All pages on the site are available in a variety of formats, either through content negotiation or a query parameter. Any URI can be looked up using an endpoint, which will return data about the resource identified by that URI.
The endpoint is at https://data.ox.ac.uk/doc/?uri=; simply append the URL-encoded form of the URI you want to look up. The site may redirect you to another page if there is a shorter URL for the descriptive page.
For example, http://oxpoints.oucs.ox.ac.uk/id/00000000 is the identifier for the University of Oxford. By appending it we get https://data.ox.ac.uk/doc/?uri=http%3A//oxpoints.oucs.ox.ac.uk/id/00000000, which redirects to https://data.ox.ac.uk/doc:oxpoints/00000000. This page then provides information about the University.
To follow links in returned RDF data, you may either resolve the URI directly (to see what the original source said about that thing), or use the URI look-up endpoint to discover information held by the Open Data Service.
Each page supports content negotiation on the Accept header, allowing you to provide a list of preferred formats for the response. For example, an Accept header of application/rdf+xml, text/turtle;q=0.9, text/html;q=0.8 would tell the server that you’d like (in decreasing order of precedence) RDF/XML, turtle, and HTML.
It’s also possible to specify the required format with a format extension (where the URL doesn’t end in a slash, or have a query string), or with a format query parameter (in all other cases). This should not be used in the general case, but is useful when using tools that don’t support setting the Accept header. For example, https://data.ox.ac.uk/doc:oxpoints/00000000.rdf is information about the University of Oxford serialized as RDF/XML, and https://data.ox.ac.uk/datasets/?format=ttl is a Turtle representation of the dataset catalogue.