Scientific and Technological Information Services in Australia: I. History and Development.

**********

Information management is a term that has been appropriated by various groups of information professionals since the 1970s and applied to a wide range of functions. It therefore suffers a variety of definitions that differ in emphasis according to the disciplinary background of the definers. Emphasis may be on systems for conveying information (of concern to those working in corporate management, information systems and content management) or on the documents that carry information (as in recordkeeping, librarianship, document management).

The various occupations that pursue their distinct visions of information management have differentiated themselves through attention to different types of documents and different approaches to information organisation. However, the prevalence of digital media, the increasingly inclusive utilisation of metadata across document types, and acceptance of information as a corporate resource, mean that a concerted view of information management is becoming more likely. Wilson is among the more prominent writers who have paid attention to the definition of information management. His thorough observations (1) are not repeated here, except to note that they encompass all types of information resources from within or outside organisations. The shaping of disciplinary understanding would be assisted by case studies of information management application. There are examples of these in the literature, (2) but they are not documented with reference to a disciplinary framework.

The following account uses an information management perspective to investigate Australian scientific and technological information (STI) services. The work is in two parts. The first part (this paper) is an examination of the history and development of the STI services, with some remarks about their continuation and necessity. The second part is a consideration of the extent to which a consolidated view of information management may be applied to provision of STI services. (3)

STI services were chosen for the study for a number of reasons. They were expected to represent many of the purposes to which information management principles could be put into practice. They each provide an example of a service that is produced by one institution principally for the benefit of many others; they were developed at the time when consciousness of information management principles was nascent; they form a relatively distinct set of cases for examination; and they appear to be a valuable resource whose continuation cannot be taken for granted, and which may benefit from exposure to further scrutiny.

Many types of services or systems that involve information management could be examined. They range from systems for inventory control or personnel management, to services that are more concerned with documents in the conventional sense such as recordkeeping or cataloguing services. The discrete group of services chosen has been maintained continuously over an extended period of twenty to thirty years. Similar services in the social sciences and humanities exist. Although many of the observations in this work may also be applied to such services, they are outside the purview of this work.

STI services themselves are sometimes differentiated into bibliographic (reporting the literature using metadata) and non-bibliographic (maintaining the type of factual information that when online is increasingly used for e-research through time series and other data compilations). Bibliographic services tend to be fewer in number but are more widely used. For example Russell and Hartwell, in a directory of agricultural information sources then available in Australia, identified 21 bibliographic databases, many of them produced outside Australia, and 62 non-bibliographic databases, all produced in Australia. (4) This work is confined to bibliographic services, and comprises case studies of six such services.

Research Method

This paper has arisen from a detailed case study of several STI services using a case study protocol, and supported by interviews with key participants, exploration and use of different versions of databases produced, and reference to literature, archives and supporting material created to support users of databases. A descriptive case study methodology (5) is applied in which the unit of analysis is a system of action--in this situation the establishment and maintenance of a service, applied over multiple cases.

Project objectives included providing an overview of development of STI services in Australia, extending this overview with a detailed investigation that takes account of public policy influences and corporate imperatives, and testing the utility of a case study procedure derived from a description of discipline formation. Information was collected via a combination of approaches requiring examination of published and archival documentation; interviewing of key figures who were involved in the creation of the national services; and study of the systems underlying, and functionality provided by each of the services. Case study questions were structured according to the context of a recently written book on information management. (6) The outcomes are documented case studies of the STI services, an overview of development reported in Part I, and an analysis of discipline formation reported in Part II with respect to operational, analytical and administrative domains.

In Part I, the characteristics of the databases are compared within the context of some commentary on national scientific publication, the use of databases that record the output, and public policy influence on their development. This leads to some discussion about the ways in which continuation of the STI services may be ensured.

Scientific Publication Output

Bibliographic STI services have performed an important role in the information life cycle. Secondary sources of information, such as specialist bibliographies on scientific subjects originated in the eighteenth century and, by the beginning of the twentieth century, had been formalised into abstracting and indexing services, such as Chemical Abstracts, that were the forerunners of many of the STI databases available today.

The future of the scholarly publication that is reported and accumulated in these databases has been the subject of intense scrutiny in recent times through conferences and numerous publications. Stakeholders such as authors, editors, publishers and research managers continue to grapple with the changes made possible in publishing models through development in information and communications technology (ICT).

Greater apparent accessibility through the interact, particularly the Web, has been facilitated by systems such as content management and cooperative work groups, together with facilities such as digital archives and e-prints servers. These have been bolstered by what is sometimes called the hidden Web--the great number of databases available via subscription through Web interfaces, though not usually available to Web search engine crawlers. Many of these databases have been available since long before the Web, at least for provision of metadata. They provide a continuing impetus for information quality and increasingly they link full text with metadata. Yet they must contend with multiple alternative avenues to the same information, as access to the same digital content is facilitated through stand-alone and aggregated portals of universities and professional associations. A case in point can be found in contributions to AARL, which are made available on the publications server of the Australian Library and Information Association (ALIA). Metadata for the contributions is provided in a number of international and national databases. Some of these, for example the ACER database [A.sup.+] Education and the NLA APAIS database, also provide links through to the full text at ALIA. As authors hold copyright for the material, contributions may also be made available via their own institution's servers in the form of preprints or postprints, thereby becoming accessible directly via search engines or more refined approaches, such as NLA Arrow or Google Scholar.

Increased access to information does not necessarily equate to improved organisation. Although a case may be made for multiple metadata descriptions to suit different contexts of use, many of the avenues to the same content may provide cursory or uncontrolled metadata and rely on full-text indexing for access. The resulting reduced ability to filter and refine search results could see the document hidden within large yields of search results.

The importance of providing access to the nation's research output was recognised long before the Web and was one of the early stimulants to information policy discussion. In the area of STI Australia's contribution to the overall literature is about 2% of the world total, though in some fields--certain branches of astronomy, medical science and agriculture--output has been disproportionately high. Recognition of the relatively small proportion of Australian literature being indexed internationally happened in the 1960s but it took some time before there were significant attempts to quantify what was not being covered. These attempts were generally undertaken as part of the process of identifying publishing that had to be inspected by institutions establishing database services. Such analyses were internal working documents. Some became public as databases were created along with guides to database coverage.

However, there were some published analyses across databases. For example, in 1981 Abbott reported on 1970s data showing the number of Australian journals covered by 13 overseas STI databases, and noted that there continued to be gaps locally, resulting from areas that CSIRO's Australian Science Index (ASI), extant since 1976, was not covering. (7) In 1983 Byrne looked into social sciences and the humanities, as well as STI. (8) His analysis showed that the coverage of literature from Australian sources varied, usually within the range of 1-3% of the global output. He also compared the international coverage of Australian STI research literature with its coverage in ASI, noting that coverage of Australian periodicals by…