An Approach to Constructing AValid Proof
Constructing a valid proof in system testing can be defined as a four-step procedure. The following sections discuss each step in detail, and explain how to construct a valid argument to support a testing conclusion.
Step 1: Define a Conclusion of the Argument
In constructing a proof, always begin by defining what needs to be proven, i.e., the conclusion. In system testing, the ultimate goal is to evaluate a software product. This is achieved by decomposing the entire functional domain into a set of functional features where each feature to be tested is a unit of a tester's work that results in one of the two possible conclusions about a feature’s testing status pass or fail. At any given time, only one of the two conclusions is valid.
The term software feature is defined in the IEEE Std. 610 as follows:
- A distinguished characteristic of a software item.
- A software characteristic specified or implemented by requirements documentation.
From a tester’s perspective, a software feature means any characteristic of a software product that the tester believes might not work as expected and, therefore, should be tested. Deciding what features should be in the scope of testing is done at the test-planning phase and documented in a test