197103

197103 197103

User Manual: 197103

Open the PDF directly: View PDF .
Page Count: 60

Download
Open PDF In Browser	View PDF

gJ!" 3630

SCIENCE & TECHNOLOG'G

•

7 2M a ..

. . ...~

...
•••

March, 1971

Vo l. 20, No . 3

~~I. ~,.mla~H~!!!:i~
.

. ..• .
•

Conversation with a Computer ---- Think Before You Speak!

~
PER I aD I l, -A,-L.-;--------,~-"
J..., ,''''_&-. _ _ _

0104
180 W SAN CARLOS ST
*D1271
SAN JOSE CA
95113

ADAPTATION AND LEARNING
IN AUTOMATIC SYSTEMS
A Volume in MATHEMATICS IN SCIENCE AND ENGINEERING
Series Editor: RICHARD BELLMAN
by VA. Z. TSVPKIN, The Institute of Automation and
Telemechanics, Moscow, USSR
Translated by Z. J. NIKOLIC
Presents a unified treatment of the theory of learning,
self-learning, and adaptation in automatic systems. Because of the complexity of many existing and developing
systems, traditional methods do not guarantee the optimal
operating conditions of such systems. The book presents
an adaptive approach to these problems using probabilistic iterative methods as a foundation. This approach
simplifies the solution of well-known problems in control
theory, reliability theory, operations research, game
theory, the theory of finite automata, etc., and permits the
solution of many new problems. A detailed survey of the
literature on adaptive and learning systems is given at the
end of the book. 1971, 312 pp., $16.50.

COMPUTATIONAL METHODS
IN OPTIMIZATION:
A Unified Approach
A Volume in MATHEMATICS IN SCIENCE AND ENGINEERING
Series Editor: RICHARD BELLMAN
by E. POLAK, Department of Electrical Engineering and
Computer Sciences, University of California, Berkeley,
California
Presents a large number of optimization, root finding, and
boundary value algorithms in the light of a unifying theory
that deals with convergence, synthesis, and computational
efficiency. Among the unconstrained optimization algorithms discussed are various gradient, quasi-Newton, conjugate gradient and variable metric methods, with proofs
of convergence and rate of convergence. Varrous constrained optimization algorithms discussed include penalty
function methods, metnods of center, first and second
order methods of feasible directions, and gradient projection methods. Proofs of convergence, efficient implementation and applications to optimal control are given. March
1971, about 320 pp., $17.50.

INTRODUCTION TO
PROBABILISTIC AUTOMATA
A Volume in COMPUTER SCIENCE AND APPLIED
MATHEMATICS
Series Editor: WERNER RHEINBOLDT
by AZARIA PAZ, Department of Computer Science and
Mathematics, Israel Institute of Technology, Haifa, Israel
This book-the first to be published in English on the subject-discusses both the practical and theoretical aspects
of probabilistic automata and sequential machines. The
first chapter, dealing with state theory from an engineering standpoint, covers the synthesis of stochastic machines, state minimization, equivalence, coverings and
input-output relations. Chapter two is devoted to nonhomogeneous Markov chains, which provide the mathematical model on which stochastic automata are based.
The final chapter presents a theory of formal stochastic
languages and events, including such topics as closure
properties, decision problems, and characterizations.
March 1971, 248 pp., $13.00.

INTRODUCTION TO THE
MATHEMATICAL THEORY
OF CONTROL PROCESSES
Volumes 1 and 2
Volumes in MATHEMATICS IN SCIENCE AND ENGINEERING
Series Editor: RICHARD BELLMAN
by RICHARD BELLMAN, Departments of Mathematics,
Electrical Engineering, and Medicine, University of Southern California, Los Angeles, California
Volume 1: LINEAR EQUATIONS AND QUADRATIC
CRITERIA
CONTENTS: Introduction. What is Control Theory? SecondOrder Linear Differential and Difference Equations. Stability
and Control. Continuous Variational Processes: Calculus of
Variations. Dynamic Programming. Review of Matrix Theory
and Linear Differential Equations. Multidimensional Contro:
Processes via the Calculus of Variations. Multidimensional
Control Processes via Dynamic Programming. Functional Analysis. Miscellaneous Exercises. Bibliographies and Comments.
1967, 245 pp., $11.50
Volume 2: NONLINEAR PROCESSES
CONTENTS: The Basic Concepts of Control Theory. Discrete
Control Processes and Dynamic Programming. Computational
Aspects of Dynamic Programming. Continuous Control Processes and the Calculus of Variations. Computational Aspects
of the Calculus of Variations. Continuous Control Processes and Dynamic Programming. Limiting Behavior of Discrete Processes. Asymptotic Control Theory.
Duality and
Upper and Lower Bounds. Abstract Control Processes and
Routing. Reduction of Dimensionality. Distributed Control
Processes and the Calculus of Variations. Distributed Control
Processes and Dynamic Programming. Some Directions of
Research.
1971,324 pp., $16.00

INTEGER PROGRAMMING
A Volume in MATHEMATICS IN SCIENCE AND ENGINEERING
Series Editor: RICHARD BELLMAN
by HAROLD GREENBERG, Department of Operations
Analysis, Naval Postgraduate School, Monterey, California
Presents all the material necessary for the understanding
and solution of integer programs. It includes examples of
the subject, the theory of integer programming, and
numerous methods to solve practical problems. Topics
covered include: applications of integer programming, a
review of linear programming, the all -integer and continuous solution methods presented in a parametric context from which the underlying principles, covergence
proofs, and algorithmic results are developed, and upper
bound variable problems. March 1971, 208 pp., $11.50.

NONLINEAR PROGRAMMING
edited by J. B. ROSEN, O. L. MANGASARIAN, and K.
RITTER, all at the University of Wisconsin, Madison,
Wisconsin
Concerned with those algorithms and related theory which
lead to efficient computational methods for solving nonlinear programming problems. One of its main purposes
is to strengthen the existing relationships between theory
and the computational aspects of this subject. Among the
more active areas of research covered are algorithms for
nonlinear constraint problems, investigation of convergence rates, and the use of nonlinear programming for
approximation. This book presents the proceedings of a
Symposium conducted by the Mathematics Research
Center, The University of Wisconsin, Madison, May 4-6,

1970.1970,502 pp., $10.50.

ACADEMIC PREss \(!J

m N E w YORK AND LONDON
111 FIFTH AVENUE, NEW YORK, N. Y. 10003

ti()s~it~llgr()~m~ti()il Srsterns~ r:.i~ m~~~et iil t~e 7()'s-

Wlt~ ~Ieilt>.:: ()r~()()rn
r()~ ileW r~ces.

Our new in-depth analysis and forecast can help you plan your moves.
From coast to coast, hospitals are rapidly discovering that automation is the
answer to keeping costs in check while, at the same time providing better care
for patients. This recognition is expected to spur hardware and software sales
to an annual growth rate of about 17% through the decade.
The 70's will see great changes in the market. Now, nearly 90% of the systems
installed are used only for "bookkeeping" functions. New functions like online complete pharmacy control, patient monitoring, medical history and laboratory reporting, diagnosis, menu control and meal distribution and teaching
are in the offing.
To help supplyir
profiles of poter
growth by numb

C(I n !:' U'1' !~ !"-1 SAN D A U'1' 0-"

provides
projects

Software and hl
ous projects an
amined for their

:lly. Vari; are ex-

Requirements fc
access and disp
1980. The relati'
progress is diSCI

t, remote
Hhrough
. R&D in

Price $295. OrdE
use this coupon.

3-1080 or

Please en!
_ _ Ch€

~y.

Please ser
reports chI

_ _ Has

-_Spe
_ _ Cor
_ _ Mi"

_ _ Dl
_ _ Pc
_ _ Pr
_ _ In
I

Name-l
Title_
Compan)
Address/
Telepho!

Vol. 20, NO.3
March,1971

computers
and automation

The magazine of the- design, applications, and implications
of information processing systems.
Editor

Edmund C. Berkeley

/l.rsistallt Editors

Linda Ladd Lovett
Neil D. Macdonald

Sojttl'are Editor

Stewart B. Nelson

/ldz Jertisillg
Director

Bernard Lane

Art Directol'S

Ray W. Hass

Computers and Education
8

USING COMPUTERS TO INDIVIDUALIZE INSTRUCTION:
[A]
ANOTHER APPROACH
by Dr. John A. Connolly
How computers are being used to adapt instruction to individual
differences in reading, mental ability, learning style, and cognitive
style, at the Conwell Middle Magnet School in Philadelphia.

[A]
LET US BUILD INTELLIGENT COMPUTER TUTORS
by Laurent Siklossy
How computers can be made to "know" many of the subject areas
that we wish to teach - with examples of the application of this
principle to elementary set theory and computer information
stru ctu res.

EXPERIENCE AND PLANS IN THE USE OF COMPUTERS
FOR UNDERGRADUATE INSTRUCTION
[G]
- CONFERENCE, JUNE 23-25, 1971
by Dr. Fred W. Weingarten, Director, Computer Services, Claremont
Colleges

Daniel T. Langdale

COlltriblltillg
Editors

John Bennett
Moses M. Berlin
Andrew D. Booth
John W. Carr III
Ned Chapin
Alston S. Householder
Leslie Mezei
Ted Schoeters
Richard E. Sprague

AdzJ;sory
Committee

Fulfillment
Manager

James J. Cryan
Alston S. Householder
Bernard Quint

William J. McMillan

Computer Programming and Software
[C/i/ol;aIOffices

Berkeley Enterprises, Inc.

[A]
PERSPECTIVE ON FLOWCHARTING PACKAGES
by Dr. Ned Chapin, Contributing Editor
Flowcharting packages are classified in eleven ways by the input
they accept, the kind of flowchart they produce, the pattern of
flow, etc., in order to provide an objective, quantitative basis for
comparing packages.

[G]
ANTI-ABM ESSAY CONTEST ANNOUNCED
by Daniel D. McCracken, Chairman, Computer Professionals Against
ABM
The insoluble problem of checking out a giant software system without any operating experience.

