On a cursory glance, The Visual Miscellaneum by David McCandless looks pretty good. The book covers a wide breath of topics and everything is very colourful. However, a more detailed inspection will reveal some serious errors in the visualizations the book contains.

Some of the data items within the information graphics have been intentionally left unlabeled. For example, “Sex Education” on page 62, presents a bar chart and the only entry that is labeled is Studio Art which has no students that are virgins. Having seen this chart before, the bar chart on page 62 emphasizes the original joke. In the introduction, McCandless asks “Can you make jokes in graphs?” Indeed you can, and this is a borderline ok example. However, the joke isn’t clear if you haven’t seen the original chart.

From here, it gets worse. The “Excuse Us: Reasons for Divorce” graphic on page 64 does not have any labels on any of the items in the graphic. Where is the actual content? Or are the data items intentionally left unlabeled?

Similarly, in the “Salad Dressings: All in proportion” graphic on page 118, why aren’t the individual dressings labeled? On the previous page, with “The Poison” and “The Remedy”, the drinks are labeled, so therefore I would expect the salad dressings on the following page to labeled as well (especially since the graphic design of these pages are so similar). However, this is not the case.

In general, it seems like there are more than a few labeling issues in the graphics through out the book, and I was left wondering if this omissions were erroneous or intentional? It wasn’t always clear.

But missing labels isn’t the only problem. The “Not Nice: Food colourings linked to unpleasant health effects” graphic on page 119 contains no mention of health effects in the actual graphic. Is that the point? That there is no unpleasant health effects linked to food colouring? I wasn’t sure after looking at the graphic.

These are just a few examples of some of the problems with the visualization found in The Visual Miscellaneum, and unfortunately these errors complete spoil the entire book. I would agree with Christian Rudder’s Amazon review that this book is an example of how not to present information, and if you’re looking for visualization excellency, consider the work by Edward R. Tufte instead.

Visualization of the Lark paper from Wordle.net.

While working on the literature review for my thesis, I came across a paper concerning “visualization pipelines” which is widely cited in the scientific visualization community. The article is part of a book, published in 1990, and proved to be difficult to locate online. The friendly people at the U of C library where able to track this one down for me–interlibrary loans strike again!–and to avoid you the trouble, here is the article.

R.B.Haber and David A. McNabb, “Visualization Idioms: A Conceptual Model for Scientific Visualization Systems“, Visualization in Scientific Computing, IEEE, pp 74-93, 1990.