This blog post is about my notes on slide set for BBST Foundations lecture 5: The impossibility of complete testing.
First read is from Doug Hoffman (2003). “Exhausting your
test options”. Interesting example how to test sqrt function.
Second reading is from Cem Kaner (1997), The Impossibility of complete testing. We are not quality assurance because we are not authorised to make all moves that will assure quality. We are doing “Good enough testing”.
Third reading is “Factors that influence
test estimation” by Rex Black. Lists all factors that must be considered when putting out testing effort estimation. “Blog: When do we stop a test?” by Michael Bolton is self descriptive. Cem Kaner (1996), “Negotiating testing resources: A collaborative approach.” Is about test plan estimations. Mike Kelly, “Estimating testing using
spreadsheets” gives step by step example how to use spread sheet for testing estimation for Windows Calculator.
Combinatorial mathematics is very important testing tool. It is used to determine testing coverage.
What is complete testing?
1. Run all distinct tests. Test are distinct if they expose bugs that other test will miss.
2. We know that there are no bugs in software.