815 Washington St.,
Newtonville, Mass. 02160
617·332·5453

/ldz'el'tisillg
Contact

THE PUBLISHER
Berkeley Enterprises, Inc.

815 Washington St.,
Newtonville, Mass. 02160
617-332·5453

Computers and Privacy
Computers and Automation is published monthly
(except two issues in June) at 815 Washington
St" Newtonville, Mass. 02160, by Berkeley En·
terprises, Inc, Printed in U.S.A.
Subscription rates: United States, I I monthly
issues and two issues in June (one of which
is a directory issue) - $18.50 for 1 year, $36.00
for 2 years; 12 monthly issues (without directory
issue in June) - $9.50 for 1 yCilr; $ 18.00 for
2 years. Canada, add SO¢ iI year for postage;
foreign, add $3.50 iI year for postage. Address
all U.S. subscription milil to: Berkeley Enterprises,
Inc., 815 Washington St., Newtonville, Mass.
02160. Second Class Postilge paid at [loston, Mass.
Postmaster: Please scnd all forms 3579 to Berkeley
Enterprises, Inc., 815 Wilshington SI., Newtonville,
Mass. 02160. (i) Copyright 1971, by Berkeley En·
terprises, Inc.
Change of address:
If your ilddrcss changes,
please send us both your new address and your old
address (as it appears on the m~gilzine address
imprint and allow thrcc wecks for the change to
be made.

[G]
DATA BANKS - A POSITION PAPER
by Prof. Caxton C. Foster, University of Massachusetts
A penetrating analysis of future likely developments of data banks,
"when every interaction of an individual with society can be
collected, sifted, and analyzed at low cost" producing erosion of
constitutional rights.

NEW YORK STATE IDENTIFICATION
[G]
AND INTELLIGENCE SYSTEM
Burroughs Corp., Business Machines Group, Public Relations Office
The current operation of a computerized record-keeping agency
which holds more than 7 million fingerprints and associated records.
COMPUTERS and AUTOMATION for March, 1971

Computer Applications
20

COMPUTERS IN COMMUNITY SERVICE:
[A]
CAN THE CULTURAL GAP BE BRIDGED? Part Two
by James F. Muench, Computer Consultant
Why a "computer person" must understand the perceptions, language, non-verbal communication, humor, and other characteristics
of people of the community before he can effectively provide computer services to the community.

INFORMATION TECHNOLOGY
AND PUBLIC DECISION-MAKING
by Alim F. Westin

[F]

Front Cover Picture

Computers and Society
23

[A]
THE DEEPER UNREST
by Dr. Kingman Brewster, President, Yale University
How an unwillingness to admit the crises which we face with regard
to conservation, over-population, politics, economics, nuclear power, etc., has prevented positive progress in dealing with those crises.

SOCIAL DATA PROCESSING CENTERS,
IN THE FORM OF TELETERMINAL TIME
by Joseph Auciello, Director, Computer Job Bank

[G]

A young visitor from Curtis Jr.
High School, Sudbury, Mass., "has
a conversation with" a computer at
Digital Equipment Corp., Maynard,
Mass. It looks as if he is much
interested - and it seems safe to
predict that sooner or later he will
be "hooked" on computers.

Computers, Science, and Assassinations
35

"THE ASSASSINATION OF PRESIDENT KENNEDY:
The Application of Computers to the Photographic Evidence"
- COMMENT

Departmen ts

I. ANOTHER VIEW
by Benjamin L. Schwartz, Ph.D.
A polemical attack on "The Assassination of President Kennedy:
the Application of Computers to the Photographic Evidence" by
Richard E. Sprague published May 1970.

II. RESPONSE
by Edmund C. Berkeley, Editor, Computers and Automation

DISTRICT ATTORNEY JIM GARRISION ON
THE ASSASSINATION OF PRESIDENT KENNEDY:
A Review of Heritage of Stone
by Neil Macdonald, Assistant Editor, Computers and Automation

[A]

62
60
57
[A]
55
56
50

Compu ters and the Philosophy of Science
6

THE NUMBER OF ANSWERS TO A PROBLEM
by Edmund C. Berkeley, Editor, Computers and Automation

Across the Editor's Desk
Applications
Education News
Research Frontier
Miscellaneous
Advertising Index
Calendar of Coming
Events
Classified Advertisements
Corrections
Monthly Computer
Census
New Contracts
New Installations
New Products and
Services
Readers' Forum

47
47
48
49
49
62
61

[E]

Computers and Puzzles
63

NUMBLES
by Neil Macdonald

[C]

PROBLEM CORNER
by Walter Penney, COP

[C]

Key
[A] - Article
[C] - Monthly Column
[E] - Editorial
[F] - Readers' Forum

The Golden Trumpet
34

REDUCING DUES FOR UNEMPLOYED MEMBERS
by the Institute of Electrical and Electronic Engineers

COMPUTERS and AUTOMATION for March, 1971

[G] - The Golden Trumpet

[G]

[R] - Reference Information

EDITORIAL

The Number of Answers to a Problem

Rudyard Kipling, well-known English poet and author
(1865-1936, winner of the Nobel Prize for literature in
1907) once wrote in a poem (In the Neolithic Age):
There are nine and sixty ways of constructing tribal
lays
And -every -single-one-of-them-is-right.
I doubt that Kipling could have defended the number 69.
If he had been a scientist, he might have said:
"The number of ways of writing national epic poems
correctly, based on my extensive investigation of this
important subject, appears to be 70, plus or minus half
a dozen."
Pleasantries aside, one of the important ways of classifying problems of all kinds is according to the number of
answers. Problems may have many answers, just one answer, or no answer.
The run of problems in the ordinary undergraduate
textbook containing problems (in mathematics, computers,
physics, or almost any of the "hard" sciences) gives a
completely misleading impression of the real world, because
the characteristic problem in the ordinary textbook has a
single answer. This is nice for the student, because when he
finds the one answer, he can stop working, and nice for the
teacher, because when he examines the single answer, he
can judge rather well how the student has worked.
But in the real world problems with a single answer are
rare. The problems that come up in the real world, and for
which you and I and other people have to gather data
before we even start to solve them, usually have many
answers - or no answers. For example:
How shall I live on an income of $120 a week dividing
it among my expenses?
In the United States in 1971 this is a reasonably
satisfactory income for a single person, and there are
literally thousands of answers to the problem, depending on
the distribution of the income among food, shelter, clothing, and other expenses and payments.
For another example of a problem:
How will 90 million people be fully employed in 80
million jobs?
The answer is, "It's impossible." And retraining, or reeducation, or computer-assisted education, or computerassisted instruction, or the Office of Economic Opportunity, etc., and all the other bright-eyed and bushy-tailed
proposals from the world of training and education simply
6

alter the identity of the persons who are holding the jobs,
as in a game of musical chairs - although in many other
ways the education and training is of course useful.
In the field of the application of computers, the design
of computer programs, and the development of computerized systems, nobody really expects a single answer, even
though a single answer has to be chosen. People expect any
one of a large number of answers to be feasible, and they
seek to choose that one which has the largest number of
apparent advantages and the smallest number of apparent
disadvantages. But it is regularly impossible to prove conclusively that a chosen answer is the best answer - usually
it is only a relatively optimal answer.
This fact has many important consequences:
1. A practical answer is regularly an approximation
to the solution of a problem, and therefore does
not entirely solve it.
2. Better approximations are possible and should be
expected.
3. As time goes on, as conditions change, as knowledge increases, as requirements alter, improved
answers should be determined.
4. Any answer that is adopted should be modifiable
and improvable without too much sand in the
gears. In· other words, it should have built-in
facilities for change.
5. To obtain better and better approximate answers
and to install them at appropriate times is always a
requirement of good judgment and a responsibility
of good management.
For many problems in the real world, not only are there
many answers, but static answers may change from good to
poor, as time goes on and as the real world with its
kaleidoscopic multiplicity of factors continues to operate.
Which is just another way of course, of expressing
Murphy's Second Law:
If left to themselves, things always go from bad to
worse.

Edmund C. Berkeley
Editor
COMPUTERS and AUTOMATION for March, 1971

READERS' FORUM

INFORMATION TECHNOLOGY AND
PUBLIC DECISION·MAKING
Alan F. Westin
From the "Sixth Annual Report", 1969-70
Harvard University Program on Technology and Society
Cambridge, Mass. 02138

This project is investigating the use of computerized
information systems as decision-making aids by public
agencies; it seeks to identify the changes that introduction
of these systems leads to in the organization and policies of
those agencies, and to explore the social implications of the
changes. The study has been under way during a time in
which much negative sentiment about technology arose
from the belief that governmental decisions that affect
people would no longer be made by traditional political
processes in which the people have a voice, but by a few
experts with computers, in ways that people would not
understand and that would threaten their privacy and their
freedom.
What has thus been seen as a threat by some, has
appeared to others as a promise of increased competence
and effectiveness in policy-making. Through the immense
data-handling capacities of electronic computers and the
speed and flexibility of telecommunication systems, it was
held, man could examine his social system in all of its
patterns and tremors, and could create monitoring and
feedback systems to achieve cybernetic levels of competence. As computer usage transformed mathematical
calculations, opened new horizons in scientific inquiry,
supplied automated systems for many basic production
processes in industry, and fostered vast space and missile
programs, many influential voices declared that this technology could be used to improve our domestic governmental process as well. The same tools that had put men
into space, it was argued, could be applied also to the
painfully intractable domestic problems confronting all
levels of American government during the Kennedy and
Johnson years.
These proposals raised a host of questions that needed to
be explored. Could information technology really deliver
more timely and useful data for program evaluation and
policy formulation? How would introduction of information technology into government agencies affect their internal structures, role relationships, and decision-making
processes? How would such a development affect legislative
and judicial oversight of executive-agency policy-making?
What would be the psychological reactions of those who
were the subjects, clients, and customers of government
programs? Would there be changes in the relationships of
COMPUTERS and AUTOMATION for March, 1971

