www.btinternet.com/~adrian.larner/review/gilb
Software Inspection
Tom Gilb and Dorothy Graham
Addison-Wesley, 1993, 472pp, softbound, £24.95
ISBN 0 201 63181 4
A Review by Adrian Larner for the Computer Journal
You may recall occasions on which you have “burned” – discarded and rewritten – your programs, and perhaps even defended the practice before an incredulous project manager. Now you can quote Tom Gilb and Dorothy Graham at them. This book has to be the definitive text on software inspection. It covers history, costs and benefits, the process, the roles, and the difficulties; and it includes six informative and varied case studies contributed by other authors.
Gilb and Graham include in the process a causal analysis of defects, with feedback to improve the process itself. Good: processes are as error-prone as programs, and making methods themselves subject to systematic criticism avoids their all-too-frequent doctrinal rigidity. Oddly enough, it seems that inspection works better for other documents (requirements, for instance) than it does for programs. But why? The book is well written, albeit at some length, and surely everything that can be said about inspections is said; except how they work (in the sense of their underlying mechanism; of course we are told how to do them).
Does this matter, to other than a few theoreticians? Yes it does: we want causal analyses of our successes as well as of our failures; and they are harder to get. Lots of methods work. We are told, no doubt on good evidence, that SSADM works. But abstract data typing and structured programming work because they reduce programmer choices (i.e. scope for error). Logic flowcharting worked because it documented structures too complex to fit in a human head. And that is what justified its abandonment: we should not program in such structures at all. Mere “working” – quality and productivity improvement – is not enough.
This book should go to a second edition, and many more. Chunky as it is, it might be allowed another fifty pages if Gilb and Graham could – please – explain their success.
Copyright © 1994, 2001 Adrian Larner. The author asserts all moral rights.
The decorative image of a key (cc004239.gif) used on this page was obtained from IMSI's MasterClips/MasterPhotos© Collection, 1895 Francisco Blvd East, San Rafael, CA 94901-5506, USA.