RDFa is a set of extensions to XHTML being proposed by W3C. RDFa uses attributes from XHTML's meta and link elements, and generalises them so that they are usable on all elements. This allows you to annotate XHTML markup with semantics. A simple mapping is defined so that RDF triples may be extracted.

The W3C RDF in XHTML Taskforce is also working on an implementation for non-XML versions of HTML. The primary issue for the non-XML implementation is how to handle the lack of XML namespaces.

History

RDFa was first proposed by Mark Birbeck in the form of a W3C note entitled XHTML and RDF , which was then presented to the Semantic Web Interest Group at the W3C's 2004 Technical Plenary. Later that year the work became part of the sixth public Working Draft of XHTML 2.0. Although it's typically assumed that RDFa was originally intended only for XHTML 2, in fact the purpose of RDFa was always to give a way to add a metadata to any XML language. Indeed, one of the earliest documents resembleing the RDF/A Syntax name, has the sub-title A collection of attributes for layering RDF on XML languages. The document was written by Mark Birbeck and Steven Pemberton, and was made available for discussion on October 11, 2004. (Note that although the document resembles the logo of a W3C note, it was never actually placed at a W3C URL.)

In April 2007 the XHTML 2 Working Group produced a module to support RDF annotation within the XHTML 1 family. As an example, it included an extended version of XHTML 1.1 dubbed XHTML+RDFa 1.0. Although described as not representing an intended direction in terms of a formal markup language from the W3C, limited use of the XHTML+RDFa 1.0 DTD did subsequently appear on the public Web.

October 2007 saw the first public Working Draft of a document entitled RDFa in XHTML: Syntax and Processing. This superseded and expanded upon the April draft; it contained rules for creating an RDFa parser, as well as guidelines for organizations wishing to make practical use of the technology.

Essence

The essence of RDFa is to give a set of attributes that can be used to carry metadata in an XML language (hence the 'a' in RDFa).

These attributes are:

about – a URI or CURIE specifying the resource the metadata is about; in its absence it defaults to the current document

rel and rev – specifying a relationship or reverse-relationship with another resource

href, src and resource – specifying the partner resource

property – specifying a property for the content of an element

content – optional attribute that overrides the content of the element when using the property attribute

datatype – optional attribute that specifies the datatype of text specified for use with the property attribute

instanceof – optional attribute that specifies the RDF type(s) of the subject (the resource that the metadata is about).

Benefits of RDFa

Five "principles of interoperable metadata" met by RDFa.

Publisher Independence - each site can use its own standards

Data Reuse - data is not duplicated. You do not need separate XML and HTML sections for the same content.

Self Containment - The HTML and the RDF are separated

Schema Modularity - The attributes are reusable

Evolvability - extra fields can be added and XML transforms can extract the semantics of the data from an XHTML file

Additionally RDFa may benefit web accessibility as more information is available to assistive technology.

Instances of RDFa

The following is an example of adding Dublin Core metadata to an HTML file. Dublin Core data elements are data typically added to a book or article (title, author, subject etc.)

However, the real power of RDFa is in how it allows you to mark up running text with semantic values:

In his latest book Wikinomics, Don Tapscott explains deep changes in technology, demographics and business. The book is due to be published in October 2006.

XHTML+RDFa 1.0 example

The following is an example of a complete XHTML+RDFa 1.0 document. It uses Dublin Core and FOAF, an ontology for describing people and their relationships with other people and things: John's Home Page

My name is John D and I like Einstürzende Neubauten.

My favorite book is the inspiring Weaving the Web by Tim Berners-Lee.

In the example above, the document URI can be seen as representing a person. The foaf:nick property (in the first span element) contains a nickname for this person, and the dc:creator property (in the meta element) tells us who created the document. The hyperlink to the Einstürzende Neubauten website contains rel="foaf:interest", suggesting that John Doe is interested in this band. The URI of their website is a resource. (If it had been rev="foaf:interest" instead, it would have suggested that Einstürzende Neubauten were interested in him.)

The foaf:interest inside the second p element is referring to a book by ISBN number. The resource attribute defines a resource in a similar way to the href attribute, but without defining a hyperlink. Further into the paragraph, a span element containing an about attribute defines the book as another resource to specify metadata about. The book title and author are defined within the contents of this tag using the dc:title and dc:author properties.

Here are the same triples when the above document is automatically converted to RDF/XML: Jonathan Doe John D Tim Berners-Lee Weaving the Web