On Wed, 29 Sep 1999, Ray Denenberg wrote:
> Three approaches to accomodating 'audience' have been considered:
> (1) Add it as the 16th element.
> (2) Wedge it into one of the existing 15 elements: subject, description,
> or coverage.
> (3) Include it in element sets that are supersets of, or which
> "import", DC.
>
> If 'audience' were a core element, (1) or (2) would be appropriate
> (which one, (1) or (2), is beyond the scope of this message) but
> assuming that 'audience' is not core, that leaves approach (3),
> which is criticized because of the possibility of interoperability
> problems caused by the potential proliferation of 'audience' elements
> in different element sets.
Let me add that option 2 has interoperability problems as well, just of a
different nature. For example, given the spatio-temporal definition of
Coverage, if you stick audience in it, you will have violated the commonly
understood semantics of the element, causing it to lose its meaning. This
diminishes interoperability.
Not every data element each of us requires, nor even every data element
many of us require, needs to be shoehorned into DC. Let's work on common
extensions rather than stretch the existing elements out of recognition.
If you have to hunt this hard for a place to stuff it, it probably belongs
in an extension.
--Robin
Robin Wendler ........................ work (617) 495-3724
Office for Information Systems ....... fax (617) 495-0491
Harvard University Library ........... [log in to unmask]
Cambridge, MA, USA 02138 .............
%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
|