reference node

Definition

A reference node is a mesh node that represents the (non-dataset) referent of the node — i.e., the thing in the world that the node stands for.

Purpose

To describe what the node refers to (person, place, concept, dataset, etc.).
To supply human/machine labels, identifiers, and minimal provenance about the referent.
To differentiate between metadata about the node itself (_meta) and metadata about the referent.

Typical Contents

Core Description

What is the node’s referent? This is what the HTML resource page uses to say:

“This is an Organization / Ontology / Rate Plan / Dataset / Whatever.”

rdfs:label (human-readable name of the referent).
rdf:type (classifying what kind of thing the referent is).
maybe skos:prefLabel, dcterms:description, owl:sameAs, etc.

Dataset-level description (if the thing is a dataset)

Optional provenance (creator, source, temporal scope).
Optional identifiers (sameAs links, external URIs).
dcat:Dataset, dcat:distribution, dcterms:issued, dcterms:modified, dcterms:license, etc.

This describes the dataset-as-thing.

Context / External Perspectives

Links to registries, catalogs, external provenance, ie. alternate descriptions imported from elsewhere.

Example Snapshot Distribution

# The referent of the node (the actual person)
<djradon>
    rdf:type foaf:Person ;
    rdfs:label "dj radon" ;
    foaf:mbox <mailto:djradon@example.org> ;

Backlinks