Category: QA

No it is not the answer.  The amount of testing we can do will always outrun the number of people who can do it.  There is always more testing that one can add to try and assure that less defects are in the system.  However, that will only take you so far.  Inserting less defects in the first place and better planning to concentrate the resources where they are required is a better solution.  Next week we will look at the demographic dividend as it relates to testing (and a lot of other industries).

If we refer to the dictionary definition of Oracle we find a couple of possible definitions that could be applied to the use made by most people of this term although a couple might lead to some questions.

One which raises a couple of questions is an ambiguous or obscure utterance – not what we want in software testing.  The second is a divine communication – very unusual in software testing.

Probably the most relevant one is the which suggests that the person who is delivering wise and potentially influential pronouncements in an orcale.    This is the definition that comes closest to the use made of this term in Software Quality Assurance and Quality Control areas.  The idea is that the person knows more and can use that knowledge to provide better Quality Control in some fashion.

The issue is that it is only as good as the person making the pronouncement.  If they are wrong, then that may lead to a cascade of mistesting and wasted time.  The second part is that it is only good as long as the person is in the place where they can claim to be oracular.  Once they leave all the information and knowledge goes out the door with them.

Quality Assurance and Process Improvement solves this problem.

We are hearing a lot (again) about Testing faster as if this is the answer to continuing increase in the volume of testing that needs to be competed to ensure system stability. However, the growth of systems will inevitably exceed the ability of people to test them and all their interconnections. There are ways around this by testing smarter and considering the impact of the software. However, first we will consider the other fallacies that have been mooted as solutions.