Joe, this is the thrust of my blog post, which started this thread:
http://kcoyle.blogspot.com/2012/05/rda-dbms-rdf.html
and I say:
" Where the goal in relational database design is to identify and
isolate data elements that are the same, the goal in library cataloging
data is exactly the opposite: headings are developed primarily to
differentiate at the data creation point rather than allow combination
within the database management system. The goal is to have each data
point be as unique as possible and to be assigned to as few records as
possible. "
I agree with you that this technique, which was useful in the card
catalog, is less so in our online catalogs today.
kc
On 5/19/12 8:13 AM, Joe M Tomich wrote:
Simon,
Many thanks for your thorough response to my record/table question. As it
happens, your response to Mac's post below addresses the concern that lead me
to raise it so I will continue the discussion from there.
Mac states:
"Some data may be unique to the manifestation represented by the record, e.g.
call number (as opposed to just class number), title,
statement of responsibility, collation. What is the advantage of having unique
information in a table rather than in the record?"
Within your response (posted in full below), you state:
"Call numbers are a very good example of how linking between records is used in MARC
today."
My new question: Is linking between Records really in accordance with RDB
principles? Is it not the ability to create interoperability between tens,
hundreds, thousands of records simply by joining the tables in which the
records are contained that constitutes a core advantage of a relational
database? While you've given many scenarios in which catalog information can be
put into an RDB environment, does doing so really produce the savings and
efficiencies we associate with other RDBs when we compare them to the
information storage and retrieval methods they replaced? This is really more of
a question to the community, as I know you may have been addressing
possibility, and not necessarily practicality.
I guess, in a nutshell, the point I'm making is that I think we'd be better of
looking at what improvements we can make to the catalog by borrowing some ideas
from relational databases rather than taking the relational database model as a
given and bending over backwards to shoehorn the libraries' holdings
information into that environment by whatever means necessary.
Joe
----- Original Message -----
From: "Simon Spero"<[email protected]>
To: [email protected]
Sent: Friday, May 18, 2012 8:52:35 PM
Subject: Re: [RDA-L] Are RDA, MARC data, and Bibliographic concepts compatible
with Relational database principles or systems?
On Fri, May 18, 2012 at 6:35 PM, J. McRee Elrod< [email protected]> wrote:
Simon Spero said:
[3] It is indeed possible to generate a full record by joining together things
from different tables.
Some data may be unique to the manifestation represented by the record, e.g.
call number (as opposed to just class number), title,
statement of responsibility, collation. What is the advantage of having unique
information in a table rather than in the record?
Beyond access points (aka entries) I see no advantage in using linked data.
I think you may be seeing some of the possible advantages, but I don't think
I'm expressing myself in the right way to let you see that you're seeing them.
Call numbers are a very good example of how linking between records is used in
MARC today. There's the call number that can go in field 050 of a bibliographic
record, but that isn't enough to deal with multiple copies of the described
manifestation.
In this situation, we might want to use MARC holdings records. These records
store, full location and copy specific call number information in 852 fields.
In order to find out which bibliographic record contains the description for
the items pointed at by a holdings record, a reference to the control number of
the bibliographic record (001) is stored in a special field in the holdings
record (004).
Simon
--
Karen Coyle
[email protected] http://kcoyle.net
ph: 1-510-540-7596
m: 1-510-435-8234
skype: kcoylenet
