Tinkertoys, Microservices, and Feature Management: How to Build for the Future
summary
Argues for migrating a monolith into modular microservices and using feature flags (LaunchDarkly) so containers stay stable while presentation changes, with concrete practices: run ring/progressive deployments, toggle one service at a time to verify status quo then the change, measure end-to-end message time, availability, accurate responses and accessibility, simulate partial outages and dependecy failures (readthedocs outage example), prefer small fast staging but test in production on a small percentage of users before full rollout.
AI-written summary · about
chapters (11)
- 0:00Start
- 3:09The future is fractal
- 3:28Why do we test?
- 4:12What do we test for?
- 5:389s don't matter if the user is unhappy.
- 16:33Modular (microservices)
- 17:47Semantics Syntax
- 19:01Testing experience, not embodiment
- 20:00Why microservices are like Tinkertoys
- 23:24Combinatoric Testing Complexity
- 24:43Do all the things in production
related talks
53mTinkertoys, microservices, and feature management: How to build for the future
27mTinkertoys, microservices, and feature management...
01h 03mThe Magic of Small Things - 10 Years of Microservices
52mMicroservices: A Utopian Mystery
40mKill "Microservices" before its too late
47mBreak Up the Monolith: Testing Microservices
45mMicroservices: A retrospective
52mMicroservice Migration Roadmap