Skip to content

Metadata

Enola uses Metadata about all of its Things when rendering them:

  • label is a normally fairly short human-readable 🏷️ text; e.g. TITLE of HTML page found at the IRI, or something similar.
  • description is a longer human-readable 📜 description; e.g. first paragraph or <meta ... description> of a HTML page.
  • image of a logo; e.g. from favicon or something like that; or an 😃 Emoji!
  • curie is a compact URI; a shorter form of the original “long” IRI of the Thing.

The info metadata sub-command is a handy tool to test this.

Label

This is read from a number of “typical” properties; in order of priority:

  1. https://enola.dev/label
  2. http://www.w3.org/2000/01/rdf-schema#label
  3. https://schema.org/name
  4. http://purl.org/dc/elements/1.1/title

If the thing itself doesn’t have any of these, it will check if its RDFS range has any.

Alternatively, the (RDFS) Class of a Thing’s (RDF) type can specify an https://enola.dev/labelProperty to specify the IRI of a “custom” property to use a label, if present; see test/metadata-label-property.ttl for an example illustrating how to use this.

This is always available; if a Thing has none of the above, then it will fallback to the CURIE, or else just the last part of its IRI.

Description

This is also read from a number of “typical” properties; in order of priority:

  1. https://enola.dev/description
  2. https://schema.org/description
  3. https://schema.org/abstract
  4. http://purl.org/dc/elements/1.1/description
  5. http://www.w3.org/2000/01/rdf-schema#comment

This is optional and may be empty.

This is read, again in order, from:

  1. https://enola.dev/emoji
  2. https://schema.org/logo
  3. https://schema.org/thumbnailUrl
  4. https://schema.org/image

and similarly to above first on the Thing itself, or if none then via its RDFS range or RDF type.

CURIE

The CURIE is determined by the current context.

This could be different than e.g. a prefix in a loaded Turtle.