Networked information resource discovery: an overview of current issues

Users need a new class of information retrieval systems to help them utilize effectively the increasingly vast selection of networked information resources becoming available on the Internet. These systems-usually called network information discovery and retrieval (NIDR) systems-must operate in a highly demanding, very large-scale distributed environment that encompasses huge numbers of autonomously managed and extremely heterogeneous resources. The design of successful NIDR systems demands a synthesis of technologies and practices from computer science, computer-communications networking, information science, librarianship, and information management. This paper discusses the range of potential functional requirements for information resource discovery and selection, issues involved in describing and classifying network resources to support discovery and selection processes, and architectural frameworks for collecting and managing the information bases involved. It also includes a survey and analysis of selected operational prototypes and production systems.<
>