Page 223 -

194 PART TWO MANAGING SOFTWARE PROJECTS

QUICK testing, test plans and procedures errors before they become defects! That is, work to
LOOK are produced. Other work prod- improve your defect removal efficiency (Chapters
ucts associated with process 4 and 7), thereby reducing the amount of rework
improvement may also be generated. that your software team has to perform.
How do I ensure that I’ve done it right? Find

SQA encompasses (1) a quality management approach, (2) effective software engi-
neering technology (methods and tools), (3) formal technical reviews that are applied
throughout the software process, (4) a multitiered testing strategy, (5) control of soft-
ware documentation and the changes made to it, (6) a procedure to ensure compli-
ance with software development standards (when applicable), and (7) measurement
and reporting mechanisms.
In this chapter, we focus on the management issues and the process-specific activ-
ities that enable a software organization to ensure that it does “the right things at the
right time in the right way.”

8.1 QUALITY CONCEPTS 1
It has been said that no two snowflakes are alike. Certainly when we watch snow
falling it is hard to imagine that snowflakes differ at all, let alone that each flake pos-
sesses a unique structure. In order to observe differences between snowflakes, we
must examine the specimens closely, perhaps using a magnifying glass. In fact, the
closer we look, the more differences we are able to observe.
This phenomenon, variation between samples, applies to all products of human as
well as natural creation. For example, if two “identical” circuit boards are examined
“People forget how closely enough, we may observe that the copper pathways on the boards differ slightly
fast you did a job —
but they always in geometry, placement, and thickness. In addition, the location and diameter of the
remember how well holes drilled in the boards varies as well.
you did it.” All engineered and manufactured parts exhibit variation. The variation between
Howard Newton. samples may not be obvious without the aid of precise equipment to measure the
geometry, electrical characteristics, or other attributes of the parts. However, with
sufficiently sensitive instruments, we will likely come to the conclusion that no two
samples of any item are exactly alike.
Variation control is the heart of quality control. A manufacturer wants to minimize
the variation among the products that are produced, even when doing something rel-
atively simple like duplicating diskettes. Surely, this cannot be a problem—duplicat-

1 This section, written by Michael Stovsky, has been adapted from “Fundamentals of ISO 9000,” a
workbook developed for Essential Software Engineering, a video curriculum developed by R. S.
Pressman & Associates, Inc. Reprinted with permission.