Do You Know How Good You Are at Test Design? We Don’t Think So. Why?
summary
Claim: many testers cannot judge their own test design skill. Evidence: a custom mutation testing framework with 15 mutants yields an average 66% bug detection from 10 exploratory testers while the author finds 80%. A new modeling technique action-state testing produces 13 model-derived tests that find 87% of faults; applying the stronger transition-pairs criterion expands to 26 automated Cypress tests killed all mutants. Tools: testdesign.org and Harmony are used to build, run and evaluate the exercises.
AI-written summary · about
chapters (12)
- 0:00Test design challenges
- 1:44What is test design
- 3:56Proper test design
- 5:04Measuring test design
- 9:04testdesign.org platform
- 12:01Rental system exercise
- 16:52Limitations of traditional methods
- 19:27Action state testing
- 24:51Modeling the system
- 27:51Refining test cases
- 30:49Results and effectiveness
- 31:55Conclusion
related talks
29mWho is testing the test?
41mMutation testing in practice
07mWho is Testing the Tests?
38mMutants to the rescue: How effective are your unit tests?
43mMutation Testing, Finally a Good Measure of Test Quality?
27mRe-thinking Regression Testing
31mCan You Trust Your Tests?
32mCAN WE WRITE PERFECT TESTS? - MAYBE!