Oracle in a Nutshell: Concise, Complete, and Consistentby Rick Greenwald, coauthor of Oracle in a Nutshell
Editor's note: O'Reilly's recent release of Oracle in a Nutshell brings order to the chaos of Oracle information by pulling together the most essential information on Oracle architecture, syntax, and user interfaces. In this article, coauthor Rick Greenwald talks about what it was like to write such an immense volume, and the methods he and coauthor David Kreines used to sift through the 13,000 pages of documentation to find all the nuggets Oracle users really need to know.
Oracle in a Nutshell: a walloping 913 pages. Are they nuts? Those might have been your first thoughts on seeing this new O'Reilly Oracle title. And, frankly, there were times during the creation of this book when we doubted our own common sense. But I think you'll see that in the final product we were able to achieve our goals and provide significant value.
When we set out to write this book, we decided to shoot for three goals: conciseness, completeness, and consistency. Our first goal was simple -- to create a concise reference that would cover all types of configurations and uses for the Oracle database. Typically, a user would have to look in several places and piece together from various sources the information they'd need to perform a specific task. We wanted to reduce the amount of material they would have to span; we wanted to give 90 percent of the useful information in 10 percent of the space. It may seem contradictory to say that a 913-page book is concise, but because Oracle in a Nutshell corresponds to over 13,000 pages of documentation, we think we hit this first target.
The second and third goals were linked. We obviously could not cover everything in those 13,000 pages without some type of compromise. We decided that one area that would not be open to compromise would be the completeness of the volume. We would strive to provide a complete span of syntax documentation in all relevant areas. This meant that we would cover, for instance, all the built-in PL/SQL packages, even though some were much more popular than others.
We would instead try to deal with each topic on a consistent level. With the occasional exception of a few sections where general concepts were explained, we would deal with each piece of syntax in a the same manner -- identification of grammar and keywords, coupled with brief descriptions of the purpose of the syntax. We obviously would not be able to provide much in the way of examples. To even attempt to do a comprehensive job at this task would double the size of the book, negating the first goal. And this book was supposed to be a reference work, not a tutorial. There are numerous other titles that cover the areas included in this Nutshell in more depth, but none that provide the scope of coverage offered by this book.
These goals laid the groundwork for the book, but we also wanted to provide some additional value. One way we did this was by the organization of the book and its specific chapters. We figured that Oracle users typically had a problem area, rather than a problem parameter or statement. We grouped related syntax together, which reduced the number of pages that a user would have to review to find the appropriate information. Additionally, this type of organization would help users to quickly understand the use of interrelated statements or parameters; it could even help users to recognize other related syntax that might also have a bearing on their particular problem.
And for those readers who might want a more direct route to a particular entry, we also included alphabetical listings of all parameters, statements, and functions in the index.
The final decision we made took another step toward reducing the number of pages a user would have to consult at any one time. One of the reasons standard documentation is so large is the need to document every clause for each statement that uses the clause. This is far from a normalized approach to documentation. We decided to group common clauses and keywords at the start of each chapter, with simple references to each common element in the syntax definition for each command. I believe this decision has created two benefits. First, it reduces the amount of text for any single command. Second, it helps to address the needs of beginning, intermediate, and advanced users. A novice user might want to learn the most commonly used clauses right away, at the start of the chapter. An intermediate user would be familiar with the common clauses, but might be able to learn new variations on them with a quick look at the common clause section. An advanced user, through experience, would already know all about the common clauses; all they would want to know is when one of these clauses can be used with a command.
We hope that the decisions we made in creating Oracle in a Nutshell provide value for our readers. Of course, we know that the task of creating worthwhile books about a large, complex product like the Oracle database is a never-ending task. Although we devoted a tremendous amount of effort to this book, we believe we can improve the book in future editions. We welcome feedback from our readers because we are always seeking ways to help our audience obtain the greatest value from our books. Please feel free to send us your comments via the Talkback link at the end of this article.
O'Reilly & Associates recently released (December 2002) Oracle in a Nutshell.
Sample Chapter 17, Performance, is available free online.
For more information, or to order the book, click here.
Rick Greenwald has been active in the field of client/server computing for more than 20 years. He has been a principal author on 10 books and has written countless articles.
Return to the O'Reilly Network.