In addition, console capabilities for "soft" (or transient) displays should provide flexible and convenient editing and control facilities including insertions, deletions, corrections, rearrangements, escale-changing, size and rotational transformations, automatic multiple copying, and the like. Still other uses of on-line reactive displays are for purposes of text editing, sorting, and printing as for example in programs under development at Bell Laboratories (Mathews and Miller, 1965) and at jethe the University of Pittsburgh. (Roudabush et al., 1965).5.1b When such techniques are applied to a complete publication cycle, by feeding the corrected tape products either off-line to automatic composing machines or on-line to a computer typesetting program, the loop is closed from original data to finished output product.5.1c

ha

5.1.1. Graphic Output and Display

n At least as early as 1953, a CRT display system was available on the Illiac.5.2 Similarly, "Whirlwind (at MIT) had a cathode ray tube and light pen in the early 1950's. A prototype of the APT system computer controlled machine tool) was programmed on Whirlwind in 1955." (Wigington, 1966, p. 86). Reactive display for purposes of air traffic control applications was investigated in the early 1950's at the National Bureau of Standards. A combination 36f the DYSEAC computer, special display equip ment, control devices including a joystick serving a Flight pen funtion, and a radar link to the Washington INational Airport provided the basis for on-line, Treal-time experimentation. With actual radar data ?being displayed on one scope, the operator was able Pto indicate targets of special interest which should be marked or brightened for him, and a second scope displayed the results of computer predictions of expected positions of target aircraft within specified time intervals. The same equipment could also be used for problem-solving simulations, such as war gaming.5.3

By 1962, an advertisement asked, with appropriate illustrations, "Would you believe a Calcomp plotter and any computer can draw pictures like these in seconds? . . . A perspective sketch of your new plant.

Statistical trend charts . . . molecular structure diagrams. . . apparel patterns, graded for sizes. and even the Mona Lisa". (Commun. ACM 10, A9 (1962).)

The development of full-scale graphical inputoutput communication and processing systems, with particular emphasis on man-machine problemsolving is of course represented first by M.I.T.'s Sketchpad. This was publicly introduced at the 1963 Spring Joint Computer Conference where Coons butlined the requirements for a computer-aided design system; 5.4 Ross and Rodriguez discussed the theoretical foundations of such systems, with emphasis upon appropriate languages and data structures; Stotz described the man-machine "console facilities, specifically including the display requirements; 5.5 Sutherland outlined the Sketchpad

graphical communication system itself,5.6 and an extension to three-dimensional drawing applications, Sketchpad III, was outlined by Johnson. The latter investigator remarked: "General threedimensional graphical communication, which deals. with arbitrary surfaces and space curve intersections, presents many difficult problems; the beginning has been modest and much work remains before the complete graphical communication problem is solved." (1963, p. 347).

Organizations neighboring M.I.T. have also been engaged in the development and utilization of graphic input-output equipment for man-machine reactive display in problem-solving applications, notably, Bolt, Beranek and Newman 5.7 and the MITRE Corporation.5.8 More recently, it is to be noted that: "Computer-made maps presenting physical, economic and social data in easily understood graphic form will be produced, along with similar diagrams, charts and graphs, in a new Laboratory for Computer Graphics at Harvard's Graduate School of Design." (Commun. ACM 9, 310 (1966).) Chasen and Seitz (1967) consider not only the early sketchpad developments, but also those Lockheed-Georgia, IBM, Control Data Corporation, and General Motors.5.8a

of

Other experimental on-line data analysis and display systems include APEX at M.I.T.'s Lincoln Laboratory (Forgie, 1965), and the similar capabilities of the SDC Variable Display system (Schwartz et al., 1965) and its Graphic Tablet Display.5.8b A West German example involves a Telefunken combination of a display console and its TR-4 computer for air traffic control applications. (Stevens, 1968, p. 9) In this system, a manually controlled moving ball permits the translation of displayed images and the multidirectional scanning of different areas shown on the display scope. At the National Bureau of Standards, developments in graphic input-output, such as ACCESS and MAGIC, are designed for differing types of requirements in U.S. Government organizations.5.9

Then there is the time-sharing facility for experiments in man-machine interaction at the Computer Center of the University of California, Berkeley, which is described by Lichtenberger and Pirtle (1965). In effect, the user has available to him a Scientific Data Systems (SDS) 930 computer with 16,000 words of fast memory, modified to exclude his direct access to input-output instructions (which, instead, are carried out for him by the system executive) and by additional software-interpreted instructions. Remote access facilities include teletypes, CRT display keyboards and a small DPP-5 processor equipped both with a CRT display unit. and a RAND Tablet.

