Catalogues will remain as important as ever in providing access to information resources in the digital era. Together with metadata repositories and as a part of integrated library systems they will evolve to suit the demands of electronic formats and will play new roles e.g. in student portals. The principles of cataloguing which result in structured, consistent presentation of bibliographic records collocating works on similar subjects and by the same authors will be required more than ever. XML, metadata formats such as Dublin Core will be used as well as MARC, and hyperlinking to full text documents and other formats (such as audio and images) will also be accessible via the catalogue. Libraries may also link their catalogue to a whole range of related information resources and to other types of organisations such as booksellers, book reviewers etc.

How are libraries providing access to the flood of electronic resources? In general they have been trying more then one approach e.g. cataloguing electronic journals as well as creating lists on their web sites. But as librarians have realised for some time, we need to provide new solutions if we are to truly serve our users and remain relevant in the future. The seas are indeed choppy and we need new flotation devices and new navigation tools if we are to have any chance of staying afloat as well as arriving at desired destinations

This paper maintains that catalogues and metadata repositories (of e.g. Dublin Core) will be the key navigation tools for the foreseeable future. Catalogues will, however, have to change, and this process of change has already started. Examples and illustrations are provided from within the university setting. While library catalogues have never been the sole pointers to resources, they have occupied centre stage because of their structured, consistent and standard features. However, more flexible approaches are required e.g. in terms of record formats, handling multiple versions, additional hyperlinks. Over time libraries will be implementing new systems to better cater for the new resources, will be reconsidering record formats for selected materials, and will be deciding what Web-based material will be included in the library catalogue.

Although non-MARC metadata repositories have not made a significant impact on library information seeking to date, they will increasingly become more important. Several subject gateways as part of the Australian Subject Gateways Forum are planning to consolidate their metadata into one easily searchable repository. Also, there is talk of a National Library initiative to investigate access provision to the nation's electronic resources across databases. Such an approach might entail searching across metadata repositories. Non-MARC metadata is likely to be incorporated into library catalogues in the near future. New integrated library systems can now accommodate other formats.

Catalogues, as well as being lists of libraries' holdings (both on library shelves and on the Web) play specific roles in serving particular needs. For example, most university library catalogues today include bibliographic records for course support, high demand material. At the University of New South Wales this role has been enhanced by a new program - MyCourse @UNSW, a student portal based on course names and codes and utilising the catalogue as central database. The library catalogue will become more and more a jumping off point for any sites of interest to catalogue users - suggesting databases on similar topics and linking to other organisations e.g. booksellers if library catalogues do not list required items.

In new integrated library systems, the catalogue will be incorporated into metasearching systems, which search across databases over and above catalogues, and citation databases- all geared to facilitating access. Libraries are aiming to provide a one-stop shop for their users.

In writing this paper the author has drawn on her work involving:

Cataloguing of print and electronic resources in the university setting;

CORC project cataloguing (UNSW Library 2000);

Cataloguing of course support material in the UNSW Library Digitisation Service (UNSW Library 2001);

Developing specifications for UNSW Library planned new system;

Metadata coordination;

Catalogue transaction log analysis studies (Talmacs 1999a, b);

Discussions with groups of reference librarians and cataloguers at the UNSW Library.

Until recently library catalogues have usually only listed material housed within the libraries. Over the last decade or so records for additional material have also been added so that, for example, electronic journals to which a library subscribes may be accessed directly via library web catalogue. All librarians have not accepted the inclusion of this new material. Some believe that Web-based resources are outside the scope of library catalogues. Others believe that other tools such as search engines are the solution. Nevertheless there has been a significant trend to cataloguing electronic journals and databases, huge and daunting as the task is. Web coordinators have often also created list of electronic journals and databases either as the sole means of access to this material or to sit alongside catalogues as an alternative means of access. Impatience with cataloguing turnaround times and with catalogues themselves, together with the immensity of the task of e.g. cataloguing e journals and /or the fear of overloading cataloguers, have resulted in these alternative approaches.

