
Core Terms: Metrics, Three Pillars, Unified Storage

Observability: the present and future, with Charity Majors
Topic 3 of 14
Published Jan 22, 2025 · Clip 13:09–17:05 (3:56)
Charity walks through foundational observability concepts: the difference between small-M metrics (generic telemetry) and big-M metrics (a number with tags), which lack contextual relationship data. She critiques the three pillars model—metrics, logs, traces—coined by Peter Bourgon, noting vendors pushed it cynically and engineers must manually correlate copies of the same data across many tools. She frames the 1.0-to-2.0 shift as moving from many sources of truth to unified storage, and touches on sampling's growing importance.