In addition to the Digital Equipment Corporation's graphic input/output equipment used in many of the pioneering and continuing responsive-output applications,5.10 a relatively wide variety of graphic output and display devices is also offered by various other suppliers. Examples include, but are obviously

not limited to Bell Laboratories' "Glance":5.11 Benson-Lehner developments; 5.12 Bunker-Ramo Datatronics Engineers, Inc.; 5.14 equipment; 5.13 Gerber Scientific Instrument Co.; 5.15 Information Displays, Inc.; 5.16 Informatics' DOCUS (Display Oriented Compiler Usage System) for the Rome Air Development Center; 5.17 ITT Federal Laboratories: 5.18 Minnesota Mining and Manufacturing Co.; 5.19 Philco; 5.20 Sanders Associates; 5.21 Stromberg-Carlson (especially the SC-1100 Display Inquiry Station and the SC-1200 Digital to Video Display for computer applications involving multiple access to data at remote locations), and the Tasker Instruments Corp.5.22 In general, as of 1967, the majority of computer manufacturers also offer graphic input-output capabilities via peripheral equipment that can be connected on-line to the central processing facilities. Obvious examples include UNIVAC, IBM, Honeywell, Control Data, and others.5.23

Some of the new applications of user-controlled graphic input and output devices include the on-line alteration of a PERT (Program Evaluation and Review Technique) network display to determine the effects of setting different target dates; 5.24 Control Data Corporation's computer-directed construction drawing system involving long distance telephone lines for transfer of problem and solution data (Business Automation 12, No. 7, 55, July, 1965), and the Comex (Command Executor) system which allows "several simulated commanders to use a subset of English to control the flight of simulated objects on the system's display console. Heading, course, range, altitude, velocity, and destination can each be controlled by English statements typed into Comex on-line." (Schwartz et al., 1965, p. 29).

Walter (1967) claims that "immediate application of graphic data processing to solve a variety of problems is both physically possible and economically desirable. Graphic data processing can be effectively applied in mathematics, engineering, banking, education, communications, medicine, management information systems, programming, and many other fields." (Walter, 1967, p. 107).

Similarly, Prince reports that "a number of industrial, government, and university laboratories are exploring various applications of graphical computer-aided design. Applications receiving primary attention include the preparation of digital tapes for numerically controlled cutting tools, trajectory studies, structural analysis, aircraft and automotive shape design, shipbuilding, flight test data reduction, circuit analysis, and printed circuit board layout. Most of these must be considered as experimental programs; only a very few are in a production status." (Prince, 1966, p. 1701).

Yet, there are many areas of continuing R & D concern with respect to character sets and remote terminal design (to be considered in the next report in this series), hardware limitations, and human engineering considerations.5.24a For example, Walter of Honeywell, Inc. (1967) asks: "Can mathematical techniques be used to improve the accuracy of a

computer graphic output device despite its hardware limitations? This developmental research problem involves discriminatory analysis, differential equations, and simulation." (p. 107).

Then we note that: "Davis has stated that because of the universality of pictures the 'improvement of display techniques has gone hand-in-hand with man's progress in every field of human endeavor.' Unfortunately, the road to good displays has not been smooth. In her paper, which surveys the history of displays, she points out how disorganized, spasmodic and even serendipitous has been the improvement in displays." (Davis, 1966, p. 236).

Mills reports that "a very thorough treatment of computer-driven displays and their use in manmachine interaction is given by Van Dam. This paper-really a mongraph-includes a brief history, a thorough treatment of display technology at a very satisfying level of technical thoroughness and depth, and a survey of some of the applications of manmachine interactive systems involving display terminal devices. Sixty bibliographic citations are included. This is one of the few papers to recognize that imaging techniques other than those based on cathode ray tubes may have an application in graphic terminal devices; some half-dozen non-CRT, techniques are mentioned. The communication problems raised by moving the display terminal to a remote location are not considered. Sutherland treats the state of the art in computer graphics and, indicates some further requirements. The paper is, not really limited to the issues of computer graphics, but considers them within the context of manmachine coupling." (Mills, 1967, p. 231).

