Why do software methods fall down?

The author's contention is that methods fall down because sufficient attention has not been paid to the problems of software testing. The focus is wrong. Much experience has shown that managers of IT in commercial organisations are unlikely to attach much importance to testing. However, testing is intrinsic to software development. If we consider the standard paradigm of software development, we always start with requirements definition and end with the production of some coded items. These may be programs in the textbook sense but often include other equally important things such as job control statements, parameters, screen and dialogue control items. The coded items have to be integrated into a system on the target hardware before they can meet the requirement. The processes of integration may be multi-layered but always involve testing. In addition to the needs of system integration there are global attributes of systems such as usability, operability and performance which cannot be assessed without testing. Most, if not all, of the above is always necessary regardless of which methods are employed.<
>