Testing As a Risk Management Activity

摘要

Managers and testers alike often describe testing as being all about finding bugs. This is a natural conclusion one might draw after observing the testing process. Look for bugs, report bugs, then ensure they get fixed. Pretty simple. But there is a nasty problem with this. As we all know, testing all the defects out of a product is impossible in nearly all cases. And as any student of logic will affirm, even if we could find and dispatch every defect, it is logically impossible to prove that no more defects exist. (You can not prove the absence of something; only its presence.) So if the objective of testing is to remove the defects from the product, we are virtually guaranteed to fail to some extent. Defects will persist. But testing has a higher objective. Yes, testing finds defects (and of course we fix the defects we know about), but finding those defects is a means to a different - although a related end. Every product may fail in some non-trivial way that would have adverse impacts on the users or the organization. This is a risk that needs to be managed, and testing enables us to manage that risk.