A lightweight situational-awareness dashboard built for the worst day — under 500KB, offline-first, designed to keep working when the server and the internet don't.
Crisis-management tools assume connectivity, servers and infrastructure will be available — exactly the things that fail first in a real crisis. A dashboard that goes dark when the network does is worse than no dashboard, because people trust it.
The design question wasn't "what features can we add" but "what must keep working when everything else stops." Resilience wasn't a non-functional requirement — it was the core product decision every other choice answered to.
The entire dashboard is client-side and under 500KB — small enough to load over 2G and to keep rendering with no backend round-trip. If connectivity dies mid-session, the last good state stays on screen.
Activepieces automations collect public reports and push them into the shared state — a data pipeline assembled from existing automation instead of custom server code that would need operating and monitoring.
Google Sheets acts as the real-time database: no server to run, nothing to bill, and an interface anyone on the team can read and repair directly. The "cheap" choice was also the resilient one.
Reports appear on an interactive map as they arrive. On constrained networks the map stays usable — the client renders what it has, and catches up when the connection returns.
Each design choice was weighed against a scenario: server down, internet down, both down. Features that couldn't survive a degraded environment were cut — not deferred, cut.
The project's core lesson: constraints like "must survive 2G" and "no budget for servers" don't block a product — treated seriously, they become the specification. Every limit produced a design decision that a well-funded build would have missed.