MHonArc v2.5.0b2 -->
emergency-msg message
[Date Prev]
| [Thread Prev]
| [Thread Next]
| [Date Next]
--
[Date Index]
| [Thread Index]
| [List Home]
Subject: RE: [emergency-msg] [Fwd: CAP Test Scenarios]
From: Art Botterell <>
To:
Date: Wed, 17 Sep 2003 20:13:27 -0700
Jerry -
That's a real good test case, thanks!
One thing I notice is that the coordinate pairs in your polygons are
in lon/lat order instead of lat/lon, which I think is more common.
(Looking at the spec, it doesn't seem like we've been explicit about
that, so it goes on the to-talk-about list.) Clearly we'll need to
agree on this format. Anyway, once I turned the coordinates around,
it worked fine with my code.
Good use of the multiple Info block structure. I noted with grim
amusement the entirely possible situation where the city of St Louis
ordered shelter-in-place but the county (which I'm assuming is
surrounding the city) orders evacuation. Such is life. Of course,
under our "rule of primacy" for Info blocks, a location-aware
receptor acts on the first Info block that applies to it... so if I
were inside the city, I'd get the shelter-in-place order and not the
county's evacuation. As it should be. (I didn't actually plot the
polygons on a map, so if there's any other subtlety there, I missed
it.)
I do notice that in the various blocks you have different <onset>
times but the same <effective> time on all. I assume the intent here
was to depict the changing conditions as the plume moved over time.
I'd suggest using the <onset> time to describe when the event will
occur, but the <effective> time to indicate when the action
recommended in the warning message should begin. (Which may be what
you did here... just thought I'd mention it.)
And for those items that appear to be basically just technical
information... e.g. the Hazard Area item... I'd suggest still
including an <instruction>, even if it just says something like
"Technical information only, no action required." One never wants to
announce something that looks like a hazard without offering guidance
on the appropriate response. (Also, one might think about the
Urgency/Severity/Certainty coding for this item... viewing it from
the public's point of view. If it's necessary to send out both
public-warning with technical-information data, that might be an
appropriate use-case for sending two separate messages, possibly with
different <scope> values.)
Anyway, a lot of that gets into issues of public-information practice
more than CAP technology. With the exception of the lat/lon order
question, that looks like real interoperable CAP to me.
- Art
At 11:56 AM -0500 9/17/03, Weltman, Jerry wrote:
>I have posted a CAP message representing a 10-gallon release of chemical
>agent VX. The release affected 3 counties around St. Louis: St. Clair, St
>Louis City, and St. Louis County. The CAP message has an info block for each
>affected county. The CAP message also has an info block representing the
>footprint of the plume risk envelope. So, 4 info blocks in all.
>
>I have attached the XML that was posted and a bitmap of what appears on my
>screen.
>
>The CAP was FTP'd to Blue292 with file name IEM_CAP_11. It was posted to
>DMI-S, with ID of IEM:11
>
>Please let me know if you are able to work with this message. If you display
>info on a GIS, I would like to know if the polygon looks the same as the
>outer envelope on the attached .bmp.
>
>Also, if you have suggestions on a better use of the CAP message fields,
>please let me know.
>
>Thanks,
>Jerry Weltman
>
>
>
>
>
>
