Risk Analysis
summary
Argues that risk is not just probability times impact but concrete stories and measurements: the Pentium floating-point bug that triggers wildly different probability estimates and leads to an eventual $475M charge and a replacement program; counting automated checks (4,000 versus 160,000 Windows Vista checks) doesn't prove quality; beta data (3 of 63 machines) scales to shipping 200,000 units and predicts thousands of support hours; recommends building risk models from detected vulnerabilities, detection methods, numbers that surprise, and narratives that show who feels the harm.
AI-written summary · about
chapters (17)
- 0:00Intro
- 1:08The Classic Approach to "Risk Analysis"
- 3:58Calculate the probability of this risk
- 4:50Given the odds, calculate the impact of this risk
- 5:54Calculate the monetary value of this risk
- 8:03The Risk Gap
- 10:49Factors that influence Risk
- 13:15Some Geometric Risk Analysis Heuristics
- 19:45You can focus risk ideas on one kind of problems...
- 20:34You can identify technical risk in a shallow way...
- 23:08"Risk-Based Testing" means organizing testing around suspected risk.
- 25:29Product Risk Story Elements
- 28:12An Example Table of Risk Elements
- 29:37Instead of calculation, try construction.
- 32:06Evocative Examples
- 33:11A Story from Perfect Software
- 36:41You can still use numbers in risk stories!
related talks
53mReady to Run the Risk? - Applying a Risk-Based Approach to Testing
46mInside Scoop on Risk-Based Testing
32mRisk-Based Test Effort Management
28mRisk Based Testing in Agile Teams
30mValue, Money, and Costs: The Ultimate Common Language for the Testing Strategy
40mRisk Based Testing in Agile Teams
56mControlling Quality in Software Development
39mHow to Talk about Coverage