Written by some of the most respected researchers in implicit surfaces (many of the authors having invented many of the algorithms discussed), this book is exactly what the title says it is: an introduction to implicits.
Since each chapter was written by a different author, the book is rather inconsistent in its laying of the material, which will indeed make it confusing to someone that is indeed a complete newbie to the subject. This is particularly bad for a book that is supposed to be an introduction to the subject. It is hard to explain how, for example, Chapter 3 (implicit patch modelling) will relate to anything else covered in other chapters. Being a chapter so early in the book, it just confuses things.
If you are already familiar with blobs or similar implicits, you will be right at home and will be able to jump to chapters you are interested in. If you've never been able to play with an implicit surface modeler, trying to read the book from cover to cover and understand the explanations, even of the first chapter, will, I think, prove somewhat hard. You will likely find better introductions to "blobbies" if that's what you are interested in on the web.
Chapters 4 and 5 are some of the most useful and practical to anyone doing any implicit software development for the first time. Bloomenthal gives a good review of all the ways of polygonizing implicits (albeit no consideration is given to taking advantages of polygonizing specific types of fields, such as point elements) while Wyvill gives also a good review on the different approaches on raytracing implicits. Both chapters do a reasonable job of pointing the benefits and drawbacks of each method presented.
Chapters 6 and 7 deal mainly with subtle issues of blending of multiple skeleton implicits. Chapter 8 mainly with morphing. And the final chapter with dynamics applied to implicits (so as to create soft objects).
It is, however, the Reference section that is one of the most important sections in the book, since it pretty much lists most if not all papers related to implicits.
Albeit the book states that it wants to be a practical book on implicit surfaces, no sample code is provided anywhere (the book is more a presentation of the material, somewhat math oriented, with discussion of the most useful and common equations for each chapter's topic) and even the reference section does not point to some of the most widely known free code available ( Bloomenthal's Gems code or Wyvill's BlobTree ).
