It strikes me as impossibly complex to construct a rule set that says
tag A means this if tag B has value such-and-such,
but tag A means something else if tag B ihas value something-else. The
Dublin Core is meant to be a simple description standard for electronic
resources.
There is a legitimate desire to use it to expose the existence of
collections of physical objects in the electronic environment (museum
collections on the Web, for example, or the photograph collection that
NYPL is digitizing). I see no profound contradiction here. The creator
is the individual who took the picture, or carved the stone, not the
person responsible for the surrogate.
What do your users want to find? Its true that we cannot always know
this in advance. Take your best shot.
On Monday, September 22, 1997 6:15 PM, [log in to unmask]
[SMTP:[log in to unmask]] wrote:
> Ricky,
>
> Perhaps, there is another reason for abandoning the approach of
using
> Resource Type as an indicator for interpreting the content of the rest
of the
> elements. If I understand it correctly, according to this approach,
two records
> would be needed, one for the original and one for the digitized (or
any
> intermediate medium). First of all, creating a separate metadata
record for the
> original would be an additional workload. Secondly, the description of
the
> original would be separated from that of the digitized. Granted, the
two records
> could be linked by URL's, but readers must be patient and persistent
enough to
> go one more step. I think this two-record (or multiple-record)
approach is
> something that the cataloging community has been struggling to avoid.
>
> Karen
>
> ______________________________ Reply Separator
_________________________________
> Subject: Re: Resource Types
> Author: "Ricky Erway" <[log in to unmask]> at ~Internet-Mail
> Date: 9/22/97 11:25 AM
>
>
|