The Good, the Bad and the Ugly solutions in test design
summary
Claims that general predicate testing reveals 100% of the predicate bugs, demonstrated on a RollerCoaster Gating requirement set. Uses a tool at test-design.org (free), comparing traditional three‑point BVA combinatorial testing (72 test cases, ~45 minutes) to GPT variants: black‑box ~25 tests (~15 minutes), gray‑box 18 reliable tests, white‑box 6 reliable tests. Presenter shows the tool builds test sets automatically, cuts design time and test count, and actually finds faults missed by coverage‑only methods.
AI-written summary · about
chapters (11)
- 0:00Metamorphic testing intro
- 1:05Inefficient test cases persist
- 2:30Test design techniques
- 4:36Roller coaster scenario
- 6:46Bad solution: 72 test cases
- 9:39Critical bug missed
- 11:03Introducing GPT
- 13:18GPT: gray vs white box
- 14:46GPT for safety-critical systems
- 15:50Key takeaways
- 16:31Answering audience questions about GPT
related talks
36mProperty-Based Testing: Unveiling the Truth on Test Reduction
02h 46mThe Joy of Testing
30mProperty-based Testing for Godly Tests
26mThe missing piece of testing
45mProperty based tests for the masses
57mDomain Invariants & Property-Based Testing for the Masses
51mProperty-Based Testing
58mProperty Based Testing: Concepts and Examples