RDF
RDF is the data model underneath the semantic web: everything is a statement with a subject, a predicate and an object, and every name is an IRI.
What it is
RDF — the Resource Description Framework — is a data model in which knowledge is written as triples: a subject, a predicate and an object. Names are IRIs, so a predicate is an identified term rather than a string; where a thing has properties but no name, RDF uses a blank node. Named graphs let a set of triples be grouped and addressed as a unit.
What it gives the knowledge layer
It gives the knowledge layer a representation that merges by default: facts from different sources can be added to the same graph without a schema agreement, producing one connected picture rather than records to reconcile.
Where the spec lives
The model is defined in RDF 1.1 Concepts and Abstract Syntax, a W3C Recommendation specifying triples, IRIs, literals, blank nodes, datasets and named graphs; the semantics are in RDF 1.1 Semantics, which defines what follows from a set of triples. Serialisations such as Turtle and JSON-LD are specified separately.
What it does not solve
RDF gives no schema and no vocabulary: it says how to state something, not what may be stated, and nothing in the model rejects a triple that uses a predicate nonsensically. RDFS adds classes, subclasses, properties and domain and range; OWL adds formal semantics and reasoning.
FAQ
No. RDF is an abstract data model: a set of statements, each with a subject, a predicate and an object. Turtle, JSON-LD and RDF/XML serialise it without changing its meaning.
A blank node stands for something with properties in the graph but no name, such as an unnamed address or an intermediate step in a process.
They let a set of triples be grouped and identified as a unit, so a store can hold several graphs side by side and a query can say which one it means.
Because statements can be added without agreeing on a document shape: two sources can contribute facts about the same subject and the result is one graph.
Sources
- RDF 1.1 Concepts and Abstract Syntax — W3C Recommendation defining triples, IRIs, literals, blank nodes, datasets and named graphs.
- RDF 1.1 Semantics — W3C Recommendation defining entailment and the interpretation of RDF graphs.