5.1.2. Machine-Aided Design

1

E

One of the most challenging areas of information processing system output is unquestionably that' of reactive display and response between man,' machine, and data banks in machine-aided design applications.5.25 Graphic input-output capabilities, together with man-machine-interactive on-line' modifications, are essential to the machine-aided'; design procedures. A pioneering type of application' to problems of hospital and architectural design,! discussed by Licklider and Clark (1962) several years before the General Motor's DAC (Design Augmented by Computers) system's claim for a' "first", (as of Fall, 1964, see Hargraves et al.) has had continuing development and expansion by Bolt,' Beranek and Newman personnel. (Fig. 6) The DAC system, on the other hand, includes rapid-response' microfilming and microfilm re-display, also evidenced in such developments as an IBM graphic1 data processing system 5.26 and the D-200 equipment offered by Strand Division of Datatronics Engineers, Inc.5.27

There are a number of isolated, but intriguing.' examples of machine-aided design applications" ranging from the design of ships 5.28 to the CADETS (Computer-Aided Design Experiment Translator)) system described by Lang et al. (1965), which has

been designed for such processing tasks as threedimensional shape description and the development of non-linear electronic circuit designs.5.29 Other areas of applicability include aerospace research,5.30 design of aircraft and automobiles,5.31 other architectural developments 5.31a and civil engineering and highway design, including the design of a super-highway interchange in a brief ten minutes by an engineer using a remote communication link to MIT'S MULTICS (Multiplexed Information and Computing Service) system.5.31b Machine-aided design has also been extended to the fields of the arts and crafts, notably in the area of textile design.5.32

In addition, at the University of California at Santa Barbara there is the Culler-Fried system for "computer displaying and of transforming mathematical functions in their pathologic intervals" (Cheydleur, 1965, p. 175) and Engelman states that "W. A. Martin. . . [is] working on input from scopes achieved by signifying with a light pen an interesting subexpression of a previously 'printed' expression, as well as anticipating using the scopes for handwritten input".5.32a Ruyle et al., provide a 1967 status report on systems developed for on-line mathematical problem-solving including AMTRAN,5.32b Culler-Fried,5.32c the Lincoln Reckoner, 5.32d MAP,5 5.32e and MATHLAB.5.32 The work at Hudson Laboratories, Columbia University, is also to be noted.5 5.328

A general survey of computer applications to design of computers is provided by Breuer (1966), who cites some 287 references. Also in the area of "shoes for the shoemaker's children", reactive display and input devices are already being applied to machine-aided programming, flowcharting and system design operations.5.33 It is noted also that Anderson has used an IBM 1620 together with a Calcomp digital incremental plotter to produce annotated flow charts on line 5.34 and commercial packages for this purpose, such as AUTOFLOW 5.35 or Autodigrammer 5.35a are now available. Then it is claimed that Bell Lab's FLOWTRACE can produce "flowcharts written in 'almost any' programming language." (Sherman, 1966, p. 845). Other examples include the IBM System/360 Flowchart,5.35b and the Usercode Documentor and Flowlister,5.35c while Abrams (1968) discusses a number of systems.5.35d

Developments have been reported in machineaided circuit analysis, schematic circuit design, and system block diagram construction.5.36 LeVier (1965) describes a program for the circuit design of a transistorized flip flop, Wall and Falk (1966) describe an IBM program for electronic circuit analysis, and Shalla (1966) reports a mapping from block diagrams of electronic digital circuits into list structures and the use of a list processing program for the Control Data 3600 to carry out circuit analysis. In particular, some examples of actual hardware for an Argonne National Laboratory computer were tested. Then there are the CIRCAL

(CIRCuit Analysis) programs available on the Project MAC facilities (Dertouzos, 1967). Ninke (1965, p. 846) describes an IBM 7094 application ir which "a system block diagram composed at the console can be used as input to a special compiler the compiled program can be run and the results viewed at the console." 5.37 However, "we are only, just beginning to explore systems where the com: puter asks questions of the programmer to resolve ambiguities in what it is told." (Sutherland, 1965 p. 11).

Hardware, software, and systems planning con siderations involving human engineering considera tions and human behavioral factors, come especially to the fore in experimental applications, such as the following:

(1) "A few users have experienced the delight of sketching and printing to the computer anc having it 'clean up the input' and then generate labeled graphs and drawings in response. (Licklider, 1965, p. 182).

(2) "In the study of computer-aided design. graphical communication should be a two-way process, for the designer wishes to enter drawings of objects into the computer for analysis and likewise he wishes to see com puter-generated or modified drawings.' (Lang et al., 1965, p. 1).

(3) "The engineer must do more than use the computer. He must actively participate in the computer solution. To do this he needs ‹ language to communicate with the computer physical accessibility to the computer, and a mechanism for obtaining engineering-orientec results from the computer." (Roos, 1965 p. 423).

(4) "We are therefore witnessing the first inroad of machine assistance into the human decision areas of printing design. It is ironical tha only a few years after printers and designer: threw up their hands in horror at the idea o machines robbing them of their traditiona role in printing practice, we should see them being actively taken back into a partnership · with the machine in which their specialised talents are being very much enhanced by machine assistance." (Duncan, 1967, pp x−xi).

