TiPS Episode 26 0204 asks the question nobody likes to ask out loud: Can a “cloud-based NG911” go offline?

You already know the answer. Yes. Of course it can.

And if anyone tells you it cannot, they are selling something.

In this episode, Fletch breaks down why “cloud” is not a force field. It is still infrastructure, and infrastructure fails. Sometimes the cloud provider has an incident, but more often the failure happens in the messy middle: the path to the cloud breaks, a dependency breaks, or your own building becomes the outage.

What “Cloud NG911” actually means (because it is not one thing)

How outages show up on the PSAP floor (spoiler: you do not get a nice email)

The four most common failure patterns:

Cannot reach the platform (fiber cut, carrier outage, ESInet transport, routing, firewall)

Reachable, but degraded and weird (maps stall, LVF/ECRF timeouts, CAD integration breaks, recording fails)

Vendor cloud region having a “moment” (shared dependencies across multiple agencies)

Your building is the outage (power, HVAC, network core failure, evacuation, smoke, flood)

“If this goes down… where do the calls go?”

Not in theory. Not in a binder. Automatically, pre-configured, pre-approved, and practiced. Because failover that is not tested is not failover. It is improv theater.

We also hit four Reality Checks:

Cloud does not eliminate outages, it changes the shape of outages

Vendor SLAs are not the same thing as your public promise

Failover that is not tested is fiction

Your weakest link is often local, and then everyone blames “the cloud”

Bottom line: Cloud can be a fantastic strategy when it is engineered with survivability. But cloud without survivability is just moving your single point of failure to a nicer zip code.