A system that requires a human to read an email and decide whether action is needed is fundamentally flawed. Software should do the interpreting; humans should be notified only when they need to take action.
## Three Valid Outputs
| Type | Definition | Human Action |
|------|-----------|--------------|
| **Alerts** | Something is happening or about to happen that needs immediate action | Act now |
| **Tickets** | System can't auto-handle, but no damage if human acts in a few days | Act soon |
| **Logging** | Recorded for diagnostic/forensic purposes | None unless prompted |
## Key Principle
Monitoring should **never** require a human to interpret any part of the alerting domain. The classic "email alert when value exceeded" pattern is explicitly called out as ineffective.
## Logging Expectation
No one reads logs unless something else prompts them to do so. Logs exist for after-the-fact investigation, not proactive monitoring.
## Cross-Domain Connections
- [[Error Budget Model]] — monitoring detects budget consumption
- [[SRE Playbook MTTR Improvement]] — what happens after an alert fires
- [[True Observability Prerequisites]] — modern evolution of this framework
---
*Source: Site Reliability Engineering, Chapter 1 (Treynor Sloss, 2016)*
*Extracted: 2026-03-25*