Software Requirements And Specifications

Software Requirements & Specifications: A Lexicon of Practice, Principles and Prejudices by MichaelJackson, Addison-Wesley 1995

ISBN 0201877120 The book is, as it says, in lexicon form, with short sections such as WhatAndHow?, EngineeringEnvy, ObjectOrientedAnalysis? and Method with (nonhyper-)links between them. It would be highly wikiable therefore. If you like this book's concept of problem frames, see also MichaelJackson's latest:

Problem Frames: Analyzing and Structuring Software Development Problems ISBN 020159627X .

When we've all learnt as much as we can from practicing ExtremeProgramming and other lightweight methods of today there will still be value in a few other books on software. This is one of them in my view.

Michael deals wittily with our DisciplineEnvy but makes the case for calling our particular discipline SoftwareEngineering. He makes the helpful point that, like physical engineering, it's far too much to expect one individual to be expert in all aspects of our field. So "I'm a software engineer" is always likely to be pretty contentless. When people start to say things like "I'm a compiler engineer" we may be getting closer to an appropriate parity with other engineers. But not through pretending that software is not software.

--RichardDrake

Seconded. This is one of my desert-island books --DavidHarvey

Dittos for sure. I am using Jackson's book in my class this semester CS665 Software Engineering I at James Madison University. The book has a series of "tours" which begin to be defined on page xiii in the preface and they are a good way to get started reading the book. Since it is a lexicon, it does not necessarily read best in a linear fashion. More insights per page than any other book except maybe FredBrooks's MythicalManMonth. --RaySchneider