Autumn 1994

The history of interactivity

Bob Cotton
Overview

Interactivity is one of the central concepts of multimedia. Bob Cotton, co-author of The Cyberspace Lexicon, traces key stages in the development of our relationship with the screen

While cognitive psychologists have shown that all media (indeed, all experiences) are ‘interactive’ in that they demand the active participation of the recipient in processing and interpreting sensory stimuli, the concept of human-computer interaction is more recent. The study of this area of interaction (also known as human factors, human-computer interface design and so on) began with the second generation of commercial computers in the 1960s – room-sized entities known as ‘mainframes’ made up of a collection of components assembled in a space frame.

The first generation of commercial mainframes were controlled by specialist operators using batches of punched card-files; this second generation could be addressed from remote terminals on a time-sharing basis. The terminal interface was based on a simple conversational model: the user would be prompted by the computer, would type in a command and would then be prompted by the computer again. In the 1980s this command line interface was superseded by the WIMP interface, a graphical representation of files, directories and applications in the windows / menu / pointer configuration familiar with Microsoft Windows and the Macintosh.

During the twenty years when the command line was still the dominant interface, several individuals were developing very different models of interactivity, laying the foundations for both the graphical user interface and the multi-dimensional interfaces that are likely to be a feature of consumer telemedia and the data superhighway. Probably the single most important contribution was that of Ivan Sutherland, whose revolutionary Sketchpad system effectively mapped out the future of interactive computer graphics. Sutherland went on to develop real-time scene generators for flight simulation and later founded Evans and Sutherland to develop military and commercial flight simulators and, more recently, game simulators.

At about the time that Sutherland was investigating real-time computer graphics, computer guru Theodore Holm Nelson published a collection of writings, drawings and ideas about how computers could be made accessible to everyone. Nelson wanted computers that were easy to use, fun, that showed pictures and played sounds. He developed the key concepts of hypertext and hypermedia and outlined a way in which these information-linking devices could be used within a global library system he called Xanadu. Like Coleridge’s ‘Xanadu,’ Nelson’s dream of digitising the world’s literature (and art, film and photography) and making it universally available through an easy-to-use interface was never completed in its original form. But his idea of a global information network pervades the thinking behind the data superhighway.

Nelson was also the first, in the early 1970s, to draw attention to the (then new) video arcade games. Here was a computer interface that anyone could use, that was colourful, animated and noisy, and that offered an intensely visceral experience combined with a suspension of disbelief that was to earn its producers billions of dollars. Nolan Bushnell, designer of the first hugely successful computer arcade game Pong, founded the Atari company to exploit it and other consumer entertainment software products. In the early 1980s, the Atari Research Lab fostered the talents of some of the most important interface and interactive software designers of the 1990s, including Brenda Laurel, Jaron Lanier (VR systems), Thomas Zimmerman (Dataglove) and Alan Kay (Dynabook, the Xerox Alto graphical user interface and Smalltalk object-oriented programming).

Kay’s work on the Alto interface at Xerox Palo Alto Research Center (PARC) inspired Macintosh interface designer Steve Jobs, who used many of Kay’s ideas in the Macintosh Finder – the best computer interface yet devised. In the late 1960s Kay had designed the Dynabook, a pad-like personal computer using a graphics-oriented interface that was easy and cheap enough for schoolchildren to use and could be connected to a communications network. Nearly 30 years later, the Dynabook has at last almost arrived, and Kay is still working with children to discover how they learn from computer software and interactive games.

Arcade games like Pong represented the first use of computers for pure entertainment. Existing arcade games such as pinball machines already demanded a high level of interaction, but the new videogames provided interaction with greater frequency, greater range and deeper significance. These were the three criteria for assessing interactivity identified by Brenda Laurel in Computers as Theater – frequency being how often you interact, range the number of choices you have and significance how much what you do affects the outcome. Games designer Trip Hawkins, founder of one of the largest consumer games software companies – Electronic Arts – and the driving force behind 3DO, defines good consumer interactive software as simple, hot and deep. For Hawkins, simple means easy to get into; hot software optimises the hardware, interactivity and media; deep refers to the levels of operation, play and the depth of information and experience on offer. Good software should fulfil, or even transcend, its initial promise; brilliant software should satisfy, surprise, enrich, excite and inspire.

But computer interfaces and video games are only part of the equation that will define the interface for the digital superhighway. There are three other important ingredients: multimedia (TV resolution stills and video and stereo audio), the third dimension, and support for multiple participants.

Computer-based multimedia arrived in 1978 with the first demonstration of interactive video by the Architecture Machine Group (later the Media Lab) at MIT. Headed by Nicholas Negroponte, the Archmac group linked a prototype Philips laserdisc system to a computer to allow users to tour the city of Aspen, Colorado. By manipulating a joystick, users could choose their direction of travel, change the season, find out more about the buildings they were driving past and so on. This first surrogate travel program proved that multimedia was the future of computing. The latest multimedia interfaces integrate animation, video, stills, sound and 3D graphics.

The third dimension was being explored through the 3D computer graphics revolution pioneered by Sutherland, through real-time scene generators (some of which were adapted to display stereo images in binocular goggles or ‘eyephones’), and through the projected artificial realities of artist Myron Krueger. By dissociating the computer image from the monitor screen, Krueger pointed the way towards fully interactive environments of the kind we may enjoy when wall-sized flat screens or videobeam technologies became ubiquitous.

Recent 3D computer graphics techniques are providing artificial realities of another kind – realities where we can morph fluidly from one object or person to another, build soft actors or ‘synthespians’, model complete terrains with fractals, and digitally compose video and computer-generated images and sounds into a seamless interactive interface. But whatever shape cyberspace media take, their unique selling point is that they are a feature of a telecommunications network, and the telemedia interface must accommodate activities – family gatherings, teleworking, games networks or videoconferencing – where many people in different locations interact at the same time. Valuable contributions to this multiparticipant interface have been made by developers of videoconferencing systems, but the most exciting are the networked virtual communities of Habitat and its successor Club Caribe.

These are virtual environments (albeit of comic strip visual simplicity) accessible through thousands of personal computers linked by modem into a mainframe. Habitat is a virtual townscape where users can start businesses, publish newspapers, run for mayor – any of the things you might do in a real township. The citizens of Habitat are the cartoon-like characters created or customised by their human operators, who can control their onscreen surrogates and engage in conversations with other citizens through comic strip-style speech bubbles. Designed by Chip Morningstar and Randall Farmer at Lucasfilm Games in 1986-88, Habitat points the way towards the kind of communities that might evolve in the real-time, high-resolution world of the digital superhighway, the net of nets that will define the future of interactive telemedia.

Bob Cotton, co-author of  The Cyberspace Lexicon (Phaidon), London

First published in Eye no. 14 vol. 4 1994