The error coding technique proposed is clearly explained, and is proving practical and robust. Far from being cumbersome, as some reviewers suggest, it is easy to implement and helps avoid code bloat.

There is a clearly specified framework for the technique, and useful guidelines about how the standards can be introduced, certified and sustained in a development team.

As a bonus, there is a short but meaty section on an architectural approach which seems to offer a number of advantages over conventional tiered approaches.

The text claims the approach has been used with success in several large commercial projects and this rings true - there is a strong sense that these are practical lessons learned at the coal-face and not in the ivory tower.

Please don't be put off this book by the negative reviews below - they are somewhat baffling as most of the criticisms are factually incorrect. For example, the author does not ignore the importance of dlls in code reuse, does not ignore error logging, does not lock users into a text field using SETFOCUS, does not recommend always using named parameters etc etc. For whatever reason, these reviewers haven't given the book a fair reading.