government agencies to the intermediate political process
(interest groups, the press, the party system) and to popular
participation through the electoral system? And how would
a trend toward such systems affect basic distributions of
power in society between haves and have-nots?
It was to explore these questions, and to provide
detailed factual knowledge about exactly what was happening in this area, that the present study was initiated in
1967.
Some of the larger issues of information technology and
democratic government that are emerging from our research
can be indicated here.
First, the anti-technology literature of the 1960's expressed grave fears that computers and telecommunication
systems would lead to the capture of decision-making by
"the machines," or at least by those technicians and their
superiors who controlled the information systems. Whatever the facts in the military and intelligence areas may be,
there has been no such take-over yet in the civilian agencies
of American local and national government. The distant
future may tell a different story, but in 1970 it requires a
powerful flight of ideological or philosophical imagination
to go from the current pedestrian uses being made of
computers to move paper and perform basic transactions to
anything resembling sophisticated, data-rich decisionmaking; nor is there the slightest sign of a displacement of
the traditional leadership elites of top and middle management in government by the information specialists.
Second, it is possible to conclude that computers have
been a factor in consolidating rather than in redistributing
governmental power; computers and their associated equipment are fearfully expensive, and the poor, the black, the
students, and the anti-war movements cannot harness computers to their causes. Also, to the extent that organs of
government might adopt restrictive policies toward the civil
liberties of dissenters or the socio-economic claims of the
poor and the black, there is little doubt that information
technology can make the execution of such policies more
efficient. While there have been thoughtful suggestions by
writers such as Robert Fano and Donald Michael that
computer systems should be designed for power-sharing
uses by the public and by the organs of criticism, no sign of
such sharing has yet appeared in the development of
computerized data banks in government.
Third, one is led to ask whether the contrast between
the shimmering visions painted by computer manufacturers,
software firms, and allied consultants in the early 1960's,
and the bleak record of accomplishment as of 1970
suggests a case of civic fraud. Enormous sums of money
from public treasuries were spent to buy computers and
(Please turn to page 60)
7

USING COMPUTERS TO INDIVIDUALIZE INSTRUCTION:
ANOTHE'R APPROACH

"When a perfect match between a student's abilities and a learning packet is
found, the student is told, 'This packet is just your style. ' When only a very
poor match can be found, the student receives the instruction, 'This packet
may be hard for you. Ask for help zf you have trouble. ' "

Dr. John A. Connolly
American Institutes for Research
135 N. Bellefield Ave.
Pittsburgh, Pa. 15213

John Lee came up with the winning name in the
name-the-computer contest - Cornelius Computer. Cornelius works with John and 250 other eighth grade students
in a school located in a poverty area in Philadelphia. A

typical exchange between the computer and a student like
John is quite simple:
CORNELIUS:

JOHN:
CORNELIUS:

P LEASE TYPE YOUR FIRST
NAME, SPACE, AND LAST
NAME
JOHN LEE
LEE JOHN
ID #0892
JOHN, PLEASE ENTER THE
FULL NUMBER OF THE PACKET YOU HAVE JUST COMPLETED
1100050005
HAS THIS PACKET BEEN SUCC E SSFULL Y COMPLETED?
TYPE IN YES OR NO
YES
YOUR RECORD NOW SHOWS
THIS PACKET WAS COMPLETED . PACKET 1100060007
ASSIGNED, WHY STUDY?
(T APE) LEVEL 1 MATCH THIS SHOULD BE EASY TO DO.

Matching Individual Aptitudes

Dr. John A. Connolly is a Senior Research Scientist
at the American Institutes for Research with previous
experience with Educational Testing Service. He received a Ph.D. from Columbia University. His previous publications are in the fields of psychological
measurement and evaluative research.
8

The final statement in this dialogue is not just a word of
encouragement. It represents three years of effort to build a
prototype system of individualized instruction. The computer's assignment of a particular learning packet results
from matching John's measured learning characteristics to a
catalog of many curricular options. The matching process is
a central feature of a computer-based system for managing
instruction currently being developed under a Title III grant
at Conwell Middle Magnet School.
COMPUTERS and AUTOMATION for March, 1971

Computers have frequently been used in programs of
individualized instruction to track students through an
instructioHal sequence on an individual basis, and to branch
the learning sequence to adjust to individual variations in
performance. This article describes a computer system
which, in addition, adapts instruction to individual differences in aptitudes. The article concludes with a review of
the tracking, branching, and adapting functions of the
system with reference to the future of programs of individualized instruction.
The Basic System Structure

In simple terms, the system under development is designed to find out what children need, and to supply the
most appropriate learning experiences to meet their individual needs. The prototype consists of three basic components: (1) a set of instruments and techniques for
assessing student needs and characteristics; (2) a bank of
learning packets which are cataloged in terms of those
needs and characteristics; and (3) a computer program for
matching the student's assessments and packet history to
the curriculum options at each step of the learning sequence.
The evaluative and curricular aspects of the model are
described first in order to set the background for a more
detailed discussion of the matching program.
1. Student Evaluation System

An individualized program begins with an evaluation of
the strengths and weaknesses ·of each student. These are
determined by means of four different kinds of assessments. Terminal tests are used to evaluate mastery or
deficiency with reference to 67 basic school objectives.
Diagnostic tests indicate which of the available learning
materials the student should be assigned. Progress tests are
designed to measure student proficiency on the materials
presented in each learning unit. Aptitude measures, the
focus of the present article, attempt to determine the most
effective way to teach each child.
Chart #1 shows the four aptitude variables which are
measured. These aptitudes were selected for experimental
use on the basis of research evidence suggesting that they
may be important dimensions of learning ability.
a. Reading. Learning may be enhanced by adjusting the
reading level of materials to the .student's capabilities.
Reading test scores are divided into three relatively
gross categories and stored in the student data bank.
b. Mental abilities. Educators tend to feel that different

MEASURES OF LEARNING APTITUDES

RELEVANT

GENERAL
DESCRIPTION

TEST

Reading Achievement

Iowa Test of Basic Skills

Mental Ability

RESEARCH

0-2nd grade
1 • 3-Sth grade
2 '" 6 th and above
3 - No information

Lorge-Thorndike Intelligence Tests
Academic Promise Tests

Keislar & Stern

Visual Learning Test
Auditory Learning Test

Ingersoll

Learning Style

DATA BANK INPUT

Kinesthetic Learning Test

O· Below average
1 .. Average or above
2 '" No information

0 .. Visual
1 • Audi tory
2 • Kinesthetic
3 .. Mixed

4 • No information
Cognitive Style

Raven Progressive Matrices
Memory for Numbers Tes t

o - Concrete
1 .. Abstract
2
No difference
3 '"' No information
lit

COMPUTERS and AUTOMATION for March, 1971

teaching methods are reqUired for teaching students
with above average and below average mental ,abilities. Although there is some doubt at the present
time that such generalized abilities will prove effective in the present model, a very gross index of
mental ability is entered in the student file.
c. Learning style. Three distinct sensory pathways - the
visual, auditory, and kinesthetic - convey most of
the information from which the student learns. Some
students appear to have strengths or weaknesses in
learning by means of one or another of these pathways. Three measures were. constructed and administered to all students in order to assess their relative
abilities in learning comparable tasks presented
through each modality. A specific learning style is
entered in the evaluation file only if a student shows
marked superiority in one of these tests. Otherwise
the "mixed" code is used.
d. Cognitive style. A student's mental processes may be
especially suited to learning either abstract or concrete relationships. For example, one student might
find that the learning of number systems is facilitated
by a considerable amount of practice with numbers,
while another student might need to be taught only
the principles involved. Again, a student is given a
particular deSignation only if he shows large score
differences on the measures of cognitive style.
These scores and all other evaluative information on each
child are maintained in a randomly-accessed disk file. The
computer system will facilitate a study of the effectiveness
of these variables as data accumulates on student progress.
Some other kinds of learning aptitude variables (e.g.,
interests) are being considered for insertion in the system
on an experimental basis.
2. Curriculum Cataloging System

Curriculum materials were prepared in order to provide
effective learning treatments for students with specified
learning aptitudes. All curriculum materials are presented in
the form of learning packets, which are brief curriculum
units involving independent or semi-independent learning
activities. Some packets were developed by local school
teachers or project staff members and others were adapted
from published curriculum materials.
Each packet attempts to teach one segment of a particular learning objective. Mastery of most of the basic
objectives requires the student to perform or comprehend a
number of component operations or concepts, each representing one segment of the total objective. For example,
the concept of place values in numbers is one component of
the objective, "Arithmetic Skills."
The structure for coding the treatment characteristics of
packets for entry into the curriculum data bank is shown in
Chart #2. Cataloging of packets relies heavily on teacher
judgment. The teachers are provided with a set of instructions for coding the treatment characteristics of curriculum
packets in a way which corresponds to the measured
learning aptitudes of children.
The structuring of learning experiences in this manner
results in 72 possible combinations which are hypothetically different ways of teaching any single learning unit.
Learning style might be varied, for example, by asking the
student to read written materials (visuaI), listen to a taped
version of the same material (auditory), or manipulate some
9

CURRICULU01 CODING STRUCTURE

COllEll CllARACnRI ST I C

CODING CATAGORIES

DATA [lANK INPUT

0-2nd grade
3-Sth grade
6 th and above

Reading Level

Difficulty Level

Simple
Average or above

r---------------+----------------~---------------

Teaching

Visual
Auditorv

~!ode

Kincsth~tic

~Iixed

Cogni tivc

~Icthod

Abstract
Concrete
~lixed

tangible object such as an abacus (kinesthetic). The other
treatment characteristics can be varied in similar ways.
Thousands of possible packet variations can be conceived
for teaching a complex subject involving many discrete
learning steps (e.g., reading). The actual preparation of so
many variations is not practical. Nor would it be desirable
at this stage in the development of an untested model.
Enough variations have been written for many topics,
however. to serve as a basis for testing the usefulness of
these variables in structuring the learning experiences of
children. The computer system permits detailed analyses of
curriculum availability and effectiveness.
3. Computer Matching System

