RDF 1.1Almost exactly 10 years after the publication of RDF 1.0 (10 Feb 2004, http://www.w3.org/TR/rdf-concepts/), the World Wide Web Consortium (W3C) has announced today that RDF 1.1 has become a “Recommendation.” In fact, the RDF Working Group has published a set of eight Resource Description Framework (RDF) Recommendations and four Working Group Notes. One of those notes, the RDF 1.1 primer, is a good starting place for those new to the standard.

SemanticWeb.com caught up with Markus Lanthaler, co-editor of the RDF 1.1 Concepts and Abstract Syntax document, to discuss this news.

photo of Markus LanthalerLanthaler said of the recommendation, “Semantic Web technologies are often criticized for their complexity–mostly because RDF is being conflated with RDF/XML. Thus, with RDF 1.1 we put a strong focus on simplicity. The new specifications are much more accessible and there’s a clear separation between RDF, the data model, and its serialization formats. Furthermore, the primer provides a great introduction for newcomers. I’m convinced that, along with the standardization of Turtle (and previously JSON-LD), this will mark an important point in the history of the Semantic Web.”

Almost exactly one year ago, we covered the announcement that Turtle had reached “Candidate Recommendation” status, and when asked about how Turtle is only now becoming a formal standard Lanthaler added, “That’s especially surprising for Turtle as it has been in wide use for quite a while now. Turtle is also more closely aligned with SPARQL. Thus, it not only supports @base and @prefix but also their SPARQL equivalents BASE and PREFIX. It’s thus easier to copy and paste between SPARQL queries and Turtle documents.”

What’s new in 1.1?

Lanthaler outlined some of the most important changes for us:

  • identifiers are now IRIs (“Internationalized Resource Identifieers” — they were “RDF URI references” before)
    Lanthaler: “In 2004, when RDF 1.0 was published, the IRI specification (RFC3987) wasn’t ready yet so it couldn’t be used. Instead, the working group back then described what they believed would become an IRI. So, to sum it up, nothing really changes for developers. It’s just an alignment with existing web standards.”
  • language-tagged literals now always have the rdf:langString datatype
  • simple literals, i.e., literals without datatype do not exist anymore
  • introduction of datasets (difference to SPARQL is that graph name can also be a blank node)
  • a couple of new XSD datatypes are now compatible with RDF, namely xsd:duration, xsd:dayTimeDuration, xsd:yearMonthDuration, xsd:dateTimeStamp
  • new rdf:HTML datatype
  • semantics uses the notion of “recognized datatypes” instead of a “datatype map”

For more specifics on the changes, see “What’s New in RDF 1.1.

 

Included in this announcement are the following standards documents:

NOTE: SemanticWeb.com is a member organization of W3C and we extend congratulations to the RDF Working Group for completing this work.