In reality librarians know very little about how users search catalogues in spite of research over the years. The author carried out a small study of telnet catalogue use at the University of New South Wales in 1999 in two parts - one in January/February during the academic break and one in May (mid session). Telnet catalogue transaction logs were analysed to determine how often particular types of searches were employed, and how successful they were. Results in both studies for external (outside the library) and internal (in the library) searches on the Telnet catalogue revealed that title searches were by far the most popular with 35% (Jan/Feb external), 31.5% (May external) and 27.7% (May internal), way ahead of author searches (13.8%, 10.9% and 11.5%) (2nd, 3rd and 4th places respectively) with title keyword also fairly popular (11.5%, 12.2%, 9.1%). Although the samples were small (5,000 in Jan/Feb, 2000 in May) the studies do show that known item searching is predominant. What we do not know is why this is so. Does it say something about subject searching? Is the subject approach inadequate or is it that the university community has that approach covered through other means? We are aware that much up to date subject searching is carried out on databases of journals, and abstracting and indexing tools have been used alongside catalogues for years. It may just mean that catalogue users are predominantly searching for known items. Further research would be needed to clarify this.

When log results were examined for success two areas were considered - large sets (150+ records) and zero hits. 2.7% of the searches resulted in large number of hits, and around half of these were followed up with alternative searching strategies. Zero hits were achieved with 5.3% of the searches, and again in 55.5% of instances were followed with alternative strategies. These results were interpreted to mean that the catalogue worked fairly well during the first search and in steering users to try other approaches.

Advanced searching strategies e.g. refining, checking for records with similar subject headings were hardly ever used and this is in agreement in with previous overseas research. Searchers, it appears, are not prepared to go beyond simple searching strategies and want quick results. Librarians often endorse more sophisticated searching methodologies, but it seems our users do not require these.

Although a more thorough study would be needed to draw any more conclusions, the Library was reasonably happy with the catalogue satisfaction level (it also matched our user surveys). We concluded that the catalogue is a fairly successful finding tool with little sophisticated use and little use as tool for subject searching.

Catalogue hyperlinking to full text digital resources has been a boon to information seekers. Searchers today want and expect quick results. Library systems provide varying degrees of sophistication with the hyperlink facility and some deep link directly to journal articles - required for course support material. In our current ADLIB catalogue links are provided from catalogue records to journals, but not yet directly to articles. The trend, however, is to provide links directly to whichever level of the resource is required.

Catalogues have for several years also been repositories for brief records for transitory material such as course support readings (which change from year to year or session to session). These records have often been created by non-cataloguers and may not be MARC. The catalogue has therefore been a database serving different needs and containing different levels of records

As well as being a searching tool for users, it should be remembered that the catalogue is also a tool for library staff in backroom and public areas. These functions are sometimes neglected but are most important and the UNSW cataloguers stressed this point in my discussion with them on the role of the catalogue. Library users can also use the catalogue to check bibliographic details as well as for finding particular items.

Catalogues are more and more often including free web-based resources, over and above the material to which libraries subscribe. Although this paper will not delve into the progress being made in creating selection criteria for these resources which are quite different to their print counterparts, suffice it to say libraries are having to face up to deciding which material to provide access to and whether this means cataloguing. Regardless of the sort of cataloguing if this is the chosen direction the June 2001 CLIR report (Pitschmann 2001:24) on this topic states that minimum identification and retrieval data would be:

Title/name of resource
Location of resource (URL)
Author or editor (i.e. creator(s}of resource and of its intellectual content)
Publisher (i.e. organization making the resource available)
Free-text description, including audience.

Other elements recommended for inclusion in the catalog record are those developed by the Dublin Core Metadata Initiative. They include the following:
subject
contributor
date (created last modified, data gathered)
type (collection, database, guide,/gateway organization, service, home page news service)
format
identifier
source
language
relation (e.g. is part of, is a version of, replaces, is referenced by, is based on)
coverage (geographic and temporal)
rights

Lots of cataloguing issue arise when processing Web resources e.g. should records be created for sites linked to, what level of granularity should the catalogue record reflect? Is MARC appropriate?

Regarding these free web resources, our library has tentatively made a start using MARC cataloguing. The UNSW Library has also compiled subject guides with links to useful resources. These subject guides have themselves been catalogued and are incorporated into our student portal MyCourse@UNSW (UNSW 2001) which is described more fully later in the section "New Roles for Catalogues".

Catalogues are continuing to carry out their traditional functions while being expected more and more to be the engine to lead searchers to whatever material they need on the Web.

In spite of their imperfections catalogues have survived very well over the years. Using internationally recognised standards, they are consistent tools created by specialists. They are predictable to librarians though not always to users (especially when switching from one library's catalogue to another).

