The pervasiveness of computers in our society has become a cliche. It has, however, done much to highlight the need to make computers easier to use, particularly for the increasing number of users who are not computer specialists. But ease of use is also an issue for computer professionals. Even experienced computer users can become very frustrated with a poor user interface. In spite of all the advances in computer science, the issue of quality human-computer interfaces overshadows all other issues which fill the computer science literature. Fancy features and algorithms matter little if they are inaccessible to the user because of a poor human-computer interface.

Lay people who hear of current research in human-computer communication are amazed to discover that this kind of work has not already been done. Why, in fact, has computer science progressed so far in other areas without solving many of the problems of human-computer in-
teraction? The problem of designing large interactive systems has always been fraught with great complexity. The still-young field of software engineering has devoted most of its attention to reducing that complex-
ity. The abilities of the systems analyst and the programmer have been taxed to the limit simply to get a system designed and built to meet functional requirements. But the problem is one of communication, not
computation.