Linked Data

Thu, 21 Jul 2022 09:38:15 GMT — Properties

---

Tim Berners-Lee came up with these four principles of Linked Data in 2006:

1. Use URIs as names for things. URIs are the best way available to uniquely identify things, and therefore to identify connections between things.
2. Use HTTP URIs so that people can look up those names. You may have seen URIs that begin with ftp:, mailto:, or prefixes made up by a particular community, but using these other ones reduces interoperability, and interoperability is what it's all about.
3. When someone looks up a URI, provide useful information, using the standards, for example, RDF, SPARQL. While a URI can just be a name and not actually the address of a web page, this principle says that you may as well put something there. It can be an HTML web page, or something else; whatever it is, it should use a recognized standard. RDFS and OWL let you spell out a list of terms and information about those terms and their relationships in a machine-readable way — readable, for example, by SPARQL queries. Because of this, if a URI that identifies a resource leads to RDFS or OWL declarations about that resource, this is a big help to applications.
4. Include links to other URIs so that they can discover more things. Imagine if none of the original HTML pages had a elements to link them to other pages. It wouldn't have been much of a web. Going beyond this HTML linking element, various RDF vocabularies provide other properties that let you say "this data (or this element of data) has a specific relationship to another resource on the Web." When applications can follow these links, they can do interesting new things.

Back to top