Pretend you see an emoticon denoting calm, stability, and moderation.
No flaming here, only earnest concern.
Like David, I also believe Jim is correct, but interestingly enough,
to my mind this supports the position that 1:1 is untenable.
It is not that I am opposed in principle to 1:1, or that I believe people
should not be allowed to create DC records according to its precepts.
Rather, I believe that 1:1 falls clearly and precisely into the realm of
cataloging rules, which, as David rightly points out, are unlikely to be
followed in DC (or at least in the DC most of us thought we were working
to develop).
Many communities already have cataloging rules, and one piece of
cataloging can serve only so many masters. If we want metadata
created in many environments to enrich the universe of DC metadata in aid
of resource discovery, DC cannot become overly prescriptive, cannot force
a "my way or the highway" approach to descriptive practice.
An earlier discussion on the 1:1 list revolved around whether every
fragment of a resource should be cataloged as a resource in its own right.
This thread made absolutely clear that there is no consensus on what 1:1
really means in practice. In the end, people will describe what *they*
want to describe, for their purposes and the purposes of their user
community. That means they may describe a TIFF of an Ansel Adams
photograph as having been created by Ansel Adams. Who's to say they're
wrong?
DC should stay out of the cataloging rules business as much as possible,
in my opinion. Because we cannot agree on what 1:1 actually means in
practice and cannot agree that DC data can, will, or must be created
according to whatever it means, I believe it should be dropped from the DC
discussions and carried forward, if at all, in cataloging rules
discussions independent of DC.
--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 .............
On Tue, 13 Apr 1999, David Bearman wrote:
> At 03:08 PM 4/13/99 -0400, James Weinheimer wrote:
> "In the case of 1:1, that can mean very different things to different
> people.
> For example, how do various people view an article in a magazine
> concerning medieval art with 10 pictures?
> a library cataloger may view the magazine (which contains hundreds of
> volumes, and thousands of articles) as 1 item.
> An indexer may believe that the article with ten pictures is 1 item.
> A slide cataloger may believe that each picture is one item.
> A specialist indexer (e.g. the Index of Christian art) may believe that
> each picture is made up of dozens of separate parts (swords, draperies,
> trees, etc.)"
>
> I believe Jim is correct and this was the reason I proposed originally the
> "1:1 rule" in Helsinki - since each of these communities will make (has
> made) metadata corresponding to the "one" that they see, it is important
> that the data in that metadata cluster refer to that one thing (whether the
> journal, the issue, the image, the iconographic subjects). Then each
> metadata cluster declare an appropriate "relation" to other metadata.
>
> The alternative, I believe, is to devise or adopt a set of cataloging rules
> that everyone must use, but 1) DC is not the kind of 'system' in which
> cataloging rules are likely to be followed and 2) we would sacrifice
> getting lots of pre-existing datasets translated into DC if we insisted a
> given cataloging practice. Far better, I believe, to recognize that every
> community wants to describe different things and to ask each to then
> describe one thing in one instance of a DC element set.
>
> But I think this discussion should be continued on the 1:1 list...
> David
>
>
> David Bearman
> President
> Archives & Museum Informatics
> 2008 Murray Ave, Suite D
> Pittsburgh, PA 15217 USA
> Phone: +1 412 422 8530
> Fax: +1 412 422 8594
> [log in to unmask]
> http://www.archimuse.com
>
%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
|