Skip to content


An EKG/IRI is an HTTPS-based1 Internationalised Resource Identifier (IRI) that forms the primary identifier of an object2 in the EKG.

According to the first---and most important---of the ten principles of the EKGF, Principle 1: Identity:

Any EKG/IRI is universally unique, opaque, permanent, non-reassignable and web-resolvable.

An EKG/IRI consists of a locator-part and an identifier-part: [locator][identifier]. Where the [identifier] is an EKG/ID. For instance, it could look as follows:

Where is the locator part that assures that any user or system in your enterprise can resolve the EKG/IRI: a user can click on the link and see a meaningful webpage representing the object, a system can fetch all the RDF triples for the given object.

The uuid:e3f0f238-81f2-4434-a441-bb5f5cf10caa part is the identifier aka EKG/ID.

There are many other acceptable forms of EKG/IRIs, besides being based on the UUID v4 standard (a random number), it can also be based on other standards:

  • SHA-128+ (hash based)
  • Web DID (but only when opaque)

See also Principle 1: Identity.

  1. Many RDF datasets are still using http://* URLs for their identifiers rather than https://* URLs. This is not secure, proliferation of such URLs across the enterprise or beyond cannot be promoted due to security concerns. Therefore, all EKG/IRIs shall be using the SSL/TLS-based variant of HTTP: use https://-urls. 

  2. There are many more or less equivalent terms for object, in some literature in the semantic technology world it's called "a Thing" or "an Individual" or "a Resource". Programmers with an object-oriented background could call it "an Instance" (although its not exactly the same).