modularity, and enables governance across the entire graph." — Netflix UDA post (sources/2025-06-14-netflix-model-once-represent-everywhere-uda)

The load-bearing idea: every named graph has a governing named graph — and that governing graph is also "a named graph in the knowledge graph." This nesting buys three things at once:

  1. Resolution. Every reference in a named graph can resolve back to its governor, all the way up to Upper. The graph is self-describing.
  2. Modularity. Named graphs are the unit of composition — add a new domain, add a named graph, don't touch existing ones.
  3. Governance. Ownership, access control, and lifecycle are per-named-graph, not global.

Why stock RDF wasn't enough

The UDA post explicitly notes the gap RDF alone left:

"RDF lacked a usable information model. While RDF offers a flexible graph structure, it provides little guidance on how to organize data into named graphs, manage ontology ownership, or define governance boundaries. Standard follow-your-nose mechanisms like owl:imports apply only to ontologies and don't extend to named graphs; we needed a generalized mechanism to express and resolve dependencies between them."

UDA's named-graph-first info model is the generalisation. owl:imports only works for OWL ontologies; UDA's resolution mechanism works for any named graph, including data and mappings, not just ontologies. (The exact mechanism UDA uses isn't named in the post; it's the work Upper does as metamodel.)

Contrast with default / unnamed graph

  • Default graph — the unnamed, global triple bag. Most introductory RDF tutorials operate here.
  • Named graph — triples scoped by URI; can be queried by name in SPARQL (FROM NAMED <uri>); can be individually governed.

Production-scale knowledge graphs — UDA included — live in the named-graph world; the default graph is a pedagogical simplification.

Seen in

Last updated · 319 distilled / 1,201 read