SV: RE: Comments to Version 0.8-Reqts Specs

["Greg FitzPatrick" <gf@medianet.org>]

Unfortunately Bob's example is overly simplified.  Though RDF is a promising
technology.  In itself it does not solve all the problems.

What happens if codes are changed at diverging times - which they are?
What happens if the same "Heaven" actually has non-matching physical
perimeters designated by different code systems?

And nation codes are really the easiest case.  What about other possible
geographic areas? And diverging statistical codes in general?

Somehow, some trusted party must take the responsibility of mapping this all
out and guaranteeing its validity and timeliness.  RDF might be a good tool
for the task, though I am not completely sure about that, but the work has
to be done.

For the International/National statistical community the potential gain of
real-time synchronization to the actual business transaction process is
(though remember statistics implies long-term alignment) a realistic
expectation of the Semantic Web - but that doesn't mean its a cinch.

>
> Consider the 'country' 'Buelahland'.
>
> In ISO3166, let's suppose the code is '777' and the name is 'Buelah Land'
> In FIPS 10-4, let's suppose the code is 'HVN' and the name is
> 'CrystalCity'.
>
> Let's further assume that the unique semantic identifier is 'Heaven'. Now,
> let's look at some metadata we might expect to find associated with the
> semantic entity uniquely identified as 'Heaven':
> 	o The ISO3166 code for Heaven is '777'
>  	o The ISO3166 name for Heaven is 'Buelah Land'
> 	o The 'FIPS 10-4' code for Heaven is 'HVN'
> 	o The 'FIPS 10-4' name for Heaven is 'CrystalCity'
> 	o Heaven is 'where GOD is'.
>
> The RDF Specification can be used to represent the above metadata, and the
> application, military or otherwise, can then find the metadata of interest
> to it.  Note that there might be lots of metadata associated with
> a semantic
> entity, only some of which may be of interest to a given application.
>
> Cheers,
>         Bob
>
>
> -----Original Message-----
> From: Jean.Kubler@unece.org [mailto:Jean.Kubler@unece.org]
> Sent: Thursday, April 27, 2000 4:36 AM
> To: Miller; Robert (GEIS); Klynsma; Steven L Mr DISC4/MITRE;
> 'ebXML-requirements@lists.oasis-open.org'; clivio@iso.ch
> Subject: Re:RE: Comments to Version 0.8-Reqts Specs
>
>
> Sorry, the comment below is not correct as the name of a country is an
> importnat
> issue and that the purpose of ISO 3166 is to align all the
> various views on
> the
> semantics of a country code AND name. ISO 3166 is aligned with UN names of
> countries. As soon as you interface with goverment regulations, the
> semantics
> and code about country become relevant. Also, from the
> statistical point of
> view
> and comparability of statistics, ISO 3166 is key for the
> collection of data.
> There might be other views on the very relevance of iSO 3166.
> Regards, Jean
>
> -------------------------------------------------------
> UN/ECE  Jean E. KUBLER
>         Trade Facilitation and Electronic Business
>         Tel: +41 22 9172774 http://www.unece.org/cefact
> -------------------------------------------------------
>
>
>
>
> ____________________Reply Separator____________________
> Subject:    RE: Comments to Version 0.8-Reqts Specs
> Author: "Miller; Robert (GEIS)" <Robert.Miller@geis.ge.com>
> Date:       4/25/00 6:11 PM
>
> MILR: The code (or name) used to represent a country is not all that
> important.  What is important is that the country have a unique semantic
> identifier; that the unique semantic identifier be accesible through
> whatever reference to it is made; and that the metadata
> associated with the
> unique semantic entity support the applications that reference the entity.
> It should be acceptable in ebXML to reference the country in any
> manner that
> yields the correct semantic identification.  The metadata associated with
> the semantic entity may include information useful in deriving other codes
> (or names)that may be used to reference the entity.
>
> Cheers,
>         Bob
>
> -----Original Message-----
> From: Klynsma, Steven L Mr DISC4/MITRE
> [mailto:Steven.Klynsma@hqda.army.mil]
> Sent: Monday, April 24, 2000 6:44 AM
> To: 'ebXML-requirements@lists.oasis-open.org'
> Subject: Comments to Version 0.8-Reqts Specs
>
>
> Below are US Army comments on the Version 0.8 Requirements Specs
>
> Para 2.3 refers to Internet RFC 1766.  RFC 1766 has not progressed beyond
> the status of a "proposed standard" despite having been around since 1995.
> Concern centers around whether it is viable for inclusion in the Spec.
>
> Same Para also requires ISO 3166 for country name codes.  There is current
> debate within DoD over acceptance of ISO 3166.  DoD currently is
> standardized on FIPS 10-4 which is not compatible with ISO 3166.  Army
> recognizes, that in a document designed to implement global commerce, ISO
> 3166 is appropriate, but conversion to ISO 3166 for the Army is
> likely to be
> economically impractical at this time.
>
> Steve Klynsma
> DISC4/Mitre
> Architecture Mgmt Div
> (703) 614-6180
> DSN 224-6180