I read the first edition of the book and was greatly pleased by it. I loved the way it described the reasons patterns were good. In other words, although it clearly explained the patterns in the book, it also explained the underlying approach used by the people who wrote the software that later became described as patterns. The book helped me think like these developers instead of just using solutions they had come up with.
I also liked the way the authors showed how to use the patterns in the real world. I had always thought you used the patterns as solutions to recurring problems. However, the authors described how the patterns were really about a new way to think of the problem you had to solve.
As good as the first edition was, however, it left some gaps. In particular, while the domain analysis approach they espouse called commonality - variability analysis looks great, not enough information on how to actually use it was presented. Also, as XP has become more popular, I had been wondering about how patterns and XP fit into things.
The book addresses these and some additional issues incredibly well. (...)
Overall, I give this 2 thumbs up and recommend reading it even if you've read the first one. It is a much more involved book than the first with new chapters and improvements on the existing ones.
