Questions
- It’s likely that alerts, dashboards, and usage reporting depend on custom telemetry and rely on conventions. How would you approach evolving conventions to prevent breaking something critical?
- Is it possible to validate that telemetry coming from some service follows defined semantic conventions?