Browse Books

Reviewer: Timothy C. Bell

Give me fruitful error any time, full of seeds, bursting with its own corrections. You can keep your sterile truth for yourself. (Vilfredo Pareto, 1848–1923) Salomon uses Pareto's words to conclude an appendix on error correction, but they could well have been used as a motto for this book. This “complete reference” on data compression provides a wonderful treasure chest of information, spanning a wide range of data compression methods, from simple text compression methods to the use of wavelets for image compression. Although it contains some vague generalizations and unfounded claims, it also provides tremendous detail, such as the full details of several file formats. It is unusual for a text on compression to cover the field so completely, with detailed descriptions of lossy and lossless compression methods for text, images, and sound. It includes methods that have only become widely known in recent years. However, the book does not quite live up to its title, as a few important methods get little or no mention, most notably vector quantization, the PNG format, the MPEG standard, and some audio compression methods. There are also some unusual definitions of terms that are not used elsewhere in the compression literature. The book's most frustrating deficiency is the lack of performance comparisons among the many methods described, although the preface includes references to Web pages that provide performance comparisons. The presentation is quirky, full of quotations (some of them relevant) and curious asides. The chatty style will make the text more accessible for some readers, while others may find the diversions distracting. For example, on page 86, the author suddenly interrupts a description of a compression method to give details of six examples of lipograms, such as books that entirely exclude the letter “e.” Nevertheless, the explanations are clear, and numerous examples illustrate the methods described. The book is suitable for an audience with a general computing background. Several appendices provide details about related topics that the reader may need, such as Fourier transforms and matrix operations. The text is interspersed with exercises, and a full set of answers is supplied. The book has been printed directly from the author's \TeX files, and while it is of a noticeably higher standard than many camera-ready books, it lacks the polish that conventional publishing can bring to a project. The text does not appear to have been copyedited, and the structure is a little unusual. For example, the bulk of the preface consists of introductory material that one might expect to find in the first chapter. Also, there are sudden changes of topic within each of the five chapters. For example, the chapter on dictionary compression methods has a section on cyclic redundancy codes, which really belongs in the appendix on error correcting codes. The index is comprehensive, though finding some topics may take several attempts. The glossary and bibliography are useful. Despite the unusual presentation, I highly recommend this book as a reference on data compression. It would also serve as a good starting point for information on a particular data compression method, and it is a very readable introduction to a number of topics relating to compression.