Skip to Content

What Is RDF?

Printer-friendly versionPrinter-friendly versionSend to friendSend to friend

Rachel Heery explains RDF (Resource Description Framework)

What is RDF?

It's the Resource Description Framework. Does that help? No?

RDF is the latest acronym to add to your list, one that is set to gain in significance in the future. At present though it is early days for RDF and little accessible information is available for the interested reader. This short summary will try to outline some key points regarding RDF and point to available further information. What is certain is that this summary will go out of date quickly, RDF is 'work in progress' and is an area which is undergoing rapid development and change.

Information for this article is largely derived from the February 1998 version of the RDF specification [1], but all interpretation is entirely the responsibility of the author, as are all views and opinions.

First to give some context, RDF is an initiative of the World Wide Web Consortium (W3C) [2], a 'members only' organisation restricted by rules of confidentiality so inevitably there is some delay in disseminating information about emerging work. A few UK universities are members of W3C as is JISC in the person of Brian Kelly, the Web Focus Officer, who is based at UKOLN. These memberships of W3C give individual UK researchers a crucial opportunity to increase their awareness of current work.

RDF is about metadata for Web resources, by resources we mean any object that can be found on the Web. Essentially, RDF is a means for developing tools and applications using a common syntax for describing Web resources. Last summer W3C recognised there were a number of connected areas which all depend on the existence of metadata for Web resources. Work on resource discovery, content ratings, intellectual property rights and digital signatures was connected in its need for a 'resource description framework'. So over the last few months a working group within the W3C have drawn up a data model and syntax for RDF.

So what is RDF? RDF is distinctive in that it is designed specifically with the Web in mind, so it takes account of the features of Web resources. It is a syntax based on a data model, and this model influences the way properties are described in that it makes the structure of descriptions explicit. This means RDF has a good fit for describing Web resources, but, on the downside, it might cause problems within environments where there is a need to re-use or interoperate with 'legacy metadata' which may well contain logical inconsistencies. (An extreme example might be the MARC format, the design of which was influenced originally by the needs of tape storage.)

The RDF data model represents the properties of a resource and the values of these properties. The model is syntax independent but can be expressed in XML, and the specification uses XML as its syntax for encoding metadata. XML (eXtensible Mark-up Language) is a sub-set of SGML. It can be viewed as a simplified dialect or abbreviated version of SGML. Whereas HTML is a non-extensible syntax and says nothing about the semantics of tagged fields, XML allows 'locally defined' tags to be created. Communities can agree on a set of tags to indicate content, as well as structure of resources, so that functionality can be built into the client to manipulate content, promising much innovative development at the client side. XML offers the possibility for building products based on exchanging structured data between applications. Some examples of such applications are given by Richard Light [3] in his recent book, they include Sun's online documentation application AnswerBook, health care industry information and electronics industry datasheets.

XML itself facilitates the creation of 'metadata' and RDF is seen as a framework to realise the potential of XML. The opportunities to build applications provided by XML apply equally to RDF. RDF provides the syntax for interworking applications to recognise and exchange metadata for Web objects. This might be metadata designed to facilitate resource discovery or metadata describing content ratings or intellectual property rights.

One of the barriers to understanding RDF is that little has been revealed of the way it will actually be used in tools and applications. Although we have an idea of the general areas of interest, little has been communicated about any future 'products'. Some of the commercial 'big players' have shown interest in RDF but where do they see RDF fitting into their products? Possibly as a component of intranet management tools or possibly as an adjunct to Web authoring tools.

RDF is a means to express properties of a resource and to associate values with these properties. It does not mandate the use of any particular properties or element names. The creator of the RDF record can choose which particular properties or sets of properties they wish to use. In order to ensure properties have unique names, and to allow for reference to an authority for the meaning and usage of the Property name, RDF makes use of the namespace mechanism, which is also used in XML. This mechanism is not yet fully defined but involves reference to a namespace URI (in practice a URL) where the particular property is defined. This URI could link to a machine or human readable definition, or the URI could merely serve as a unique identifier with no real link to information content. A collection of properties describing a resource is called a Description. The properties and values used in a Description also are defined by a schema identified by a namespace URI. The specification of RDF schemas is under discussion and has not yet been made available publicly.

One of the significant issues for those interested in providing managed access to the internet is whether future implementation of RDF will offer sufficient flexibility to use it as basis for 'low cost' resource discovery. Will RDF tools use properties based on Description schemas decided by the product vendors? Or will communities be able to agree on their own Description schemas tailored to their particular searching requirements? So, for example, will Web browsers and authoring tools allow a variety of RDF Description Schemas according to the choice of the user, or will these tools mandate a particular Description schema?

The RDF 'association models' between the metadata and the resource are similar to HTML. The metadata can be embedded in the resource, it can be associated to the resource, it can be independent from the resource, or the resource can be contained within the metadata. Some of these methods of association will not be available for certain resource types, e.g. it is not possible to embed metadata in a gif image.

How accommodating will RDF be to 'legacy' metadata formats which are not based on data models and do not fit easily with the logical structure of RDF? Will the namespace mechanism be used within a system of registries? Or will the namespace URI merely serve as an identifier with no data behind it? If the latter then it seems likely that 'unauthorised' identifiers will proliferate.

For RDF to reach its potential for resource discovery it seems likely that namespace registries with machine readable schemas will need to develop in tandem. But even without this development, RDF will still offer considerable progress for syntactical interoperability.

So there are still a lot of questions to answer. Once again service providers need to make critical decisions as to whether they wait for the answers to emerge, or use existing technology. Certainly within the Dublin Core community there is acknowledgement of the potential significance of RDF, and there is considerable activity centred on accommodating Dublin Core to the RDF model. However alongside this activity various projects are beginning to identify the real issues involved by implementing Dublin Core using the existing HTML technology.

This summary provides a snapshot. In order to keep in touch with progress refer to the W3C home page for RDF at URL: http://www.w3.org/RDF/

References

[1] RDF specification February 1998
http://www.w3.org/RDF/Group/1998/02/WD-rtf-syntax-19980206/

[2] World Wide Web Consortium
http://www.w3.org/

[3] Light, Richard. Presenting XML. Sams.net, 1997.

Further information

Some example of RDF records provided by Andy Powell:
URL: http://www.ukoln.ac.uk/metadata/resources/rdf/examples/
Lassila, Ora and Swick, Ralph (editors). Resource Description Framework (RDF) model and syntax. W3C, February 1998. Latest version at:
URL: http://www.w3.org/RDF/Group/WD-rdf-syntax/
Lassila, Ora. Introduction to RDF Metadata, W3C., Note 1997-11-13
URL: http://www.w3.org/TR/NOTE-rdf-simple-intro
Frequently asked Questions about RDF, W3C.
URL: http://www.w3.org/RDF/FAQ
Frequently Asked Questions about the Extensible Markup Language,W3C.
URL: http://www.ucc.ie/xml/#FAQ
THE XML Specification, W3C. Proposed version December 1997.
URL: http://www.w3.org/TR/PR-xml.html
The navigational home page of XML at
URL: http://www.w3.org/XML/

Author details

Rachel Heery
Metadata Group
UKOLN
University of Bath
Email: r.m.heery@ukoln.ac.uk

Date published: 
19 March 1998

This article has been published under copyright; please see our access terms and copyright guidance regarding use of content from this article. See also our explanations of how to cite Ariadne articles for examples of bibliographic format.

How to cite this article

Rachel Heery. "What Is RDF?". March 1998, Ariadne Issue 14 http://www.ariadne.ac.uk/issue14/what-is/


article | by Dr. Radut