The common fallacies on platform building
summary
Platform discussions focus on concrete tradeoffs not slogans: one company reports 30% faster delivery after a single platform team for hundreds of apps, but that’s an anecdote. Common proposals include locking provisioning behind use Jira to create instances as a first-step experiment, offering services via a Postgres operator, or building a CLI to save costs versus shipping telemetry or just talking to users. Watch for hasty generalizations, slippery slopes and authority claims; prefer short experiments with success criteria.
AI-written summary · about
chapters (9)
- 0:00Platform engineering fallacies
- 1:45What is platform engineering
- 6:44Hasty generalization in decisions
- 10:56Slippery slope fears
- 15:28Strawman arguments
- 19:21Future consequences in design
- 23:14Other fallacies
- 24:58Authority in choices
- 25:41Closing thoughts
related talks
28mConvergence on Platforms
39mPlatform Engineering as a (Community) Service
41mPlatform as a Product
50mBuilding an Application Platform: Lessons from
33mPlatform engineering: Build your own PaaS
47mPlatform Team Structures and Concerns
50mWhy Your Platform Monolith Is Probably a Bad Idea
35mPlatform Engineering Is Not About Tech