[EAS] GMC

[Harold Price]

Richard,

I'm not sure what specific proposal you are asking about, give me a pointer
and I'll take a look.

In general, though, there are individual EAS devices, and there is 
the EAS system.

The devices can be shown to be compliant with an existing rule 
set.  They were in 1996/7, and all of the new combined devices on the 
market now were tested more recently.

The problem is, individual box testing does nothing to show that the 
system will function as desired - something that was demonstrated 
last week.  The system includes procedures, processes, humans, and a 
lot of parts that have nothing to do with the EAS protocol, like 
whatever it was that caused a force tune to channels that weren't 
carrying EAS details, or whatever it was in the delivery pipe that 
cause the mix in the alert audio.  (There were also issues with EAS 
devices as I've discussed elsewhere, but I'm making the point that 
there are plenty of opportunities for non-EAS device issues to cause problems.)

So, what is it the FCC can test?  Individual boxes, clearly.  Can the 
FCC, on their own, ever make a tight enough specification so that 
there is no room for multiple interpretations of the rules, so the 
larger system will work with no system testing?  I don't think so.

Another issue is with the type of testing done.  Correct protocol 
data is sent to the devices during those tests, to make sure the 
devices act according to the rules.  For the Part 11 tests, data with 
bad syntax is not sent, and good data with bad semantics isn't sent 
either.  For CAP tests, some alerts that are valid CAP but should be 
ignored by IPAWS/EAS devices is used, but in general, only proper 
data is sent.

Someone needs to do the system testing, in the lab, and in the 
field.  FEMA did do lab system testing, I've seen the equipment, and 
I've talked to the people.  In hind sight, additional lab testing 
needed to be done.  Likewise, the Alaska tests identified a class of 
problems, but those tests did not have all PEP stations involved.

In the field, we need a way to do national testing that doesn't 
require notifying the public in advance, and without post event 
hearings on the hill discussing why a particular location didn't get 
the warning.  The next one or two tests should be nationally originated
RMT or NPTs so the national dissemination part of the system can be 
shaken out.  We need to decide what the fix is for the issues that 
came up with EAS devices, (effective time, multiple headers, upcut 
audio, among others).

It would be good if there was a way to test parts of the EAN that are
different, with a new code that says "test in the crawl", but does 
all the other EAN
things - no audio limit, automatic relay, etc.  This is not possible with
the old legacy devices in the field.  If the FCC had closed the loophole
that allows for the continued existence of legacy devices, it would be a lot
easier.  The old hard (or impossible) to update devices could have died a
natural death.  I'm not talking about discarding legacy EAS at this time,
just old legacy hardware.

I've wandered off topic, I'm afraid.  Am I in favor of more testing?
Always. Is "FCC testing" the right way to do it?  Maybe not, but someone
needs to do it, and the FCC is the only sheriff in town that can force it to
happen.  We don't need box testing, per se, we need system testing.

Harold

At 07:43 PM 11/16/2011, Richard Rudman wrote:

>As you and many others know Comments were filed related to the Part 
>11 rewrite that the FCC should do their own of tests of EAS devices. 
>The purpose: To make sure that all new EAS devices will "play nice" 
>with all legacy devices still in service, and that all EAS devices 
>(new and old) in the installed base will comply with all provisions 
>of Part 11. FCC testing would go beyond what has been done by FEMA 
>for their CAP conformance testing.
>
>I think the group would be interested to know your position on this 
>proposal at this time.