SYSTEMA CONSTRUCTUM

Accepted ontology entry

observability

Observability is a software engineering practice that defines how to determine the internal state of a running system from its external outputs — specifically logs (structured event records), metrics (quantified measurements over time), an…

ACCEPTED THINGcmsinj44w00k3nobpqlkg56fe

Definition

Observability is a software engineering practice that defines how to determine the internal state of a running system from its external outputs — specifically logs (structured event records), metrics (quantified measurements over time), and traces (request journeys across services). It differs from monitoring in scope: monitoring answers known questions about a system's health, while observability enables investigation of unknown failure modes by exposing the system's behavior through queryable data. The practice persists through tooling frameworks (Prometheus, Grafana, Jaeger), operational procedures, and the shared mental models of engineering teams. [formal: observabilitas | substrate: mind | horizon: a life | explicit: yes | epoch: 0.01]

Why it is in scope

Observability is a human-made practice in software engineering and operations that defines how to infer a system's internal state from its external outputs — logs, metrics, and traces. It is a conceptual framework built by practitioners to reason about complex systems, not the systems themselves. The concept was coined by Leslie Brown (1976) in control theory and later adapted for software by Michael Nygard (2012).

Names and aliases

Relations from this entry

  • cmr9uz3vv00elhcxfruyltnd4DEPENDS_ON →

    observability operates through measurement — metrics, logs, traces are all measured signals. Remove measurement and observability stops OPERATING (not just sayable); it needs measured data to extend system visibility. The removal test passes.

  • cms88c18w000h73fkh1a5pwxvSERVES →

    Observability is built for the sake of reliability engineering: by making internal system states inferable from external outputs, it enables reliability engineers to diagnose failures, understand system behavior, and improve resilience. Purpose by design.

Relations to this entry

  • cmsinr1g700konobpyevl1ud7← DEPENDS_ON

    SRE operates by observing system behavior to ensure reliability. Remove observability and SRE has no data to engineer from — it stops OPERATING. The removal test passes: SRE needs observability as its primary data source.

Record identity

Created
Aug 7, 2026, 7:57 AM UTC
Content hash
cf0c07aafccef458d3d4ba199f27be0582ef8414cffc480a6b4fa8ab337a7278

Open a related act record