With the ratification of RDF and OWL earlier this year, there's now a framework for a massively distributed, worldwide database connected by ontologies: the Semantic Web.

There's no killer application for the Semantic Web to show the public what the framework can do. As the Semantic Web enters what Tim Berners-Lee calls its second phase this year, developers must start building applications and make working code available to the public.

The Semantic Web is a common framework that enables data to be shared and reused across applications and between enterprises and organizations. It is based on the Resource Description Framework, which integrates a variety of applications using XML for syntax and URLs for naming purposes, and the Ontology Web Language, which is used for defining structured, Web-based ontologies that enable richer integration and interoperability of data across applications.

The potential for the Semantic Web is great. While most Web content today is designed for humans to read, the Semantic Web framework will bring structure to the content of Web pages so computers can begin to understand the data they display.

This is why eWEEK Labs was pleased to see Berners-Lee, who developed the WWW 15 years ago, challenge fellow developers at the Thirteenth International World Wide Web Conference to work on applications for the Semantic Web. The World Wide Web Consortium, headed by Berners-Lee, has championed the Semantic Web by promoting it to the Web community.

During his keynote at the conference, held in New York in May, Berners-Lee applauded his colleagues for their work on RDF and OWL and encouraged them to build applications as the second phase of the Semantic Web's development.

"We're going to have to bootstrap and justify the Semantic Web in the short term by producing applications," he said.

However, Berners-Lee told his colleagues to forget about looking for a killer application for the Semantic Web: Proof of the technology will emerge when new links among information begin to emerge, he said.

A killer app for the Semantic Web may not be necessary, but we saw a very promising Semantic Web application at the WWW show. Haystack, a client-side Java application, applies a consistent interface to different types of user data, such as e-mail, addresses and Web bookmarks.

The Haystack tool, developed by the Massachusetts Institute of Technology's Haystack Project, enables individuals to manage information in a way that makes the most sense to them by removing barriers created by applications handling only certain types of information.

Although e-mail and instant messaging have similar properties, they can't interoperate effectively because they require different applications. Haystack allows users to apply messages to either type of communication.

There is a caveat, however. Although Haystack shows promise, it's a large (40MB to download), somewhat slow application that takes up 512MB of memory.

A beta version of Haystack can be downloaded from haystack.lcs.mit.edu.

Also at the WWW conference, Hewlett-Packard Co. presented Jena 2, a Java framework for writing Semantic Web applications.

Senior Writer Anne Chen can be reached at anne_chen@ziffdavis.com.

This article was first published on eweek.com.