Telemetry that survives the dead zone
Cellular and satellite links drop in transit, at sea, underground, and at remote sites, and telemetry goes dark with them. Here is how to keep the stream alive.
Deep dives on mesh networking, offline identity, and coordination without a server, plus field notes and perspective from running the protocol in production. Written by the team building Offline Protocol.
Cellular and satellite links drop in transit, at sea, underground, and at remote sites, and telemetry goes dark with them. Here is how to keep the stream alive.
Modern software treats the network as a fact of nature rather than a dependency. Everywhere the work actually happens, that assumption quietly fails, and the case for offline-first is really the case for declaring it.
The delay-tolerant machinery underneath Offline Protocol: how a message waits, deduplicates, and moves the moment a link appears, instead of failing when one does not.
How devices on Offline Protocol prove who they are to each other with no certificate authority, no directory, and no live network to check against.
The smallest useful thing you can build on Offline Protocol is a message that reaches its recipient with no server in the path, over an encrypted mesh.
Inside the transport layer of Offline Protocol: how DORS runs BLE, WiFi Direct, and internet concurrently and moves traffic between them in under a second.
Downtime is usually measured in hours. The more useful unit is dollars per hour, which reframes offline-first from a nice-to-have into a line item.
How devices on the mesh advertise, discover, and invoke capabilities with no DNS, no registry server, and no internet connection at all.
A record 313 internet shutdowns, $19.7B in blackout losses, and 4.62 billion people affected. The numbers behind building offline-first.
We shipped fernweh, a messenger that works without internet, and watched the mesh under it grow to 350,000+ devices across 80+ countries.
No posts match your search.