The computer performs a variety of functions in the
overall operation of the instructional management system.
The following discussion is limited to the computer's role in
prescribing a learning packet for the student corresponding
to his individual needs by matching the student's assessment file to the curriculum catalog.
a. System configuration. A remote computer terminal
(IBM 2740-2) is located in the Center for Individually
Prescribed Learning Activities at Conwell School. Telephone lines link the terminal to a 2701 Transmission
Control unit at the Philadelphia Board of Education Building. The control unit is attached to an IBM System 360,
Model 30, central processing unit with core storage partitioned for multi-programming. The foreground partition,
with 14K core storage locations, is dedicated to the present
system. Data files are stored on a disk pack mounted on an
IBM 2311 disk drive. The operating programs reside on a
second, permanently mounted, systems disk pack.
The system operates mainly in the on-line processing
mode, with provision for remote batch opera1.ion as well.
The major on-line request is for new packet assignments.
Students request their own packet assignments by interacting with the computer through the terminal and receive
virtually immediate decisions. The remote batch mode is
used when the full computer system is needed. This occurs
when a new student is added or a new packet is entered
into the curriculum file, and in other situations. These
requests are transmitted during the school day, queued on
the disk pack, and processed during the night.
10

h. Matching algorithm. The rapid selection of the
appropriate packet for a particular student at a given point
in time is one major reason for automating the system to
utilize the speed of a computer. The packet choice is based
on a matching algorithm which follows the logic shown in
Chart #3. The algorithm consists of six procedural steps, as
follows:
(I) Search for a perfect (Level 0) match between the
student's measured learning characteristics and
the coded packet characteristics for an instructional unit. The computer first relates reading
level, aptitude level, learning style, and cognitive
style to the coded dimensions of the available
packet variations in an effort to find an exact
match. If a packet matching all of the student's
measured characteristics is available, the instruction to take that packet is transmitted to the
student. If not, the computer moves to the next
step in the process.
(2) Search for an imperfect (Level 1) match. There is
an implicit hierarchy in each of the four dimensions used for coding packet characteristics.
Packets involving little or no reading, or reading
at the 3rd to 5th grade level, for instance, are
well within the capabilities of a student who
reads at the 6th grade level or above. Similarly,
the student with average aptitude can handle
materials at a simple level; students with a
strength in any of the three learning modes
should prove capable of working with mixed
learning modes, and the abstract learner can be
taught using either a concrete or a mixed presentation. (The formulas referred to in the chart
involve a detailed development of these general
principles.) If a packet is available that is within
~
COMPUTER HATCIIING ALGORITHM'
stART HERE LOOKING AT FIRST PACKET IN SEGMENT.

(MATCH LEVEL ¢ IS ON.)

COMPUTERS and AUTOMATION for March, 1971

the student's capabilities on all four variables,
that packet assignment is given to the student. If
not, the computer search continues.
(3) Search for an imperfect (Level 2) match disregarding the measured cognitive style of the student. At this point the computer is searching for
a packet which is within the student's reach using
three variables only - reading, aptitude, and
learning style. Cognitive style has been eliminated
first since its validity seems least firmly established. Again the student gets an assignment if
one can be found.
(4) Search for an imperfect (Level 3) match disregarding both cognitive style and learning style.
The search is continued with only two variables
brought into play.
(5) Search for an imperfect (Level 4) match disregarding cognitive style, learning style, and aptitude level.
(6) Search for an imperfect' (Level 5) match disregarding all four variables. In effect, assign anything that is available on the requested topic.
The result of this process is an instruction to a student
to take a particular learning packet. The level of the match
which was obtained is shown along with a statement to the
student about the packet he was assigned. For example,
when a perfect match is found, the student is told, "This
packet is just your style." When only a very poor match can
be found, the student receives the instruction, "This packet
may be hard for you. Ask for help if you have trouble."
No student has had serious difficulty learning the operating procedures required to "talk" with the computer.
Their apparent enjoyment of this process is believed to be
an important motivational factor. The average student
completes the entire exchange in about two minutes. The
terminal's response time is about two seconds for each
transmission. The interactive language and the equipment
configuration are currently under study in an effort to
make the exchange more effective and more efficient.

A New Role for the Computer?

In what sense does this approach offer a new role for the
computer? What are the current functions of the computer
in the individualization of instruction? What changes may
occur in the future?
All forms of individualized instruction, no matter how
diverse in other ways, involve individual as opposed to
group pacing of students. Computers are sometimes used to
keep track of large numbers of students who are moving at
different paces through one or more learning sequences.
The collection, storage, and retrieval of student progress
data is one basic approach of computer systems designed to
individualize instruction.

vals, and prescribing different learning assignments depending on where a particular score falls. Computer branching of
a learning sequence to adjust to variations in individual
performance is another approach to individualized instruction.
Adapting Techniques of Instruction

The approach described in this article is an effort to
adapt instructional techniques to individual differences in
children. It assumes that students will learn best when
taught in a way which conforms to their particular learning
abilities. A computerized matching algorithm is used to
prescribe a learning assignment which represents the best fit
between the student's learning abilities on the one hand and
the available variations in ways of learning on the other.
Computer adapting of ways of learning to individual abilities may be a new direction for future programs of
individualized instruction.
Computers can be used to track, branch, and adapt
instruction within a single system. These are complementary functions and all seem to hold promise for delivering
improved instructional techniques to students. It is virtually
certain that computers will also play an important role in
the collection and analysis of research evidence on the
effectiveness of these various techniques for the individualization of instruction.
Efficient computer programs do not by themselves
produce good programs of individualized instruction. Effective tracking, branching, or adapting of instruction can be
done with or without computers. The real justification for
the use of computers in education lies in their ability to
help educators in their efforts to solve the complex and
urgent problems which they face.
0

References

Baker, Frank B. Computer Based Instructional Management
Systems: A First Look, Review of Educational Research,
In Press.
Connolly, John A. A Computer-Based Instructional Management System: The Conwell Approach, Interim Report. Silver Spring, Maryland: American Institutes for
Research, 1970.
Cronbach, Lee J. How Can Instruction be Adapted to
Individual Differences? in Robert M. Gagne (Ed.), Learning and Individual Differences. Columbus, Ohio: Charles
E. Merrill Books, Inc., 1967, pp. 23-29.
Edling, Jack V. Individualized Instruction: A Manual for
Administrators. Corvallis, Oregon: Oregon State University, 1970.
Ingersoll, Gary M. The Effects of Presentation of Modalities
and Modality Preferences on Learning and Recall. Pennsylvania State University: Doctoral Dissertation, 1970.

Prescribing Alternate Learning Paths

Computers are also used to prescribe alternate learning
paths for students. The system typically reviews
performance data on tests administered at periodic intervals
through the learning sequence in order to reach a decision
about the next learning assignment. The prescription is
often based on table look-up procedures which involve
dividing the test score distribution into several score interCOMPUTERS and AUTOMATION for March, 1971

Jensen, Arthur R. How Much Can We Boost IQ and
Scholastic Achievement? Harvard Educational Review,
Vol. 39, No.1, Winter 1969.
Keislar, Evan R. and Stern, Carolyn. Differentiated Instruction in Problem Solving for Children of Different Mental
Ability Levels, Journal of Educational Psychology, Vol.
61, No.6, Part 1, December 1970, pp. 445-450.
13

LET US BUILD INTELLIGENT C OMPUTEIR TUTORS

"The advances in artlficial intelligence indicate that computers can perform
activities at a level of sophistication that we would be quite happy to see
many students (and even teachers) attain. "

Laurent Siklbssy
The University of Texas
Department of Computer Sciences
Austin, Texas 78712

The numeric capabilities of computers can render many
services in the educational system. Besides services such as
record keeping planning that go with any sizable operation,
computers can schedule classes and help instructors with the
tasks of grading and evaluation. We shall be most interested
here in the tutorial capabilities of computers. CAl (Computer-Assisted Instruction) has seen the development of
drill-and-practice sessions handled entirely by the computer. But many workers in the field are quick to point out
that computers are only a help to the teacher and should in
no way be considered a threat to his position. On the
contrary, their function is to relieve him of unpleasant
chores.
A "Good" Teacher

Why has the computer been condemned so soon to not

becoming a full-fledged tutor? The advances in artificial
intelligence indicate that computers can perform activities
at a level of sophistication that we would be quite happy to
see many students (and even teachers) attain. Perhaps it is
felt that the computer can never become a "good" teacher,
because a teacher understands what he teaches, and the
computer does not understand anything, but only does
what it has been told.
What is understanding? We shall not answer the question,
but instead shall quote a paradox, attributed to the English
physicist J. J. Thomson: "We never understand anything;
we only get used to it!" We have felt that it was more
productive to try to design seemingly intelligent computer
tutors than to argue about the possibilities of such a design.
The extent to which our early results indicate hope of
future progress is for the reader to judge.
General Schema for a Computer Tutor

There are many tasks that the computer can become
After an early interest in Mathematics and physics
(B.A., Yale Univ.; M.A., Harvard Univ.), Laurent
Sikl6ssy received his doctorate in Computer Science
at Carnegie-Mellon University. He has been on the
faculty of the University of California at Irvine, and is
presently in the Computer Sciences Dept. at the
University of Texas at Austin.
Mr. Sikl6ssy's principal research interest is artificial intelligence, particularly in the areas of learning,
problem solving, speech processing, and the development of intelligent computer tutors.
Later this year Prentice-Hall will publish Representation and Meaning, which Siklossy coedited with
H. A. Simon.
14

used to performing. From among such tasks we have
selected two different areas: the manipulative capabilities
necessary for solving problems in set theory, and the
capabilities that go with understanding and getting used to
information structures used in computer science (such as
stacks, trees, queues, list structures, etc.). Although we shall
give examples from the two systems that we have programmed, that teach elementary set theory and computer
information structures, we shall try to stress the generality
of our approach.
Programs That Perform Tasks

