[EAS] Single point entry of CAP messages

[Alex Hartman]

Since we're proposing systemic changes to an already drafted system,
let me spell my ideas out again.

CAP is not a bad idea, not by any means, it's the methodology the
gov't is using for deployment i take serious issue with. Having the
entire country rely on a single aggregation point is pretty damn
stupid, even by today's government standards. Expecting 30,000+
machines in the country to connect to a single point, BAD IDEA. This
is where the IT people speak up and point out the "backwards" design
of it all.

IT SHOULD BE in the hands of local government, period. Since that's
where the "emergency" will be. Pick your favorite non-weather related
activation. Train derailment, chemical spill, any civic emergency. If
the local sheriff activates, sends to the state server, then is sent
to the federal server, and my box then goes and checks the federal
server for a message once a minute, and hey, the message got there 2
seconds after my last check, my alert is now delayed by at least a
minute. Not counting the processing time required to generate the TTS
audio and message retrieval. Has anyone been able to answer what the
mandated poll interval is between state and federal servers? I haven't
seen anything of the sort yet.

HERE'S HOW IT SHOULD BE DESIGNED:

The STATE has a server... or more (no less than 3), that is on the
states already installed infrastructure (even in MN, there's fiber
backhauls that're state owned and operated interconnecting the
government agencies) that are the point of contact for the decoders,
NOT a federal aggregation system. And then with those 3 servers,
scattered throughout the state, that are sync'd to eachother so they
ALL have the same data, and using a basic round-robin DNS, all the
state decoders LOOK TO THEM instead, thus reducing the load, all using
existing infrastructure. This also has built-in redundancies. Any of
the servers can fail, and my machine/decoder will still get the
messages. In a single server system, backhoe fade is very real. What
if DC gets wiped off the map? How do we know short of the obvious
(keeping CAP in mind here)? The server is unreachable, we never get
the message. Or god forbid a router upgrade goes wonky and takes them
offline, or any other inherent problem using the public internet.

This is where the federal server should be nothing more than an entry
point into the system. The state servers can poll the feds for
messages and that's ALL it should do. Instead of having 30,000+
machines polling every 10s-1m, you could have, what? 300 machines
polling 3 different servers at a time (distributed load remember),
seriously, an old pentium 100 could handle 100 clients at a time NO
PROBLEM.

The current design spells disaster IMO. It's not redundant, it's
poorly planned, and from their own website will "reach 90% of
Americans"... EAS reaches 100% right now, why did we settle for less?

These are the VERY questions that need to be answered by the FCC/FEMA.
I think we as broadcasters have a right to know why we're taking a hit
for such a flawed design. And make no mistake about it, any failure
(and there will be, there always is) will be blamed on the
broadcasters, because government doesn't make mistakes, do they? ;)

If the foundation of the system is shaky at best, we as broadcasters
CANNOT make it any better than it's weakest link, which has already
been spelled out on line 1.

EAS at least has the ability to be improved upon in it's current form,
CAP is the right direction, i just disagree with it's fundamental
design as it stands today.

--
Alex Hartman

On Thu, May 19, 2011 at 10:06 PM, Richard_Rudman <rar01 at mac.com> wrote:
> The purpose of this list is to provide an open forum for all EAS questions and discussions. We have certainly seen this play out in current threads.
>
> Tom brings up an interesting solution to wireless CAP distribution.
>
> I would like to propose another solution.
>
> The responsibility for issuing warnings rests with government, not broadcasters. Broadcasting, however, has been, is and will be (for the foreseeable future)  the best broadband channel that government has to reach the public in real time. This holds true for warnings, and the ongoing delivery of post-warning information.
>
> It is my fervent hope that local and state governments will work with the broadcasting community to supply broadband wireless links for CAP messages that can bypass the public internet. So, how can they do this and not break the bank?
>
> Here's one way.
>
> Local government can identify existing low priority UHF repeater systems they hold licenses for that could be connected so CAP-EAS messages can interrupt what these repeaters normally do. This puts the access point for wireless CAP message distribution where it needs to be - under government control and at a place where CAP messages can be relayed to all the broadcast studios in a given region without touching the public internet.
>
> Each broadcast entry point installs a UHF receiver and CAP interface for that channel.
>
> Cost to implement? Government: An interface box for the input side of the repeater fed by CAP equipment at government warning centers. Stations: A UHF receiver (could even be a scanner) and a box to convert CAP messages back into IP that can connect to a CAP decoder.
>
> Downside: CAP data will be in "broadcast" one way mode -- no return path. May not be a problem for local/state CAP messages.
>
> Funding: Some states have spent DHS and other grant money on far more costly projects.  If there is any DHS communications project grant money left, why not ask them to use some of it it to improve warnings by building wireless CAP relay systems?
>
> We should also ask the Department of Justice because of their involvement and dedication to the AMBER program if DOJ grant dollars could be used to build out wireless CAP relay paths that bypass the public internet.
>
> I would like to ask our EAS vendor members on this list to comment on this approach as to their view of its practicality and costs. I do know from personal experience from a project I worked on back in the 80's that "broadcasting" AX-25 packet data on an LA County low band VHF channel does work, and works well.
>
> Meanwhile, here's to an open EAS Forum where we can all ask questions, learn and explore all innovative solutions.
>
> Richard Rudman
>
> On May 19, 2011, at 7:03 PM, Tom Spencer wrote:
>
>> Sudden thought -
>>
>> With the proliferation of updated telemetry and control systems, perhaps
>> there are some few stations that could make a co-operative effort and
>> re-purpose their old 450 TSL links - CAP data would fall under
>> "operational communications", I think, maybe...
>>
>
> _______________________________________________
> This is the EAS Forum Discussion List
>
> Please invite your friends to join our Forum!
> http://lists.radiolists.net/mailman/listinfo/eas
>
> And, remember the main page: http://eas.radiolists.net
>