As we celebrate twenty-five years of knowledge capture research we can view it from a short-term perspective as a substantial component of the sixty-year development of digital computing technologies, or from a long-term perspective as part of the most recent segment of the hundred millennia evolution of recorded knowledge processes that have shaped our civilization. We can trace the development of knowledge capture processes similar to those we now study: from the Neolithic origins of our civilization; through the Babylonian development of mathematics and writing; Greek innovations in logic, ontology and science, and their medieval elaboration; the development of formal logics, metaphysical systems and sciences stemming from the scientific revolution; to the computational implementation of knowledge representation, capture, inference and their ubiquitous application in our current information age. This presentation outlines major events in the trajectory of knowledge capture processes over the millennia, focusing on those relevant to where we are now and where we may be going. It encompasses: the evolution of civilization from archeological, economic, sociocultural and systemic perspectives; highlights in the formalization of knowledge capture processes through the ages; trajectories of the development of knowledge technologies supporting its representation, capture and use; to projections of expected major issues and advances in the next quarter century.

Group facilitation is an important element of group approaches to requirements engineering (RE). The facilitation in traditional face-to-face groups is challenged by the increased globalization of the software industry. Thorough empirical investigation of human facilitation in computer-mediated requirements meetings is needed. This paper presents findings about the facilitation of distributed group settings in a controlled environment. Three professional facilitators mediate 15 three-person groups negotiating software requirements. Facilitation in face-to-face meetings is contrasted with four group settings in which the facilitator is physically separated from the group or co-located with key stakeholders. Rich qualitative and behavioral data enables an understanding of differences and similarities in the facilitation of the distributed groups and of aspects that were detrimental or beneficial to their facilitation. The empirical evidence indicates a reduced richness of social behaviors in computer-mediated group settings which: made the group facilitation problematic; but also enabled certain facilitation support in the medium itself.

The Internet/World Wide Web has grown very rapidly to become a major resource supporting collaborative activities in a wide range of groups, disciplines, and communities. However, the growth of the net/web creates problems of information overload and of maintaining awareness of activities at other sites relevant to one's own tasks. This article discusses awareness issues in collaborative group; develops a conceptual framework for studying and supporting awareness on the web; describes CHRONO, a tool for supporting awareness of changes at another site. Finally a preliminary sketch of an overall conceptual framework for the Internet is presented and explored. This CyberOrganism framework is used to analyze and classify chronological awareness support systems and general communication systems that support awareness maintenance on the Internet. © 1997 John Wiley & Sons, Inc.

The Internet (the net) and World Wide Web (the web) have grown rapidly in the past decade and have come to play a major role in supporting discourse and publication in scholarly communities. The development and application of new services has been very rapid with little central planning, and despite the widespread use, there is little information as yet on the human factors of the use of the net and web. In particular, models of the human factors of individuals interacting with workstations have to be extended to take into account the essential social aspects of computer-mediated discourse and publication. This article provides a framework for analyzing the utility, usability, and likeability of net and web services, and illustrates its application to significant aspects of supporting scholarly communities. The utility of the net and web are measured in terms of the growth of usage, and the different services involved are distinguished in terms of their specific utilities. A layered protocol model is used to model discourse through the net, and is extended to encompass interaction in communities. An operational criterion for distinguishing different communities is defined in terms of the types of awareness that resource providers and users have of one another. A temporal model of discourse processes is developed that enables the spectrum of services ranging from real-time discourse to long-term publication to be analyzed in a unified framework. The dimensions of awareness and time are used to characterize and compare the full range of net services, and model their unification through the next generation of web browsers. © 1997 John Wiley & Sons, Inc.

No abstract available.

Knowledge modeling perspectives on knowledge acquisition suggest that it is reasonable to analyze knowledge bases as collections of models. This article focuses on those parts of the knowledge base that model the practical reasoning processes of human experts, and asks what properties those models might be expected to have. It surveys the general notion of a model and its connotations in information systems science. It analyze the structure of practical reasoning as a control process with information flows involving essential uncertainty and adaptivity, where robustness is more significant than optimality. Systemic considerations suggest that models of the relevant knowledge will consist of a collection of isolated productions, many of which will be concerned with avoidance rather than goal achievement. the relations of such models to role-limiting methods, generic tasks, and deep knowledge are discussed. Finally, the implications of the knowledge modeling perspectives for developments in knowledge acquisition are discussed. © 1993 John Wiley & Sons, Inc.

