The purpose of this study is to examine the information needs and behaviour of students, lecturers and researchers, in the APTIK universities as they performed their learning, teaching and research tasks. Causal models have been developed to represent the factors which influence each aspect of information needs and behaviour. In these models, determining factors have been grouped into input and process factors. A path analysis was used to test the postulated information needs and behaviour models, and to estimate the magnitude of the direct and indirect effects of factors taken as causes on factors taken as effects, in the models.

This thesis identifies the major problems involved in the use of computers in library and bibliographic information systems in Brazil, examines motives for and results of automation, and offers an in-depth description of existing automated systems. The researcher personally visited the thirty-one most significant automated systems in five localities in Brazil, and systematically collected numeric and descriptive data that could help characterise automated processes and identify problem areas. Questionnaires were sent out by post to institutions which could not be visited. Attitude tests were also applied to fifty-five senior staff of these institutions, to determine their ranking of problems, motives and results, their views of other relevant factors and their educational background. This same attitude test was also applied postally to a group of senior librarians in institutions which did not use computers; eighty-six replied.

The main objective of this work is to try to identify the common grounds existing between archives, library and information studies from which a core curriculum could be derived for the professionals in the three fields. The underlying concept in building up such a core is that the above areas must be viewed as a whole within the information spectrum. The historical development of librarianship is taken into account as well as the early education provided for librarians in England, United States and Brazil up to modern times when technological and social innovations have brought about changes in the information services, practices and procedures which have obviously had a direct effect on the educational patterns of members of the profession.

The economic importance of the contribution of information services of the Nuclear Information Centre of the National Commission of Nuclear Energy (CIN/CNEN) in Brazil is carefully analysed from the point of view of economic analysis.

Archives began when man first started to write and archival repositories were often found in archaeological excavations. Modern archives started after the French Revolution and are a nineteenth century achievement. The twentieth century, however, has brought revolutionary changes to the archival world: new materials, new techniques, together with new problems. Archives have re-assumed their pre-nineteenth century administrative role and re-established the link with record-producing agencies: modern records management and records centres have appeared on the scene. A wider public have started to use archival records. Archives have become information agencies. In developing countries, however, the economical difficulties that affect all aspects of life also affect archives. Governments with more urgent matters to solve do not give enough attention to these institutions. Archives are usually under-staffed, under-funded, lack adequate buildings, equipment and trained professionals.

The digital computer has greatly increased our capacity for storing and accessing information. The internet, and particularly the World Wide Web, have vastly increased the amount of information available to us. Information retrieval and hypermedia research has greatly reduced the time and effort required to fulfill a searcher's information need; however, problems still remain. To access <EM>multimedia</EM> information, text metadata must usually be assigned to the multimedia objects, requiring an (often prohibitively) large amount of time and effort. Alternatively, some systems use low-level features of the media to allow a searcher to find objects similar to a query object. Such features cannot always identify two media objects depicting or representing the same real-world concept; in the case of images, different camera angles and lighting conditions mean that semantically similar images are visually very different. Additionally, semantic relationships between objects may not be expressed. This thesis presents a way of addressing these problems by drawing on the field of semiology, in which a symbolic object has two parts: That which is being represented (in the <EM>plane of content</EM>), and the symbolic object doing the representing (in the <EM>plane of expression</EM>). A semantic layer is introduced holding real-world concepts, and connections to the multimedia objects that represent them. Knowledge about these concepts can be introduced by connecting them with semantic relationships. A prototype multimedia information system incorporating a semantic layer feature, the <EM>multimedia thesaurus</EM>, is introduced. The construction and use of a museum application is described, and used to evaluate the semantic layer technique. Finally, some reflections on these findings and some suggested future directions for the work are presented.

The study investigates the information needs and seeking patterns of rural people in the context of rural development. The research population consisted of respondents from selected villages in Botswana, Malawi, and Tanzania. To establish rural information needs the study looked at economic activities, problems faced in income generation, perceived barriers to community development, direct questioning of rural people and extension workers.

