This book has benefited greatly from the input received from many sources.
First and foremost, I must thank the technical reviewer, Hadley Wickham, of ggplot2 and plyr fame. I suggested Hadley to No Starch Press because of his experience developing these and other highly popular R packages in CRAN, the R user-contributed code repository. As expected, a number of Hadley’s comments resulted in improvements to the text, especially his comments about particular coding examples, which often began “I wonder what would happen if you wrote it this way. . . .” In some cases, these comments led to changing an example with one or two versions of code to an example showing two, three, or sometimes even four different ways to accomplish a given coding goal. This allowed for comparisons of the advantages and disadvantages of various approaches, which I believe the reader will find instructive.
I am very grateful to Jim Porzak, cofounder of the Bay Area useR Group (BARUG, http://www.bay-r.org/), for his frequent encouragement as I was writing this book. And while on the subject of BARUG, I must thank Jim and the other cofounder, Mike Driscoll, for establishing that lively and stimulating forum. At BARUG, the speakers on wonderful applications of R have always left me feeling that writing this book was a very worthy project. BARUG has also benefited from the financial support of Revolution Analytics and countless hours, energy, and ideas from David Smith and Joe Rickert of that firm.
Jay Emerson and Mike Kane, authors of the award-winning bigmemory package in CRAN, read through an early draft of Chapter 16 on parallel R programming and made valuable comments.
John Chambers (founder of S, the “ancestor” of R) and Martin Morgan provided advice concerning R internals, which was very helpful to me for the discussion of R’s performance issues in Chapter 14.
Section 7.8.4 covers a controversial topic in programming communities—the use of global variables. In order to be able to get a wide range of perspectives, I bounced my ideas off several people, notably R core group member Thomas Lumley and my UC Davis computer science colleague, Sean Davis. Needless to say, there is no implication that they endorse my views in that section of the book, but their comments were quite helpful.
Early in the project, I made a very rough (and very partial) draft of the book available for public comment and received helpful feedback from Ramon Diaz-Uriarte, Barbara F. La Scala, Jason Liao, and my old friend Mike Hannon. My daughter Laura, an engineering student, read parts of the early chapters and made some good suggestions that improved the book.
My own CRAN projects and other R-related research (parts of which serve as examples in the book) have benefited from the advice, feedback, and/or encouragement of many people, especially Mark Bravington, Stephen Eglen, Dirk Eddelbuett, Jay Emerson, Mike Kane, Gary King, Duncan Murdoch, and Joe Rickert.
R core group member Duncan Temple Lang is at my institution, the University of California, Davis. Though we are in different departments and thus haven’t interacted much, this book owes something to his presence on campus. He has helped to create a very R-aware culture at UCD, which has made it easy for me to justify to my department the large amount of time I’ve spent writing this book.
This is my second project with No Starch Press. As soon as I decided to write this book, I naturally turned to No Starch Press because I like the informal style, high usability, and affordability of their products. Thanks go to Bill Pollock for approving the project, to editorial staff Keith Fancher and Alison Law, and to the freelance copyeditor Marilyn Smith.
Last but definitely not least, I thank two beautiful, brilliant, and funny women—my wife Gamis and the aforementioned Laura, both of whom cheerfully accepted my statement “I’m working on the R book,” whenever they asked why I was so buried in work.