I'll be honest, I wasn't attracted to this book by its title, but there aren't all that many books around on the semantic web that are also recent enough for me to be confident that they are up to date.
I found this somewhat like A Brief History of Time in that the first half is reasonably straightforward, but somewhere in the second half there's a step-change and the text drives off into another world. So the chapters on RDF, SPARQL, RDFS and RDFS-Plus and basic inferencing were clear and informative, but whilst the discussion of OWL makes sense at one level, it becomes rather unclear how to apply the concepts. The chapter on anti-patterns sounds like a good idea, but unfortunately the solutions are rather weak or even self-acknowledgingly non-existent (effectively saying "this isn't the way the Semantic Web works"). I think this is a limitation of the authors' philosophical approach: it doesn't necessarily apply (and certainly not in the same way) when a system is the owner of the data (and a lot of Semantic Web data is owned by some system), rather than a client of some cloud database. There are some good ideas and clear explanations in the book, but I do think it's let down by some of the later chapters.