Cataloguers are specialists with a knowledge of standards such as MARC, AACR2, national systems such as ABN/Kinetica, local systems, authority tools such as LCSH Mesh, LCNA etc. Standards are international with some local variations. The application of standards by trained cataloguers has resulted in well-formed records and consistent headings, and the collocation of material by the same author or on the same subject. Authority systems with references aid users to unravel such complexities as variant author names and related subject headings and subheadings.

Conversations held with UNSW Library cataloguers prior to the writing of this paper focussed on how the complexities of catalogue and web searching will never be able to be reduced to simple formulas. They maintained that because research material is not simple, users will always have to make decisions in the searching process as they interface with Library web sites. Catalogues have therefore been devised to cater for these complexities. Often, however, it seems that the complexity is catered for from a librarian's point of view and this does not always suit users e.g. multiple records for multiple formats.

A big question then is how to cater for the complexities of materials catalogued while catering for a wide range of users with varying levels of expertise in using the catalogue. In theory the standards used to make up catalogues should be used with new technology to provide for more user-friendly flexible tools.

While recognising that catalogues have much to offer in terms of standards and consistency, they will need to be adapted to suit the new resources and new searching requirements of users.

The UNSW reference librarians I met with pointed out the necessity of user education to enable university staff and students to find desired resources. One librarian explained how she took first year students aside and advised them to forget all the searching methodologies they use on the Web, reminding them that they now had to use a completely different array of specialist library tools. Another reference librarian explained how she had seen students enter assignment questions on the library web home page and expect immediate answers. One librarian encapsulated her view in the expression "speak and print" - this is what students want the system to do. What they see and hear at the library desks shapes the reference librarians' point of view. They realise that users want a quick and easy access to resources, and that search tools need to be more intuitive.

Where does that leave libraries as they ponder the new sorts of systems they will be acquiring in the near future? They will most certainly be assisted by the availability of new systems with improved catalogue features e.g. accommodating additional formats, improved linking devices.

Subject access via catalogues has always been problematic. Many catalogues of large research libraries use Library of Congress Subject Headings. Catalogues have both benefited from and been constrained by LCSH. Keyword access on the other hand has opened up other possibilities, although there is more scope for improvement in this area. Research has suggested that a combination of controlled vocabularies and keywords is ideal for subject access.

Another area of confusion for users is when multiple formats exist for the same work. Catalogues have been inconsistent in the treatment of multiple versions. There have been a variety of solutions provided even by similar types of libraries. How has your library catalogued hard copy, microform copies and electronic versions of the same journals? Consistency and helpful standards are needed. A new approach is required to allow different formats to be attached to a core record for an intellectual work.

Catalogues have also accommodated records of varying standards and often quality, for course support/reserve/high demand material. New systems will allow greater flexibility in this area with the possible use of metadata such as Dublin Core for this material.

Catalogues nowadays often accommodate records from a variety of sources. Even if Kinetica is the main source, there may also be records from RLIN, OCLC or other sources. Records may often be created by agencies not just by member libraries. This makes for a broader range of sources. With an increased range of records, copy cataloguers have to be even more vigilant in checking for accuracy. This comes at a time when staff available for cataloguing are being squeezed even further.

Catalogues are now playing new roles in new library or institution programs. For example, the MyCourse@UNSW student gateway at UNSW uses the catalogue as central database through inclusion of the course codes as linking device. The non-MARC records for course material e.g. reprints, book chapters, journal articles, past examination papers include the course code composed of 8 characters (4 alpha and 4 numeric e.g. SOCW1001 for Introduction to Social Work) and as well as MARC bibliographic records are displayed after a search by course code.

Web-based subject guides produced by the UNSW reference librarians are fully catalogued by cataloguers. They include the relevant course codes, which they glean from the metadata in the HTML head section of the subject guide pages. The subject guide creators have added this metadata. As well as demonstrating added roles for the catalogue, this also shows a convergence of MARC cataloguing and Dublin Core metadata.

MyCourse@UNSW is a single access pathway to course resources. It relies on the catalogue as the linking database containing the relevant course codes. It can be accessed from the UNSW web page or from the Library's web page, and demonstrates a new emphasis on assisting our university community to more easily find material relevant to them.