At the heart of our teaching systems are programs that
can perform the tasks that we wish our students to learn to
perform too. When we teach set theory, we wish the
COMPUTERS and AUTOMATION for March, 1971

student to learn about the union of two sets. Thus, our
teaching system includes programs that can calculate the
union of two sets. This, of course, assumes programs that
can recognize when something, whether produced by some
program or by the student, is in,fact a set. When we wish to
teach about binary trees, we first need programs that act as
if they understood binary trees. Such programs should be
able to answer questions about particular trees, such, for
example, as: Which node is the father of node ABC? How
many sons does node ABC have? What is the length of the
path from node ABC to the root of the tree?
In the areas that we have considered, elementary set
theory and information structures, the type of performance
programs that we have mentioned are easy to write. To the
core of our tutor, we must add input/output routines that
permit communication between tutor and student. Input
routines preprocess the student's message, and protect the
heart of the tutor. Our routine to calculate the union of
two sets would fall sick if it were not given two sets as
input. The output routines transmit answers, comments and
suggestions to the student.

inputs from the student, or they can be generated by an
internal program, a set generator that can produce an
extremely large number of different sets. Programs can
calculate the union but, more importantly, they can compare "intelligently" the student's answer with the calculated union. The tutor will verify that (V T A S) is an
acceptable answer, while (B S T) is not. Moreover, it will
determine that (B S T) is not acceptable because part of set
(A S) is missing in the union, as well as part of set (V T).
(In each case, the tutor also identifies the particular missing
parts.) Furthermore, the element B in the student's answer
should not be there, since it belongs to neither of the two
given sets. If necessary, the tutor can take these two sets
and go step by step through the method it uses to calculate
their union. The student is then presented with a working
model that works in front of his eyes, and which he can
imitate.

Learning by Practice and Imitation

The goal of the tutor is to transmit the knowledge,
stored in his "heart", to the student. We try to achieve this
by getting the student used to the subject matter. The
student solves problems, that he himself has dreamed up, or
that have been suggested by the tutor. The tutor will
carefully compare the student's answer with an answer that
it has calculated and base the future course of the tutorial
dialogue on its diagnosis. The tutor can also show explicitly, by running through a problem, the methods that it
uses to. obtain solutions. The student therefore learns by
practice and imitation. We shall give examples of diagnostic
capabilities in the next section.
IF YOV WISt< TO STOP ,

TYPE YES

Figure 2. The tree before and after the addition of node
FF.
Dynamic Capabilities

WHICH tlODE DO YOV WISH TO

~DD

Figure 1. At the student's request, the tutor draws a tree.
Diagnostic Capabilities

A simple example demonstrates the diagnostic capabilities of our computer tutor. Suppose that we ask the
student to calculate the union of the sets (A S) and (V T).
An answer is the set (A S V T), but since the order of
elements in a set is immaterial, so is (S T V A). For that
matter, there are 4! = 24 correct forms of the answer. It is
not appealing to store all these 24 answers! So suppose that
we follow the multiple-choice road:
Which is the union of (A S) and (V T):
(a) (A V), (b) (A S V T), (c) (), (d) (A S T) ?
Now suppose that our student somehow calculates his
answer. What happens if he finds (V T A S), or if he finds
(B S T) ? The multiple-choice road does not help at all.
In our system/ the sets are not even stored. They can be
COMPUTERS and AUTOMATION for March, 1971

Unlike the human instructor, the computer is not limited
to the use of a blackboard. In teaching information structures, it is essential to make the student familiar with the
transformation in time of these structures. The blackboard
does not lend itself to dynamic manipulations. On the other
hand, the graphic display or CRT of a computer provides a
superior medium for showing the changes in structures.
Figure 1 shows the CRT during a lesson on binary trees. 2
The student has already built a tree with nodes labelled
AAA, ABC and ART. The nodes have (optional) contents
and left and right links. The program is building the tree at
the student's request. Figure 2 shows the state of affairs
somewhat later. The tree has grown further and the student
has just requested that node FF, with contents C, be added
as a right node to node UU2. The tutor shows how the
OLD tree looked, and how the NEW tree appears. (Since
the new tree is bigger, smaller boxes were used in drawing it
by the general tree drawing routine!)
It is hard to deny that our tutor knows something about
1

The set theory teacher is programmed in LISP.
15

PERSP,ECTIVE ON FLOWCHARTING PACKAGES
"The SFL (Symbolic Flowchart Language) family is the only one that currently offers to the user any really effective degree of control over the level
of detailin the flow diagram. "
Dr. Ned Chapin
1190 Bellair Way
Menlo Park, Calif 94025
The computer user today has a wide choice of packaged
programs available for doing flowcharting. To see one's way
through the welter, it helps to classify the offerings. This
article attempts to make useful classifications, based upon a
historical survey and an intensive study of a sample of
currently available packages.

languages most commonly supported have been COBOL,
FORTRAN, and symbolic languages (such as IBM-360
Assembly Language). For some languages, such as
SNOBOL, LISP, and APT, the author has been unable to
find any flowcharting packages at all. Some languages are
supported by only a limited number of packages, such as
PL/l and RPG.

Families of Flowcharting Packages

The flowcharting packages available can be divided into
three convenient categories based upon the input they
accept. Historically the oldest and currently the most
popular is the family that accepts the programer~written
source code. That is, it accepts as input an ordinary source
program written in a programing language. In practice the

SFL Input

A second family of flowcharting packages accepts SFL
input. SFL (Symbolic Flowchart Language) was devised by
F. David Lewis and his colleagues at IBM. This family has
enjoyed considerable activity over many years. The input
language is convenient and easy to use, but does require
some way of converting the source language in the program
to input language.
The two alternatives commonly available for this are
manual translation and machine translation. The manual
translation permits introducing various degrees of summarization and provides the high degree of control over the
format of the resulting flowchart. Machine translation
typically proceeds by a process over which the user has
little control, and hence produces a relatively standardized
translation and flowchart.
Other Inputs

Dr. Ned Chapin has been a data processing consultant with
InfoSci, Inc. in Menlo Park, Calif. since 1965. He has been in
the consulting field since 1954, and his broad range of
experience includes: EDP systems analysis and design; file
and data base design and implementation; programming of
computer applications; software design, implementation, and
evaluation; and evaluation of computer manufacturer bids.
Dr. Chapin earned his Ph.D. in Business and Economics at
the Illinois Institute of Technology. He has taught many
college courses, and has published several books and more
than 100 articles, papers, reviews, etc. His professional
activities include: ACM National Lecturer; Member, DPMA
Curriculum Committee; Accreditation team member for the
National Association of Trade and Technical Schools; and
Program Chairman for COMPSO. He is a contributing editor
for Computers and Automation.

The third family of flowcharting packages accepts other
inputs. Typically these inputs are specialized languages
designed for particular user needs and used only in one
package, or used only in sharply modified form with other
packages. Historically in terms of the volume of flowcharts
produced, this has been a very important family, since it
includes the AUTOCHART series of programs widely used
within IBM for software documentation. The family, however, has little acceptance outside of its regular users
because of the difficulty of learning the specialized languages which have no wide adaptability.
Because this subdivision into families based on input
proVides a comprehensive and operationally practical classification basis from the users' point of view, it is the one
featured in the chart shown in Figure 1. A full historical
review can be found elsewhere. 1 Detail on a selection of the
most common packages available currently is given in
Figure 2.
1 Ned Chapin, Flowcharts (Princeton, N.J.: Brandon/Systems Pr~ss,
Inc., 1970), Chapter 6.

COMPUTERS and AUTOMATION for March, 1971

Among the packages which produce flow diagrams, a
classification can be made in terms of the level of detail
shown in the flow diagram. Most of the packages create a
flow diagram at the same level of detail as the level of detail
in the source program. That is, the level of detail obtained
by reading the source language is the same as the level of
detail obtained by reading the flow diagram produced by
the package.
Two exceptions can be observed. One is in the SFL
family, since SFL permits the user to control the level of
detail. The user with SFL can achieve any level of summari-

Flowcharts Produced

Another basis of classification is in terms of the kind of
flowchart the packages produce. The two main varieties are
flow diagrams for the algorithm used within a program, and
system charts showing the interrelationship by input and
output of a series or sequence of programs. In practice the
latter are all members of the SFL family and are represented by only one or two packages. As a de facto matter
therefore, the primary emphasis is upon flow diagrams in
the available packages.

Figure 1. Classification of Flowcharting Packages
YEAR

1957

1959

1958

1962

1961

1960

1963
7070
AUTOCHART

::J

1964

1965
1966
1967
1968
1969
1401
360
AUTOcHART·AUTOcHART-:o:-------------

r----------------,-----FLUSH--

FLOWRITE----

FLOGEN
PRECHART

t-----------------SBDS-----~I-·R.oberts

L- Hain

r------~~~CRKT-COMCHART---
\ - - - - - - - - - F L O B OL-N~V:~t-

~~~!------

IDYNACHART----

'"'"::J
c
'"

~~:~~~~,-----------~~;~~~~7-

a:
w

. If not delighted,
I ~e~;~\;(l\~~ir ~~e~n~~W~~. 11sd~y~r~d:e~~~~i
ace\

~~~~ tiC~~nfg;: Jl ~~r~e~~~~Zr~~~~:~~ri~~;

Address ..............................••.......

I
I
I

City .•.................. State •...... Zip ....... .
nooks bought for professional purposes may be a
tax-deductible expenS(l. (Olfer good in Continental
U.S. :md Canada only.)
7-353

II Each
~i~in~~if~~~~ ;g~o ~~~~in!d ~c~~~~~t~dm~~~
month, I will receive advance reviews deI ~~~~~~fenro~~~~mf~rg r~~~~~lf~~' aNeO~~teWistSec~
I ~i~~csh~;er I b~~y ~ho~l;~ !O~o;~~rybo~k~el(i~~~~~
I ductory olfers counts as first selection).
3 books for only $1 each
I
(write in numbers)
I
I
I Name ••................••........... ····.·· ..•

~----------------~

DATA BANKS

A POSITION PAPER

Prof. Caxton C. Foster
University of Massachusetts
Amherst, Mass.

