This book disturbs me. No, it is not the book itself so much as the need for it. Let me quote a couple of sentences from the preface:

The Java Developers Almanac is like a map of the Java class libraries. It's a compact and portable tool that covers almost all of the libraries, if only from a bird's-eye view. It's great for reminding you of things like method names and parameters. With today's class count at 3000, you're bound to forget a few details now and again.

There is a footnote that explains that 'almost all':

Due to size constraints, the javax.swing.plaf.* packages are left out of volume 2.

As each volume is already around the 1000 page mark, that is perhaps understandable. However I am not sure that compact and portable is an appropriate description of a two-volume work that spans over 2000 pages.

Note that I am not complaining about these books, they do indeed meet a real need in the Java community. I am expressing concern that Java has reached the stage when 2000 pages considered necessary even though it acts as no more than a reminder and, by the authors own admission, you will need more books if you are using a class that is new to you. Worse still, you will probably need to replace these books in a couple of years. So much for the claim made by some that Java is free.

If you are a serious programmer in Java, you already know that you need this book (or one like it). The alternative is to spend more of your time re-inventing classes that already exist or delving away for information about classes that you vaguely remember.

This is a well planned but weighty tome (well I have only got volume 1 on my desk). The examples are just about enough to help your memory when dealing with classes that you have already studied.