@netwerk-digitaal-erfgoed/network-of-terms-catalog

Network of Terms Catalog

This is the catalog of terminology sources that can be queried through the Network of Terms.

Data model

Each terminology source is modelled as a Schema.org Dataset. The catalog directory contains all datasets.

Most of the dataset’s properties match the NDE Requirements for Datasets, with the following additions specific to the Network of Terms:

• schema:url is used for the HTTP URI prefix of terms in the dataset, e.g. http://vocab.getty.edu/aat/ for Getty resources. This prefix is needed when clients look up terms by their URI in the Network of Terms: the Network then has to know which source to consult to retrieve the term;
• schema:inLanguage is a required property;
• schema:mainEntityOfPage is a required property;
• each distribution has two or more schema:potentialActions:
  • a schema:SearchAction and a schema:FindAction, both with a schema:query property that points to the queries directory;
  • optionally, a number of schema:Actions that configure the features that the Network of Terms provides for the distribution, such as Reconciliation.

The queries directory contains SPARQL queries for retrieving terms from the datasets. There are two types of queries:

• search queries find terms matching a textual string query input;
• lookup queries retrieve a single term based on its URI.

Contributing

Adding a dataset

• Create a your-dataset.jsonld file in the catalog/datasets directory and add a description.
• Create a your-dataset.rq file in the queries/search directory and add your SPARQL search query. A SPARQL lookup query goes into the queries/lookup directory.
  • If your SPARQL server supports fulltext search relevance scores, you can return them as vrank:simpleRank values to have search results ordered by rank instead of the default, alphabetical order.
• Run the tests to make sure your dataset description conforms to the dataset SHACL.
• To try your queries locally, you can run the GraphQL API with your catalog.