National Library of Australia

Australian Committee on Cataloguing Workshop - September 1997

National Cataloguing Conference, Canberra, 11-12 September 1997

Summary prepared by Pam Dunlop and Julia Trainor, ACOC

The Australian Committee on Cataloguing (ACOC) conducted the two sessions of Workshop 4, focusing on issues relating to the International Conference on the Principles and Future Development of AACR, to be held in Toronto in October 1997. The International Conference, organised by the Joint Steering Committee for the Revision of AACR (JSC), will re-examine the fundamental principles of AACR. Since AACR2 was published there have been many advances in online catalogues, and the emergence of publications in a variety of new and mixed formats and multiple versions has raised a number of issues. The International Conference will ask is AACR still valid in this environment, or is there a need for change in the cataloguing rules to encompass these advances?

There are seven Australian delegates invited to this conference, including Australia's representative on JSC, Ann Huthwaite. Australia is therefore in a good position to have some influence on whether there is a need for change, or on any new directions likely to be taken. The workshop sessions held at the National Cataloguing Conference provided an opportunity to listen to opinions from participants, our current Australian cataloguing practitioners.

The International Conference identified 6 major issues for consideration, and 9 papers have been mounted on the Conference home page, to serve as the basis for discussion. The ACOC workshops chiefly focused on six of these papers, although, as can be seen below, there are recurring themes across papers. The overwhelming evidence from these workshops was that cataloguers wanted a clear statement of principles within the rules. This would help cataloguers to focus on the forest not the trees, and fewer rule revisions would then be needed.

Following is a brief summary of issues raised relating to the papers discussed. The first three papers below were presented and discussed on Day 1; the last three on Day 2. Generally, different participants attended each session. This accounts for certain inconsistencies in summaries.

AACR & catalogue production technology

Giles Martin outlined Rahmatollah Fattahi's paper which considers AACR in an online environment. Does the online catalogue serve users as well as the card catalogue in all instances? Is there a need to analyse assumptions underlying the cataloguing code in describing and providing access to works? Authority control, particularly for uniform headings, and online displays are central to good user access to information. This paper also introduced the concept of a 'super record'; for example:


Shakespeare, William, 1564-1616


This work includes the following editions available through this catalogue:

Editions (by date, by editor)
Translations (by language) 
Versions (by physical form) 
Adaptions & arrangements 
Works about this work


Such a record would then provide hypertext links, for the user to retrieve the required subset of data. In discussion the group was not entirely convinced that online display should be an integral part of AACR. There was also the thought that some online displays were better than others, and that more sophisticated software may well be the answer. However, most liked the idea of the super record concept, and grouping of retrieved items by type, although they felt that most of the required data was covered by the rules and already existed in catalogue records. They thought that there could be some argument for a rule change to encompass the 'work' being catalogued first; and that there are useful data elements which AACR does not stipulate as access points (eg. language). Specifying all relationships would be important in a 'super record' environment.

Bibliographic relationships

In considering this paper, Ann Huthwaite stated that bibliographic relationships are the 'glue that sticks the catalogue together', and that they help the user to navigate the bibliographic universe. The paper emphasises the need to display the relationship between the work and the item (referring back to the IFLA's Functional Requirements for Bibliographic Records). We decided in discussion that cataloguers are good at relationships! However, this paper reinforced that perhaps we didn't think sufficiently about the underlying principles. The paper also makes reference to multi-level records for display, and workshop participants felt that there was a need for more detailed user studies to identify search and retrieval needs before substantial changes to rules were proposed.

Access points for works

This paper commenced life as access points for personal names and corporate authorship, but was subsequently expanded to cover 'works', in which the need for 'work identifiers' is explored. Susanne Moir explored the author's argument that main entry is currently the means for identifying a work, although author and title is really required. There are inconsistencies, particularly in how MARC links (or does not link) uniform titles to main entry for retrieval. In this instance, the workshop group agreed that the card catalogue did it better than the online catalogue, where the relationship between author and uniform title is often lost in display. Workshop participants were divided over the value of main entry. Perhaps the concept of an authority record for a work, consisting of author and title would solve the problem. Is this the way to go? Other issues were: Are authority files part of AACR, and is there scope for incorporating the concept of 'work' in AACR?

What is a work?

Giles Martin summarised Martha Yee's concerns about the difficulty of defining a work in AACR, especially for works of mixed responsibility, works intended for performance, and works which are adaptations or representations. However, our workshop participants were sceptical. The answer to the question 'What is a work?' was 'Who cares'. The workshop participants thought AACR2 'ain't broke', so why fix it? There is no correct way to define a work, so let the system display the work in the terms requested by the searcher. For example, in the case of a photograph of a building, display it as the photographer's work if that is how it is retrieved, or as the architect's work if the user searches by the architect's name. In summary, the rules are good enough, our data is good enough, it is the system's online display which is deficient.

Content versus carrier

Although this paper has not yet been written, Helen Hoffmann described for us the concepts which it will address. AACR2 asks us to catalogue the item in hand, the physical manifestation of the work. However, the Library of Congress catalogues reproductions as versions of the original work. Catalogue users search for works, and they don't have any trouble retrieving these works in our catalogues. The problem is that they retrieve too much. We looked at informative ways of presenting multiple versions in online displays, including the 'super record' which lists the elements common to the 'work' before the elements distinguishing the 'manifestations'. How different from our usual online catalogues which repeat the common data, and hide the data which distinguishes one version from another.

Participants in this workshop wanted change. Many confessed to 'tweaking the rules' in their local systems. Two libraries described how they had tried to improve the online displays in their local systems. They had changed the data not the system. The consensus was 'change the rules to catalogue the content not the carrier'.

Issues related to seriality

Just as we learned earlier that 'super record' was the phrase of the future, we also learned that 'seriality' might become a phrase of the past, loaded as it is with lots of 'cultural baggage'. The new term 'ongoing publication' was accepted as the third dimension in cataloguing. The practical paper by Jean Hirons and Crystal Graham was not contested by the workshop participants, who also supported fewer meaningless title changes by emphasising identification rather than description of ongoing publications. They also agreed that single issue updated publications, such as loose-leaf and databases, should be described in terms of the latest issue rather than the earliest. Yes, change the rules.

Evolving solutions

It is interesting to observe how the problems with AACR, and the proposed solutions, have been a theme of National Cataloguing Conferences over the last few years. In Fremantle in 1993, ACOC conducted a seminar to examine multiple versions and how to catalogue them. In Sydney in 1995 we heard Barbara Tillett speak about the concept of 'content versus carrier' in the way we try to catalogue in the electronic environment. This year we are looking at changing AACR to provide solutions to the problems described above. Perhaps in 1999 we will be discussing the implementation of the new AACR?

Implementing rule changes

The benefits of rule changes are attractive. They will bring increased efficiency to the catalogue user, who will benefit from improved online displays of retrieved data. They will also improve the efficiency of the cataloguers who will work to principles rather than detailed rules. But rule changes cost money. AACR2 was costly for Australian libraries to implement. Hopefully computer catalogues will be cheaper to change than card catalogues. We need to plan carefully any change to the cataloguing rules. We may require changes to the MARC format, and we will certainly need to 'sell' the changes to our managers and system vendors. We need to start preparing now, by getting on top of the issues, and ensuring that we are part of the solution not part of the problem.