|
|
|
|
|||
Shirley Sullivan, Collection Management Librarian (Physical
Sciences)
February 1996, revised May 1996
The collections were assessed using criteria laid down in the Australian Conspectus Manual:
Computer science studies the theory, analysis, design and application of the algorithmic processes that describe and transform information. It is a discipline broad in scope and heterogeneous in nature. Within the discipline there is no general agreement as to its boundaries and purpose. To some, it is closest to mathematics, to others, engineering. Some emphasise its connections with philosophy and logic. It can be divided into a number of subfields, including algorithms and data structures, programming languages, architecture, numerical and symbolic computation, information retrieval, operating systems and networks, software methodology, artificial intelligence, and visualisation and robotics. Computers play a major, and ever increasing , role in the practice of many disciplines, and their use is now essential in areas such as engineering, medicine, banking and finance.
Since the close of World War 2, English has become the lingua franca of the sciences. Eighty per cent of all information stored in the world's electronic retrieval systems is in English; for example, INSPEC in 1990 was 83% English. In Russia and other European countries the tendency is to publish in the English language. Authors seek to publish in English in order to gain wide dissemination, and academia and publishing houses ensure that journals are published in translation. Translation journals provide either cover-to-cover or selective translations of the foreign language journals, often published by the original publishing house or by major American scientific publishers or professional associations. Japan is an important source of information in this field, but provides much of this valuable material in English translation. There is, therefore, no necessity for computer scientists to be able to read in other languages.
Much computer science literature is of recent vintage, reflecting the youth of the discipline, which began to take shape after World War 2. More than in other disciplines, a large portion of it quickly goes out of date. Some literature used in computer science research, however, remains useful almost indefinitely. For example, access to both old and recent mathematics journals and monographs is crucial for some computer scientists.
The literature of computer science comprises mainly journals, newsletters, technical reports, dictionaries and encyclopaedias, directories, preprints, trade and product information, standards, books, proceedings of conferences and symposia, and indexing and abstracting databases. Like other sciences, computer science relies far more heavily on serial than on monographic literature.
From an academic point of view, the most important journals are those produced by learned and professional societies, and commercial and university presses, and research journals from manufacturers. These are research oriented and contain original papers or authoritative reviews. They report advances in understanding of the technologies or report new insights into the fundamental science underlying the technologies. Other journals produced by the same organisations are for rapid dissemination of information, and are not refereed.
Commercially produced journals are valuable sources of current product information, as are company technical journals. Journals aimed at computing professionals offer a technical approach, and include regular reviews of new products. Popular computing magazines are likewise a useful source of product reviews. Popular science journals, such as New Scientist and Scientific American, provide a selective and often useful introduction to various aspects of computer science. Some journals covering other industries or professions also provide occasional articles on new developments in computer peripherals. Accountancy, for example, has published several useful articles on computer hardware and software aimed specifically at those working in the financial sector.
Preprints are becoming, if they are not already, the primary information dissemination method in computer science. Many computer scientists rely on colleagues' preprints, often communicated via email, because this gets them the most recent information. For working researchers, particularly in theoretical computer science, practically all the information dissemination takes place through early preprint exchanges and interactions at conferences. For example, the article of Cisler in Serials Review 1992 quotes a computer scientist as saying, "If I read it in a journal, I'm not in the loop."
Technical reports from university departments and from industrial research and development laboratories, such as those at IBM, are useful sources of information. Though researchers are primarily interested in recent reports, older ones, particularly those with an emphasis on theory, are sometimes requested.
Standards are an important part of information technology. Most of the industrialised countries have a standards body which coordinates the development of standards for all areas.
Trade literature provided by companies includes product catalogues and brochures, house journals and annual reports. Trade catalogues are a valuable source of information on products, as are trade exhibitions.
Patents are concerned with technical concepts which may still need development into products. An argument against using patent literature is that any worthwhile information will be reproduced in more accessible form in journal articles; however, patents and journal articles are written for different purposes and at different stages of the lifetime of a product.
Conferences range from highly specialised theoretical deliberations of academics and researchers up to more pragmatic and general events geared towards the layperson. They are useful sources of up-to-date information. For working researchers, particularly in theoretical computer science, practically all the information dissemination takes place through early preprint exchanges and interactions at conferences. In the words of Joan Feigenbau, in theoretical computer science, "If it didn't happen at a conference, it didn't happen." The conference system provides computer scientists with rapid evaluation and feedback, and also opportunities for extensive personal interaction. Conference organisers include professional institutions, academic institutions, government agencies and commercial conference organisers. Conferences are sometimes used as the venue for important announcements, both of theoretical advances and for new product launches. Conferences organised by learned and professional institutions often have a formal refereeing process analogous to the process used for learned journals. Presenting a paper at the more prestigious conferences is preferred to publication in a leading journal. Many institutions, such as IEEE and ACM, publish conference papers in their "Proceedings".
Conference material is useful as a compact source for obtaining a picture of what is going on in a particular field and who is working in it, but the published proceedings may already give an out-of-date picture.
The Department now forms part of the Faculty of Engineering, where it moved from the Faculty of Science. The Bachelor of Computer Science aims to develop skilled computer scientists with the technical background, knowledge, and adaptability to develop well-designed, robust, computer-based solutions to a range of problems. In the first two years students are introduced to concepts of computer systems (both hardware and software), acquire an appreciation for the scientific and engineering principles underlying the development of computer based solutions to problems, and develop skills in programming in a variety of programming languages and environments. The third year of the course builds on the core technologies introduced in earlier years and allows students to focus on advanced applications in their areas of interest. At all year levels there is a focus on the cultivation of practical skills together with assimilation of the relevant scientific and engineering principles.
