Review

I hate to down-rate this book, because it is very well researched and well written, but I just wasn’t able to be as excited about this as I have been with Edward Tufte’s books, to which this one is compared by one of the blurb writers. Part of the problem is that this book deals solely with visual representations of networks - an interesting genre, I’m sure, but one that stands more in the realm or art than science unless you drill down pretty far into the details of the network analysis and the techniques used to produce the visualization.

Lima does at one point present a ‘syntax’ of network representations - interesting, I suppose, but presented too superficially to be of actual use to the would-be practitioner.

By the latter part of the book I was just flipping pages, eager to get to the end, hoping to see something that would engage my attention. Didn’t happen.

Update: after reading the other reviews of this book on goodreads, I want to add that, like them, I did not get any deeper knowledge or insight into the underlying data by looking at the visualizations. Contrast that with the graphics, and the approach to visualization, offered by Tufte, and you can understand my disappointment with this book. Tufte generally shows before and after representations of data, and has clear, consistent principles on which to evaluate or create a visualization. The foremost principle is that of conveying as much information as possible; aesthetics are important, but secondary, and artistic merit barely deserves consideration. But what are we to make of a network graph that allegedly represents millions of nodes and tens of millions of connections? Where is the information content? How does such a graph (represented visually) advance our understanding? I would claim that in nearly all cases, it does not.

Metadata