Re: [tm-pubsubj-comment] Relevance of dc:subject to PSIs

[Murray Altheim <m.altheim@open.ac.uk>]

[sorry to take so long to get back to you - a bit busy at this end.]

Steve Pepper wrote:

> At 15:54 25/04/02 +0000, Murray Altheim wrote:
[...]
>> The exact content of "*****" would be a subject indicator
>> string from a known vocabulary such as LoC, DDC, WorldCat, whatever.
> 
> OK, now I'm starting to understand. "Subject indicator string" is not a 
> term that has precise meaning as yet, so I'll assume you mean "a value 
> from a controlled vocabulary or formal classification scheme" (Dublin 
> Core wording).
> 
> If I understand correctly, you're saying that a dc:subject metadata 
> property on a published subject indicator would be used to state 
> identity relationships between the subject indicated by that PSI and 
> subjects represented in other controlled vocabularies. By extension this 
> means you would be establishing equivalence relationships between the 
> published subject *identifier* (the "Identifier" field, above) of the 
> PSI in question and the value taken from the controlled vocabulary.


Yes, precisely.


> I think the code for Norway in DDC is "914.81", so that could be the 
> value of the dc:subject property of the PSI shown above, right?


Yes.


> And if OCLC were to publish a PSI set for DDC following the PubSubj TC 
> recommendations, "914.81" might instead be something like 
> "http://www.oclc.org/ddc/914.html#81";, right?
> 
> Assuming my interpretation is correct, I have the following comments.
> 
> (1) In general, establishing mappings like this would (obviously) be 
> very useful. If the purpose of Murray's examples was to demonstrate 
> that, then I fully agree.
> 
> (2) However, I'm uncertain whether we should encourage people to 
> establish such mappings in a formal manner through metadata attached to 
> the PSI, since this involves making assertions and goes beyond the mere 
> act of "indicating a subject". It would be better to encourage people 
> who want to establish mappings to do so formally via a topic map, thus:


I think this might depend on authoring and application requirements,
but I don't disagree with your essential point.


>     <!-- excerpt from a topic map whose purpose is to establish mappings 
> between
>          ISO 3166 and DDC subject identifiers for countries -->
>     <topic id="no">
>       <subjectIdentity>
>         <subjectIndicatorRef
>           xlink:href="http://www.topicmaps.org/xtm/1.0/country.xtm#no"/>
>       </subjectIdentity>
>       <subjectIdentity>
>         <subjectIndicatorRef
>           xlink:href="http://www.oclc.org/ddc/914.html#81"/>
>       </subjectIdentity>
>     </topic>


I'm certain you probably mean:

      <topic id="no">
        <subjectIdentity>
          <subjectIndicatorRef
            xlink:href="http://www.topicmaps.org/xtm/1.0/country.xtm#no"/>
          <subjectIndicatorRef
            xlink:href="http://www.oclc.org/ddc/914.html#81"/>
        </subjectIdentity>
      </topic>


But, yes. This could also be done using <association> elements, if

we had an "equivalent" PSI (which I do in various incarnations of
my logic topic maps in development).

Murray

......................................................................
Murray Altheim                  <http://kmi.open.ac.uk/people/murray/>
Knowledge Media Institute
The Open University, Milton Keynes, Bucks, MK7 6AA, UK

      In the evening
      The rice leaves in the garden
      Rustle in the autumn wind
      That blows through my reed hut.  -- Minamoto no Tsunenobu