Too Darned Big to Test

Abstract

Testing large systems is a daunting task, but there are steps we can take to ease the pain. T he increasing size and complexity of software, coupled with concurrency and distributed systems, has made apparent the ineffectiveness of using only handcrafted tests. The misuse of code coverage and avoidance of random testing has exacerbated the problem. We must start again, beginning with good design (including dependency analysis), good static checking (including model property checking), and good unit testing (including good input selection). Code coverage can help select and prioritize tests to make you more effi cient, as can the all-pairs technique for controlling the number of confi gurations. Finally, testers can use models to generate test coverage and good stochastic tests, and to act as test oracles.

On the Effectiveness of Manual and Automatic Unit Test Generation

“…It computes the percentage of source code lines actually visited during tests execution. This is necessary but not sufficient to determine tests quality, because it "merely measures that a statement, block, or branch has been exercised" [10] giving no information about the meaningful execution of the code. Also 100% code coverage cannot guarantee exhaustive testing, since all the code would need to be executed with all possible combination of variable values, which is likely to be impossible.…”

Shedding Skins: Metaphors of Race and Sexuality in the Writing of Tatamkhulu Afrika∗

“…The reader's impression of a complex sexuality and sense of belonging is confirmed by a reading of Mr Chameleon and the semi-autobiographical novels and novellas. Elsewhere I comment on vexed same-sex intimacies mentioned in Mr Chameleon and Bitter Eden (Stobie 2005); here I will limit myself to a brief overview of characteristic homoerotic and homophobic preoccupations in Afrika's autobiography and his fiction, using for illustrative purposes the novella "The Vortex" in Tightrope. In Mr Chameleon Afrika remembers his child-self as being highly sensual and erotically charged, although living in a sexually repressive home.…”

A Tool to Automatically Map Implementation-Based Testing Techniques to Classes

“…Software systems are becoming larger and more complex, while software testing remains a challenge. In support of this claim Stobie [33] states that although computation power doubles every 18 months and the size of the software code doubles every seven years testing of the software does not appear to be keeping pace with the size and complexity of the software. In this paper the term software testing refers to: (1) the use of techniques and methods to generate test cases, and (2) deciding whether or not the test cases adequately cover some predetermined test criteria [11,41].…”