Architecture
FrogNet is the Private Internet: a distributed networking fabric designed for environments where connectivity is unreliable, bandwidth is constrained, privacy is non-negotiable, and operational continuity matters more than centralized convenience.
Executive summary
It enables local-first services and edge intelligence to remain available during upstream failures, while still interoperating with conventional networks when they exist.
A Private Internet fabric: locally authoritative naming, resilient routing, deterministic convergence, and controlled propagation of state changes across connected nodes.
The system is designed to keep working when cloud assumptions collapse: DNS outages, upstream loss, high latency, intermittent links, and network partitions.
Not a VPN. Not a cloud service. Not “security policy on top of the public Internet.” FrogNet is private because it avoids public exposure by design.
It runs on commodity Linux hardware and treats wired, Wi-Fi, and radio-class links as transport options for the fabric.
Problem statement
Modern stacks assume persistent upstream infrastructure: global DNS, external auth, centralized APIs, and always-available cloud resources. Under degraded conditions, those assumptions collapse.
Core design principles
FrogNet is built around five architectural principles that make a Private Internet viable in the field.
Nodes identify neighbors, resolve service names, and continue operating without upstream infrastructure.
Topology changes converge to a stable state through bounded reconfiguration without central coordination.
Wired, Wi-Fi, and radio-class links are interchangeable carriers for the fabric’s control and service planes.
State changes propagate as minimal, deduplicated signals—no broadcast storms or loop amplification.
When policy requires local-only behavior, FrogNet fails safely rather than leaking into unintended paths.
System overview
FrogNet operates as a self-organizing fabric: changes are detected, stabilized locally, then harmonized across connected nodes.
Nodes identify candidate peers and only commit next hops after reachability is validated over a specific interface. No phantom paths. No wishful routing.
A merge rebuilds the node’s operational state: peers, routes, naming/resolution, and default egress decisions. Single-flight execution prevents thrash under churn.
Nodes emit minimal “re-evaluate” signals. Waves are deduplicated and bounded, yielding convergence without broadcast storms.
Egress & resolver anchoring
If an off-fabric exit exists, FrogNet can route toward it deterministically. If it disappears, local services keep working.
Security & privacy posture
This page intentionally omits proprietary enforcement mechanisms, but the architectural posture is clear.
To protect proprietary implementation details, the public write-up is architectural. If you are evaluating a deployment, partnership, or pilot, request the PDF and current demo materials by email: john@fawcettinnovations.com.
If you want the practical deployment story, start here: FrogNet overview.
