THE DATA ADMINISTRATION NEWSLETTER – TDAN.com
ROBERT S. SEINER – PUBLISHER

Authors: C.J. Date and Hugh Darwen
Addison-Wesley, 1998
ISBN: 0-201-30978-5

I could be described as an avid reader. I have dog books, car books, books on roses, gardening, planes, trains, history, science fiction, mysteries, action novels... Every room in my house has books!

I have amassed quite a collection of books on database technology over the years. None of them are as intense as the Foundation for Object/Relational Databases, The Third Manifesto by C.J.Date and Hugh Darwen.

This book is written for those interested in a firm understanding in the underlying foundation of relational and object database technology. It focuses on what the authors believe to be the logically correct approach to integrating relational and object databases.

Although the authors do concede that object technology will make important contributions to the database management field, they do not believe it will ever replace relational database technology. Date and Darwen feel that the relational model is and will be highly relevant to database theory and practice. But they do not believe that SQL should be the language used to access data stored in relational databases.

You did not incorrectly interpret the previous sentence. Date and Darwen feel that most of the limitations of current relational technology can be attributed to the use of the SQL language and the incorrect implementation of relational databases by database vendors. They believe that a new language could (and should) be created to overcome the limitations inherent to SQL. The authors create a mythical language, called D, that serves as the basis for examples used to reinforce their proposals contained in the book. They dedicate an entire chapter to comparing the new language to SQL3. Another chapter focuses on comparing their proposals to those advocated by the Object Database Management Group (ODBMG).

The book is divided into four major parts: Preliminaries, Formal Specifications, Informal Discussions and Explanations and Subtyping and Inheritance. There are also several appendixes that provide various discussions to reinforce the author’s proposals.

The authors concede that the book is not an "easy read." This is not a book that allows readers to skip a chapter and then return. The authors provide advice on how to read each chapter and it is in the reader’s best interest to follow it.

I am usually quite opinionated when I discuss database technology. In this case, I have decided to let the readers decide for themselves if they agree with the proposals put forth by Date and Darwen in The Foundation for Object/Relational Databases, The Third Manifesto. I will say that the authors explain the theories behind relational and object databases extremely well. Although the book’s intention is to convince readers that the relational model is superior to object and obji-relational hybrids, readers benefit from the author’s extensive background in relational and object database theory.

If you are looking for a book that helps you tune a database or construct a data model, this is not the book for you. What this book will do is provide readers with a firm understanding of the theoretical foundations of relational and object databases. I strongly recommend it for those seeking a deeper knowledge in relational and object theory.