Simple and Powerful Things That Work For Me
summary
Shows a concrete invoice-import feature: upload XLSX, parse with Apache POI, and decide architecture by a 15‑minute cost/maintenance analysis that finds roughly €20k/month saved versus ~€10k–30k implementation plus ongoing ops. Recommends splitting into modules or separate microservices, shipping an MVP uploader first, driving requirements with BDD and Spock reports, using baby‑steps/backward programming drafts, adding logback logs and metrics for every execution branch, and treating high‑order functions as observability hooks.
AI-written summary · about
chapters (18)
- 0:00Introduction and context
- 1:04Simple practices that work
- 2:49Case study invoice import system
- 8:09Feature value and cost analysis
- 10:13Refining requirements with scenarios
- 11:16Cognitive load in microservices
- 15:51System failure and attack surface
- 17:39Microservices vs modules trade-offs
- 20:49Minimum viable products approach
- 22:13Behavioral scenarios and feedback
- 25:44One hour of design saves weeks
- 26:25Coding is not about perfection
- 36:55Techniques for complex problems
- 39:04Drafts and code deletion
- 42:55Exploring solutions before coding
- 45:02Observability in production
- 50:15Logging and reasoning in production
- 53:03Summary and additional tips
related talks
57mThings That Work So Well for Me I Can't Believe You're Not Using Them
27mTradeoffs on the road to Observability
18mDemonstrating your Architecture
46mLightning Talks
23mA peek into observability from a tester's lens
55mThe Troubles of Automating “all the Things”
55mSurviving your First Big Customer as a Startup
29mBackstage Revealed: Bright spots and dark corners