xmlhack: New book on XML and databases

New for 2001 is Professional XML Databases from Wrox.

A big (1000+ page) book, Professional XML Databases (Kevin Williams and nine co-authors) attempts to provide developers with comprehensive information on "how to integrate XML into their current relational data source strategies." Overall, it seems to do a pretty good job of fulfilling that goal -- especially in the chapters that don't rely on details about specific applications.

No part divisions other than chapters are provided in the table of contents or in the book itself. But if you happen to read the book's Introduction, you'll discover that its 21 chapters are actually grouped into five parts: Design Techniques, Technologies, Data Access, Common Tasks, and Case Studies. Also included are five appendixes: primers on XML and on relational databases, references to W3C XML Schema datatypes and SAX, and instructions on setting up a virtual directory for SQL Server 2000.

The best and most broadly useful chapters in the book are:

all four chapters in the 140-page Design Techniques section, which provide conceptual and how-to information on topics ranging from representing data in XML to migrating existing databases to XML to creating database structures for existing XML to broad design guidelines. Most readers should find this first part of the book useful regardless of any specific implementation they may have in mind.

the chapters on W3C XML Schema ^[1] and the XML Query language in the Technologies section (though the chapter on the XML Query language is understandably not completely up to date). The chapter that closes out the Technologies section -- on moving data between flat files and XML -- is also good, though some readers might not be thrilled with the author's choice of VBScript as a language for the code examples.
all three chapters in the Common Tasks section, which treat data warehousing, data transmission, and marshalling and presentation

Most of the other chapters either treat various W3C technologies that are already covered better and in more detail in other books, or discuss implementation details in terms of proprietary technologies (for example, ADO and SQL Server 2000) -- with two exceptions: a chapter that discusses implementation of platform-neutral, device-independent access to data using the JDBC API, and a chapter on the open-source DB Prism framework for dynamically generating XML from a database.

Overall, because of the book's focus on Microsoft technology in the Data Access chapters, it's probably most useful to developers who are using that technology. However, other readers may find it worthwhile for the chapters mentioned above, which should be useful regardless of implementation plans. Developers looking for an open-source approach to some of the same material may also want to take a look at Liam Quin's Open Source XML Database Toolkit.

[1] The W3C XML Schema chapter not only it provides better W3C XML Schema overview information than you'll find in most general XML books, it also does a good job of relating W3C XML Schema to typical database schema. And it includes a note acknowledging that in addition to W3C XML Schema, "a number of other proposed schema languages have arisen." Unfortunately, instead of discussing independently developed schema languages such as Schematron or RELAX, the writer mentions only Microsoft's XDR:

...Microsoft took the initiative to implement schemas with their XML parser before Schemas became a recommendation... XDR schemas have allowed people to work on projects that they would have been unable to otherwise had they not implemented them, although Microsoft is also committed to supporting XML Schemas when the W3C release them.

A book of this kind doesn't seem like the appropriate forum for a writer (even if he happens to work for Microsoft) to highlight Microsoft's "initiative" or official commitment to anything.