Abstract: "Because of its wide availability and open architecture, UNIX[superscript TM] provides a suitable and convenient environment for command user interface research. We believe a formal, yet practical, knowledge representation scheme for UNIX utilities is a necessary tool to build higher-level online help and active help systems. We propose a hierarchy of knowledge levels to represent knowledge of UNIX: static objects, command syntax, command synopsis, command semantics and UNIX semantics. We describe how to implement the first three levels using taxonomic classification technology. We illustrate their usefulness by showing how they can be used to correct some non-trivial mistakes made by novice users. The ultimate goal of this research is to realize a UNIX command user interface (UNIX shell) that is knowledgeable enough to give intelligent advice to the user. By intelligent advice, we mean advice of a kind previously available only from human experts."

In this international collection of papers there is a wealth of knowledge on artificial intelligence (AI) and cognitive science (CS) techniques applied to the problem of providing help systems mainly for the UNIX operating system. The research described here involves the representation of technical computer concepts, but also the representation of how users conceptualise such concepts. The collection looks at computational models and systems such as UC, Yucca, and OSCON programmed in languages such as Lisp, Prolog, OPS-5, and C which have been developed to provide UNIX help. These systems range from being menu-based to ones with natural language interfaces, some providing active help, intervening when they believe the user to have misconceptions, and some based on empirical studies of what users actually do while using UNIX. Further papers investigate planning and knowledge representation where the focus is on discovering what the user wants to do, and figuring out a way to do it, as well as representing the knowledge needed to do so. There is a significant focus on natural language dialogue where consultation systems can become active, incorporating user modfelling, natural language generation and plan recognition, modelling metaphors, and users' mistaken beliefs. Much can be learned from seeing how AI and CS techniques can be investigated in depth while being applied to a real test-bed domain such as help on UNIX.

Access Versus Ownership to Word Formation in Language and Computation

Innovations in adult learning are a complex mix of pedagogy, technology, organisation, strategy and vision. In this book, 17 cases representing state-of-the-art design and practice from nine different countries are presented, grouped around the themes of I. Innovative Instrumentation. II. Innovations in Learner Collaboration and III. Innovations in Practice.While all of the cases deal with innovative instrumentation (software or combinations of software and communication technologies), the six cases in section I offer a detailed look at software packages designed for some aspect of the innovation of adult learning, such as reducing information overload by intelligent tools or using a World-Wide Web environment for communication and learning.In section II, a variety of ways to innovate adult learning through collaborative activities are described, including learning scenarios that make effective use of collaboration and the technology and instrumentation that make collaboration at a distance possible.Section III describes innovative learning situations that have been successfully integrated into broad scale field settings, each focusing on a particular situation in which innovative technologies play a part.