|
This seminar will explore the how and the why of software measurement. In particular, the area of most interest
is the determination of the nature of change in evolving software systems. One of the central issues in this
measurement problem is that there has heretofore been no way to measure the change activity. While it is clear
that a software system will typically change massively during its development, it is not clear just what the
specifics of these changes really are. We will explore a variety of different software metrics and then explore
the dimensionality of the actual measurement space. As software systems change through successive builds the
complexity characteristics of the individual modules that make up the system change in a variety of different
ways. This makes the task of measuring the precise changes that have occurred from one build to next a very difficult problem.
A methodology is presented employing principal components analysis that will extend the notion of software complexity
domains across sequential builds. Changes to software systems are then measured on a set of factor scores. to provide
leading indicators of potential problems introduced by the changes. Also, the notion of establishing a measurement
baseline is presented. This permits the comparison of a sequence of successive software builds to one another. A
specific software measurement example will be presented using measurement data from the Space Shuttle Primary Avionics
Software System.
|
|