identifier
Semantic Flow resources can be identified via two types of identifiers:
Guidance
- Prefer relative or site-root-absolute paths inside the mesh; do not hardcode full base IRIs so the mesh remains portable across hosting locations (see Reference Iri Choices).
Identifier Senses
Content Identifiers
Identifiers that denote concrete information resources (files on disk or over HTTP):
- Distributions → materialized datasets, e.g.
test.ttl,djradon.jsonld, etc. - Resource pages → e.g.
index.html - Other documentation resources → e.g.
README.md,CHANGELOG.md
These are retrievable representations (materialized content), i.e. when dereferenced with a request to a Semantic Flow site, the content itself is returned.
Concept Identifiers
Identifiers that refer to concepts, entities, or abstract things, including:
- bare node identifiers → Organizational containers
- reference node identifiers → denotational
- payload node identifiers → Concepts that are datasets
- Abstract flow identifiers → Dataset-as-persistent-concept
- Concrete dataset identifiers → Specific dataset snapshots
- Handle identifiers → Mesh node themselves
When dereferenced with a request to a Semantic Flow site, concept identifiers return content, but they still denotation a concept.
Identifier Pattern Semantics
| Identifier Type | Trailing Slash? | Refers to… | Example |
|---|---|---|---|
| Content identifier | No | A fetchable document or asset | https://example.org/ns/foo/index.html |
| Concept IRI | Yes (/) | A real-world or mesh concept | https://example.org/ns/foo/ |
Even though you might be tempted to think of datasets as concrete things, the IRIs for payload nodes, flows, and snapshots all refer to concepts, i.e., non-retrievable entities. Only Distribution IRIs refer to downloadable data, i.e., dataset distributions.
Children
Backlinks