Visualize This is a book about designing visualizations for data ("graphs" more or less, although there are visualizations which are not, strictly spe...moreVisualize This is a book about designing visualizations for data ("graphs" more or less, although there are visualizations which are not, strictly speaking, graphs). The focus of the book was not what I expected; given that the author is a graduate student in statistics, I expected the book to have more of a scientific focus. Instead, it is mostly focused on designing visualizations for websites and/or newspapers and magazines. While there can be a lot of overlap between these tasks and more directly scientific output, the focus made the book less useful and interesting to me than I had hoped it would be.
This is not to imply the book does not contain useful information, it certainly does. It introduces a decent variety of graph styles, many of which are likely unfamiliar to readers. It does discuss valuable resources of which the reader may be unaware, such as Inkscape (a free, open-source alternative to Adobe Illustrator) and 0to255.com (a website for helping choose color schemes). When it does slip into visualization theory the advice is solid and to the point (e.g., how to scale bubble plots so the area of the bubbles properly reflects relatives sizes).
Although I was clearly disappointed in the text and have many criticisms, in the end it has enough advantages to be a worthwhile read, particularly for those who are complete newcomers to visualization or who have a particular interest in data visualization for the web.(less)
Before going into the review itself, a comment on a slight oddity of the book (which will become important in the review): The copy I read is the 7th...moreBefore going into the review itself, a comment on a slight oddity of the book (which will become important in the review): The copy I read is the 7th printing (March 2011) of the second edition (originally published in 2001; the first edition was published in 1982). The reason I bring this up is a discrepancy not mentioned anywhere in/on the book or on any website I could find. At least one chapter has been rewritten (or added) since the second edition was originally published. Chapter 8 contains examples and data from 2003 and 2004, as well as ending with a data point from 2009! Generally, one would consider the rewriting of a chapter (rather than the correction of typos and other errors) to be, if not a new edition, at least worthy of mentioning in the description of the book, but no such mention has been made. Without a copy of an earlier printing of the same edition, I cannot comment on whether other changes are present or absent; this was the only chapter where I noticed data or examples that post-date the original second edition printing. --- This book by Edward Tufte is considered a classic in graphics design and its easy to see why. Filled with examples of both good and bad graphic design, he eloquently argues for rules of design that maximize information and "truth" to avoid misinformation and distraction. Truly a masterwork, it is not without flaws. He has a tendency toward hyperbole, declaring certain graphs to be the "worst ever made" or "best ever made". He accuses many graphical designers of deliberately lying to the consumer, when the reality in many (although certainly not all) cases is likely more a case of incompetence or ignorance. He states rules to follow with a certainty that are sometimes themselves not backed up with data, but rather simply reflect opinion.
And finally, some of his principles lack any real discussion or acknowledgement or context, something which stands out in particular with the recognition that most of the examples are rather dated. For example, he has a large discussion focused on the size of printed graphics, a discussion which completely ignores any context in which a graphic might be presented beyond the printed page. When the second edition was originally printed in 2001, most scientific presentations were still using overheads or physical slides; the widespread use of Powerpoint for presentations did not take off until a few years after the book was published. However, the chapter with this discussion is the one mentioned above that has clearly been updated since 2009! This enhances the dated feeling of some of the discussion, making one wonder if there is a bit of statistical and graphic Luddite influence to the writing.
I should say that he does mention Powerpoint and Excel at one point with clear disgust, and I am the first to agree that using either for designing *printed* graphics is a very poor choice. However, for for visual presentation in a talk or lecture, Powerpoint is better than many of its competitors. Unfortunately, some of his design discussion simply doesn't translate to non-printed publication, including presentations and graphics to be found on the web, which may have very different design considerations, not the least of which is the potential for consumer interactivity. (He has written additional books, some of which I plan on reading, and some of which may get into these issues, although I suspect not).
Despite all of these flaws, I truly believe the book is incredibly well done and its influence since it was originally published cannot be understated. Scattered throughout the book, often (although not always) recapped at the end of chapters, are "rules" of design that are so striking in their statement, I plan on collecting most of them onto a single piece of paper to hang on the wall by my desk as a visual reminder of what to do and think about when designing my own graphics going forward. That, more than anything else, illustrates my feelings toward this book.(less)