Enterpriseleadership.org recently sat down with Dr. Hendler, a professor of computer and cognitive science, and the assistant dean for IT at Rennselaer Polytechnic Institute (RPI), to learn more about the Semantic Web. A pioneer with Berners Lee on the development of the Semantic Web, Dr. Hendler also serves as the RPI's constellation professor of the Tetherless World Research Constellation, a program to access information at any and place without being tied to a specific computer or device.

Developed by Tim Berners Lee, the World Wide Web, as it was first called, made it possible for us to get information, communicate, do business, and entertain ourselves via the Internet. Because the Web contains boundless amounts of information, we rely on search engines to find what we want across the Web and within specific Web sites. However, search engines cannot always accurately interpret what we seek. We wind up having to examine the results to see if we have a correct match. For example, a search on soap could produce everything from soap we wash with to soap operas to SOAP, which stands for Simple Object Access Protocol. A May 2001, article in Scientific American, called The Semantic Web, discussed an  innovative semantic technology or agents that would be able to distinguish the relevance between pieces of similar and unlike pieces of information.  So, if you put in Camay and Dove, you would get soap not people named Camay and Dove. The authors of that article included Berners Lee, Dr. James Hendler, and Ora Lassila.

EL. How does the Semantic Web differ from the Web we know today?

JH. The first Web was about documents and then pointing the documents at each other. Web 2.0 had added the social aspect. Humans can quickly generate things on the Web and share them with other people. Twitter is a good example. I can very quickly put something out there and many of my friends will see it. Facebook is the same sort of thing, but it is a little slower and designed for larger communities. The Semantic Web does those same types of things with the data in your life. For example, how do I, as a company, make my catalogue something that can get out there? My database and your database know they talk about the same things. How can we share our data?  How does a publisher make it clearer what defines an article? The Semantic Web is all about adding more information to the Web in a way that computers can better process that information, and use it to help people do a better job on the Web.

EL.  Can you give an example of how an organization you work with uses the Semantic Web?

JH.  You will not come to a Web site and see that you are immediately using the Semantic Web. The site using Semantic Web technology can do better things than traditional Web sites. For example, I am on the board of Bintro, an emerging company that does job matching. You want to look for a nanny position in New York City. Bintro might match you with someone looking for a childcare provider, not a nanny. By using the Semantic Web, Bintro understands something about the location, such as New York, and about nanny being a type of childcare provider. Bintro brings that type of information to the Web, and uses it in Web applications so you do not have to do many of these tasks through key word searches. The site will automatically use matching technology and other types of technologies to make the Web better.

Small companies you have never heard of supply the tools to other small companies, such as Bintro, that want to leverage the Semantic Web. These companies are trying to figure out how to take this new technology and make it available to people.

EL. Can you explain the role of ontology with the Semantic Web?

JH. Ontology is a term used in the semantic Web. It is a simple idea. If I have a database with the number 17 in it and if you have a database with the same number, we might want to know if they stand for the same attribute. Now if they both represent someone's age, then we know our databases are the same thing. On the other hand, if one database represents the data as age and the other database, as an interest, then the databases are talking the same language. To tell a computer they are both the same, I need some kind of structure that says there are things called people, that people have things called ages, and that people have things called addresses. Ontology defines how to develop that kind of vocabulary. The Web has many different levels of that. They range from simple to complex. The first generation of Semantic Web products had simple ontologies. If I know that you are a radio person or a journalist, then I know you must be a person. It does not sound very exciting. If I am looking for people's pages and I find some databases that say you are journalist, then I know you are a person. A publisher is one type of person and a journalist is another. Now we can share standards. It becomes a way for computers to see what you think in your terminology and then pull it to other people's computers.

Many companies now work on how to help people turn their data into this format by building tools for manipulating this format, and bringing the formatted data to the Web.

EL. Is CERN involved in the Semantic Web?

JH. Jim Berners Lee worked at CERN when he developed the World Wide Web. He still has some connection to CERN. It, of course, has an interest in putting data on the Web for collider projects that generate huge amounts data for large groups of diverse global scientists to analyze. CERN wants to learn how these people work together.  CERN has an interest in the Semantic Web, but it is not a key developer of the Semantic Web.

EL. Who are the key developers of the Semantic Web?

JH. Many standards organizations have been involved in creating standards for the Semantic Web. The funding for the early research 10 years ago came from the U.S. Defense Department and the European Union. Later research has come from work done by universities, and emerging companies. Large companies, such as Microsoft and google.com, see the Semantic Web in some of their operations. Oracle supports many of the Semantic Web standards directly. Some of the search sites use it.

