Re: [emergency] Groups - CRSRegistry Document by Galdos.doc uploaded

[R. Allen Wyke <emergency-tc@earthlink.net>]

Carl: apologize for just now getting back to this, but I needed to take 
some time to let some of the GIS-related comments and discussions that 
have been floating around to settle into my head. And while I admit, as 
I have done many times before, I am not GIS expert, I *think* I am able 
to understand the purpose and idea behind this document. As such, I had 
a question.

In terms of how the Units of Measures are created, which if I 
understand correctly are used to create a given CRS, I am curious as to 
why these units (meter, degrees, etc.) where not defined as derived 
data types? It would seem that by having a UOM dictionary, which would 
be essentially a given CRS, that was nothing more than a list of these 
data types, one could simply reference, through Namespaces, that UOM 
dictionary in instance documents to therefore accomplish the same thing 
- rather than defining your own referencing scheme (srsName, codeSpace, 
etc.). Or, if the author of a given XML dialect, such as our authoring 
of CAP, wanted to explicitly specify which CRS' that were allowed, we 
would only need to import the applicable UOM dictionary(ies) to do 
that.

Just wondering if I was misunderstanding something, as it seemed to me 
the same thing could be accomplished by simply using some of the 
features of XML Schema.

Allen

On Apr 10, 2004, at 4:42 PM, creed@opengis.org wrote:

> The EM TC requested more information about how the OpenGIS Consortium 
> specifies coordinate reference systems (CRS). This request was in 
> relationship to CAP. This document provides basic information about 
> the prototype OGC CRS registry and how an application or encoding can 
> reference this registry.