Far from the Shallows: The Value of Deeper Incident Analysis
summary
A claim: shallow metrics mislead — histograms from over 10,000 public reports in VOID (Verica Open Incident Database) show heavily skewed durations so that mean time to recover is misleading and duration doesn’t correlate with assigned severity; root‑cause labels flatten system causality. The work shows qualitative artifacts — incident stories, near‑miss analyses, a rotating incident commander program, and examples like trading disk buffer for tiered storage (losing ~20h replay) — that reveal safety boundaries and pockets of expertise.
AI-written summary · about
chapters (6)
- 0:00Deeper incident analysis intro
- 1:33Beyond shallow metrics
- 13:22From metrics to incident stories
- 22:52Safety boundaries explored
- 31:04Near misses and averted incidents
- 36:00Community via incident stories
related talks
27mTales from the VOID: The Scary Truth about Incident Metrics
37mIncident Management Metrics That Matter
41mAre We Getting Better Yet? Progress Toward Safer Operations
33mThe Problem with MTTR: Learning from Incident Reports
24mThe Un-Incident: Extracting Value from the Gray Area of...
14mWhat dashboards don't tell you
42mThe Fault in Our Metrics: Rethinking How We Measure Detection & Response
34mTaking Human Performance Seriously in Software