EL. Does the Semantic Web have applications in certain industries such as pharmaceuticals?

JH. Most new technologies first find a foothold in some particular vertical area. Healthcare and life sciences were the first ones to realize the importance of the Semantic Web. Within their individual systems, they were doing some of it. Financial services companies also have an interest in the Semantic Web. Now we are seeing search engines, such as Google.com, getting interested in it.

EL. What specific applications for the Semantic Web do you see in some of these verticals?

JH. People are now looking at the Semantic Web in several ways. Data integration within the enterprise is one area. Many companies in vertical market segments have many different databases and want to pull that information together and provide it to people. Social networks enable us to create people talking together, but they cannot see the data, use the data, or change the data. For example, a pharmaceutical company has many different chemical databases and many different drug test databases, the Semantic Web could pull together all of the compounds that have certain properties.

Cross-enterprise data integration is another area for the Semantic Web. For example, a company wants to publish some of its data so that it is integrated with other people's data. The U.K. government and others are interested in transparency that can come about by publishing government data. They envision people building applications that will help citizens analyze that data and, in the process, derive some trust in the government. Citizens will be able to say that too much money is going into one place when it should be going to another place. To do that, you need to integrate information from all sorts of different government agencies, all of whom have data in different formats. If they want to expose it on the Web in a way that is integrated, that needs Semantic Web technologies. That is something is happening now in a big way.

Large-scale Web systems are adding functionalities, such as the semantic search engine, and the semantic match engine. Many publishers want to do this to allow them to better track things in new media.

Many different players have an interest in the Semantic Web for different reasons. The tools have started to become available. Some people refer to this as Web 3.0. Web 2.0 brought people together in a conversation. Now we are trying to bring people and machines together in those conversations in doing different things. Web 3.0 is very exciting for early adopters and large entrepreneurs who are trying to create the next Google.com.

EL. What is RPI's role in the Semantic Web?

JH. The research lab I run does some development work. For example, with the government datasets, we have been turning them into these Semantic Web forums and building demos to show people how easy it is to do integration in this new way. As a researcher, I have to be early into a technology, help make it happen, and then evangelize it. I have been doing the Semantic Web for a long time. In some ways, my lab is about figuring how we can start building on top of the Semantic Web to create the future of it. You might say my job is to create Web 4.0. We do much work with companies and government agencies that are trying to learn how to use these technologies in new ways.

EL. You have the nickname father of the Semantic Web?

JH. Tim Berners Lee is the father of the Semantic Web. Our article in Scientific American 2001 was the first use of the term in a widely read popular place. It was attributed to us as the originators. Many people had been working on the Semantic Web before us. I am certainly one of the people who helped to make it popular by making people understand the vision and how to apply it to the Web.

EL. What should CIOs know about the Semantic Web because it is going to affect the types of applications they build?

JH. There are several answers to that. With any new, potentially disruptive technology, the people who understand early what is coming and how to use it can provide much value to an organization. Nowadays when companies are just starting to figure out how to use enterprise social networks, such as twitter.com, CIOs and CEOs really need to track technology trends. Because the Semantic Web has passed the potential technology phase, CIOs and CEO need to understand how to use it in their enterprise.

Companies that depend on their Web presence need to consider ways to improve their visibility. Search companies have started to generate Web pages with certain information. As a result, when you do a search, you see an organized presentation of the information. Currently, when you Google your company's name, for example, you get a couple of random sentences that have you search words. It would be nice if you could say: 'If someone looks for my company name, I would like them to see the company name, logo, and location.' Right now, there are many ways of doing that, but there is no way you can give that information to Google to make sure it gets it right.

Large companies make these metadata deals. They tell the search engines how they want their information put out and displayed.  Now the search engines are opening that up to smaller companies, new companies and individuals through these Resource Description Framework front ends. That has generated much excitement. CIOs and CEOs must know about this.

EL. Could the semantic Web have an affect on Google.com?

JH. It will affect Google in a couple of different ways. In the past few months, Google has adopted some of the semantic Web standards. It enables people to do a better job of showing what they have and getting Google to display it. Google has an interest in new technology that has to do with a search engine. For a long time, Google said it was not interested in the Semantic Web because it did not see how broadly to apply it. Google now sees how apparent this is to do, and as a result, has become interested in the Semantic Web.

Elizabeth M. Ferrarini is a technology writer from Boston, MA. You can contact her at elizabethferrarini@yahoo.com.