Data banks are not a new problem. Hamurabi kept
an extensive clay tablet library of legal transactions several thousand years ago in Babylon. What
has happened with the advent of computers is that
the cost of operating such data banks has dropped
by several orders of magnitude. Consider a clerk
who can perhaps examine one record per second and
is paid $1.80 an hour. He (or she) can then examine about 2000 records at a cost of $1.00. Now consider a computer that can scan records at a rate of
500,000 per second and rents for $250 an hour. For
the same cost of $1.00 it can examine 7,200,000 records. Done by hand, a comparable search would cost
$3,600 instead of $1.00. This drastic reduction in
cost has made economically feasible things that
could only be dreamed of yesterday. New and exciting services can be provided to individuals, organizations and researchers. We need not extoll their
virtues here. Many others are doing that elsewhere.
Data banks are inevitable and essential in a complex society. Our civilization would quickly strangle in the flood, of paper work if computerized files
were eliminated. Although the potential for good
inherent in data banks is very large, we are afraid
that the potential for evil is infinite.
When every interaction of an individual with society can be collected, sifted and analyzed at low
cost, he will be paralyzed by fear -- fear that today's innocent behavior, recorded indelibly on the
data bank, can become tomorrow's subversive activity. As we see it the most serious problem related
to the establishment and growth of data banks is
the erosion of our constitutional rights. Once surrendered these rights will be most difficult to regain. "Big Brother" armed with computerized data
banks is very big indeed and will ~e almost impossible to displace. If this is indeed the case, and
we firmly believe it is, then this country should
be very careful of allowing such a possible instrument of repression to be forged.
In what follows we will attempt first to show
that the above described danger is very real, indeed inevitable, unless strong action is taken immediately. Second, we will try to outline some of
the "strong action" we think might be helpful.

form. By themselves, these separate data files are
reasonably innocent, but unfortunately, they don't
stay by themselves; they tend to agglomerate -- in
the name of efficiency, economy and service. Agency
A and Agency B each have computerized files, and,
to some extent, each could utilize some of the other
agency's information. So they set up a joint operation, halve their operating costs and 1984 is one
step closer.
This "gossiping" among supposedly benign agencies
could presumably be prevented by appropriate laws.
The second way that data banks grow is by gathering more and more peripheral information. It
seems that every time one fills out a new form it
has more questions on it than the last one; more
personal and more irrelevant questions. While not
yet asked to list one's grandmother's maiden name
on a form. it may be expected any day. To some of
us this is not sensitive data. To some it might
well be.
Other information is more sensitive. What may
be very pertinent in one's family physician's files
would be impertinent in the extreme in the hands of
a direct mail advertiser.
The individual should be free to refuse to answer
any question on any form that he considers personal,
and it should take a court case for the interrogating agency to deny to him, on this basis alone, the
services it purports to deliver.
Now let us turn to more malign groups - a wouldbe dictator, or perhaps organi~ed crime.
Some computer experts say that data banks can be
made secure. At best these people can be called
over-enthusiastic. As many husbands discovered at
least as long ago as the crusades, any lock caribe
opened by a skilled smith with enough time and
enough incentive. Let us look at some of the 'ways
this might happen.

The Development in the Near Future

Machine Failure - We have all heard of computers
printing out checks for $1,000,000. What one does
not hear of, unless he works in or around a machine
room, is the number of times a day programs are rerun because of transient hardware errors; but the
number is not insignificant.

More and more organizations are finding it advantageous to develop data banks. Motor vehicle registries, income tax agencies, armed forces, police departments, credit bureaus, banks, insurance companies, direct mail advertisers, and many others are
collecting, or are about to begin collecting, little
bits and pieces of our lives in machine readable

Logical Errors - Although most logical errors
are caught in the program development stage, some
persist for a long time. Every large program has
undiscovered flaws in it. One very famous example
concerns FORTRAN - perhaps the most'widely used of
all computer languages. The manuals describing the
system said that an attempt to take the square root

'28

COMPUTERS and AUTOMATION for March, 1971

of a negative number would result in a value of zero.
It was many years after FORTRAN was designed and had
gained world-wide acceptance that someone discovered
that the square root routine actually returned, not
zero, but the square root of the absolute value of
a negative number. If this supposedly glaring error
could pass undetected for years, what subtle flaws
could there not be in a data bank supervisory program?
Wiretapping - If a data bank is remotely accessible, then an enterprising group could soon have
copies of most of its data by tapping one or more
of the phone lines leading between the central computer and the remote terminals. Of course "scramblers" are available, but that just brings us back
to the problem of locks again.
Unintentional Compromise - When a request arrives
for data out of the bank, there must exist some kind
of validation procedure that establishes the requestor's right to the data in question. Now, if
there are many different people requesting information, the validation procedure cannot be too complicated if the system is going to be useful. Generally speaking, the computer industry has relied
on the use of one or more "passwords" for gaining
access to a system. If a person signs on the system, says he is Mr. Smith, then presents the secret
code word that "only" Mr. Smi th knows, we generally
let him in. This, of course, can be elaborated into
many levels with new and different' passwords (secret
codes) being required at each.succeedingly more sensitive level. We can change the code words every
day; every 15 minutes if necessary. But eventually,
if the user presents the right keys at the right
times, we open the locks. If Mr. Jones should somehow get hold of Mr. Smith's secret codes, we would
have to believe he was Mr. Smith (if he said he was)
and let him into the system.
Pseudo Sign-On- A couple of enterprising high
school students in Amherst wrote the following program for their own "amusement." We have several
teletypes in the computing center that are permanently connected to the computer. Operation over
permanent leased lines is quite common and would
be equivalent. These boys wrote a program which
mimicked the normal sign-on procedure of the system
and requested (as usual) the user to give his user
number and secret code. They would leave the terminal running with this program in the machine and
an unsuspecting user coming up to the terminal in
question would think he was talking to the system,
while actually he would be talking to their program.
Once the user number and secret code were captured,
the program would write them out onto a file so the
students could examine them later, and then it would
fake a "disaster" (system crash) so that the user
would be forced to sign-on again - this time to the
real system, and never know that his code was no
longer secret. By the time they were caught these
students had "secret" codes of over 100 users. Fortunately, we don't keep classified information on
our time-sharing system.
Eavesdropping - We have a second generation machine (CDC-3600) which has write and execute protection but no "read" protection. With third generation machines this situation could be simulated
by judiciously grounding out the read protection
circuits. Since reading "out of bounds" is not a
usual activity, this might go undetected for a long
time. How many systems have incorporated in them a
check of that circuit? In any event a computer
science graduate student wrote a fairly simple program-that would eavesdrop on the I/O buffer of any
COMPUTERS and AUTOMATION for March, 1971

other terminal and reproduce a copy of all transactions on his own terminal. The programming consultants at our center now use this program regularly
to help distant users debug their programs.
With a bit of modification this program could
scan all the ports looking only for sign-on codes
and list them as they appeared.
Mini Computer Code Crackers - Mini computers of
high capability are now available for under $5000.
For an additional $1000 one can purchase about 10
telephone adapters (data-sets or modems) and set up
a lucrative business cracking codes. The mini computer calls up the time-sharing system, presents a
user number, and "tries" a particular secret code.
If it doesn't work the mini calls again and tries a
new one.
Now on a typical system a secret code is any
combination of four letters, say: ABCD. There are
roughly 360,000 possible codes (26 4 ). If each try
takes 10 seconds and we have 10 data sets working,
we are making one try per second or 3600 per hour.
With average luch we should hit on the proper code
after trying half the possible codes, or in about
50 hours of trying. If the time-sharing system is
operating 16 hours a day, that represents an average of two codes cracked per week.
Suppose the entire mini computer plus data sets
costs as much as $10,000. At typical leasing costs
this would rent for about $200 per month, or $50
per week. This makes the cost of cracking a code a
ridiculously low $25.
Subversion of Operation Staff - Perhaps the
weakest point in the security of any system is the
staff that operates the machine. Peter Ustinov in
the movie "Hot Millions" has explained this fully.
Bad Data - Erroneous data entered into a bank
unintentionally is bad enough, but the possibilities
for character assassination available by entering
intentionally false and derogatory information into
a data bank are staggering.
Authorized Misuse - So far, we have mentioned
hardware, software and personnel problems, against
which some safeguards can be provided. What will
be the most difficult to control will be the overzealous administrator who can, and must, because of
his job, have access to the data banks at will. I
don't think we will ever appreciate how lucky we
are that Hitler and Company were defeated before
computerized data banks became available.
Suggested Paths of Action

