Here is a pretty good explanation of the Lincoln Index, which is a simple way to estimate the total number of errors expected in a software project. All you need are 2 independent testers.
Interestingly, when the testers find completely distinct errors, the estimate is infinite. Unfortunately, a typical bug may tend to crash the whole program in many different tests until it is corrected. Testing does not really find bugs independently.
A better application might be 2 independent code reviews.
The index actually finds its primary use in ecological population estimates, where its derivation is an obvious result of capture-mark-release-and-recapture.
Subscribe to:
Post Comments (Atom)
No comments:
Post a Comment