Oh, boy. This really points out the problem. Carl is right -- by
considering metadata to be a resource in its own right (which I
thought we did), we acknowledge that its characteristics as an
information resource can be described in turn.
What David describes is trying to shoehorn the description of the
metadata into the metadata itself, something that the 15 DC elements do
not provide for. Using Relation to point, not to another information
resource, but to *this* resource's creator is clearly a kludge. We
should resist the temptation, because IF Relation can be used, in effect,
to define new elements, there will be no end to the kinds of Relations
we can concoct.
This is not exactly the same case, to my mind, as the 1:1
problem -- or at least the 1:1 problem that I care about. The
information resource and its metadata are a different case than
instantiations/transformations/versions/what-have-you of an information
resource. What I thought 1:1 was did not involve new data elements, or
even new qualifiers. 1:1 is troublesome enough without bringing the
relationship between the description of content and the description of
description into it.
--Robin
"I guess this wouldn't be good time for me to suggest we all
join hands and sing Kumbaya?"
--Babylon 5
On Fri, 22 Jan 1999, David Bearman wrote:
> Dear Paul,
> As you know, this is why we had the 1:1 rule.
> If the DC metadata set describes the book, the relation element might point
> to a metadata author.
> If the metadata set describes the cataloging, its creator is the metadata
> author.
> Both are possible. But in the ook metadata set, we have no other way to
> record the metadata author now.
> David
>
> At 03:21 PM 1/22/99 +0000, Paul Miller wrote:
> >On Fri, 22 Jan 1999, David Bearman wrote:
> >
> >> If I create DC metadata for a book, the creator element, according to our
> >> definition, references the book.
> >
> >True. But Carl's right as well, in that the line between 'metadata' and
> >'resource' is rarely as clear as in the catalogue record/book example you
> >give.
> >
> >A museum collections management record, for example, is metadata for an
> >object in the collection. It's also a creative work (of fiction, in some
> >cases...! ;-) ) in its own right.
> >
> >In the fuzzy, hazy, world in which we operate, where _I_ certainly would
> >be hard-pressed to definitively label any of my stuff as a 'resource' or
> >'metadata', it is, perhaps, easier to consider the single concept of a
> >creator (or agent, or whatever) of 'stuff' (to steal terms from Godfrey).
> >
> >If I, Paul Miller, am describing a BOOK by Carl Lagoze, the Creator is
> >Carl Lagoze. If I, Paul Miller, am describing the catalogue record for
> >that book by David Bearman, then the Creator is David Bearman.
> >
> >I've not had to decide whether or not I'm describing 'data' or 'metadata'
> >(because both book and record are in some cases 'data' and in others
> >'metadata', depending upon perspective), and use of Title and other
> >elements allows unambiguous identification of that which I'm describing
> >('Carl's great book' or 'Dave's Catalogue record of "Carl's great book"').
> >
> >Paul
> >
> > -- dr. paul miller - interoperability focus - [log in to unmask] --
> > u. k. office for library and information networking (ukoln)
> > tel: +44 (0)1482 466890 fax: +44 (0)1482 465531
> > ---------------------------- http://www.ukoln.ac.uk/interop-focus/ --
> >
> >
> >
> >
> 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
>
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 .............
%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
