I actually don't think that including an option to describe the original
with reproduction information in a note will keep RDA from being acceptable
to the Dublin Core and other metadata communities. For one thing, it is
just an option, not a rule. For another thing, I am not really advocating a
MULVER approach here.
If I understand it correctly, MULVER is when you represent two different
formats on one bibliographic record. We do use a MULVER approach here at
the University of Chicago if we have (and are keeping) the print version.
The catalog record represents the print version with access to the
electronic version added via 530 and 856. There is no subfield h
[electronic resource] in the 245 and the 2nd indicator of the 856 is 1 for
version of the resource.
What I am talking about for this option is a record which represents the
electronic version, but the date and place of publication, publisher,
standard numbers, etc. are for the original. There is a subfield h
[electronic resource] in the 245 and the 2nd indicator of the 856 is 0 for
the resource itself instead of 1 for version of the resource. There is a
533 which says that this is an electronic reproduction and gives
information on who digitized it, where, and when.
The reason I think this is important is that I believe users who are
looking for digitized resources are going to want to search by the original
date, publisher, standard number, etc., rather than the date it was
digitized, where it was digitized, by whom it was digitized, etc. We ran up
against this same issue when we were developing our Guidelines for Use of
Dublin Core in University of Chicago Library Projects, available at:
http://www.lib.uchicago.edu/staffweb/groups/metadata/guidelines/
We established a local qualified Dublin Core element publisher.original for
digital reproductions because publisher had been defined as "An entity
responsible for making the resource available", which had been interpreted
here to mean whoever digitized it. We did not have to establish a local
qualified Dublin Core element date.original because the date element had
always been used here to refer to the date of the original, as opposed to
the date it was digitized, in spite of the fact that the actual Dublin Core
date element had been interpreted here to mean the date the resource was
digitized.
I would like to understand your objections a little better, though. Does
this have to do with the 1:1 principle? If so, the problem I see is that
sometimes we are not keeping the original, such as when we are digitizing a
book for preservation purposes. In that case it does not make sense have a
record for the print version in our catalog. If it were, theoretically
someone could search using the date or place of publication, publisher, or
standard number of the original, pull up the record for the original which
would then link to the record for the digital version. However, since we
don't have a record for the print version, we want to have the date or
place of publication, publisher, or standard number of the original in
fields that are easily searchable in the record for the digital version.
Basing the description on the original, with information on the
reproduction relegated to a note, does that.
Renette
At 04:14 PM 1/12/2006, Diane I. Hillmann wrote:
It's been a long time since I agreed with Mac on much regarding
cataloging, but I'd like to agree strongly with him on this one ...
I copied everyone on Renette's cc: list, though I'm sure some of
them will not thank me for that.
My reasons for agreeing is that I have been spending the last five
years of my life working to aggregate metadata from a lot of
different sources, very little of which came from libraries. I am on
this list representing the Dublin Core community, and I suppose, by
extension, the other metadata communities who lack content standards
and are hoping that there will be something in RDA that can be useful
in their world.
It seems to me that retaining the convention of multiple versions in
the way Renette suggests is one sure way to keep RDA an exclusively
library content standard. It continues the hoary tradition of working
around the limitations of library catalogs (and library vendor
software) by kludging the data, which makes any re-use outside of
libraries extremely problematic. Yes, I agree (before somebody nails
me on this) that there are problems expressing to users the important
context of the original work in a traditional catalog environment
where reproductions are described separately, but I have some hopes
that the expanding use of FRBR models in catalogs will assist us in
this effort.
I would hate to see an option to create multiple version records--it
should be viewed as an aberrant practice, with terrifying health
warnings.
Diane Hillmann
Research Librarian
Cornell University Library
I am sending this to several other lists besides RDA-L. Apologies for
duplication. I am also cc'ing to some people who have been involved in
discussions of the Digital Library Federation's Registry of Digital Masters
Record Creation Guidelines and who I haven't seen on the RDA list. If you
wonder why in the world you got this, that's probably why! My comments
are below Mac's original post with RD beside each paragraph.
Renette
At 11:19 AM 12/30/2005, Mac Elrod wrote on the RDA-L list:
The LCRI turning AACR2 on its head for reproductions fortunately (from
my point of view) did not make it into RDA. In 2.2.2 (choice of prime
source), 2.3.1.5 (title), 2.5.0.4 (edition statement), 2.10.9.4
(series), as well as 2.12.0.4 and 4.10.2 we are told to choose the
facsimile in hand for transcription, rather than the item reproduced.
This nicely avoids the problem created by reproductions of
reproductions, attempting to follow the LCRI.
RD - I would actually like to see an option to describe the original and
include the reproduction information in a note, as in the current LCRI
1.11A, particularly for electronic reproductions. I believe that users are
most interested in the date, place of publication, publisher, etc., of the
original rather than when and by whom it was digitized. An ISBN in a note
(as specified in 2.12.0.4) is not as easily searchable as an ISBN in an 020
field.
RD - We are using a single record approach here at the University of
Chicago, adding access to the electronic version on the record for the
print version. However, if we don't have the print version (or we are
discarding the print version), we create a separate record for the
electronic version. It is much easier to derive a reproduction record from
the print version record, than to create a new record describing the
electronic version. As we move into more digitization for preservation
purposes, I think this is an important consideration.
RD - In addition, the Digital Library Federation's Registry of Digital
Masters Record Creation Guidelines available at
http://www.diglib.org/collections/reg/DigRegGuide.htm recommend describing
the original version with information on the reproduction in a 533 note.
