Software evaluation. A semantical approach to measures of information (Q4023947)

There are a lot of software metrics for the evaluation of software in the direction of its psychological complexity. Unfortunately, every software metric or software measure begins with the individual point of view relating to this complexity: the number of decisions in a program, the nesting level, the number of data definition and data use relations etc. The author tries to define a ``neutral'' measure that is based on the domain theory of the denotational semantics of programs. The book is a description of this approach in terms of \(\bullet\) some critical remarks about existing complexity measures (chapters 1 and 2), \(\bullet\) methods of formal descriptions of the semantics of programs (chapter 3), \(\bullet\) foundations of the domain theory (chapter 4), \(\bullet\) the application of this theory for a quantitative measure of programs (chapter 5) and \(\bullet\) possibilities for further applications and extensions of this theory (chapter 6 and 7). The author formulates the possibility of the quantification as a special form of an abstract program interpretation --- as quantitative information transformation. In chapter 6 a simple example is given of a software measure on the basis of this approach. This book includes only the description of a new approach to the evaluation of the complexity of programs but it makes the reader curious about possible concrete applications.