Semantic Web for the Working Ontologist Effective Modeling in RDFS and OWL is a new book authored by Dean Allemang and Jim Hendler. I can offer a quick summary by teasing apart the title. For starters, the Semantic Web it discusses is generally in line with the current consensus view of the developer community, though with a lean in the direction of the ‘O’ word. The emphasis in the book is very much on Working and Modeling. It is practically oriented, and while it covers most of the technologies associated with the Semantic Web, its focus is on how to describe things using RDF, RDFS and OWL.
There’s a serious shortage of approachable books in the Semantic Web space – if you check the ESW Wiki list, there are only a handful that aren’t heavy duty academic works. Aside from the issue of convincing publishers there’s a market for such material (a problem that’s no doubt evaporating), there’s the difficult problem of what to write about. In the 2003 book Practical RDF, Shelley Powers used the parable of the Blind Men and an Elephant to suggest how RDF has many different aspects and can mean different things to different people – and RDF is just one Semantic Web technology (though arguably the most important). What’s more the elephant changes over time and is lavishly decorated: while the core standards solidified in 2004, since then we’ve seen various auxiliary specifications come along: the SPARQL query language, Turtle/N3 syntax, RDFa, GRDDL and so on. Ideas on best practices have also developed considerably over the years. This book is scoped to modeling with RDF, RDFS and OWL, and covers that ground admirably.
Allemang and Hendler are known experts, well-versed in the subject matter, but what’s more they have spent considerable time teaching courses on the Semantic Web, and this experience shows. The writing is clear and the book’s full of well-illustrated examples, along with a short but very handy FAQ at the end. The practical side is hinted at in their decision to devote significant space to the SKOS Simple Knowledge Organization System and FOAF Friend of a Friend vocabularies. The syntax used throughout is N3/Turtle, which makes a refreshing change from the eyestrain of RDF/XML.
There aren’t any programming (as in running code) examples, and the coverage of things like HTTP and the use of these technologies on the Web is really confined to illustrated prose. I must admit I was disappointed by the limited coverage of SPARQL, I do think this has relevance to modeling decisions. Given the rise of Linked Data in the wild, I would also have expected maybe a chapter devoted specifically to this approach (the ideas are all there in the text, but they don’t jump out).
On the other hand the coverage of reasoning with Semantic Web languages is excellent, material that can be very hard to get a handle on is here presented in an easily digestable form. Similarly the fundamental theory is explained in simple terms without recourse to arcane notation, and common misconceptions around the Semantic Web are disposed of without malice.
Contents
- What is the Semantic Web?
- Semantic Modeling
- RDF – the Basis of the Semantic Web
- Semantic Web Application Architecture
- RDF and Inferencing
- RDF Schema Language
- RDFS-Plus
- Using RDFS-Plus in the Wild
- Basic OWL
- Counting and Sets in OWL
- Using OWL in the Wild
- Good and Bad Modeling Practices
- OWL Levels and Logic
- Conclusions
- Frequently Asked Questions
RDFS-Plus is RDFS with the addition of some handy bits of OWL (IFPs etc).
In conclusion, this is an approachable book for anyone with interest in the field, and gives excellent coverage of the Semantic side of the Semantic Web, as it pertains to modeling the real world. With the caveat that this is the scope of this book, I’d personally strongly recommend it. I do intend to read this book cover to cover thoroughly, it is insightful writing, and as an occasional OWL user I’ll be keeping it on hand for the recipes.
See also: Henry’s [p]review
