XSLT 2.0 Programmer's Reference (Programmer to Programmer)

Rating:
Summary: 5 Stars to Kay, 0 Stars to Wrox
Review: As with the previous editions, Michael Kay has written a book whose content is of the highest quality. Being the editor of the W3C XSLT and XPath recommendations and the developer of one of the only XSLT 2.0 implementations in town makes Kay the ultimate authority on XSLT. He also writes in a style that is accessible to developers of almost any level (although this is not ideal as a beginner's book).

My big beef with the book is likely not Kay's fault. Being an author myself, I know how stubborn and pig headed publishers can be about their "style guidelines". Well, Wrox, your guidelines stink because this book is virtually impossible to use as a reference. Your font usage makes information impossibly hard to find by flipping pages. Your use of page headings is lame and unhelpful to the developer needing to find info fast.

In the end I have to recommend this book to XSLT 1.0 developers that need to get up to speed fast on XSLT 2.0 but it is too bad most of the profits are going to Wrox and not Kay.


Rating:
Summary: Can't lookup things; It ain't a reference
Review: C'mon, if you can't lookup information in a reference book, then it is not a reference. Thus, it is not any good. Don't give something 3 stars if it is a failure!
This book is practically unusable as a reference book. Its predecessor, covering version 1.0, has great page labels and CAN be used as a reference. Of course, the index still stinks and thus it deserves your 3 stars.
I bought this book expecting the same page labelling as the 1.0 reference. It is criminal how poorly labelled the pages of this 2.0 book are.
Also, this 2.0 book requires the XPATH 2.0 book, also called a "reference" book and fails similarly.
Basically, do not buy Kay's 2.0 books.

Rating:
Summary: Difficult reference to use
Review: I have three main problems using this book for the last few weeks on my first serious attempt to use xslt on a non-trivial problem.

The first is the most minor and is that the graphics are really poor. This is true for almost every one in the document, but if you have a chance to look at the book, check out page 56 for example. They are not what I would consider production-quality graphics. The extreme amount of aliasing makes the small font used in these diagrams almost unreadable. I don't understand how this could have been judged acceptable by the publishers.

Second is that it is very hard to find things in this book. Chapter 5 is the alphabetical reference for the xslt elements. The header at the top of the page does not list the element that is being described on the page. Also the font and style for the element headings are no different for their subheadings. This means there is no easy way to navigate this 300+ page section by flipping through it to find what you need. Whenever you want to find something, you have to go to the index first...

which is the second problem with trying to find things. The index itself is poor. When I first got the book, I read it from cover to cover (except for chapters 5 and 7 which are alphabetical reference sections). As I have been trying to use this book as a reference, I remember paragraphs or tables that I want to look at again, but I can't find them from the index. For example, I knew there was a table somewhere that listed all the different axes and I wanted to find it to get the exact name of an axis I wanted to use. The word "axis" (or "axes") is not in the index at all.

I am using this book daily and am finding myself frustrated every day with similar problems trying to find something that I know is in the book, but can't get to directly from the index. More than once I have resorted to flipping page-by-page through the book to find what I am looking for. At nearly 900 pages, it gets old really fast.

My third problem with the book stems from his statement in the introduction that "Since XSLT 2.0 has such a strong dependence on XPath 2.0, you really need both books..." where he is referring to his XPath book. He is not joking when he says that. It doesn't say that explicitly on the cover or on the web page description. But you can't go far in the book without finding a statement that what you are looking for is explained in a chapter of his XPath book. Maybe that is more true for me, as a newcomer to both, but it is different from his previous edition and something you should be aware of.

On the positive side, the book is comprehensive (within the bounds of its purposeful exclusion of XPath). I do not recommend it as an introductory text. The introduction acknowledges it is not meant as a tutorial. From my experience, it is downright unfriendly to xslt newbies.

I was able to get "The XSL Companion" by Neil Bradley from my library and found it to explain things better. I have read good things about Jeni Tennison's books and am waiting to check out her upcoming "Beginning XSLT 2.0".


Rating:
Summary: GREAT For Experienced Readers, TERRIBLE For Beginners
Review: It is difficult to give a numeric rating to this book because it depends on what kind of reader you are:
- If you have done little or no XSLT, and you want a book to efficiently explain how to start doing XSLT this is a TERRIBLE choice.
- If you are a beginner who wants to know every tiny detail of XSLT and has plenty of time to learn it, then this is a good choice.
- If you've already read an XSLT book, you are already comfortable with XSLT, and now you want to learn all the extra details the other authors thought wasn't important enough to include, then this is a GREAT choice.

This book almost reads like a specifiation. Although to be fair, I've read some specifications, like the EJB specification, that are more focused than this book. Any good trainer or training author knows that you have to organize your material to first put the emphasis on the central introductory concepts. Once your audience understands the basics, then you can build upon that foundation to explain the advanced topics. Along the way, you should always put the main focus on the most important topics, and just briefly mention extraneous details. Unfortunatley, this book does not organize the material for learning and covers everything with approximately the same emphasis. As just one example of this, Chapter 1 spends eight LONG pages on the history of XSLT including details like when so-and-so joined the specification team or presented a paper at a conference. What Chapter 1 does not do is give you any idea of how to write an XSLT sheet. I plodded my way through the first two chapters wondering when we'd get past all the gory details to a description of how to write an introductory XSLT sheet. Finally, I had enough and looked through the book trying to find how far I should jump ahead to find the introductory section, and realized it didn't exist. All the basics are interspersed with endless details throughout the book.
To be fair, the book calls itself a "programmer's reference." So one could argue that it shouldn't be designed to learn XSLT. However, trying to use this book as a reference would be equally probelmatic because its too hard to find the important information among all the extraneous details.
So if you already know XSLT well and want to know all the extra details, I truly do highly recommend this book. But if you want to learn XSLT in a resonable amount of time, I strongly recommend against this book.

Rating:
Summary: No other book on XSLT is as good
Review: There is no better book on XSLT. No matter if you want to learn XSLT from scratch, or have a long-term reference for the XSLT syntax. This book is the bible for XSLT users. The original book covered the 1.0 standard. This book extends that to support both 1.0 and 2.0 in the same book.

There are two large pieces to the book. The first piece covers the basics of XSLT, with a tag-by-tag reference to the language, as well as XPath in the same manner. Both of these start with introductions that make it easy for novices to pick up the language. The second piece covers design patterns for templates then brings the whole work together into case study sections.

XSLT is an invaluable tool for anyone who works with XML. Once you understand it you will never want to be without it. This book is the best way to learn XSLT, and the best reference for those who already know XSLT. Period.

Rating:
Summary: For the wannabe guru.
Review: This book aims to provide complete coverage rather than a tutorial and the author explains in both detail and depth. A chapter on stylesheet design patterns is followed by 3 comprehensive case studies that explore 3 different types of problems and their solutions. The book is written in a clear and comprehesive style. The differences between 2.0 and 1.0 specifications are also also discussed.

The code examples and solutions to the use cases are downloadable from the wrox website. This book is destined to be at the elbow of most XLST programmer and is definitely worth owning.