In the previous section we have tried to point
out that any data banks at all contain dangers that
can only be partially guarded against. Yet data
banks already exist and will undoubtedly proliferate.
In this section, we wish to suggest some possible
paths of action that might prevent some of the most
serious abuses of these systems. These fall into
two broad categories: Protection of the individual,
and control of the agencies maintaining data banks.
Protection of the Individual - There are a number of steps which can be taken to protect individual rights. Some might require new laws, some, new
interpretations of existing laws. All might possibly be contained in a new amendment to the constitution. The right to refuse to answer non-pertinent
or impertinent questions. The 5th Amendment protects
29

the citizen in court. Military codes of ethics require prisoners-of-war to give only name, rank and
serial number. The private citizen needs to be protected from being forced (however subtly) to answer
questions he considers to be personal. An agency
collecting data should be required by law to accept
a "refuse" answer as fulfilling all its requirements
for information until it can be established in court
that the desired information is indeed necessary to
enable it to do business properly. Obviously, loan
agencies need to know how deeply one is in debt before deciding whether they will lend money or not.
Other examples are legion.
Right of Access - A citizen is guaranteed the
right to face his accusers in court. He should also
be permitted to examine any files, anywhere, that
relate to him. Semiannually copies of files should
be automatically mailed to each individual. Other
copies might be provided at cost.
Right of Challenge - If an individual feels that
a file contains erroneous information about him, he
should have the right to 1) add explanatory information, and 2) challenge the validity of the file.
This second course of action should have the effect
of removing that file from the data bank until the
challenge is resolved in some legal fashion. The
provision of local Ombudsmen will simplify these
actions for the average citizen.
Right of Restriction - An individual should have
the right to restrict, at his discretion, the circulation of a file concerning him. Perhaps a more
positive approach to this would require his cooperation (signature, for example) before the file could
be opened and/or copied.
Regulation of Data Banks At the other end of the scale there must be control and regulation of those maintaining data banks.
The problems concerning those data banks which are
necessarily secret must be examined carefully with
the interests of the private citizen in the forefront.
Regulatory Commission - Congress should establish
a regulatory commission with full powers over the
collection, use and dissemination of personal information. This should be autonomous, but subject
to Congressional control lest these very guardians
be left unguarded. They should be provided with
the very best of technical counsel on their own
payroll. Some areas within their spheres of interest might include:
Disclosure of Distribution - All questionnaires
should list the agencies that will have access to
data supplied by the individual thereon.
Disclosure of Access - The individual should be
notified whenever anyone examines his file. This
notification should include the date, the agency
examining, and the reason for the examination.
The Cost of Protection of the Individual - Let
us take as a starting point the cost of a bank
check. This is typically around 10¢. For this
price a bank will take a piece of paper and convert
it into money and provide you with a monthly record
of all transactions. Particularly if the check must
go from one bank to another, there are several humans involved in the processing so one cannot assign
more than a penny to the computer processing involved. This would seem to be a small cost to impose on say a credit bureau which charges several
dollars for a credit reference.
30

Flying Squads - The Commission should maintain
one or more "flying squads" of programmers and engineers who, without prior notice, appear at a data
bank and" take over" to test its compliance wi th the
law. This will help to keep the computer room personnel honest.
Devil's Advocates - The Commission should retain
several astute, devious, highly trained programmerengineers whose job it is to try to find ways to
"crack" the various data banks. These people should
be rewarded sufficiently well as to make selling out
unprofitable, or at least unnecessary.
Mergers - Data banks should not be merged without
the permission of the Commission, and then only for
compelling reasons .. Keeping data banks sufficiently
segmented is a built-in safeguard that will make it
much more difficult to misuse total information on
an individual. Bonafide research in Public Health
or Sociology would be helped, not hindeLed, by
strict control of data banks, because it would increase the people's confidence that data collected
would be treated confidentially.
Conclusions

Perhaps the problem of data banks could be
greatly alleviated if the concept of "property"
were to be extended to cover information about an
individual. There is probably some precedent for
this in common law, where a man's "good name" is
his most precious possession. If the law recognized that information about John Smith belonged
to John Smith to do with as he chose, many problems
could be easily handled. Maintaining a data bank
would then be a privilege, not a right, and hence
could be licensed and controlled. John could lend
information about himself, or he could sell it-rrhe desired, but it could not be forcefully, or covertly wrested from him. Name, address and social
security number are required by law. All other information is at the owner's discretion.

EXPERIENCE AND PLANS IN THE USE OF
COMPUTERS FOR UNDERGRADUATE INSTRUCTION
- CONFERENCE, JUNE 23-25,1971
Dr. Fred W Weingarten, Director
Computer Services
Claremont College
Claremont, CA 91771

A grant of $50,000 has been awarded by the National Science Foundation to Dartmouth College for a
Conference on Computers in the Undergraduate Curricula. The Conference is to be held at Dartmouth College in Hanover, New Hampshire, June 23, 24, 25, 1971.
The purpose of the conference is the dissemination
of actual experience and plans in the use of computers in undergraduate instruction. Referred submitted
papers will comprise the bulk of the conference,
which will have parallel sessionsj invited papers,
panel discussions, and demonstrations will rou~d out
the meeting.
The scope of the Conference will be national.
Content will be broad enough to encompass most academic fields at two and four-year colleges as well
as at undergraduate schools of universities. For a
similar conference held last summer at the University
of Iowa, nearly 200 papers were submitted.
COMPUTERS and AUTOMATION for March, 1971

NEW YORK STATE IDENTIFICATION AND
INTELLIGENCE SYSTEM
Burroughs Corporation, Business Machines Group, Public Relations Office
Second A ve. at Burroughs
Detroit, M I 48232

•
,

In the State of New York, fewer innocent citizens
are detained for a significant length of time and
more criminals are arraigned more quickly as the
result of efforts by the state supported criminal
justice agency located in Albany, New York.
The New York State Identification and Intelligence System (NYSIIS) is a computerized recordkeeping agency which houses more than seven million
criminal and non-criminal fingerprints and corresponding identification and case history data. The
agency makes the data available through a communication network to over 3,600 agencies of criminal justice in New York State. NYSIIS was created by statute in 1965, as an agency within the Executive Department of the New York State Government.
Objectives

The ultimate objectives of NYSIIS are twofold:
(1) improvement of criminal justice administration
through systemic computerized information sharing,
and (2) protection and enhancement of civil liberties. NYSIIS has made significant strides toward
these objectives since its establishment over five
years ago.
Here is an example of how NYSIIS can protect
civil liberties.
A person is arrested and fingerprinted by the
Nassau County Police Department. The Nassau County
Police Department (NCPQ) has one of the forty-one
facsimile installations located throughout the state
which can electronically transmit to NYSIIS a copy
of the set of fingerprints in a total of 14 minutes.
Following receipt at NYSIIS and classification
of the prints, the NYSIIS computer, a Burroughs
B5500 soon to be replaced by a larger B6500, searches
more than two million stored classifications in no
more than 35 seconds to ascertain if there is a
prior criminal history on file in the agency for the
individual. If there is, a record of it will be
sent in 4.5 minutes by facsimile to the Nassau County Police Department. The entire process takes no
more than three hours.

Gallati, "More rapid and positive identification of
perpetrators of criminal acts means that the innocence of possible suspects is established sooner
and with more certainty. Victims may be more readily compensated for their suffering and financial
loss and they are spared undue harassment from
lengthy and traumatic investigation. Offenders'
rights are likewise better protected for they may,
in appropriate cases, receive summonses in lieu of
arrest, or if arrested, be discharged on their own
recognizance. JUdicial officers will have the necessary data to fix bailor sentence, or otherwise
rapidly dispose of cases, thereby preserving the
civil liberties of offenders and according them due
process, while at the same time, knowledgeably protecting society from those offenders who are a
threat to the community."
If the arrestee in the above example had a
record, the Nassau County Police Department would
recei ve a '''rap sheet" defining hi s criminal hi story.
This information would then be used by agencies
along the criminal justice continuum to see that his
case is handled appropriately.
These summary case histories contain such information as: personal description, arrest charges,
dates and places of arrest, arresting agencies,
court dispositions, sentences, institution data and
information concerning parole and probation. It is
planned to expand these case histories to include
considerable additional data as the system evolves.
NYSIIS began to produce "rap sheets" directly
from the computer during 1969, becoming the first
agency in the world producing summary case histories
by computer. The NYSIIS computer prints out "rap
sheets" at the rapid rate of 1,040 lines of record
information per minute from a vast computerized
data base containing mor& than 400 million bits of
criminal history information.
Fingerprints sent to NYSIIS during calendar year
1969 via the Statewide Facsimile Tramsmission Network reached a total of 57,847 compared with totals
of 46,956 in calendar 1968 and 25,062 in 1967.
Facsimile Network

Time for Fingerprint Search

Previously, the fingerprint search routine took
from seven to ten days during which time many persons were unnecessarily detained without bail, or,
in the alternative, released when the record would
warrant their being held.
Aocording to NYSIIS Director, Dr. Robert R. J.
COMPUTERS and AUTOMATION for March, 1971

With the completion of new installations during
1969, the facsimile network now covers 41 locations
of which 32 are in law enforcement agencies and the
remaining nine are used to service the courts, probation departments and district attorney's offices
in the five boroughs of New York City.
The heaviest contributor of facsimile prints
during 1969 was the New York City Police Department
31

with a total of 34,483, followed by the Nassau
County Police Department which submitted 7,517 sets
of fingerprints.
Although budgetary constraints make it impossible
for NYSIIS to install facsimile equipment in all of
the criminal'justice agencies throughout the state,
wherever such equipment has been installed it is
with the agreement that the equipment be 60nstantly
available for the use of all criminal justice agencies in the immediate area. Agencies that do not
have facsimile equipment available on site should
bring fingerprint cards to the nearest facsimile
installation, with the request that such prints be
transmitted to NYSIIS at the earliest possible moment.
Fingerprint Identification and Coding

Approximately 60 percent of the fingerprints received by NYSIIS (currently averaging 1,850 per day)
are not "identified" as a result of the name search
that is conducted in the Name Index Section. This
means that every day about 1,100 sets of fingerprints -- or 11,000 individual prints -- must be
carefully analyzed and fully classified before a
technical search is conducted by the computer or,
in certain cases, by manual means. The purpose of
the technical search, of course, is to determine
whether the individual does have a prior record
under a different name, which would explain why the
search in the Name Index Section had proven negative.
During the course of one year~ a technical employee assigned to the Classification Section will
carefully study and analyze approximately 200,000
fingerprint patterns. The task of a classifier is
to study the ridges in each pattern and then make a
determination, based on specific rules contained in
the American System of Fingerprint Classification
textbook, as to whether the pattern meets the loop,
whorl or arch requirements.
'
There are small loops and large loops, small
whorls and large whorls -- and varying ridge counts
are actually the basis of the formula that the computer uses in selecting one particular classification from among the hundreds of thousands of other
similar classifications within its memory.
Latent Fingerprints

Another area of fingerprint record-keeping that
NYSIIS makes available to the New York State criminal justice agencies is the Scene-of-the Crime (Latent) Fingerprint File. This latent file, one of
the largest in the nation and continually being expanded, contains scene-of-the-crime latent prints
and the fingerprints of several thousand selected
individuals well known to the police -- individuals
who have been convicted of homicide, burglary,
robbery, car theft and similar felonies. Fingerprints developed at the crime scenes (latents) are
submitted to the NYSIIS Latent Fingerprint Section
for searching and comparison. Latents not identified are added to the base file and are continually
compared against incoming arrest prints.
In 1969, trying to develop an even more effective
Latent Fingerprint Processing System, NYSIIS and
the law enforcement agencies in Onondaga County
participated in a pilot research study called, "The
Latent Value Field Study."

