© Copyright 2006 Texas Digital Library. All Rights Reserved.
This paper explores the idea of dynamically adding multi-destination links to Web pages, based on the context of the pages and users, as a way of assisting Web users in their information finding and navigation activities. The work does not make any preconceived assumptions about the information needs of its users. Instead it presents a method for generating links by adapting to the information needs of a community of users and for utilizing these in assisting users within this community based on their individual needs. The implementation of this work is carried out within a multi-agent framework where concepts from open hypermedia are extended and exploited. In this paper, the entities involved in the process of generating and using "context links" as well as the techniques they employ to achieve their tasks, are described. The result of an experiment carried out to investigate the implications of linking in context on information finding, is also provided.
View full text: HTML
Existing library classification systems, such as the Dewey Decimal Classification (DDC) system and the Library of Congress Classification system, are generally considered effective at bringing similar documents together. For example, the DDC system is widely used to classify books in libraries and increasingly is being used for online applications, such as browsing Online Public Access Catalogues (OPACs) and Cooperative Online Resource Catalogs (CORCs), and will probably see increasing use in digital libraries. Based on theoretical considerations, a classification system consistent with the Gray code provides an optimal organizing principle for documents; it can be used to classify knowledge, as can the classification systems commonly used in both paper and digital libraries. What is the relationship between a theoretically optimal Gray code-based classification system and the existing classification systems? We suggest that the non-reflected Gray code provides a basis for ordering documents that is more consistent with existing classification systems than is the more frequently discussed reflected Gray code. This provides both a theoretical basis for existing techniques and a standard by which document organizing systems in digital libraries may be compared, evaluated and improved.
View full text: HTML
Hypertext linking is regarded as a key element of identifying or building hypertext structure. However, links provide a nonlinguistic excess that has generally been under-theorised in hypertext criticism. When combined with the role of teleological contextualisation in interpretation links have been largely misunderstood. This paper utilises the example of cinema in relation to hypertext to explore alternative conceptions of links and the relation of meaning in hypertext to narrative teleology suggesting connections to other practices of argument that may be relevant to structure.
View full text: HTML