Plaza — a gallery of self-describing datasets

▸ Open the gallery → — "a gathering place for self-describing semantic datasets": a searchable grid of every published .rete graph, each shown as a live-read card rather than a hand-written blurb.

The rete plaza page, 'a gathering place for self-describing semantic datasets', with a search box, a row of vocabulary/tag filter chips (ChEBI, DBpedia, GeoSPARQL, OWL, Wikidata, ontology, scholarly, ...), and a grid of dataset cards each showing a schema-pyramid thumbnail of classes (Paper, Person, City, literal), a title (scholar, history, nomisma, linked-jazz), a description, and stats like size, triple count and term count with the vocabularies used. — Every card on this page is read live from its dataset's own `.rete` file — nothing here is typed by hand.

What it is

Plaza is not a catalog file someone maintains — it's a page that opens a list of .rete files and asks each one to describe itself. Two small HTTP range reads (the header, then the metadata section) pull back the dataset's Dataset Card: its title, triple count, term count, the vocabularies it uses, and a schema-pyramid sketch of its classes. That sketch is drawn on the spot from the card's own class list, so it changes the moment the underlying data does.

How to use it

Type in the search box to filter by name, tag, or vocabulary, or click any of the chip filters (ChEBI, Wikidata, GeoSPARQL, OWL, …) to narrow the grid to datasets that actually use that vocabulary. Click a card to open its detail view and explore further — the query panel, schema graph, and file links all read from the same live card.

The key idea

Because a .rete file carries its own card, a directory like this needs no server and no maintained index — pointing the page at a list of files is enough. That's the same Dataset Cards mechanism the playground uses to show a dataset's stats before you load it: the file is the catalog entry.