Observation model
StackRadar records infrastructure signals as reduced time-window observations. Each observation is associated with a service surface, region group, signal class, confidence score, and review window.
Alpha documentation
Technical notes
Operational notes for the StackRadar alpha observation model, public visibility rules, and restricted workspace access.
Probe cadence
Probe cadence is assigned per workspace during alpha onboarding. Cadence can differ by service surface, region group, and review objective. Public summaries show only coarse timing markers.
Snapshot delay
Public status data is delayed and sanitized. Workspace telemetry is not exposed on the public surface. Snapshots are reduced for public visibility.
Sanitized public surface
The public surface presents only component-level state, reduced regional markers, and confidence ranges. Workspace identifiers, service names, internal routes, and reviewer notes remain restricted.
Workspace access
Workspace access is limited to invited alpha testers, infrastructure partners, and selected technical reviewers. Activation requires an assigned onboarding link, workspace identifier, and invite token.
Regional sampling
Regional samples are grouped by observation role rather than exact deployment topology. Current public groups include EU-West, EU-Central, RU-Edge, TR-Gateway, KZ-Transit, and Baltic Relay.
Resolver variance
Resolver variance compares lookup timing and answer-set stability between observation windows. Elevated variance can reduce confidence even when API availability remains operational.
No personal data collection on public pages
Public pages are designed for product orientation and alpha access flow. They do not collect personal data, do not include third-party analytics, and do not expose workspace telemetry.