It arrived today, actually.
It can't seem to make up its own mind about whether its an implementation book or not. On the one hand, it sometimes goes into detail about how to make minor visual tweaks in R (on page 98, it spends two pages describing, with screenshots, all the optional parameters to barplot()). On the other hand, pretty much every chapter of this book uses a completely different tool to do the work, literally everything from R to Adobe Flash to Python, and it takes each of these from scratch, so you can imagine the lack of thoroughness for any one of them.
On the other hand, it covers a lot of visualizations with, I'm sensing, a particular emphasis on spatial data. Which should be right up your alley, Paul :) But the variety is stunning; traditional charts and graphs, treemaps, heatmaps, scatterplot matrices, Hans Rosling-style bubble plots, spider charts, and various animation trickeries all get meaty sections, with quite pretty full-color illustrations.
I view both you and Lisa as somewhere in the pantheon of data visualization Gods, so I don't know how much of it will be useful to you. But if you can get over the sections where he (literally) teaches you how to get Adobe Flex running in Eclipse, it looks like a pretty interesting survey of techniques.
Hope that helps. I can answer specific questions too now that I have it in my hands.
Cheers,
Adrian