The Semantic Web and Your Intranet
Paula Gregorowicz
SemanticWeb.com Contributor
Much like in the music industry where there are so-called “overnight sensations” that toiled for years to create their success, web technology seems to sprout the “next big thing” out of nowhere. At least that is how it feels to me with the semantic web.
What is the semantic web, you say? According to Wikipedia, the semantic web is an evolving extension of the World Wide Web in which web content can be expressed not only in natural language, but also in a format that can be read and used by software agents, thus permitting them to find, share, and integrate information more easily. It is a vision of information that is understandable by computers, so that they can perform more of the tedious work involved in finding, sharing, and combining information on the web. The semantic web derives from Tim Berners-Lee’s vision of the Web as a universal medium for data, information, and knowledge exchange.
What does that really mean?
There are basically two types of information — data stored in some sort of database and information stored in unstructured ways (think — desktop applications, audio, video, etc.). Currently web applications can query databases effectively to bring back information stored in databases. If the query matches the data, viola, you have the right results. Unstructured data is more difficult for machines. Sure we have a variety of search technologies that help to manage that informational space, but anyone who follows the developments in the search game knows that there remain many challenges to a strictly machine-based approach. (After all, if it were simple and accurate for machines, why would a company like Mahalo www.mahalo.com even get started?)
With the semantic web, in theory, technology will be able to bridge the gap between all forms of data. The secret is in the metadata and the namespaces that tell the machines what the markup in another document really means.
What technology is involved?
The foundational pieces for the semantic web include URI’s (which identify resources) along with XML and namespaces. These drive the engine behind locating, marking up, and interpreting information so that machines can read and understand the data.
The OWL Web Ontology Language provides the construct for defining and instantiating web data models that need to be read and presented to machines rather than humans. What that means is that it provides the vocabulary for describing objects and how they relate.
Resource Description Framework (RDF) provides the metadata model which is a major component of the semantic web. Think of RDF as an XML-based language to describe resources. Using a subject — predicate — object format (called triples) it is used to express relationships between objects. Since the semantic web is all about defining relationships between disparate objects that can be read and interpreted by machines and presented to humans in a meaningful way you can see how RDF will drive the success (or lack thereof) of this new web.
What is the business case for semantic web?
In the article “The Business Case for Semantic Web,” the author talks about how organizations have tons of information but often don’t know how to find it, interpret it, or make meaningful connections with other, related information. This is not news as anyone involved with the web and with complex corporate intranets knows how challenging it can be to find something you need.
The goal of semantic web is to help solve these challenges. By providing a standard way to markup various forms of information, it should not only be able to put more of the right information at your fingertips but also tag that information in such a way as to aid in decision-making. Whether you talk about customer relationship management, financial analysis, or business strategy, the possibilities for the semantic web could be endless if it can deliver the desire results.
What are the obstacles to widespread success?
Like any common standard in technology, what you get is only as good as what gets fed into the system. Garbage in/garbage out stands the test of time and holds true once again. Only information and systems adhering to semantic standards will reap the benefits of the technology. As I see it, the only way benefits can truly be achieved (especially for the data the average business user doesn’t even knows exists) is through widespread adoption both internal to an organization and across the World Wide Web. That is quite a hurdle to overcome. We managed to clear the hurdle when it came to moving from a text-based Internet to the graphical web we know today so, nothing is impossible.
The technology to support the semantic web is evolving slowing but surely. Over the last several years we’ve seen a shift in the industry from proprietary standards and extensions to common standards and a more open source approach. This bodes well for further adoption of XML and RDF.
Another key consideration is the manual effort involved to optimize information for the semantic web. If an employee has knowledge of something in their head, unless it is stored properly in a system, that information walks out the door when the employee leaves. In this day of tight deadlines and overworked employees, what is the motivation and reward for people taking the time to digitize their knowledge? What business processes need to be built around the semantic web to make it truly useful and complete?
Metadata and a common vocabulary are the engines that truly drive semantic web. Anyone who has ever worked with metadata or attempted to create a lexicon for document management knows that there are people-intensive steps that are necessary to make the machines do their magic. How can you make that happen within your organization?
What’s next?
Much of the practicality of the semantic web is still in its infancy. Since it is built on open standards, however, the best thing you can do is to become knowledgeable of what the semantic web entails so you can plan and prepare as appropriate for your organization. Consult some of the following resources to get started:
The 
Eric Franzon
VP Community
Jennifer Zaino
Contributor
Angela Guess Contributor
semanticweb.com Twitter feed loading...