The Department undertakes research in a wide range of areas, including:
Note: There is no separate computer science branch library. Items are located primarily in Maths, Engineering, Baillieu and ERC. The collection comprises in total roughly 10,000 monographic items. The collection does not have a separate bookvote, but is incorporated into that of the Faculty of Engineering. The budget is divided roughly between the 6 teaching departments of the Faculty. At the end of 1993, the Faculty applied for $60,000,00 to take over computer science subscriptions formerly paid for from departmental funding. The faculty has received this amount over 1994-1996.
5.1 Audio-visual collections
There is an AV collection held at the Eucation Resources Centre. The collection totals about 265 items, and is almost all computer files. It is used chiefly by students within the Faculty rather than by the wider university community.
A number of lists were used to assess the collection strengths. The list checking took 10 hours.
The most suitable and thorough list for monograph checking was Information Sources in Information Technology, despite its date of publication (1990). It was comprehensive, particularly for serials, dictionaries and historical literature. Anecdotal evidence from the academic staff that the collections are missing the latest editions of monographs, dictionaries and handbooks was substantiated by the list checking.
Table 1: Monographs
| Title | No. | Checked | No. held % |
| Haynes | 731 | 321 | 43.9 |
| Hurt | 43 | 21 | 48.83 |
| Malinowsky | 45 | 21 | 46.66 |
| OCLC | 180 | 39 | 21.66 |
| Overall | 999 | 402 | 40.24 |
As can be readily seen from the table below, the rating of the serials collection depending upon the list checking. From Malinowsky, the Library rated 78.6%, whereas from INSPEC it rated 30.32%.
Table 2: Serials
| Title | No. | Checked | No held % |
| INSPEC | 3833 | 1162 | 30.32 |
| Malin | 14 | 11 | 78.6 |
| SCI | 83 | 58 | 69.78 |
| Overall | 3930 | 121 | 31.32 |
A shelf scan was undertaken concurrently with list checking. This took 6 hours. Libraries on campus holding items in the relevant DDC numbers were examined. This provided a valuable overview of the monograph coverage particularly. As mentioned previously, holdings are scattered throughout a number of libraries. The impression gathered was of a haphazard collection scattered across campus, forming no coherent strengths and a source of grave dissatisfaction to users.
Note: As Michalak remarks in a 1994 article from Journal of Library Administration, the increasingly interdisciplinary nature of much teaching and research has resulted in the housing of library collections by subject specialty in branch or departmental libraries becoming increasingly inconvenient for users. They need to go to several campus libraries to pursue one research strategy. While members of the campus community appreciate the specialised and individualised service which the branch library provides, economic considerations and the goals of providing increasingly multi-disciplinary information services, or gateways to information rather than repositories of hard-copy collections, may move campus library stems in the opposite direction from departmentalisation by subject.
A number of academic staff members from the Department were interviewed and assisted the Conspectus Officer to assign levels to worksheets. They further provided advice on areas which need to be built up to support newly established fields of teaching and research. Specific information is provided under sections 11 and 12.
The collection is poor, with only 50% of the items listed in Hurt and 64.7% listed in Malinowsky held in the collection. Forty percent of the titles from Hurt were held in an earlier edition only; the figures similar for Malinowsky. It will be necessary to upgrade the collection during 1996.
Although Haynes (1990) declares that a collection of roughly 4000 titles is considered "respectable" in an academic institution, the existing collection of roughly 10,000 items does not appear to satisfactorily meet the needs of academic staff in their areas of teaching or research. The collection is consistently referred to by the academic staff as poor, in some areas overall and some only for recent work. The list checking confirmed the impression gained from the shelf scan that the collection is poor in most areas of computer science, a level 2bE.
The journals collection in computer science is poor. An increase is essential to support the expanded areas of teaching and research, particularly as this is an area in which the undergraduates are expected to use the serial literature. The overall level of the collection is 2bE.
The University of Melbourne Library holds or provides access to the major indexes and abstracts in the discipline in print and/or electronic formats.
These include
Many academics and graduate students make use of the information available through the world wide web. The following sites are examples of the diverse sources available for research:
These sites and the information they contain cannot yet be seen as substitutes for the more traditional print, microform and electronic sources, but as supplementing them.
The following are examples of refereed electronic journals available in computer science:
There are no areas of substantial research strength.
The collections are not adequate to meet the research needs of the Department. They need strengthening in most research areas with an increased volume of monographs and new journal titles. The following areas are given as examples of research areas needing strengthening.
11.1 Logic Programming
The existing collection is described as "reasonable" but is required to be research level.
11.2 Software Engineering
The collection is poor in this area and needs significant upgrading to reach the required research level.
11.3 Computer Vision
The Collection is poor in this area and needs significant upgrading to reach the required research level.
Thanks are gratefully offered to the many people who gave freely of their time and advice. I thank particularly Juliet Flesch, Elisabeth Garran, and Graeme Murphy from the Library, for their support and assistance, and the academic staff who generously gave their time and advice to assist me to build a picture of the current collections and the needs that they are required to meet.
Last modified: August 18 2004.
| | Find a Library | | About ALG | | Pathways to Australian Information | |
| | For Libraries | | Online Exhibitions | | Help | | Contact ALG Administration |