I thought it safe to buy any O'Reilly book sight unseen, until I bought this one. It's padded by a factor of 4 or 5. (For example, instead of explaining once that it's unwise to use features deprecated by the HTML standard and then simply marking deprecated features with a symbol, the authors tell you over and over what "deprecate" means: such-and-such a tag is still popular, still supported by the popular browsers, but will go away some day, yada yada. They waste an entire paragraph on this as late as page 439.)

The book is incomplete (for example, the description of the MIME multipart/form-data type doesn't tell you how to get the boundary from the initial "Content-Type" line). It's disorganized ("pixel" is defined in a footnote on page 119, in connection with the horizontal-rule tag, which is hardly the first place one would look for a definition of a fundamental term) and poorly indexed ("pixel" doesn't appear in the index) so that it's hard to find anything.

Most disappointing, after you've read it, you will find yourself looking at moderately complicated web pages and realizing that you still have no idea how the authors managed to accomplish them using HTML. One problem is that the most complicated examples in this book are simple to the point of being crude (the notion of using tables to align form elements is considered by the authors to be a "trick", and doesn't appear till page 481, just 8 pages before the end of the text proper and the beginning of the appendices). Another problem is that the authors don't describe any real tricks (e.g., emulating the nonexistent vertical rule tag with a tall narrow table cell whose background color is set to black). And part of the problem is that real-world pages use Javascript, a description of which would have taken less space than all the needless repetition.