software engineering

My criticism of the book is not with its content. The content is fine for a text book. My criticism is on the price: $$?! The information is basic software engineering material found in numerous sources. The fundamentals that students need can be found in other texts just as well written and significantly more economical. If you are an instructor you may want to look at Pfleeger or even the Systems Analysis and Design book in the Cashman series.

Again, this is a respectable reference and text book - the price is too much to ask of students though!

I gave up on this book when I reached the extended example of object-oriented analysis, design and implementation. The analysis was ok; the design dropped a few elements without explanation, but was largely coherent.

The implementation was a nightmare. It looked like procedural C++, with practically no relationship to the analysis and design.

I think the book does a good job of conveying the time-tested key concepts behind software engineering. It should not be taken seriously as a discussion of object-oriented methodology.

Schach's Object-Oriented and Classical Software Engineering is a textbook in the traditional sense of the word. The book is divided into two parts. The first part, Introduction to Software Engineering, deals with software life-cycle models, teams, software engineering tools, and a few other general topics. The second part, The Phases of the Software Life Cycle, then takes a more detailed look at requirements, design, implementation, and so forth.

The book has undergone a number of revisions during its lifetime, and it shows. Schach discusses both structured and object-oriented methods, but the coverage isn't unified enough; the book feels like a quickly-made patch. I also wish that iterative development and agile methods had received more attention.

To make matters worse, Schach's writing style doesn't impress me much. Ineffective passive constructions abound and modifiers always are put before the verb, even if the sentence contains a modal verb or the verb is be. This actually gets annoying after a while! What's more, Scach's approach is very, very prescriptive, and at least I frequently found myself objecting vehemently to the advice presented.

It's sad that to my knowledge there are no really good general books on software engineering. Sommerville's book suffers from the exact same defects as Schach's. Can a software engineering book not be made accurate, thought-provoking, and fun to read?