HUSTEF 2014 - Stephan Schulz
summary
Claims that MBT auto‑replaces testers collapse into specifics: a model of the system must be written, test targets and test generation metrics must be chosen, and generated suites (often 10 000 tests) still require validation because oracles are only as good as the model. Practical wins come from modeling use cases to limit complexity, running models as virtual systems, and advanced techniques like symbolic model checking or constraint solving to produce oracles and meaningful coverage rather than raw test counts.
AI-written summary · about
chapters (8)
- 0:00Start
- 1:16Model-Based Testing
- 3:10When Can We Start Calling Something Model-Based Testing
- 6:48Test Generation Metrics
- 9:25Test Execution
- 14:33Change Is Painful
- 15:04Number of Tests
- 19:34Discourage the Use of Design Models for Testing
related talks
11mDiamond Partner of HUSTEF 2019: Márton Mátyás, HTB
01h 05mMicrotesting
58mTesting Strategy - New Model, Better Outcomes
28mYour Testing Strategy is broken - lets fix it!
43mCan you Test it
23m„Complete” In-Sprint Testing Using Collaborative Models
55mShifting the Testing Role Pendulum
48mTest Automation - A Systems Thinking Approach