On Sat, 4 Dec 1999, Stuart Sutton wrote:

> QUALIFIED DC.description: Of course, the example of qualified 
> DC.description.agelevel somewhat alleviates the problem I see with 
> unqualified DC noted above; however, even when we reach agreement on 
> appropriate qualifiers (agelevel, etc.), I still think it is problematic.
> We 
> have watched the discussions around "agent" and the justifications for 
> "creator", "contributor" and "publisher" (all agent forms).  So why not use
> a
> single DC.agent element with a role qualifier?  Why elevate three agent 

Hello all,

I've been meaning to post a mail to this list regarding the 'place' of 
audience ... whether as a qualifier to description or as a a separate
element.... or whatever else it might be! Apologies if some of this was
covered at DC7, I wasn't able to be there.

I'm chairing the Subdesc WG and at present and we have a proposal on the
table for audience as a qualifier of description. This proposal is
motivated largely, I think, by the need to put it somewhere and reluctance
to take the step of adding 'a new element'. I think there needs to be some
liaison between our groups :-)

Personally I think that it is significant that the content (value) of
'Description' is unstructured, it is typically a piece of text which acts
as a summary, abstract or other account of the content, describing the
content in a way that the metadata creator thinks is useful. I imagine
that such a description would usually be indexed as 'uncontrolled
keywords'. It seems odd to me to qualify this unstructured value with a
piece of information such as audience where the content (value) of would
usually be chosen from a controlled vocabulary.

On the other hand I'm not sure I want to argue for a 'new element' and all
that implies...

Is there any way we can explore using the IMS metadata structure here, is
there some way a DC description can just add on the IMS element
'IntendedUserRole'?? (this is what I thought might have been discussed at
DC7??)

>From the DC viewpoint this would mean audience was dealt with by 'an
extension' taken from another metadata element set. How big a problem
would this be for software tools? presumably most software deals with
'local extensions' now ..... such as date metadata created, ownersip of
metadata etc so IntendedUserRole could be like one of these. 

Presumably this 'odd element' would not make much sense for an 'IMS
aware' software tool. Presumably if one had created a DC description of a
resource  one would want to be able to(automatically) map all DC
elements to their equivalent IMS elements and add that one extra so that
IMS tools/systems could make sense of it??

Anyway I'm not sure that gets us very far just now, when we are trying to
decide on qualifier proposals by Dec 10th :-(

For info this is the relevant note regarding audience extrated from the
SubDesc proposal [1] 

<quote>
There has been discussion within the dc-subdesc list and on dc-general as
to whether an 'audience' (IntendedUser) qualifier should be introduced,
and if it should then where it fits best: with Description or Subject or
Coverage. Various implementations have chosen differentsolutions. For
example OCLC CORC project qualifies Description with 'Audience', EDNA uses
a local extension 'EDNA: Userlevel' , GEM uses its own local extension
'GEM:Audience'. In addition there is an IMSelement 'IntendedUserRole'. The
suggestion that an IMS element might be used seems problematic at this
stage. Taking into account current technology and the lack of deployed RDF
tools, it is not easy to see how singleelements from other schemas (e.g.
IMS)  can be'adopted' and used in conjunction with DC elements. With
deployment of RDF tools in the longer term this will be a possible
solution. An alternative, now, is to use a 'local extension' as GEM has
done with GEM:audience. However as several implementations need to
describe audience it would seem better for the sake of interoperability to
agree on a DC element qualifier.

</quote>

Rachel

1. http://www.mailbase.ac.uk/lists/dc-subdesc/files/wd-subdesc-qual.htm
 
---------------------------------------------------------------------------
Rachel Heery
UKOLN (UK Office for Library and Information Networking)
University of Bath                              tel: +44 (0)1225 826724
Bath, BA2 7AY, UK                               fax: +44 (0)1225 826838
http://www.ukoln.ac.uk/






%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%