Challenges: Speed Bumps Ahead For Semantic Web
XML has become the standard language to describe data for the Semantic Web. But companies are still hammering out XML standards for their industries, and that could be a problem.
The creators of the Semantic Web have the interesting challenge of forging a new computing environment that's based on a language, XML, that itself is still maturing as a practical business tool.
Designers have generally settled on XML as the lingua franca of the Semantic Web. It will almost certainly be the bottom-level language used to identify a number as a "cost" or "price" in documents, and accompanying data dictionaries will help programs understand that those are similar concepts.
- Comparison of Avaya and ShoreTel Unified Communication Solutions
- Don't Get Stuck on Your Virtualization Journey: Where to Focus Next
- Strategy: Building and Enforcing an Endpoint Security Strategy
- 10 Emerging Threats Your Company May Not Know About
XML is increasingly used in business to describe a piece of data. About half of 375 IT managers InformationWeek Research surveyed earlier this year say they're using the markup language. But companies still are hammering out XML standards that define common terms for their industries. "As long as the XML standards are still immature, it's premature for the Semantic Web," says Alexander Linden, director of emerging trends and technology for Gartner. "It's the horse before the wagon. First we have to have an XML standard; then, we can build a Semantic Web."
The nonprofit organization leading the Semantic Web effort, the World Wide Web Consortium, is developing RDF, or the Resource Description Framework, as a middle layer in the Semantic Web, where it will define ontologies, the relationships between objects. It's RDF that will help your computer understand that a "price" takes the form of "dollars" or that "Hawaii" is part of the "USA." Even after that, the consortium will need to create yet another language that will allow the expression of logical concepts and let people ask questions of semantically tagged data.
The consortium has gotten the effort off to a good start on these standards, but businesses soon will need to become part of the process, says Prabhakar Raghavan, chief technology officer at knowledge-management software maker Verity Inc. "For someone like the W3C, this is too big a bite," he says. Raghavan suggests that industry groups will need to step in and define ontologies that relate to their businesses. The chemical, mining, and software industries all have relationships that outsiders will overlook. "The standards out of the W3C seem to have solidified significantly, but the sticking point is the ontology aspect," agrees Ron Kolb, director of technology strategy for infrastructure software vendor Autonomy Corp.
These obstacles explain why computers aren't likely to talk via a Semantic Web for five or six years.
Return to main story, "The Next Web."