The purpose of this research was to examine the effect of information cues, presented via a hypertext system, on the fiction reading activity of public library readers and the practical implications of the effect. A three-group post-test experimental design was applied for this purpose. The groups were formed from readers in two public libraries near the university by random assignment. The experimental treatments were three versions of a fiction searching and browsing system, differing in the complexity of information cues and the hypertext features. Data for the experiment were gathered by an observation schedule and a self-administered questionnaire and were analyzed with the Kruskal-Wallis one way analysis of variance and the Mann-Whitney test in SPSSA. The three groups showed significant difference in the amount of their book selection mid borrowing, the extent to which they made use of the searching system and the extent to which they relied on their own experience for book selection. They also differed significantly in the types of book they borrowed. They did not differ significantly, however, in their fiction searching pattern, the conformity of their book selection to their general reading tastes, their subjective feelings of well-being and their cognitive experience during reading. It was concluded that within the typology of information cues proposed in this research, the amount of readers' book selection and borrowing, the types of book they borrowed, their reliance on the system or their own experience for the decision making are significantly influenced by the level of information cues they have been exposed to. However, readers' searching patterns, conformity of book selection to their general taste, and emotional and cognitive experience do not relate significantly to the level of information cues they have been exposed to. It was suggested that detailed categorisation or classification of fiction should be a priority in processing fiction. 'The provision of adequate information cues should have more professional attention in promoting fiction reading, and the policy of fiction services should not be too high-brow.

The thesis has grown out of an extensive survey of the literature pertaining to the county of Rutland. The survey was an original idea to produce a county bibliography which had both a short-title catalogue of conventional references and full physical descriptions of the books of the county. Two types of bibliography, analytical and systematic, are combined in one survey. The thesis is an account of this project together with a rigorous review of the theoretical background against which it was conducted. The practice of bibliography, and specifically of bibliographies of counties, is reviewed. A new definition of county bibliography is offered, together with its appropriate objects of study (book, pamphlet, leaflet, ephemera and publication are some of the specific concepts defined here). The research methods employed for the Rutland project are described and a plan outlined showing how a comprehensive survey of any English county's literature could be achieved. Especially important here are the lessons learned from on-line searching. There is a detailed discussion of inclusion and exclusion policies appropriate for a county bibliography, and an account of how this material should be presented and described. A new classification system for a county bibliography is outlined, and older solutions to the classification problem reviewed. The lessons of the Rutland project are discussed, including presentation of a statistical breakdown of the Rutland material, and whether the way in which the Rutland project was conducted could be copied for other counties. The Rutland statistics are contrasted with those for other counties. Several appendices present bibliographical information, lists of categories of literature encountered and samples from the Rutland survey.

This thesis examines the human issues underlying the design and usability of electronic text systems. In so doing it develops a framework for the conceptualisation of these issues that aims to guide designers of electronic texts in their attempts to produce usable systems. The thesis commences with a review of the traditional human factors literature on electronic text according to three basic themes: its concern with perceptual, manipulatory and structural issues. From this examination it is concluded that shortcomings in translating this work into design result from the adoption of overly narrow uni-disciplinary views of reading taken from cognitive psychology and information science which are inappropriate to serve the needs of electronic text designers. In an attempt to provide a more relevant description of the reading process a series of studies examining readers and their views as well as uses of texts is reported. In the first, a repertory grid based investigation revealed that all texts can be described in reader-relvant terms according to three criteria: why a text is read, what a text contains and how it is read. These criteria then form the basis of two investigations of reader-text interaction using academic journals and user manuals. The results of these studies highlighted the need to consider readers' models of a document's structure in discussing text usability. Subsequent experimental work on readers' models of academic articles demonstrated not only that such models are important aspects of reader-text interaction but that data of this form could usefully be employed in the design of an electronic text system. The proposed framework provides a broad, qualitative model of the important issues for designers to consider when developing a product It consists of four interactive elements that focus attention on aspects of reading that have been identified as central to usability. Simple tests of the utility and validity of the framework are reported and it is shown that the framework both supports reasoned analysis and subsequent prediction of reader behaviour as well as providing a parsimonious account of their verbal utterances while reading. The thesis concludes with an analysis of the likely uses of such a framework and the potential for electronic text systems in an increasingly information-hungry world.