(This book has a lot of "highlights". For example, its author, Rod Johnson, has 10 years of experience in writing Java programs and is now a member of the Servlet and JDO2.0 JSR expert groups; For another example, Spring, Hibernate, and WebWork, which are introduced in the book, are all popular open source structures. IoC, AOP, and so on are all popular concepts. The biggest highlight is on the cover of the book: "without EJB". We have seen in many books and articles that EJB is one of the core technologies of J2EE; But this book by Rod Johnson claims that the vast majority of J2EE uses the underlying layer without EJB. This almost provocative posture makes it difficult for any responsible J2EE architect not to have a final idea -- no matter whether you agree with him or refute him. But none of this is the book's greatest value. What is the basis for selecting an architecture and technology? Rod Johnson believes that it should be based on the basis of practice, experience from previous history projects or personal experiments, rather than any mode of idolatry or opinions of the family. The book talks about the problems and solutions used by enterprises in all aspects, and none of these solutions is not the product of this "evidence-based approach". In addition to delivering these solutions to the reader, the more important message that Rod Johnson hopes to convey through this book is just the "evidence-based" office approach - which should have been the way programmers work.)