________________________________________
>From: J. McRee Elrod [m...@slc.bc.ca]
>Sent: March-15-12 11:01 PM
>To: Brenndorfer, Thomas
>Cc: RDA-L@listserv.lac-bac.gc.ca
>Subject: Re: [RDA-L] Card catalogue lessons


>>What's great about the lists of relationship designators in RDA is
>>that the cataloger can at a glance see the nature of the types of
>>relationships that have historically been made.

>The RDA relationship lists are not in one alphabetical order, but
>broken up among Appendices I, J, and K aren't they?   I seem to
>remember we had to look several places to find the terms to give as
>one alphabetic list in the MRIs.  I find RDA very fragmented.

The order is neat and tidy:


Appendix I is for Group 2 to Group 1 (relationships of persons, corporate 
bodies, families to works, expressions, manifestations and items)
[in MARC covered by main and added entries]


Appendix J is for Group 1 to Group 1 (relationships of works, expressions, 
manifestations, items to other works, expressions, manifestations, and items)
[in MARC covered by added entries, structured or unstructured descriptions 
(notes, linking entry fields, etc.), or by See Also references in authority 
records]


Appendix K is for Group 2 to Group 2 (relationships of persons, corporate 
bodies, families to other persons, corporate bodies, and families)
[in MARC covered by See Also references in authority records]



>>The problem with those GMDs is that they are the same GMDs regardless
>>of the Content Type.

>A compound [carrier : content]  GMD would have dealt with the problem
>more easily.  I don't see that the media type term (337) adds much,
>particularly since one is so misleading ("computer" for "electronic"),
>and one patrons don't understand ("unmediated").


"Computer" is straight out of AACR2. AACR2 uses "computer" when describing 
computer carriers (AACR2 9.5B1), as in:
computer chip cartridge
computer disk
computer optical disc
computer tape cartridge
computer tape cassettee
computer tape reel

These are _computer_ media types.

"When new physical carriers are developed for which none of these terms is 
appropriate, give the specific name of the physical carrier as concisely as 
possible, preferably qualified by _computer_." AACR2 9.5B1


>>The 336-337-338 divide enumerates all the historical choices into
>>clean, neat categories.

>Except for kits, equipment, and large print among others.


"Large print" is a characteristic of the manifestation, and is a value under 
the Font Size element. It is not a Content Type. It is not an intermediation 
device. It is not a general form factor for a carrier. The 
square-peg-in-round-hole method of cataloging is the mess that RDA is trying to 
dig us out of. One doesn't fill out a government form by arbitrarily filling in 
boxes by disregarding the purpose of each element and using whatever data 
subjectively looks better.

Kits and equipment don't fit the categories either. "Object" is a List 1 GMD 
(AACR2 1.1C1) that is carried forward in RDA. It makes more sense than the 
listing of a handful of 3D object types for a GMD value to the exclusion of 
thousands of hypothetical objects.


>Quoting RDA Ch. 18:

>"Relationship Designator
>      The term relationship designator refers to a designator that
>      indicates the nature of the relationship between persons, families,
>      or corporate bodies represented by preferred access points and/or
>      identifiers"


The quotes and numbers are mixed up. The most recent version of this definition 
belongs in Chapter 29, not Chapter 18.


Actual quote:
RDA 18.1.6 Relationship Designator
"The term relationship designator refers to a designator that indicates the 
nature of the relationship between a resource and a person, family, or 
corporate body associated with that resource represented by an authorized 
access point and/or identifier."

These are the relationships between creators, contributors, and so on and the 
resources they are responsible for or have some assocation with (not just 
creating-- a person could be related to an item with a designator like 
"curator", a corporate body could be related to a manifestation with a 
designator like "film distributor", or a person could be related to an 
expression with a designator like "writer of added text").



whereas...

RDA 29.5 Relationship Designator
"A relationship designator is a designator that indicates the nature of the 
relationship between persons, families, or corporate bodies represented by 
authorized access points and/or identifiers."

These are the relationships between persons, corporate bodies, and families. 
These are coded in MARC as See Also references in authority records. There is 
an established practice for comparable designators with $w, which produces 
display constants like "Earlier heading" and "Narrower term". the new RDA 
designators are coded in $i following $wr as in the following examples:

100 1# $a Pei, I. M., $d 1917-
510 2# $wr $i founder of $a I.M. Pei & Partners


The other category of designators, between works, expression, manifestations, 
and items (RDA Chapters 24 to 28), cover a wider variety of conventions for 
identifying entities (not just access points and identifiers), and in MARC 
appear in a wider variety of situations, including See Also references between 
works in authority records. Here is an ideal utilization of designators:

100 1# $a Stoppard, Tom. $t Rosencrantz and Guildenstern are dead
500 1# $wr $i based on (work) $a Shakespeare, William, $d 1564-1616. $t Hamlet.


>I seem to recall from my schooling decades ago, that a deffinition is
>not supposed to simply repeat the term being defined.  How about:

>"'Relationship designator' refers to a term indicating the role of a
>person. family, or corporate body in creating the resource being
>described, attached to the name used as an access point."?


The conventions for showing the relationship vary, but follow this basic 
pattern:

<entity> <relationship designator> <entity representedy by identifer or 
authorized access point>

This is the underlying data logic. In the Linked Data discussions, each 
component could be a registered value. In standard database design, each value 
could be the target of a query, and utilized in displays in different ways-- 
from facets and limits, to a report or listing, to a formatted display that 
organizes information starting with one or the other entity. It's not always 
about the AACR2 flat-file structure of headings, with designators "attached" to 
names. The data should be captured and recorded in anticipation of the modern 
usages.

Thomas Brenndorfer
Guelph Public Library

Reply via email to