IT Professor rubbishes texting as an emergency alert system

Now this is an interesting one. A very interesting one. You know how I’m all in favour of text systems for emergency or company/organisation-wide alerts?

Well, that’s based on the system actually working as expected. The ability to be able to send a message to all 5,000 staff immediately (e.g. “Hello, come and empty your desk please, Mr & Mrs Lehman Brothers”) can only be a good thing, surely?

But as we’ve reported before, some systems haven’t entirely worked when they’ve been tested. Witness, if you will, the Louisiana State University text system screwing up during a test.

There are obvious implications here for the companies that have grown up specialising in emergency text services for colleges, universities and organisations.

The research by this Computer Science Professor is pretty direct. And it’s not the service providers to blame per se — it’s the actual operator infrastructure that is, typically, simply not built to send 10,000 messages through one or two cellular masts in one location, immediately.

The research conducted for the paper indicates that there are serious limitations in third party Emergency Alert Systems (EAS). In particular, because of the general architecture of CDMA, TDMA and GSM cellular networks, such systems will not be able to deliver a high volume of emergency messages in a short period of time. Through discussion, modeling and simulation, Traynor demonstrated in the paper that current systems not only cannot widely disseminate such messages quickly, but also that the additional traffic created by third party EAS solutions may disrupt other traffic such as voice communications, including that of emergency responders or the public to 9-1-1 services.

Cellular News has the full story.