RE: [emergency-rim] The value of "Unknown" for Urgency, Severity, Certainty in CAP and Elsewhere

["Paulsen,Norm [Ontario]" <Norm.Paulsen@ec.gc.ca>]

I totally agree with Art's comments on this in CAP. Without USC being mandatory, it would make it hard for CAP to be truly interoperable. For that reason "unknown" is necessary. I also agree that considering USC values to be a ranked set of values is a poor practice. In our implementations we use the following...
 
If Urgency is "Immediate" or "Future" then do this
 
and not
 
If Urgency is equal or greater than "Future" then do this
 
Its been difficult to communicate this to some. But we need to rather than change "unknown", or worse remove it and make USC optional.
 
Its also poor practice to just use "unknown" always in USC to get around the mandatory use of USC and then go create a <parameter> to do a similar thing. Nothing wrong with the <parameter> itself, the problem is in ignoring the reason for mandatory use of USC in the first place. They are mandatory for very good reasons.
 
Norm
 
 

________________________________

From: emergency-rim@lists.oasis-open.org on behalf of Rex Brooks
Sent: Fri 2/8/2013 2:46 PM
To: Art Botterell
Cc: Eliot Christian; Doug Allport; rex.brooks@ncoic.org; emergency-rim@lists.oasis-open.org; Jacob Westfall; Christian, Eliot; Elysa Jones
Subject: Re: [emergency-rim] The value of "Unknown" for Urgency, Severity, Certainty in CAP and Elsewhere



Thanks Art,

It is good to be reminded about the issues considered and decisions
taken in previously work on the USC issues.

Would it be possible for you to join us on the next RIM call on the 21st
at 12:00 p.m. Eastern?

We are considering putting the the USC elements into the CommonTypes
collection to be used consistently by any new or future versions of
standards.

Cheers,
Rex

  On 2/7/2013 1:35 PM, Art Botterell wrote:
> Definitely agree, Eliot... we don't want to impose a few users' bad
> practices on everyone.
>
> My sense is that some folks get confused by the fact that in each of
> the USC 5-level codings the first four values represent levels on a
> continuum, while the fifth ("unknown") really isn't a level at all,
> but indicates a lack of knowledge.
>
> The underlying philosophy was that the alert format should never
> require the author to lie.  It was deemed better to enable alert
> originators frankly to admit that they didn't know something than to
> force them to make up pseudo-information... or else worse, to withhold
> or delay a warning while they tried to perfect their knowledge.
>
> Of course it might be argued... and in fact it was, very early on...
> that a semantically purer way to deal with this might be to give each
> of the USC elements just four levels and make them optional, with no
> instance being equivalent to "unknown".  But that was viewed as an
> invitation for lazy implementers and/or originators to sidestep
> specifying USC values at all, which would break a lot of content-based
> decisions downstream about which particular delivery mechanisms
> (sirens, EAS, etc.) should be used to disseminate a given alert.
>
> So we went with what appeared the lesser of two evils. We made the USC
> elements mandatory, while still empowering folks to admit explicitly
> that they don't actually know the correct value if that's the case.
>
> While I'll acknowledge that some folks seem to have difficulty
> accepting that "unknown" means exactly what it says, I'd recommend
> educating them in preference to imposing their error on everyone else.
>
> - Art
>
> On Thu, Feb 7, 2013 at 12:21 PM, Eliot Christian <echristian@usgs.gov> wrote:
>> Perhaps it would be useful to just add a caution against some
>> common misuses of the value "unknown". Here's some text to
>> consider: "For a value coded as 'unknown' there is no basis to
>> treat it as having a minimal, a maximal, or any other value".
>>
>> Eliot
>>
>> At 02:56 PM 2/7/2013, Doug Allport wrote:
>>
>> Rex et al
>>
>> In CAP 1.2 we added "to no known" to the statement associated with CAP
>> Severity Minor; it now reads "Minimal to no known threat to life or
>> property". I believe we have a clear value for "insignificant".
>>
>> I look at the use of "unknown" as A. Lazy or B. Managing issuer risk, as it
>> may be easier to defend the ongoing practice of using unknown than it is a
>> stated value that proves to be less or greater than it should have been. If
>> we are to add a value, we might need "Not willing to state one".
>>
>> My two cents...
>>
>> Doug
>>
>>
>>
>> From: Rex Brooks <rex.brooks@ncoic.org>
>> Reply-To: "rex.brooks@ncoic.org " <rex.brooks@ncoic.org>
>> Date: Thursday, 7 February, 2013 1:04 PM
>> To: " emergency-rim@lists.oasis-open.org" <
>> emergency-rim@lists.oasis-open.org>, " art.botterell@west.cmu.edu" <
>> art.botterell@west.cmu.edu>, Jake Westfall <jake@jpw.biz>, "Christian,
>> Eliot" <echristi@usgs.gov>, Elysa Jones <ElysaJones@yahoo.com>
>> Subject: [emergency-rim] The value of "Unknown" for Urgency, Severity,
>> Certainty in CAP and Elsewhere
>>
>> Hi Everyone,
>>
>> In today's EM RIM meeting we discussed the issue of how the value of
>> "Unkown" for Urgency, Severity and Certainty is commonly taken, how the
>> weather service uses it and how it is viewed by other domains in
>> Emergency Management. We are thinking of changing it or defining it more
>> emphatically or removing it in favor of other value terms which might be
>> better understood.
>>
>> So we are asking your opinions on and experience in this.
>>
>> Best Regards,
>> Rex
>>
>> --
>> Rex Brooks
>> GeoAddress:
>> 1361-A Addison
>> Berkeley, CA 94702
>> Phone: 510-898-0670
>>
>>
>> ---------------------------------------------------------------------
>> To unsubscribe from this mail list, you must leave the OASIS TC that
>> generates this mail.  Follow this link to all your TCs in OASIS at:
>> https://www.oasis-open.org/apps/org/workgroup/portal/my_workgroups.php
>>
>>
>
>


--
Rex Brooks
GeoAddress:
1361-A Addison
Berkeley, CA 94702
Phone: 510-898-0670


---------------------------------------------------------------------
To unsubscribe from this mail list, you must leave the OASIS TC that
generates this mail.  Follow this link to all your TCs in OASIS at:
https://www.oasis-open.org/apps/org/workgroup/portal/my_workgroups.php