JSON-LD is designed around the concept of a "context" to provide additional mappings from JSON to an RDF model. The context links object properties in a JSON document to concepts in an ontology. In order to map the JSON-LD syntax to RDF, JSON-LD allows values to be coerced to a specified type or to be tagged with a language. A context can be embedded directly in a JSON-LD document or put into a separate file and referenced from different documents (from traditional JSON documents via an HTTP Link header).
The example above describes a person, based on the FOAF vocabulary. First, the two JSON properties name and homepage and the type person are mapped to concepts in the FOAF vocabulary and the value of the homepage property is specified to be of the type @id, i.e., it is specified to be an IRI in the context definition. Based on the RDF model, this allows the person described in the document to be unambiguously identified by an IRI. The use of resolvable IRIs allows RDF documents containing more information to be transcluded which enables clients to discover new data by simply following those links; this principle is known as Follow Your Nose.
By having all data semantically annotated as in the example, a RDF processor can identify that the document contains information about a person (@type) and if the processor understands the FOAF vocabulary it can determine which properties specify the person’s name and the homepage of the person.