Synopsis
This book is a collection of 16 papers that trace the author’s exploration, understanding, and implementation of artificial intelligence concepts and design principles in decision-support software systems. The papers were presented over a 20-year period (1997-2017) at an annual series of international symposia (InterSymp) organized by the International Institute for Advanced Studies in Systems Research and Cybernetics (IIAS) in Baden-Baden, Germany. The work was performed initially in the Collaborative Agent Design Research Center (CADRC), which was established by the author and a colleague (Art Chapman) at California Polytechnic State University (Cal Poly, San Luis Obispo) in 1986 and expanded into the commercial realm with the formation of CDM Technologies Inc. in 1994. In 2010, CDM Technologies was acquired and merged into Tapestry Solutions, a non fully integrated subsidiary (NFIS) of Boeing. The author started his career as an architect in the early 1970s after completing a Bachelor of Architecture degree at the University of Melbourne and Master of Building Science and Ph.D. in Architectural Science degrees at the University of Sydney in Australia, but quickly recognized the profound changes that computing capabilities would bring to all professions and society as a whole. Since the principal benefits of computers were initially seen to be their ability to perform mathematical computation with unimaginable speed, they were applied exclusively to rote data-processing tasks. While this role of computers was universally accepted by computer users, it appeared to be unnecessarily restrictive to the author. The vision that began to formulate in the author’s mind was a collaborative computer-user partnership in which the computer would have some understanding of the user’s intent. The question that became an intense pursuit for the author was; - How could this be achieved? After a great deal of thought and some experimentation it became clear that a major difference between user and computer was that human users understood the context of their undertakings, while the computer was simply following the software instructions without any understanding of the purpose of those instructions. Over the next 30 years, the author together with colleagues pursued the goal of representing subject domain context in software applications. The virtual context model had to be machine processable so that the computer would have some level of understanding of the intent of the planning, monitoring, analysis, and execution functions that the user was attempting to perform. The papers also portray a parallel struggle. During this period developments in information technology were moving ahead at such a rapid pace that they were outstripping the ability of the user to fully comprehend the opportunities for computer-based decision-support tools with orders of magnitude greater capabilities. There was a need to explain the concepts involved in the new technology so that they would inspire the user community to increase their expectations. During the 1990s both military and commercial users were still struggling with the effective utilization of the first generation of rote data-processing applications. Concepts related to information representation, embedded intelligence (i.e., software agents), and service-oriented architecture design principles were generally not understood by users. Although tools for implementing a new generation of information-centric systems had become commercially available, the user community was still largely satisfied with more mundane data-processing capabilities. Therefore, the reader will detect a secondary intent in many of the papers; - namely, to communicate the difference between data and information. The notion that computer software could have at least some limited understanding of the context of its application domain and how this feat could be accomplished had to be explained in intelligent layperson terms.
"About this title" may belong to another edition of this title.