As digital resources flourish they are more and more often being incorporated into catalogues. Cataloguing of electronic journals is prevalent and many records appear in Kinetica for sharing. Electronic databases are also being catalogued. Electronic versions of journals are either being catalogued separately or are being added to records for hard copy journals. As print version cease publication the records are spilt accordingly. Other web-based resources are being catalogued as they are "selected" by librarians. Some of these are later editions of earlier hard copy versions.

The UNSW Library cataloguers had a good opportunity to experiment with new ways of cataloguing of Web-based resources during the CORC project in 2000. Together with a handful of Australian libraries, they catalogued in the free OCLC pilot project to check out the features - DC/MARC mapping, record export, pathfinder creation and the huge resource it is in itself - over 500,000 records for web based resources. OCLC with its huge research commitment is exploring ways of creating a union catalogue internationally, thereby providing all the conventional access points to Web based material in the same way libraries have been providing access to hard copy material on their shelves. While Australian libraries with a strong Kinetica focus will not be rushing to join CORC, it is a significant initiative on a global scale and is well worth a look and a try if possible. The mapping Dublin Core to MARC and vice versa is very valuable, Libraries can harvest data including metadata from web sites and this data is then converted to MARC with the possibility of upgrading by cataloguers.

Online course support material is on the increase. With demand from students and their teachers for 24-hour access from anywhere for teaching material, university libraries are digitising material more and more. They are doing this in accordance with the Copyright Amendment (Digital Agenda) Act 2000 and are thereby attempting to satisfy the groundswell of demand. Though all material cannot be scanned because of the stricter copyright restrictions (when compared to those for hard copies) it is a definite improvement in access over traditional reserve section activities in libraries. The catalogue continues to provide links to teaching material with hyperlinks to material on Library servers. There is also the possibility that library catalogues might become databases of records for copyright purposes for material on servers in parent institutions for example the whole university.

Images for the College of Fine Arts courses have started to be added to the UNSW Library server. We are also planning to add audio material online i.e. a replacement for the lecture cassette service. The technology for the latter is being investigated, with the optimum results requiring much consultation and cooperation amongst units on campus. Library users and the University community are coming to expect multiple formats online with the 24-hour access from anywhere. The UNSW Library Digitisation Service needs to increase its range of services to meet expectations. Traditional print services for course support are still being provided so staff resources are feeling the now familiar strain as they attempt to maintain two strands.

A further broadening out of the scope of library catalogues is already occurring with links to related databases within the catalogue searching process and links to book reviews and booksellers. Libraries can decide more and more what the catalogue provides in the way of links. This should be seen in conjunction with the concept of metasearching systems (discussed in a later section). Some integrated systems e.g. Innovative's Millennium allows staff and library users to book any sort of library materials from books to rooms or whatever the library chooses via the Web OPAC. This represents an opening up of the functionality of the catalogue to be a community service point.

Until recently MARC has dominated record keeping in libraries. With the coming of the Web and its use of HTML, XML and Dublin Core, the MARC supremacy has been challenged.

Miller (2000: 1) for one has promoted the replacement of MARC by XML. He argues that "If we do not act quickly to integrate library resources with mainstream web resources, we are in grave danger of becoming marginalized". XML is described as" the de facto Internet standard for representation of information content (not format) - optimized for web delivery". Miller advocates that XML not only replaces MARC but that XML replace formats for other functions e.g. ILL, patron data, circulation transactions, orders, checkin data etc.

New integrated library systems do support XML (as well as other formats), and no doubt libraries will use XML more extensively in the near future. Whether it will replace MARC is another question that will not be explored in this paper.

Dublin Core and other metadata formats are also being supported by new systems and will used as appropriate for material not requiring full MARC cataloguing.

Just how will metadata be used to provide access to electronic resources? First, we need to look at how metadata is being used now. It is used in subject gateways, in Web pages and in various other databases. So far metadata is not usually integrated into library catalogues. This will change as new systems, accommodating metadata are implemented in library catalogues.

Subject gateways such as AVEL, Agrigate and others depend on metadata that is stored in their databases and searched by gateway search engines as users select various access points. These search engines fully support the metadata that is mostly based on Dublin Core with some additional elements e.g. from AGLS, and some gateway specific elements. Searching is usually either by browsing using a thesaurus hierarchy or searching using specific terms - subject, author etc.

