Summer Reading for the Working Ontologist

By SemanticWeb.com on July 14, 2008 5:25 PM

Jennifer Zaino
SemanticWeb.com Contributor

This summer saw the publication of Semantic Web for the Working Ontologist, co-written by Jim Hendler, who is the Tetherless World Senior Constellation Chair at Rensselaer Polytechnic Institute and associate director of the Web Science Research Initiative at MIT, and Dean Allemang, chief scientist at semantic web technology company TopQuadrant Inc. Hendler recently responded by email to some questions Semanticweb.com about the new book, what readers can expect from it, and what it indicates about the maturing Semantic Web.

Semanticweb.com: What will this book provide its readers that they can’t get from reading up on semantic web specs or picking up one of the academic works in this area? That is, in what ways do you think it will serve as a more practical handbook to readers?

Hendler: This book is full of examples based on real-world use cases and users’ needs. We even include a FAQ that is based on what users are trying to do, rather than the language features used to solve them. The book also shows how to combine the features usefully. Thus, it differs from the references in helping users find the answers to their own specific problems, and it differs from the more academic tomes in that it focuses on using the language, not the underlying theories thereof.

Semanticweb.com: You talk in the book about creating semantic models or ontologies that are not only understandable and durable but can “perhaps even be beautiful.” Beautiful how?

Hendler: There are many different ways that the same objectives can be reached in any complex language (think about all the different programs you could write for even a simple problem in a traditional programming language). When we see a well-written program with an elegant solution, serious programmers think “beautiful!” We mean it in the same sense–learning how to use RDF, RDF Schema, and OWL in ways that are really helpful, can make it so those who desire to use someone else’s ontology can understand it more clearly and intuitively.

Semanticweb.com: Reading some of the reviews of the book, some people were thrilled that XML/RDF is not mentioned once. Why leave it out, and why do you think people are so happy that you did?

Hendler: There is often confusion between the syntax of a language and its power or utility — us old AI guys (and I guess Dean and I both qualify) remember the years of trying to explain Lisp to people who couldn’t see past the parentheses. In this book, we focused on what people were trying to say, not on how they were formatting it. All of our examples use formats that can be easily turned into RDF/XML for machine use, but we felt we didn’t want to spend a lot of time explaining when to use “&” vs. “:” vs. a full URI–that’s what machines are for. Let them do the work! (We note that we are working on a web site that will have all the examples in RDF/XML for people who want to master this, or need to use “cut and paste” for their own documents).

Semanticweb.com: Equally, some readers wish there’d been more on SPARQL. Why did you choose to limit that?

Hendler: We started our conceptualization of the book before the working group that created SPARQL was formed, and we were in final edits before SPARQL became a recommendation. If we go to a second edition, you can be sure there will be a SPARQL chapter, and we are thinking about doing an online “appendix” about SPARQL.

Semanticweb.com: How did your experiences teaching about the semantic web to students and professionals influence your approach to writing this book?

Hendler: It taught us that the people who will bring this technology to their companies may be neither experts in AI nor experts on Web architectures and languages. Rather, they are people who know some field well, and are trying to exploit this technology as it develops. The W3C working groups that created the Semantic Web languages, and continue to evolve them, are largely formed by a mix of Web and AI experts, and thus the documents being produced often missed the needs of the users who will make the Semantic Web happen. We wrote the book to try to meet exactly this need.

Semanticweb.com: What does it say to you about the maturation of the Semantic Web
that the time is right to publish a book for the “working ontologist”?

Hendler: Tim Berners-Lee once told me we’d know the Semantic Web was real when people stopped asking “Why?” and started asking “How?” I think this book helps prove we’ve reached that point.

Semanticweb.com: Similarly, what does it say to you about the maturation of the semantic
web that you were able to get a publisher interested in publishing a book on this topic?

Hendler: More importantly, we did not have to “twist arms” to find a publisher, but rather had a number of publishers interested. I think it is clear that the field is still in its early stages, but this is yet one more sign that it is on its way to maturing into an important technology area.