This is the first book of its kind on software metrics - the measurement of software products and processes used for establishing quality assurance thresholds and in predicting quality levels. This authoritative review of the subject discusses a number of weaknesses which have been discovered over the last ten years and describes a method for the derivation and validation of metrics. This is the first work to describe a methodical derivation process for software metrics and is a major improvement over the current ad hoc ways of developing metrics making this an important and much needed contribution to software engineering.