Visualizing Data
O'Reilly Media, Inc, (2008)Description
This book is a hands-on introduction on how to visualise data, using the Processing language developed by the author.
The main part of the book presents a self-contained visualisation project per chapter, and the author has put on his website (see URL above) the Processing code for each project. Each chapter starts with a very concrete problem (e.g. comparing the consumption of milk, tea and coffee in the US over the last decades) and illustrates a particular visualisation technique (e.g. bar charts). The introductory chapters explain the basics of Processing (the author assumes some knowledge of programming concepts) and the 7 stages of data visualisation (acquire, parse, filter, mine, represent, refine, interact). Not all visualisation problems need all stages, and almost always one has to iterate over several stages until the final visualisation is achieved. The project chapters then show us the author 'thinking aloud' about each problem, following the stages and iterating over them, discussing the rationale for each change to the visualization.
The book's main message is that each particular problem requires proper planning of what one wants to visualise and then developing a bespoke solution: one-size-fit-all approaches (e.g. Excel charts) are usually limited. I especially liked the remark that although many data sets have a natural network-like structure, visualising trhem as networks is often not the best solution, as it doesn't scale up to large datasets.
Overall, a very good book, systematically going into the nitty-gritty details required to produce appropriate visualisations for the data at hand.