N88-16418 Intelligent Man/Machine Tnto_faco _ on the S Dace 8tation

The computer systems that will be resident on the space station will necessarily be large, complex configurations of software and/or hardware which must function in concert to maintain the operations of the mission. However, much of the complexity should be transparent to the user of those systems, allowing the user to concentrate as much as possible on the purpose of his/her interaction with the computer, rather than unnecessary detail. This paper addresses some important topics in the development of good, intelligent, usable man/machine interfaces for the space station. These computer interfaces should adhere strictly to three concepts or doctrines: generality, simplicity, and elegance (just as the programming language code below these interfaces should follow). Generality refers to the commonality of usage and the similarity of form and function that should predominate all the interfaces, so that the user is provided with computer working environments that appear and perform in much the same way. Simplicity is obviously desirable, but not a concept to be taken lightly. It is very important that the interfaces simplify operations wherever possible (and desirable), and make intelligent inferences about the intent of the user to save time and unnecessary attention to detail. Finally, the elegance of the interfaces should be of great concern. The interfaces should be as concise as possible, exhibiting the "principle of least astonishment". The author will also discuss the motivation for natural language interfaces and their use and value on the space station, both now and in the future. AI provides an extremely powerful tool with which to think about and develop software applications to run on the space station. But, without well-thought-out, intelligent, truly usable man/machine interfaces to harness this asset, much of this power will be lost