Web left out of emergency alert testing

At 11 a.m. Pacific Time, 2 p.m. Eastern today, for the first time ever, FEMA activated the Emergency Alert System nationwide. Everyone watching television or listening to the radio got the familiar alert tone on their show, and had to sit through the "this is only a test" message for 30 seconds.

And everyone on the Web had no idea that the EAS was being tested.

The broadcast-based Emergency Alert System is "decades old," FEMA's Rachel Racusen told me, and it's never been tested nationwide all at once. All tests so far have been regional or local. It is critical, she said, that the government finds out if this old system can work on a national scale.

Meantime, we have a population that is spending increasing amounts of time glued to mobile devices and the Internet, communication platforms that the EAS does not address.

The official Twitter page of the US Federal Communications Commission
FEMA
A social-media guide to dealing with Hurricane Irene
ReadydotGov

However, FEMA and the Dept of Homeland Security are working on a more contemporary system to augment the EAS. The Integrated Public Alert and Warning System (IPAWS) will, it's hoped, be able to send emergency messages to mobile devices and to Web sites.

Pieces of the system, in particular mobile alerts, are already being tested in some cities.

The exact mechanisms that the IPAWS system would use to break into a Web platform to send an emergency message are still being worked out. For TV and radio, the old EAS system sends a code to stations that authorizes it to break into the broadcast stream. PAWS will likely use a similar system to get a site's attention: Web site owners will have to set up their servers to recognize and authorize the incoming emergency code. Sites will also have to be configured to display the incoming message in some way. In talking to Racusen, I did not get the impression that there was any intention to intercept actual Web services at the DNS level.