The danger of using axioms in software metrics
The authors contrast axiom sets used in mathematics with those used in software metrics research. There are striking differences in the way the two types of axiom sets are derived and their intended purpose. In particular, axiom sets in software metrics are primarily used for purposes of definition, whereas in mathematics they are used for logical deduction. In addition, software metrics axiom sets have been developed without a consensus as to valid example measures and without a common understanding of the data to which they apply. The authors conclude that there is a non-negligible risk that inappropriate axiom sets will not be detected which may lead to an incorrect assessment of the validity of some software measures.
