Unplug and Go — Hub-Spoke Topology, Role Promotion, and Five-Minute Failover in xGrid
Any Spoke can become a Hub in minutes. A failed Hub gets replaced in five. Every Raspberry Pi ships with the full stack pre-installed — the role is just a config file. This is how xGrid handles topology changes in disconnected environments.
Offline-First Is Not a Fallback — How xGrid Runs Without the Internet
Most medical systems treat offline mode as degradation. xGrid treats it as the default. When your deployment site has no cell towers, no routers, and no IT staff, every design decision starts from zero connectivity.
When the Wall Is Breached — Designing Medical Systems with Safety-II
Safety-I builds walls against failure. Safety-II asks how people succeed despite failure. In disaster medicine, the difference determines whether a forced evacuation loses patients or saves them.
One Database, One File, Zero Configuration — Why Simplicity Wins in Disaster Medicine
Enterprise database servers are the right choice for most hospitals. On a portable device running disaster medicine software with no IT staff, simplicity is not a limitation — it is the most important feature.
Walkaway DR — How a Phone Rebuilds a Dead Server
Your Raspberry Pi just died mid-surgery. Every patient record, blood product, and medication log was on that device. A nurse plugs in a fresh $80 board, and her phone restores everything in under three minutes. Here is how Walkaway Disaster Recovery works.
The Walkaway Test — Designing Software That Outlives Its Creators
What happens when the development team disappears? We formalized the software industry's 'bus factor' problem into five rigorous acceptance criteria — and built a medical system that passes all of them.