Rip It Up And Start Again?
summary
Argues that microservices only deliver if organization and ownership change: a ticketing case shows 60 services → one story touches 6 services creating a distributed monolith; continuous delivery correlates with faster releases and lower failure rates; collective ownership is viable at small scale — collective ownership works up to ~20 people — but breaks as teams grow; strong team ownership plus SLOs and ops support enables independent deployability; cites Amazon (EKS, Fargate, Lambda, SQS) as an extreme autonomy-first, duplicated-product outcome.
AI-written summary · about
chapters (20)
- 0:00Start
- 1:59Monolith to Micro Services
- 5:06Types of System Decomposition
- 5:48Horizontal Styles of Decomposition
- 12:36Low Cohesion
- 19:55Organizational Structures
- 22:37Future Based Team
- 23:11Feature Based Teams
- 27:38Continuous Delivery
- 29:58Collective Ownership
- 33:06Collective Code Ownership
- 35:44Challenges
- 38:09Autonomy Mastery and Purpose
- 39:01Micro Services
- 41:15Aws Lambda
- 42:55Bottlenecks
- 43:57Pull Request
- 46:40The Roving Custodian
- 47:30Local Optimization versus Global Optimization
- 55:18Resources
related talks
01h 00mMicroservices, Where Did It All Go Wrong?
21mAutonomous Teams
44mIs It Time Already? Microservices Are Dead; Long Live Microservices!
57mEmbracing Microservices
47mMicroservices? Monoliths? An Annoying Discussion!
01h 03mThe Magic of Small Things - 10 Years of Microservices
45mContinuously delivering containerized, highly available microservices
45mMicroservices: A retrospective