Route studies
Repeatable measurements of path changes, handshake timing, packet loss, and the boundaries between local and upstream networks.
Independent infrastructure practice
Arc Field studies how small, dependable network systems behave outside the lab—under real latency, imperfect routes, and ordinary operational pressure.
We favour legible architecture over clever architecture: fewer moving parts, measurable failure modes, and documentation that survives the person who wrote it.
Practice
Repeatable measurements of path changes, handshake timing, packet loss, and the boundaries between local and upstream networks.
Small reference deployments with explicit health checks, renewal paths, and recovery procedures that can be tested before they are needed.
Short records of decisions and incidents, written to preserve context rather than create ceremony around routine maintenance.
Field notes
Selected technical notes from ongoing experiments. Entries are intentionally concise and revised when assumptions change.
A useful baseline includes more than reachability. Connection time, certificate verification, and first-byte latency each isolate a different part of the path.
A fallback should be ordinary, complete, and independently maintainable—not merely an error page waiting behind the primary service.
Two successful requests may cross different interfaces. Capturing the active route prevents a proxy or tunnel from becoming an invisible test variable.
Service status
This page is served from the same edge used for public connectivity checks. No external analytics, fonts, or third-party scripts are loaded.