Been teaching visualization for some years now and while the book offering used to be really narrow (i.e. Tufte + some on stat graphing), things have gotten better now.
I divide the books to three different groups.
Information Graphics
- More high level visualizations, usually hand drawn. There are good (Tufte) and not so tood (Wurman) examples of this genre. You'll mostly learn genre techniques, prettifying, get plenty of inspiration and ideas, but also learn a lot of bad habits hard to get rid of later on, if you start with these (Tufte being the exception). Recommended for illustrators who need to to info-gfx and need to SELL their illustrations to magazines/books, where visualizing properly is usually not the 1st priority
Data visualization books (how to)
- These are books about graphing, plotting, displaying data points - so a lower level representation than the information graphics approach, which already has a story to tell. Data viz is about displaying data and it's underlying structures properly, even when there's no hidden story to tell (yet, may not be discovered). A really good primer in this field, imho, is Harris' Information Graphics: A Comprehensive Illustrated Reference. It's not pretty, sexy nor step-by-step how to, but it show you most of the graph types and studying them you can learn a lot. If you are comfortable with some maths (not heavy) and more technical approach, then Handbook of Data Visualization by Chen (ed.) is a good start as an anthology. For more maths I'd reach for Introduction to Scientific Visualization by Wright, but as I'm not into that myself I must say that this recommendation is based on cursory glance.
Visualization & perception
- These are books that go one additional step back and show you how humans see and make meaning. These are imho, expert books, but great for anybody who is serious about getting into the field and really wants to make good stuff and break the rules with a purpose and succcess. My standard recommendation used to be Colin Ware's 'Information Visualization' in 2nd ed now. Another more recent recommendation would be 'Introduction to Information Visualization' by Mazza (Springer, 2009). Don't be fooled by its name, it is 90% data visualization (using the terminology I defined in this post). Ware's more recent 'Visual Thinking for Design' is more perception oriented, so less about visualization than about how people see visualizations.
There's also a class of books about language specific visualization, like 'Beginning Python Visualization' and 'Visualizing Data - Exploring Data with Processing'. Haven't had time to more than glance at those, but it's good to be aware of those, if one wants to get in via the programming route and is a beginner in programming.
Then there are more specific books depending on your needs like 'How Maps Work' by MacEachren (understanding/making maps), 'Graphics of Large Datasets: Visualizing a Million' by Chambers (maths and algorithms), 'Graphing Data' and 'Visualizing Data' by Cleveland (how to draw statistical graphs properly) and 'Information Dashboard Design' by Few (basics of designing business information dashboards) + many, many others.
If I'd start with one (for a programmer), I'd start with the Mazza book, or if you just want to look at examples, then the Harris book. If your really into the visualization, then read all Tuftes, Cleveland books, Harris and Wares for a start.
And visualize A LOT, show your stuff to others, kill your darlings and learn from criticism. After all, visualization is about conveying meaning. If people don't understand your visualizations, it's time to go back to the drawing board.
Best of luck with your learning! Visualization can be fun, but really knowing it can also be hard work at times :)
P.S. The Bertin's now out-of-print classic Semiology of Graphics is a must for any *serious* student in the field, esp. if one is not adept in graphic design. It's such a pity it has not been reprinted. Not worth $400 though, unless you are really dedicated and have read/understood most everything else :)