- Edition
- ISBN 9780596100162
Reviews

If you've taken the time to read Tufte's Envisioning Information then I'm afraid you won't get much additional value from this. The author writes like an intelligent technician, thorough and wooden. The first 188 pages of the book are primarily a catalog of visual principles learned in a first year design class. It's not until the last few pages of the book, after having waded through a swamp of definitions and an exhaustive, exhausting cataloging of chart types, that the author broaches the subject of making your own dashboard. Disappointing. I give points for correctness and thoroughness of this book, but it's neither useful in practice nor engaging from a theoretical subject. The author took an interesting, vital subject and made it boring. The most valuable chapter was the list of 13 dashboard anti-patterns, which could have just as well been a blog post: (view spoiler)[ Thirteen Common Mistakes In Dashboard Design 1. Exceeding the boundaries of a single screen 2. Supplying inadequate contest for data 3. Displaying excessive detail or precision 4. Expressing measure indirectly 5. Choosing inappropriate display media 6. Introducing meaningless variety 7. Using poorly designed display media 8. Encoding quantitative data inaccurately 9. Arranging information poorly 10. Highlighting important information ineffectively or not at all 11. Cluttering the display with visual effects 12. Misusing or over-using color 13. Designing an unattractive visual display (hide spoiler)] I cannot recommend this book to anyone spending their own money. The web contains better resources for making your own dashboards which are more fun, engaging, and interactive than this very dry book.

While the underlying principles of effective data visual communication are well expressed and remain unchanged, the rest of this book—from 2006—is seriously dated. In this field, no text can go without a complete catalog of chart types, and complaints of why radar graphs are obscure and hard to read; this book is no exception. Advances in design systems and component libraries make many of the tactical recommendations moot. Interesting from an historical perspective, I suppose. Not recommended.