Some Web pages have also been metatagged in the head section of the HTML. The searching advantages are less clear for web pages with mixed support from the search engines. This support is so far minimal although Infoseek Ultra and others (see the Dublin Core site (DCMI 2001) for examples) do take metadata into account in their Web indexing.

As mentioned above, the National Library of Australia is investigating how access to electronic data might be provided across databases and repositories nation wide. Would this be through a national metadata repository? Or would searching be distributed? This might provide an Australian answer to CORC. But investigations are still at the early stage so how it will work is not clear.

Libraries embarking on using metadata will need to draw up guidelines for what material they will catalogue using metadata and what material they will continue to catalogue using MARC. One category of material that might be catalogued by metadata such as Dublin Core is course support readings that are currently catalogued in a variety of ways - in local systems in brief records, and which cataloguers often do not handle. Non cataloguers are often required to create records for material such as journal articles, book chapters, lectures, tutorial notes, lecture cassettes or examination papers which either do not require in depth analysis or are only temporarily required for specific courses. Categories of Web material may be best catalogued using Dublin Core and may take advantage of metadata already embedded in the pages.

Libraries are now exploring ways of providing metasearching for users. For example, the University of New South Wales Library in its Expression of Interest document for procuring a new Library system states in the section on an Information Access System:

"The IAS shall provide a Web based interface for simultaneous searching and retrieval of metadata across heterogeneous networked information resources, including the Library Resources Database, catalogues of other libraries, externally hosted databases (including abstract/indexing/citation databases, document delivery services, online learning resources, publisher full text services and aggregator services), Internet search services and internally hosted databases and digital object repositories" (UNSW Library 2001b)".

Metasearching tools such as MetaLib (ExLibris), MetaFind (Innovative) are beginning to appear. They perform parallel broadcasts of searches to distributed catalogues and databases and promise concise intelligible result sets. These tools will aid wide ranging, precise searches and obviate the need for users having to make as many choices as they make their way around web sites. For example, the current UNSW Library Information Resources page (UNSW Library 2001c) has a bewildering array of choices - the catalogue, databases, electronic journals, course materials internet links, subject guides, new electronic resources and digital theses. There is considerable overlap in these categories and users must make a stab at one or other and try to find some material of value. This pattern is not unique to the UNSW Library Web site. Similar choices are required on other library web pages

What metasearching will do is to search across databases simultaneously and produce broader result sets. There will still be a need to make choices, and users may well set defaults in their systems once they are familiar with the software. Searchers will hopefully benefit from the umbrella approach. As Ex Libris advertises in their publicity blurb, libraries will be moving "From Library systems to information systems". Several years ago talk about the hybrid library began and since then libraries have been grappling with the issue. New generation systems may be able to help us out but have yet to be put to the test.

Taking into account the inherent complexities in data being searched and the need to simplify searches for busy catalogue/database users, librarians have an obligation to devise systems that streamline searching. In an era of staff cutbacks and doing more with less we can no longer expect our users to rely heavily on librarians to show them around web tools. Specialist advice will always be required and flexible schemes are being developed to cater for this, but we need to enhance user self-sufficiency wherever possible.

We will need to make good choices when creating bibliographic records e.g. which format is best? Do we catalogue material into our library catalogues or do we point to lists and include these databases in the metasearching capabilities? Libraries must also decide guidelines for what is to be included in their catalogues or local databases.

So what will new catalogues look like? They will increasingly provide access to more than just what a library has on its shelves. They will provide access to all material regardless of location, and to both free and licensed material. Catalogues will include even more types of material than now. They will need to accommodate a variety of records - levels and formats and to provide appropriate access. Non cataloguers and resource creators may create records that could perhaps be upgraded by cataloguers. Library course support staff will continue to create records for course material (increasingly digital), but will do it in more standard ways (with the help of new and improved systems). Current UNSW course support records are not compatible with MARC records but meet the needs of the Reserve and Digitisation sections and their users. Improved templates will be provided in our new system. If our library is to become record keeper for copyright purposes then campus departments and faculties might possibly enter data into a template indicating material stored on their servers. Librarians might then enrich records as necessary to fit in with other catalogue records. The format of these records could be Dublin Core metadata. Similarly Dublin Core might be used for some web sites and created by specialist non-cataloguer librarians. These records could use the same authorised headings from the authority system maintained by the cataloguers (which may in turn have been derived from Kinetica or Library of Congress).

