Position Statement for the
ACM Workshop on Strategic Directions in Computing Research,
Working Group on Human Computer Interaction

Jonathan Grudin

Computers, Organizations, Policy and Society (CORPS) Group
Information and Computer Science Department
University of California, Irvine
Irvine, CA 92717

The End of Exile


The figure depicts several major foci of computer systems research and development and approximately when they first became prominent. Despite our tendency to think of "the computer field," each has inspired different research communities (and development organizations): software engineering, management information systems, office systems, human-computer interaction, and computer-supported cooperative work. Their principal research conferences were initiated in the same sequence: ICSE(1974), ICIS (1980), COIS (1982, now COCS), CHI (1983), and CSCW (1986).These communities have interacted less than one might have expected, although opportunities for synergies exist. And in fact, CSCW, a strikingly robust conference despite having no host organization, drew from the others, most heavily from the three preceding it. This is significant, and by no means happenstance.

What may be happenstance is that the field called human-computer interaction enjoyed a decade of relative isolation, marooned quite happily on the island called Stand-Alone PC. It is recorded that the Apple Macintosh was by conscious design difficult to network; it was designed to be a single-user system. One can understand the logic of that decision, although in hindsight...Whether or not by design, the IBM PC was also notoriously difficult to network, more so even than contemporary 8-bit machines.

With individual users working in isolation and exchanging results as hardcopy, the focus of design and development was wholly upon the tight interaction of one human and one computer. Perceptual and cognitive psychologists and graphic designers were able to focus with little distraction on much-needed improvements in the interface. Perhaps of equal importance, individuals retained a discretionary control of software acquisition in a way not possible before... or, very likely, in the future.

Current status.

Networking has brought PC users and those doing associated research or development out of private rooms and into the offices, organizations, and agencies that have otherwise always dominated the use of computers. The computer as a computer had given way to the computer as a container (which also did some computing), and is now yielding to the computer as a window (which may also hold some objects, and do some computing, for us). Human-computer interaction must turn and face the social and organizational issues that flow around and through our networked machines.

We need not all conduct ethnographic studies of workplaces, or build groupware. Productive research still emerges from most of the traditional subject areas of computer science; likewise, work on perceptual and cognitive aspects of interface design, on design methods and development tools, and other threads of human-computer interaction continues. However, they are all affected by the changing playing field. For example, one's choice of system and application is of increasing consequence to one's colleagues in a networked organization. Pressures build toward collective purchasing decisions; standards, de facto or otherwise, dominate more quickly. This in turn leads to reduced opportunities for innovation in established niches. We must avoid designing the equivalent of better keyboard layouts than QWERTY, oblivious to a world that has moved on to new problems. Human-computer interaction at an abstract level will always be significant, but any individual or collection of practitioners (of research or development) could.

Most or all aspects of human-computer interaction research will proceed, but we would be remiss not to observe that computer-supported cooperative work, groupware (however you choose to define it), and other social aspects of computing are in ascendancy, as we start to see organizations that are completely intranet worked and tens of millions of people internetworked. The CSCW conference is dominated by CHI members, yet it draws almost as many research paper submissions as the parent conference.

Strategic issues.

I am torn between identifying the strategic issues that most concern me personally, and those that concern groups such as computer-supported cooperative work, participatory design, or society and the future of computing that are not strongly represented in this gathering.

How do we design and introduce software to be usable and useful when its use will be centered on groups rather than in individual interaction? Laboratory evaluation was at its peak effectiveness when isolated PC use very much resembled typical laboratory situations. Methods to design and develop in the new circumstances of socially situated use are an area of intense focus today, and will remain a strategic issue for the future.

The relationship of standards to design and innovation is of critical importance in human-computer interaction (as well as elsewhere). Standards both stifle and enable innovation. Along with copyright, this is an area of intellectual as well as practical significance; it is unfortunate that it has not become a more active focus of research in its own right, as the alternative is to move forward in relative blindness to the consequences of our choices.

In my view, the single most important issue is the cognitive and social consequence of the quantum leap in the immediacy and visibility of activity that our technologies are handing us. We are able to see aspects of the world that we could not see before. This might enable us to organize better, but tensions develop around disjuncts between what we see and what we thought was there. In his book Bill Gates noted that we have the technology to insure total compliance with speed limits, and now we must decide how we really feel about traffic law enforcement.

Like other communication media, computation and networking enable immense creative possibilities. Like other communication media, they also have tremendous constraining effects, cutting off creative possibilities. Understanding the interplay of the enabling and constraining effects of new technologies is essential.

Relations with other disciplines.

As of this writing, the human-computer interaction working group may be delegated to cover a broader expanse than we usually embrace. There are workgroups on education and on electronic commerce and digital libraries, but none appear to cover graphics, multimedia, hypermedia, CSCW, privacy, or other social impacts of computer technology. We must be proponents for building bridges to the groups focused on these topics; in many cases there is membership overlap with ACM SIGCHI and the conferences it sponsors. This gathering cannot address strategic directions in computing research at the dawn of an era in which computation will have a substantial, direct effect on hundreds of millions of people, and fail to support consideration of the full range of effects our technology will have on these groups, organizations, and societies.

As computer use becomes an integral part of the work of every discipline, human- computer interaction will almost by definition be one logical meeting-place for any discipline and computer science.