Hello all,

I am also somewhat unclear as to why these new qualifiers are needed.  I
think I can see how they do something that can't be achieved purely by
relying on encoding schemes, but I'm not sure it's enough to merit their
inclusion.

Classification is defined in the draft as "Term or code from a controlled
vocabulary or scheme designed to aid browsing and search by subject matter"
Keyword is defined as "Terms describing the specific subjects of the
resource"

The comments make it clear that classification is intended to facilitate
broad searches, while keyword is intended for narrow inquiries.

The use of an encoding scheme would be beneficial in both cases, and it is
tempting to see that the structure of the encoding scheme would provide the
meaning as to whether the value is a specific or broad term.  (ie if we
know that a particular thesaurus is used, we can tell where a given term
sits in the hierarchy of broader and narrower terms).  However, this will
only indicate whether the term is broad or narrow in respect of that
particular scheme, not in respect of the resource itself.  Depending on the
controlled vocab used, a given term can either be a top level term or
bottom level one.

As I understand the proposal, it is aiming to provide a means of
contextualising the descriptor as broad ("Classification") or precise
("Keyword") with respect to the resource, rather than the encoding scheme.
A resource may be about a subject much more precise than the encoding
scheme - so the narrowest term available could only be considered a
"classification", or equally, a resource on a broad topic could use a high
level term as a keyword.

That said, I'm not convinced that there is a real need for such a
refinement, and I think the issue will generally be dealt with within a
particular implementation by standardising the encoding scheme(s) to use
and exploiting the structure inherent in them.

Maewyn, perhaps you could confirm my understanding of how it is envisaged
that the two qualifiers be used?


John

John Roberts
Senior Archives Policy Analyst
Statutory Regulatory Group
Archives New Zealand

Phone: +64-4-496 1392
Fax: +64-4-495 6210
Mail: PO Box 12 050, Wellington New Zealand
Email: [log in to unmask]
WWW Page: http://www.archives.govt.nz





Palle Aagaard <[log in to unmask]>@JISCMAIL.AC.UK> on 22/03/2002 19:28:29

Please respond to This list is intended for discussion of the uses to which
      the Dublin Core E <[log in to unmask]>

Sent by:  This list is intended for discussion of the uses to which the
      Dublin Core E <[log in to unmask]>


To:   [log in to unmask]
cc:
Subject:  SV: Subject Qualification [Was: Re: SV: DC Gov Application Profil
      e]

All,

There are now two almost different opinions about Subject qualification. Or
more exact, there is one major question to the need for subject
qualification.

THERE IS NOW A NEED FOR MORE COMMENTS OR DISCUSSION ON THIS TOPIC. ANY
CONTRIBUTION WILL BE APPRECIATED.

Palle









-----Oprindelig meddelelse-----
Fra: Chris Croome [mailto:[log in to unmask]]
Sendt: 21. marts 2002 16:23
Til: [log in to unmask]
Emne: Subject Qualification [Was: Re: SV: DC Gov Application Profile]


Hi

On Thu 21-Mar-2002 at 02:43:31 -0000, Cumming, Maewyn - e-Envoy
e-Government - wrote:
> The main reason we wanted a separate 'Keyword' qualifier is to
> distinguish between terms taken from controlled vocabularies and those
> that aren't.

I thought that a unqualified Subject is one that is not taken from a
controlled vocabulary, for example:

  <dc:subject>Formate Dehydrogenase</dc:subject>

> I know that an encoding scheme can tell you which vocabulary a term is
> from, and allow you to search on terms from a particular vocabulary,
> but it isn't that clear if you simply want to distinguish between
> controlled and uncontrolled terms.

It seems fairly clear to me, the above example might look like this when
qualified:

  <rdf:Description>
    <dc:subject>
      <dcterms:MESH>
        <rdf:value>D08.586.682.075.400</rdf:value>
        <rdfs:label>Formate Dehydrogenase</rdfs:label>
     </dcterms:MESH>
   </dc:subject>
 </rdf:Description>

This example is taken from
http://dublincore.org/documents/2001/11/30/dcq-rdf-xml/

What exactly is the difference between the proposed 'Subject Keyword'
and a unqualified Subject?

And what exactly is the difference between a 'Subject Classification'
and a qualified Subject?

Chris

--
Chris Croome                               <[log in to unmask]>
web design                             http://www.webarchitects.co.uk/
web content management                               http://mkdoc.com/
everything else                               http://chris.croome.net/