(5) "An engineer may, for example, substitut different values of components in a circuit and observe the effect in the form of a picture o the output wave that would be produced b the circuit in question under any combination of conditions. He may also record the variou: changes on microfilm and observe them late via a projection device at the same console He may magnify sections of a diagram fo detailed study and manipulation, retaining the ability to return the original picture to the screen at any time." (Silveira, 1965, p. 37)

Next, therefore, we may ask how conveniently the otential system client may thus interact with the machine.

What processing operations are automatically vailable to the user as he watches and wishes to modify the data pertinent to his problem displayed or him? In systems such as Sketchpad he may apply caling and rotational transformations of the image isplayed,5.38 in CALCULAID5.38a or MAP* he may all directly for regression or Fourier transform perations,5 5.39 and, in general, he needs to be able o request as directly as possible the performance of specified operation, a display of the consequences of his proposed modifications, and the re-input of pecifications of the next processing operations to be performed which he has determined on the basis of the previous results.5.40

Finally we note that utilization of direct outputs, whether by man or machine or by both, involves Iso the possibility of reactive control and of inerruptibility and re-direction of source data utomation, intermediate processing, intrapolation nd extrapolation of results. Feedback of various ypes is an important concern.5.40a The problems of nan-machine interactive situations at the levels of erminal design, user-oriented languages, human ngineering, and behavioral or attitudinal factors vill thus require continuing R & D attention.5.40b

.1.3. Computer-Assisted Instruction and Problem

Solving

The problems of computer-aided instruction present, at least in the general literature surveyed, somewhat less in the way of continuing R & D mplications than is the case for machine-aided design or machine-aided inference. Nevertheless, he field of CAI represents an important area of development and application.5.41 In its computer directory issue of June 1968, Computers and Autonation provides a roster of 40 CAI centers and aboratories. Silvern and Silvern emphasize that: 'CAI, computer-assisted instruction, developed rom dissatisfaction with simple text formats and eaching machines which did little more than automatically turn pages for the student. Early nvestigators saw, in the digital computer's capability, the seeds of two-way, learner-machine lialogues which would lead a student along a path which more primitive methods failed to travel.” Silvern and Silvern, 1966, p. 57),5.41a

Examples of CAI systems include the PLATO Programmed Logic for Automatic Teaching Operations) system developed by the Coordinated Science Laboratory of the University of Illinois,5.42 he CLASS (Computer-Based Laboratory for Autonated School Systems) facilities developed at System Development Corporation,5.43 PLANIT Programming Language for Interaction and Teach

*See note 5.32a.

ing) 5.43a and the Socratic System as described by Muerzeig in 1965. IBM has developed the 1500 Instructional System 5.44 and Engvold and Hughes of IBM describe (1967) adaptation of the 7044 Graphic System to a teaching system which can be interleaved with programming problems to be handled in a variety of languages at the display console.

Still other examples include the simulation by students of experiments in chemical analysis 5.45 and a Bolt, Beranek and Newman development where the computer deliberately introduces ambiguities into an interactive system designed to teach medical diagnosis. (Science, Oct. 29, 1965, pp. 572-576). Byproduct advantages of CAI techniques include automatic performance measurements of student progress.5.46 Then it is pointed out that "remote terminals used for CAI can also be used to provide instructions for leading the student to appropriate supplementary material stored on microform." (Burchinal, 1967, p. 4).

Crowell and Traedge, 1967, provide a review and bibliography of developments in computer-assisted instruction. In addition, they suggest areas for continuing R & D efforts. They conclude, in particular, that "before assigning system functions to either man or computer. . . much research is needed in analyzing the decision-making process. When such information is added to our present store of knowledge concerning man-machine factors, greater efficiency in CAI systems will be realized." (p. 423).

Other commentators similarly suggest that a great deal of work remains to be done (Adams, 1966), that, "without discredit to the good work now under way,... the total level of the effort is unacceptably small" (Mills, 1967, p. 237), and that "although man-machine interaction may eventually revolutionize problem-solving, even the leaders in the field are just taking their first hesitant steps into the deep unknown of joint human-computer problemsolving techniques." (Davis, 1966, p. 243).

5.2. Multiple Output Modes

For information processing systems generally, output, like input, should be available in a variety of modalities-visual, as in printing and in graphic displays; voice and other audio signal outputs; control signals, mechanical displacements of pointers and plotting devices, and the like. Output results may be presented immediately to the operator or user, recorded photographically, or fed to various types of communication links, including telephone, coaxial cable, and facsimile transmission lines. Before discussing the more conventional forms of output, such as the use of printers and plotters, and hard copy requirements generally, let us consider in this section audio outputs; three-dimensional, color, and motion picture outputs, and some examples of systems using various combinations of output techniques.