MULTIMEDIA DEVELOPING

Grady and Caswell (1986) offer a good description of Hewlett-Packard's software metric program, including both the primitive metrics and computed metrics that are widely used at HP. Primitive metrics are those that are directly measurable and accountable such as control token, data token, defect, total operands, LOC, and so forth. Computed metrics are metrics that are mathematical combinations of two or more primitive metrics. The following is an excerpt of HP's computed metrics:

Source: Grady, R. B., and D. L. Caswell, Software Metrics: Establishing A Company-Wide Program, pp. 225–226. Englewood Cliffs, N.J.: Prentice-Hall. Copyright © 1986 Prentice-Hall, Inc. Used with permission to reprint.

Average fixed defects/working day: self-explanatory.

Average engineering hours/fixed defect: self-explanatory.

Average reported defects/working day: self-explanatory.

Bang: "A quantitative indicator of net usable function from the user's point of view" (DeMarco, 1982). There are two methods for computing Bang. Computation of Bang for function-strong systems involves counting the tokens entering and leaving the function multiplied by the weight of the function. For data-strong systems it involves counting the objects in the database weighted by the number of relationships of which the object is a member.

Branches covered/total branches: When running a program, this metric indicates what percentage of the decision points were actually executed.

Defects/KNCSS: Self-explanatory (KNCSS—Thousand noncomment source statements).

Defects/LOD: Self-explanatory (LOD—Lines of documentation not included in program source code).

Defects/testing time: Self-explanatory.

Design weight: "Design weight is a simple sum of the module weights over the set of all modules in the design" (DeMarco, 1982). Each module weight is a function of the token count associated with the module and the expected number of decision counts which are based on the structure of data.

NCSS/engineering month: Self-explanatory.

Percent overtime: Average overtime/40 hours per week.

Phase: engineering months/total engineering months: Self-explanatory.