|
This book contains painstaking research work in the field of software architecture, and gives a competent treatment of its subject. Unfortunately it is rather solid, and lacks those Eureka moments that make a good book into a great book. The approach is biased towards careful, methodical progression, rather than flashes of inspiration. If you're involved in managing a software architecture and feel that you're in over your head, or if you are responsible for a software architecture and don't know where to start, then this book could be just what you need. If you're a software architect looking to raise your game then this book will help you, but Lakos, "Large Scale C++ Software Design" and Gamma et al., "Design Patterns" will help you much more (although both these books have a lower level view of the software development process). Good points: it has a consistent, integrated, method that uses UML as its notation. The four in-depth case studies are a good reference for how the process works in practice. Debating points: it uses four views of a software architecture that are different from the RUP views - this gives you a choice, but expect the RUP views to be more mainstream. Bad points: the connectors idea isn't well explained, and it seems to be a tool for making simple structures difficult to understand. The flow of the text tends to the soporific.
|