[EAS] The words we use

[Botterell, Arthur@CalOES]

>From: EAS <eas-bounces at radiolists.net> on behalf of Sean Donelan <sean at donelan.com>

>It's easy to create requirements for systems when you don't pay for them.

It's also easy to filter and rationalize requirements to fit systems you don't need to pay for.  At least NYC gave some serious thought to what they actually operationally needed, and didn't self-censor that knowledge just because it was technically inconvenient.

> NYC OEM would likely scream bloody hell about the cost, and come up with all sorts of reasons why it shouldn't be forced to geo-target 0.1 mile areas if it needed to pay for the system.

A better analogy would be if NYC or any other local jurisdiction were suddenly required by law to provide warning services at all.  But no such obligation exists, so how could they justify spending money on something that wasn't required?  Especially with EAS there as a handy fig leaf for whenever questions about local public warning are raised.

Anyway, caricaturing NYC for giving an honest answer to a direct question about local requirements seems maybe a bit off-target.

> What is needed for....
>A national alerting system
>A multi-state alerting system
>A state-wide alerting system
>A metro-area/regional (multi-county) alerting system
>A single municipality alerting system

Another semantic pitfall is framing the issue of warning in terms of political boundaries, like those on which FIPS codes are founded.  (Note that the list ignores the all-too-common case of alerts that need to go to a limited area comprising portions of multiple adjacent jurisdictions... which is NOT the same as the "multi-state" or "multi-county" cases.)  Among other problems, that semantic baggage reinforces the political overtones of warning, which are one of the main reasons responders tend to wait till too late to issue warnings.  I'd suggest two alternate questions:

1) How can warning systems be devised that are indifferent to scale and adaptable to hazard boundaries?  (CAP was a first step in this direction... now the main challenge is enabling reliable authentication of alert messages while leveraging the wide variety of delivery systems available.  Or to put it another way, to devise a warning architecture that is genuinely technology-agnostic.)

2) Where warning technologies are limited in terms of 1), how can they be applied, selectively and seamlessly, to just those alerts to which their technically-inherent scales are a good match?  (Note that this is a fundamentally different problem than the classic EBS/EAS challenge of allowing the President to preempt ALL broadcast signals.  We continue to ignore that fact at our peril.)

Art