Report to the Cataloging Advisory Committee
from Elizabeth O’Keefe, MARBI Liaison to the Committee

Although there were some papers on non-RDA issues at both the annual and midwinter conferences, much of the summer and winter MARBI sessions was devoted to considering changes to MARC required for the implementation of RDA in MARC bibliographic and authority formats. Discussion papers at the summer session laid the groundwork, offering background information about RDA, summarizing principles that the RDA MARC Working Group considered in its deliberations, and making specific recommendations for new fields and subfields; the ideas floated in these papers came back as proposals at the winter session.

Some overarching RDA-related themes at both sessions:

Coded versus non-coded fields. International MARC users and also systems people tended to favor coding, which is language independent and easier for machines to process. But RDA usually specifies controlled vocabularies rather than codes, so this is a problem. How to indicate the source of the codes or terms is also an issue.

Linking: the Germans are very into linking records (bib to bib; bib to authority) and linking fields within records; RDA also sets greate store by linking, either within records or between records. So almost every new field added to the authority format got a subfield $0 defined for it.

Balancing RDA needs against needs of non-RDA users within the MARC community. It is important to maintain an environment where RDA terms can co-exist with non-RDA terms. MARC isn’t just for RDA, and has to maintain relator lists that contain terms that may not fit into the RDA vocabularies. The same goes for descriptive information and for relationships between bibliographic and authority records, and the various FRBR entities—MARC has to support users who will not be using RDA or will not be implementing full relational databases. The evolving authority record: there is a growing desire to exploit the wealth of information in authority records. Currently recorded in unstructured form in note fields to justifiy the formulation of a heading, information about persons, corporate bodies, and works could, if structured, transform the authority record into an authoritative source of metadata about the entity itself, as well as facilitating retrieval and display. (But it was noted: “The maintenance aspect of this extra data should be considered when agencies are implementing”)

Discussion Paper 2008-DP05/3: Treatment of controlled lists of terms and coded data in RDA and MARC 21

RDA proposes much more detail about carrier attributes. This prompted the drafers of this paper to suggest new values for the 007 to describe production method. Among the possibilities relating to visual materials: engraving, etching, photoengraving, lithograph, woodcut. Codes to distinguish between manuscripts on the basis of production method were also suggested: e.g. manuscript, holograph, typescrpt, and printout. It was noted that the 300 $b would have to be redefined as repeatable since it will be defined for a number of other RDA carrier attributes.

A more granular 340 field might include subfields for:

Some one asked about the GMD, and was told that RDA has done away with it. It was suggested that media and/or carrier information from the 3XX or 007 fields could be displayed in the title area, if you thought it was important (and if your system allowed it).

Proposal 2009-01/3: Identifying work, expression, and manifestation records in the MARC 21 Bibliographic and Authority formats

Since RDA is organized around FRBR enitites, catalogers would find it helpful to be able to identify which type of entity is involved. This could be done by defining a new field for bibliography and authority formats to identify whether the record is for a work or an expression. The field would be optional. But the idea foundered on the issue of why the field should be defined for both formats (the answer is because not everyone can implement the full relational database required to fully realize a truly FRBR’ized catalog, so the option to use the bibliographic record has to be left open). There was also the question of record maintenance--when a work suddenly acquires new expressons, who goes to the authority record and creates records for the expression(s) it has whelped? It was eventually agreed not to pursue the definition of a field for this purpose, until there is more guidance from JSC on implementation.

Discussion Paper 2008-DP05/2: New data elements in the MARC 21 Authority Format Proposal 2009/01-1: New Data Elements in the Authority Format to accommodate RDA

Alignment with FRAD has introduced a number of new elements that were not covered in AACR. For example:

The discussion paper and followup proposals identified new fields where most of this data could be encoded. A lot of the discussion centered on why different fields were defined for different types of date, instead of just defining a single date field and using indicators to identify the type of date. It was also objected that date information often has to be repeated in the 670; this leads to a lot of redundancy; Even worse if the source for each data type has to be repeated, as would often be the case, where birth and death information comes from a single source.

Some non-RDA related papers:

Proposal 2009-05: Defining $u for the 510 field

The 510 field is used for citing standard bibliographies (in print or electronic form) and abstracting and indexing services that cover the item being described. This proposal focused chielfy on the former, but will include examples for other types of material in the published version. During the discussion, it was agreed that the positioning of the $u field is significant; it should follow whatever subfield it references (for example, if the URL points to a description of the resource, it should follow the $a, while if the URL points to an entry within the bibliography or database, it should follow the $c).

Proposal 2009-02: Definition of new codes for legal deposits in 008/07 (Method of Acquisition) in the MARC 21 Holdings Format

As written, the proposal defined two new codes for deposits in the n subfield in the holdings format. In the ensuing discussion, two broad categories of deposit arrangement were identified: deposits that involve a transfer of custody, but may or may not involve a transfer of ownership (when e.g. someone puts personal papers, a personal library, or a collection on deposit) and depository programs where there is a tranfer of both custody and ownership (though ownership may apply only to the physical items, and not the intellectual content). The original proposal suggested defining new codes for legal and non-legal deposit (v?) arrangement, in addition to the existing code d, which covers deposits of all kinds. It was decided to redefine d, to include all deposits that involved a transfer of custody but which were not legal deposits; and define a new code,l, for legal deposits. (Personal note: This discussion encouraged me to think about using this code. We don’t currently use it, and it would make it much easier to track items that are on deposit.)

Discussion Paper DP 2009-02: Definition of field 588 for metadata control note in the MARC 21 Bibliographic Format

This paper discussed defining a new field for information about metadata (metadata metadata?). Examples given of this were "Description based on" and “last issue consulted” notes (bot crucial in serials cataloging); also information about the source of the information and its reliability. The paper argued that the ordinary user is not interested in this information; if it were sequestered in a separate field, institutions could choose not to affront their users by displaying it in the public catalog. But this point of view was challenged; the information is often valuable not just to catalogers but to researchers. The issue of display could be handled by an indicator similar to the public/private indicator used for the Source of Acquisition (541) field; this would enable catalogers to decide on a case by case basis whether the information in any given record was of interest or not. (Of course the issue of retrospective implementation is a problem). Participants felt more granularity was desirable in the field, and that more structure, to permit the generation of common introductory phrases (cf. with the 246 field use of indicators and $i), was desirable. The authors agreed to go off and write another draft.

Proposal 2008-04: Changes to Nature of entire work and Nature of content codes in field 008 of the MARC 21 bibliographic format

A new code, 5, was defined in 008/24-27 for Books and fields 008/24 and 008/25-27 for Continuing resources for Calendar; code 6 was defined for “Comics/Graphic novels"; the phrase "sequential art" will be added to its definition. At Sherman Clarke’s suggestion, the Books field 008/33 (Literary form) code “c” (Comic strips) was redefined as obsolete in order to reduce redundancy.

Elizabeth O'Keefe
eokeefe@themorgan.org


... go to more ALA reports ...