A study is reported of the use of ripple-down rule induction to develop a metamodel of ten years of clinical data captured as part of the development of an expert system for thyroid diagnosis. It is shown how the suitability for inductive knowledge discovery from such real-world data can be characterized in terms of its stationarity, and how the best error rates achievable and the amount of data necessary to achieve them can be estimated.>

The knowledge acquisition bottleneck impeding the development of expert systems is being alleviated by the development of computer-based knowledge acquisition tools. These work directly with experts to elicit knowledge, and structure it appropriately to operate as a decision support tool within an expert system. However, the elicitation of expert knowledge and its effective transfer to a useful knowledge-based system is complex and involves diverse activities. The complete development of a decision support system using knowledge acquisition tools is illustrated. The example is simple enough to be completely analyzed but exhibits enough real-world characteristics to give significant insights into the processes and problems of knowledge engineering.>

No abstract available.

A decision support system is under development for planning in flexible manufacturing, using a consortium of knowledge-based systems utilizing expert system, database, and simulation techniques. An object-oriented approach is incorporated. There are six basic modules: machine selection optimizer; database; production flow analyzer; rapid modelling techniques system; FMS simulator; supervisor.

The author presents a conceptual framework for complex systems of computers and people. Distinctions between technology and people, and between computers and nonprogrammed technology, are analyzed. This analysis is used to show how various forms of analogy and abstraction may be used to derive design principles for person-computer interaction. The analysis is extended to include relations between system structure and behavior, and used to develop a hierarchical model of the protocols in person-computer systems.>

It is shown that experts may be modeled as managing the inductive dynamics of knowledge acquisition in the knowledge process of society, and that who becomes an expert may be modeled as a random process under the influence of strong positive feedback loops in the social mechanisms giving access to knowledge. These models have implications for the design of expert system.>

We have entered an era of intelligent system development in which fifth and sixth generation computing systems are expected to emulate and enhance natural intelligence. However, human knowledge processes are not well-understood and it is tempting to base the technology of intelligent systems on formal reasoning rather than human psychology. In particular, the attempt to base knowledge processing in fifth generation computing systems on logic programming suggests that we should examine the rationale behind standard logic. It was developed as a foundation for arithmetic and psychological considerations were ruled out as irrelevant. We should consider that there may be systems of logic more appropriate to knowledge processing. Logical and psychological system theories based on an axiom of unrestricted predication are plausible and attractive alternatives to standard logic. It is suggested that the design of logical inference processors should take such systems into account, and that investigation of the logical, psychological and systemic foundations of knowledge are fundamental to the development of intelligent systems.

A recent paper in this TRANSACTIONS1on an automatic inference procedure for probabilistic grammars contained major anomalies between the technique proposed and the computational examples given. These would mislead others attempting to reproduce the results and also undermine the apparent validity of what is in fact a sound technique. This letter details and corrects the anomalies.

It is shown that an adaptive threshold logic element (ATLE) whose weight vector is perturbed reconverges without error provided its threshold is large compared with the perturbation. This result is generalized to a criterion of convergence, based on admissibility, for the nonseparable case. It is shown that conventional ATLE's cannot cope with nonseparability even according to this very weak criterion.

It is argued that a principle of casuality is fundamental to human thinking, and it has been observed experimentally that this assumption leads to complex hypothesis formation by human subjects attempting to solve comparatively simple problems involving acausal randomly generated events. This correspondence provides an automatatheoretic explanation of this phenomenon by analyzing the performance of an optimal modeler observing the behavior of a system and forming a minimal-state model of it.

In conventional digital computation, continuous quantities are quantized and represented by binary words in which the number of bits determines the precision of representation. In stochastic computation, as described in this paper, continuous quantities are represented as a sequence of one-bit binary words (i. e., as a pulse stream or sequence of logic levels) in which the probability of an ON logic level is a measure of the quantity. Since probability is a continuous variable in the range O≤p≤1, this removes the effects of quantization. However, a probability cannot be measured precisely, only estimated subject to random variance, and hence there is an effective random noise in the output of the computer.