I'd thought I'd add my two cents on this topic.
I agree with Bronwyn Lee about the
linkingIntellectualEntityIdentifier likely pointing to a descriptive
cataloging record. Since the intellectual entity is an abstraction,
one can't actually point to it; one would have to point to a
description of the entity. I think this is a sensible, practical
approach that should work pretty well. Such a use of existing
metadata would save the repository staff the trouble of creating its
own description of the intellectual entity as such.
I do think this raises some issues as that descriptive record, if it
is a traditional library catalog record or like thing (bibliography
entry, discography entry, etc.), will be a description of another
physical object (digital or not) that also is a generalized
description of the edition (I'm thinking of published
entities--books, articles, scores, prints, etc.; unique or
unpublished entities would be another matter) as a whole, and that
serves as a description of the work as a work (by naming the author,
artists, etc. and giving a title for the work, and other information,
etc.) that provides a relatively clear or recognizable citation for the work.
Another approach would be to use an existing description (e.g. a
library catalog record) as a starting point for the creation of a
more work (FRBR-sense) oriented record by extracting just those
elements that pertain to the work--artist, author; title, titles,
form of content (image, text, score, map, etc.), date of creation,
and, specific matters for certain classes of material, for instance
the language of the entity if it is textual material. This would
generate a less ambiguous description of the intellectual entity as
an abstract concept of the work, but it would not be possible to make
converting a traditional description into a FRBR work citation fully
automated. So, the cost of the "better" solution leads me to think
that simply pointing to an existing description is sufficient.
At 6/28/2006 04:33 PM, Zhiwu Xie wrote:
>A question on the use of the linkingIntellectualEntityIdentifier:
>Does the Intellectual Entity an abstraction of the articles, books, etc,
>that the same Intellectual Entity will have multiple representations
>such as one digital object in my repository, another one in your repo,
>and one hard copy in a journal residing on the basement of the science
>library on campus?
>If this is the case, the value of the
>linkingIntellectualEntityIdentifierValue shall be something like a DOI
>number, an ISBN number or alike, instead of an object identifier in some
>repo since that's the ID of just a digital object but not the
>abstraction of that digital object, am I right?
>If this is the case, the usage notes on p.2-70 confuses me:
>"... This may be a link to descriptive metadata that describes the
>Intellectual Entity or some other surrogate for it that can be
>I had thought the identifer is for the abstract intellectual entity but
>not one of its representation, such as the descriptive metadata. Also:
>"This link will likely be to an identifier of an object that is at a
>higher conceptual level than the object for which the metadata is
>provided, for example, to a collection or parent object."
>Again here the identifier may be for a concrete digital object.
>If this is to link two concrete object, isn't it better to just describe
>it in the "relationship"?
>Also, is there a way to describe the nature of the link, e.g., this link
>is from a MODS metadata object to the intellectual entity, that link is
>from a full-text pdf representation to the intellectual entity? Is there
>a way to convey the descriptive metadata and/or the full-text nature of
Yale University Library
130 Wall Street P.O. Box 208240
New Haven, CT 06520-8240
phone: (203) 432-4947
fax: (203) 432 7231
e-mail: [log in to unmask]