The path towards personal computing was not taken by chance. Alan Kay directed the Learning Research Group (LRG) at PARC. His master and doctoral thesis at the University of Utah about the Flex Machine (cf. 3.1.4) became the research program for the group. Many aspects of the Alto computer are directly influenced by Kay’s vision of the Dynabook (cf. 3.1.4), A Personal Computer for Children of all Ages [Kay 72a]. Alan Kay and Adele Goldberg took the Alto computer as an Interim Dynabook into the classroom. The article Personal Dynamic Media [Kay/Goldberg 77] shows that school kids were able to program the computer with the newly developed object-oriented programming language Smalltalk (cf. 3.1.5). The work with children revealed many interesting insights into the field of human-computer interaction.
in Vision and Reality of Hypertext and Graphical User Interfaces
«Machines which do one thing only are boring», says Alan Kay in The Reactive Engine [Kay 69]. With the Flex Machine Kay aims in the late 1960s for a tool that can support human cognition processes. His starting point is the human language. People describe the world with words. They use language to communicate with each other. Language is most basically used to think about the world. It follows a special form – it’s syntax. And it has a meaning, carried representations of the world and of abstract concepts – the semantics of language.
These considerations call for a computer, that should participate in an interactive dialog with a human, that the interface needs to be flexible enough to express new ideas; as Kay formulates, «it has to be able to form the abstractions in which the user deals» [Ibid.].
The Reactive Engine outlines the system design of the Flex Machine. The computer is equipped with a standard keyboard as well as a five-finger chording keyset like Engelbart’s NLS system. A graphic tablet is installed for 2-dimensional input. Output is to be drawn on a calligraphic 1024 by 1024 pixel CRT.* * A calligraphic cathode ray tube is a special vector display. Bitmapped displays are to be invented at Xerox PARC about four years later. Like Sketchpad, the actual drawing area is much bigger than the screen size. The drawing algorithms of the Flex Machine can map any rectangular area of a virtual screen to the monitor. And the monitor can even be divided between several virtual screens. However window controls like scrollbars to manipulate the region of the virtual screens are still not employed.
The Flex Machine ought be so compact to fit on a desktop. An utopian concept for the computer technology of the late 1960s. And this goal could never be reached for the Flex Machine, although various components have been implemented on several mini computers. For example, compilers for FLEX have been developed for the Univac 1108 and Doug Engelbart’s SDS 940 at SRI.
The FLEX language is integral part of the machine (cf. 2.1.5 Flex and Smalltalk). The user interacts with the system entering FLEX statements that get executed immediately and the results are displayed on the monitor.
Fig. 3.5 Alan Kay’s vision of the Flex Machine, 1969
With the Flex Machine Alan Kay lays the foundation for Personal Computing. With the Dynabook his vision takes full shape. The Dynabook is inspired by the invention of the flat-panel display, although the prototype of 1968 is just a 1 square inch piece of glass. Alan Kay, now at the Xerox’ newly founded research laboratory in Palo Alto, starts to ponder about a conjunction between the Flex Machine and the flat-panel display. Two papers present the resulting concept, A Personal Computer for Children of All Ages [Kay 72a] and A Dynamic Medium for Creative Thought [Kay 72b].
The Dynabook is the distilled vision of a «personal, portable information manipulator» [Kay 72a, p. 1]. It is a device useful and powerful enough to support adults in their daily work, and it should be simple enough to be used by children for playing and learning.
Fig. 3.6 (a) Children playing with Dynabooks and (b) a mockup of the Dynabook from 1972 [better resolution: ChronoLink | Internet Archive]
Fig. 3.6b shows a mockup of the computer. The shape of the case should be similar to an ordinary notebook. The Dynabook has a keyboard, and a stylus is used as a pointing and drawing device directly on the surface of the display. Alan Kay would have liked to have a black and white flat-panel display of 8H by 11 inch with a resolution of at least 100 DPI. This high resolution is necessary to compete with paper for legibility of text. The option of several different bit-mapped fonts and font faces is crucial to make the Dynabook a personal device.
To illustrate how a Dynabook can be used, Alan Kay portrays a scenario with two children. They have connected their Dynabooks to play the multi-agent game Spacewar. After they lost interest in playing they discuss how to incorporate the attractive force of the sun. They know how to reprogram the game, because they have written it themselves. But they have no idea about the mathematical formulas for gravity. They decide to ask their teacher for help. The teacher connects her Dynabook to the local library to look for information about our solar system. She copies articles that match her criteria to her Dynabook and gives some hints to the waiting children how they can improve their Spacewar program.
The scene touches on many important aspects for personal computing. First of all everyone has its own computer. Kay foresees a price tag that is in the same range with a television set. This alone is pure science fiction in the early 1970s when minicomputers start at $10,000. But a low price is necessary to make the Dynabook affordable to everyone. Dynabooks should also be portable and independent from wired networks for data transmission and power supply. All this is merely a matter of time and technological progress. Really important are the implications of the personal computer as a new medium. Alan Kay defines [Kay 72a, p. 3]:
What then is a personal computer? One would hope that it would be both a medium for containing and expressing arbitrary symbolic notions, and also a collection of useful tools for manipulating these structures, with ways to add new tools to the repertoire.
The children in the scenery above use their Dynabooks as a personal medium in the way that they use it for reflexive communication. They are author and audience of the medium and use the Dynabook like a conventional notebook – although as a slightly magically enhanced one.
Alan Kay continues that it is essential for personal computers to be «superior to books and printing in at least some ways without being markedly inferior in others» [Ibid.]. The teacher connected her Dynabook with ease to a local library. Entire books can be transferred to the Dynabook for study and leisure. This is also an indication for a new kind of medium as Marshall McLuhan argues in Understanding Media: The Extensions of Man [McLuhan 64], that new media adopt the content of previous media first. But a computer can give access to the content of a book in a more flexible way than its printed counterpart can do. Kay contends, «It need not be treated as a simulated paper book since this is a new medium with new properties» [Kay 72b, p. 3]. From here to Ted Nelson’s vision of a global form of hypertext is just a small step.
Why children? The reason why Alan Kay took two children for his introductory scene is more profound than just to accent the Dynabook’s easy and straightforward user interface. Alan Kay is impressed by the psychological works of Jean Piaget of the 1920s and Jerome Bruner’s of the 1960s. Both scientists have studied children to figure out how they learn. The result is a model of human learning and cognition processes that forms the theoretical background for Alan Kay’s Learning Research Group at Xerox PARC (LRG). Kay and his team manage to develop principles for human-computer interaction that are in accord with these psychological models of human cognition (cf. 3.2.2 Three Stages of Human Development).
For a free PDF version of Vision and Reality of Hypertext and Graphical User Interfaces (122 pages), send an e-mail to:
mprove@acm.org I’ll usually respond within a day. [privacy policy]