REPLY TO 09/16/97 08:57 FROM [log in to unmask] "khsu": SOURCE element
Karen,
I have come to accept that the web implementation of Dublin Core
metadata, in most instances, is intended to describe an original
"document" embodied in web pages and does not address the issue of
reformatted originals. The Source element was a bone tossed to the
reformatting community and would need work to be made useful.
I now believe the surrogate problem can be minimized without much
disruption by loosening the definitions (especially those with the
phrase "in its present form"). All implementations could describe
the intellectual content of the original (or first manifestation).
The implementation that is being worked on by a small group of
librarian and museum representatives (heading for sending something
to Helsinki) involves repeating the whole set of elements to describe
the surrogate or "present form". Simply repeating individual
elements would do the same job for simple retrieval, but it would not
be clear when one is repeating an element, say, for co-authors or to
describe the person who made the text available on the web. By not
requiring a hierarchy, it would have no additional requirement on the
software that interprets the data.
The Web implementation will drive the development of software tools.
By requiring something more complicated, we do ourselves a disservice
(though if one wanted to use a hierarchy for clarity that did not
affect the access to or utility of the data, I don't see a problem).
Currently we are working on a small set of minimal element definition
changes, a narrowing of the Source element, some Resource Type
values, and a set of guidelines for extending the use of the Dublin
Core for non-web resources.
Stay tuned,
Ricky
To: [log in to unmask]
cc: [log in to unmask]
|