Cataloguing of electronic journals and databases is a whole topic in itself and a lot of expertise is being applied globally to come up with solutions as electronic versions sit alongside, and then often replace hard copy journals. Versions need to be brought together for easy user selection but each format needs full description. Multiple versions is another whole topic which cannot be tackled here. If these issues are not resolved catalogues may become sidelined and that would be a great loss. Cataloguers need to solve record problems and provide user-friendly solutions to users.

Catalogues will be central to metasearching systems. Librarians will have to decide what material goes into the catalogue and what material can remain in separate databases available through cross searching. This approach will entail a completely different strategy to access provision. Also, will libraries have to devise their own individual solutions? (I'm sure they will do this anyway) or will they be able to make use of a National Library searching umbrella system (currently being investigated)? The UNSW Library planned metasearching system will include the catalogue as only one of the databases searched. Most initial searches will, we imagine, start at the broad level, with the ability to narrow searches to the catalogue or other specific databases. Once libraries have had more experience with these technologies we will be in a better position to refine our tools and further improve searching. In the meantime we can imagine that such systems will aid library users through simultaneous searching and sifting of results.

Catalogues with all the benefits of authorised access points, standardisation, collocation of like resources, have much to offer library users. The structured features of catalogues bring order into the searching process, a far cry from the searching on the Web. Non MARC metadata will be a suitable addition as formats such as Dublin Core become more and more common when full cataloguing is not required. Catalogues will play new roles in portals like MyCourse@UNSW. Catalogues will continue to include data such as records for high demand material e.g. journal articles and book chapters. In addition, the catalogue will hyperlink to more and more digital material e.g. journals and, more specifically, to journal articles. Such records might possibly be in a non-MARC format such as Dublin Core. All the benefits of the catalogue, including authority control will aid resource retrieval for such material. The catalogue will also be integrated into metasearching systems, which will provide one-stop shops which librarians have been striving for over the last decade or so. Other databases searched in such systems will include metadata repositories, which may be subject gateways, or other databases such as for theses e.g. Australian Digital Theses Program. These non-MARC metadata repositories will allow for a controlled approach to bibliographic access but at a level more suitable for some material. Cataloguing standards will often be incorporated into metadata systems and cataloguers will continue to provide guidance in their establishment and in the maintenance of standards. New library systems coming onto the market are now providing features to achieve this integrated approach and to improve access to digital material. Catalogues will inevitably be central databases within any such system. In this way the benefits of catalogues utilised for decades will still be available to information seekers in the digital era. The features will be enhanced and modernised but will be similar in principle. In short, catalogues will enable searchers to find their way around the turbulent seas, which are currently shaking up the library world.

References

Dublin Core Metadata Initiative. (2001) [Online] Available: http://au.dublincore.org/
[2001, September 24]

Miller, Dick. (2000) XML: libraries' strategic opportunity. [Online] Available:
http://www.libraryjournal.com/xml.asp [2001, September 24]

Pitschmann. Louis A. (2001) Building sustainable collections of free third-party web resources. [Online] Available: http://ww.clir.org/pubs/reports/pub98/contents.html
[2001, September 24]

Talmacs, Kerrie. (1999a) Transaction log analysis of the UNSW catalogue Jan/Feb 1999. Unpublished.

Talmacs, Kerrie. (1999b) 2nd Catalogue transaction log analysis May 1999. Unpublished.

University of New South Wales. (2001) MyCourse@UNSW [Online] Available:
http://www.library.unsw.edu.au/mycourse/mycourse.html [2001, September 25]

University of New South Wales Library. (2000) CORC Project [Online] Available:
http://www.library.unsw.edu.au/~eirg/corc/corc.html [2001, September 24]

University of New South Wales Library. (2001a) Digitisation Service [Online] Available:
http://www.library.unsw.edu.au/%7Egsd/digihm.html [2001, September 24]

University of New South Wales Library. (2001b) Information & Resource Access Management Systems [IRAMS]: Call for Expressions of Interest, 7 May 2001 [Online] Available: http://www.library.unsw.edu.au/eoi.html [2001, September 26]

University of New South Wales Library. (2001c) Information Resources [Online] Available: http://www.library.unsw.edu.au/elec_resources.html [2001, September 25]

Kerrie Talmacs
Metadata Coordinator and Digitisation Librarian
University of New South Wales Library