W3C takes first step toward RDF spec

The Internet standards group releases a working draft of an evolving technology that promises to help categorize, organize, and search Web-based information.

3 min read
The World Wide Web Consortium standards group released today the first public working draft of an evolving technology that promises to help users and content developers categorize, organize, and search through the ever-growing morass of Web-based information.

The Resource Description Framework, or RDF, is a "metadata" architecture--in other words, it provides a way to describe and organize supersets of data across otherwise incompatible applications. For example, this Web page can be labeled an "article" and Alex Lash can be labeled a "reporter." RDF would then establish an "author" relationship between Alex Lash and the article, and users would be able to search across a variety of documents, not just Web pages, for such relationships.

The specification is not yet a standard, but Netscape Communications has already promised to ship a product based on the evolving draft in the first half of 1998. Code-named Aurora, the software will let PC users customize the way they view information on their desktops.

"If Netscape is preparing the community for a quick adoption of RDF as the metadata exchange mechanism when RDF becomes an approved W3C recommendation, that can only be a service to the Web," W3C metadata project manager Ralph Swick recently told CNET's NEWS.COM.

Some Web developers complain that both Netscape and Microsoft often build browsers with their own versions of a technology before it becomes standardized, as both have done with dynamic HTML. By making works in progress commercially available, the companies risk having to change their product to support the final specification; developers who start to use the technology face the risk of having to rebuild their content.

Others, including W3C director and Web pioneer Tim Berners-Lee, have defended the practice of adding nonstandard extensions to Internet software, saying that it encourages the development and adoption of new technologies.

For this first RDF draft, the working group, which includes representatives from IBM, Microsoft, Netscape, Nokia, Reuters, SoftQuad, and the University of Michigan, has combined several proposals from member companies.

One thing that's certain is that RDF will use the Extensible Markup Language (XML) as the underlying syntax. XML, itself an evolving standard, is an extension of HTML that allows for the creation of new sets of tags without the approval of the W3C.

"It gives us a very extensible syntax with which people can go off and invent their own tags," said R.V. Guha, Netscape principal scientist.

As an Apple employee, Guha was a creator of the Meta Content Format, which the W3C is incorporating into the RDF spec.

"RDF takes good ideas from both [Microsoft and Netscape] but is unique," Tom Johnston, a product manager for Microsoft platform marketing, said in a statement.

The original RDF design was based on the PICS framework, which has been used to rate Web sites according to content.

Netscape and Microsoft have recently jockeyed to establish a lead role in the shaping of RDF, with each claiming before today's release to have their own technology at the heart of the evolving standard. Meanwhile, the standard is unlikely to be finalized before 1998. A spec becomes a standard when the W3C gives it a "recommendation."