Can Chaotic Methods Improve Software Quality Predictions?
Jeffrey M. Voas · IEEE Software · 2000
To have 100% software quality predictability, we must know each fault's consequences. This requires the impossible task of knowing each fault's location and how each fault interacts with each software input. Because 100% predictability is impossible, the question becomes, “How close to it can we come and how?” By matching an arbitrary-randomness technique to the problem of the arbitrary randomness of software faults, we could increase predictability about future software output behaviors. While not perfect, this approach lets us increase predictability when faced with unknown future events that we can't model probabilistically because of our inability to define them