[GILS Logo]

gils: Re: GILS Profile version 2

gils: Re: GILS Profile version 2

Re: GILS Profile version 2

Turner Fay (fst@ITS.NLC-BNC.CA)
Fri, 06 Sep 1996 12:10:00 -0700 (PDT) 

Date: Fri, 06 Sep 1996 12:10:00 -0700 (PDT)
From: Turner Fay <fst@ITS.NLC-BNC.CA>
Subject: Re: GILS Profile version 2
To: GILS Forum <gils@cni.org>
Message-Id: <3230774E@its.nlc-bnc.ca>


In response to Eliot's comments:

> > 3. Controlled Subject Index: the structure of Controlled Subject Index
> > should be realigned to parallel the structure of Place in Spatial 
Domain.
> > They both do the same thing - cite a thesaurus and provide terms from 
the
> > thesaurus.  Except Controlled Subject Index has the added data 
subelement
> > Subject Terms Controlled which I think is redundant.
>
> Consistency would be nice, and I do think we should follow whatever is
> the prevailing convention. If I understand your proposal correctly, we
> should have:
>
>   Controlled Subject Index (Repeatable) This element is a grouping of
>   subelements that together provide any controlled vocabulary used to
>   describe the resource and the source of that controlled vocabulary.
>
>     Subject Thesaurus (Not Repeatable) This subelement provides the
>     reference to a formally registered thesaurus or similar authoritative
>     source of the controlled index terms, plus the terms themselves.
>     Notes on how to obtain electronic access to (e.g., a URI) or copies of 

>     the referenced source should be provided, possibly through a Cross
>     Reference to another locator record that more fully describes the
>     referenced resource and its potential application to locating GILS
>     information.  Each term is provided in the subordinate repeating
>     field:
>
>       Controlled Term

Yes, this is what I had in mind.  However, may I further suggest that the 
definition for Subject Thesaurus be made clearer by changing the wording to: 


This subelement provides the *name* of a formally registered thesaurus or 
similar authoritative  source of the controlled index terms (omit 'plus the 
terms themselves').  Notes on how ...

This is closer to the definition for Place Keyword Thesaurus.

A definition for Controlled Term can be adapted from the one for the now 
defunct Subject Terms -Controlled element: My suggestion:

Controlled Term: Term(s) drawn from a controlled vocabulary source, such as 
a thesaurus, to describe the resource.

> > 4. Available Linkage, Linkage Type.  Structuring this element and its
> > compnents posed a real problem when drafting the DTD.  Available Linkage 
is
> > repeatable and contains the subelements Linkage (repeatable) and Linkage 

> > Type (repeatable).  We suggest that Linkage Type be not repeatable.
> > Instead, for each different Linkage Type value, you create a new 
instance
> > of the Available Linkage data element.  This allows you to couple the
> > Linkage with the Linkage Type.
>
> I think this will work for Available Linkage. What do you propose to do
> with Cross Reference Linkage?
>
I would think that the same rules should apply. There can be more that one 
Cross Reference Linkage (i.e. more than one URL) for every instance of Cross 
Reference Type.  And there can be more than one Cross Reference Type (html, 
MSWord, WordPerfect)  for a Cross Reference Title.  I am not sure though on 
how to allow for  repeats of this couple (Cross Reference Linkage and Cross 
Reference Type) for each Cross Reference Title without having to repeat the 
complete Cross Reference  grouping for the same title.  Any ideas?

> Eliot Christian
> <echristi@usgs.gov>
>
Fay

Fay Turner
Information Technology Services
National Library of Canada

e-mail:  fay.turner@nlc-bnc.ca
tel:        (819) 994-6830

[CNI Home Page]