A small defined area of Onondaga County, with a
high burglary cr~me rate, was selected for the study
site. All burglary crime scenes within this designated area were searched for latent prints by the
investigating police officers. The Criminalistics
Research Bureau of NYSIIS used experimental semiautomatic coding techniques to prepare a base file
for the area against which incoming latent fingerprints were rapidly checked.
As a result of this pilot study, 69 latent fingerprint cases were submitted to the NYSIIS Latent
Fingerprint Section for searching. Six perpetrator
identifications were made on the basis of only sceneof-the-crime prints without other clues or any other
reason to suspect the previously unknown individuals
who committea the crimes.
"The latent fingerprint records are one of the
most important areas in which NYSIIS is involved,"
Gallati said. "Because criminal mObility from ci ty
to city is quite frequent, and because our criminal
and latent fingerprint files are two of the largest
in the nation, we have become a valuable service to
all agencies of criminal justice throughout the
State of New York.
"If law enforcement agencies take the initiative
to detect latent fingerprints at crime locations and
send them to NYSIIS, we know that we have an excellent chance of making a 'hit' on rapists, murderers,
arsonists and other criminals," Gallati stated.
Computer Capacity

NYSIIS was originally computerized in 1968 and is
in the process of installing a new, larger computer
to handle their rapidly growing data files. A Burroughs B6500 computer will replace the present B5500
which is operating at virtually 100 percent of capacity around the clock every day of the year. Its
data storage already exceeds maximum load for a
single processor.
Grants

"One of our greatest contributions to the criminal justice system has really been from a scientific technology point of view," Gallati said. "Thanks
to confidence displayed in NYSIIS by the Governor
and the legislature and hundreds of thousands of
dollars in Federal Grants, we have been communicating with most leading research and development people throughout the country. We have also visited
major research laboratories'around the country and
we are continuously relating new technologies to our
needs in the interest of new c?pabilities, improved
efficiencies and lower costs."
NYSIIS received $425,000 in Federal Grants for
new and ongoing projects in 1969. These projects
include: a personal appearance study; an auto related crime study; an evaluation of organized crime;
a concept evaluation of modus operandi (methodology
used in committing crimes); a nationwide computerized retrieval system for criminal history; research
and development on modus operandi; a probation and
parole system requirements study, and development of
a name search technique.

COMPUTERS and AUTOMATION for March, 1971

ANTI-ABM

ESSAY

CONTEST

ANNOUNCED

Daniel D. McCracken, Chairman
Computer Professionals Against ABM
4 Inningwood Road
Ossining, New York 10562

This is to announce an essay contest on the topic, 'Would you trust the lives of your children to
a highly complicated computer system that cannot be
checked out?"
The contest is inspired by a letter in the January, 1970, issue of Modern Data, signed by Dr. John
S. Foster, Jr., Director of the Office of Defense
Research and Engineering in the Department of Defense. Dr. Foster's letter was in response to the
statement of Computer Professionals Against ABM, in
which we argue that a complex real-time system can
only be brought to reliable operational status
through an evolutionary process of testing and development under actual operating conditions. This
is inherently impossible with the ABM (the Anti
Ballistic Missile system). We believe it is extremely dangerous to turn over the control of hundreds of nuclear missiles to a highly complicated
computer system that cannot be checked out.
Dr. Foster's reply was that all the individual
parts of the Safeguard Anti Ballistic Missile system (radars, missiles, and computers) have been
tested separately, and, "thus, the only real task
that the Safeguard system has, is to integrate all
of these functions in the computer programs and to.
check thoroughly and test out the programs before
the system is made operational." He explained that
the testing would be done with simulation tapes "so
that the system is exercised just as it would be in
a real battle."
If statements like these were made by a student
in his first course in computing, you'd give him a
C-minus and forget about it; in his second course
you'd flunk him and forget about it. But when they
are made by a person who controls an annual military
R&D budget in excess of $10 billion, you have to
think about trying to educate the man. Hence, the
contest.
I think 500 to 700 words would be about right. Please
send copies to me, Computers and Automation, Dr.
Foster (The Pentagon, Washington, D.C. 20310), and
(very important) your Senators (The United States
Senate, Was-hfngton, D.C. 20510). If you can't
think of better ones, here are a few points to use
for Dr. Foster's edification.
-- The full-scale operating system for the IBM
System/360 is of the same general order of magnitude of size and complexity as the ABM software.
It was thoroughly tested by IBM before release,
using simulation tapes as it happens, but after it
was released customers discovered literally thou-

COMPUTERS and AUTOMATION for March, 1971

sands of programming errors when they used it
under actual operating conditions. ·(Five thousand, to be precise, when my highly-placed source
told me the story two years ago.) This isn't a
slap at IBM; that's just how it is with big computer systems.
-- American Airlines has spent, in round numbers,
ten years and $100 million developing a passenger
reservations system. It was a near-disaster the
day it first went on line with real passengers although it had been thoroughly tested, but after a
process of development under actual operating conditions it was brought to a state of high reliability and usefulness. This might qualify as an
example of a system that had been tested by itself. So what happened recently when two other
airlines tried to do essentially the same job?
Well, one result is a $70 million suit by TWA against Burroughs, after cancellation of the contract because the system wouldn't work.
-- NASA, which Dr. Foster cites to show that
complex computer systems can be made to work,
freezes program changes 120 days prior to a launch,
then devotes 30 to 45 days to a fUll-time simulation based on all the lessons learned in previous
Mercury, Gemini, and Apollo flights. The computers are run during the simulation by the programmers who wrote the programs in the first place,
and later during the real thing the computers are
also run by the same programmers. Does Dr. Foster
expect the Russians (or the Chinese, or the Portugese, or whoever he's afraid of this week -- it
changes from time to time) to give 120 days notice
of a planned attack? Landing men on the moon
doesn't prove that we can make the ABM computer
work -- on the contrary, it demonstrates all the
things that would have to be done that cannot be
-- ---done.
-- A RAND study shows that when a single instruction is changed in a large program, the program
works less than half of the time when it is first
tried after the change. Since the ABM software
would be in a constant state of flux to keep up with
changes in the offensive threat as well as modifications for other reasons (such as the complete
change of computer now under study), how would it
ever be possible to get the programs stabilized long
enough to get the errors out? And you can be sure
that the offensive threat would change: since the
Russians have an ABM system of their own, which
their scientists privately say is useless, they already know that the simplest way to incapacitate
ours would be to keep forcing us to change it.
(When NASA sends men to the moon, nobody jerks the
moon out of orbit just as they start their descent.)
33

REDUCING DUES FOR UNEMPLOYED MEMBERS
CONTEST PRIZE

Computers and Automation will award a prize of
$100 to the best essay in this contest meeting the
requirements mentioned above and below.
The closing date for receipt of essays (copies of letters) is May 31, 1971 in the office of
Computers and Automation, 815 Washington St.,
Newtonville, Mass., 02160. The judges will be
Daniel
McCracken and Edmund C. Berkeley, Editor of C&A; their judgment will be conclusive.
It is expected that the winning essay(s) will be
published in the August issue of c&A.

The essay should be written in such a way as
to be convincing to a senator or representative
in Congress. The length of the essay may be
less than 500 words. The essay should meet ordinary requirements for being publishable. All
entries will become the property of Computers
and Automation. Two copies of the essay should
be submitted. The author's name and address
and three or four sentences of biographical information about him should accompany the essay.
In case of ties, the prize will be divided.
The prize will not be awarded if in the opinion
of the judges no sufficiently good essay is received.
.
Let me close on a positive note, with a constructive suggestion: I urge the ABM computer system designers to provide a thorough system trace,
writing the inputs and the system's responses on to
tape units deep underground, so the units could
survive a nuclear holocaust. Then if the Safeguard
system is ever used, the designers of the ABM system for the next civilization -- if any -- will
have the benefit of one debugging run.
The group -"Computer Professionals Against the ABM"
includes (organizations are listed for identification only):
Executive Committee
Daniel D. McCracken, Chairman, Consultant
Paul Armer, Stanford University
Prof. Joseph Weizenbaum, Mass. Inst. of Technology
Gregory P. Williams, Phoenix, Ariz.
Sponsors (incomplete list)
John W. Backus, International Business Machines, Corp.
Prof. Richard Bellman, Univ. of Southern California
R. W. Bemer, Phoenix, Ariz.
Howard Bromberg, Information Management, Inc.
Prof. Fernando J. Corbato, Mass. Inst.of Technology
Phillip H. Dorn, Union Carbide Corp.
Prof. William S. Dorn, University of Denver
Prof. Robert M. Fano, Mass. Inst. of Technology
Prof. Edward A. Feigenbaum, Stanford University
Robert B. Forest, Datamation
Prof. George E. Forsythe, Stanford University
Prof. Walter Hoffman, Wayne State University
Prof. Alston S. Householder, Univ. of Tennessee
Prof. Harry D. Huskey, Univ. of Calif., Santa Cruz
Prof. Donald E. Knuth, Stanford University
Prof. J. C. R. Licklider, Mass. Inst. of Technology
Prof. John McCarthy, Stanford University
Prof. Marvin L. Minsky, Mass. Inst. of Technology
Prof. Allen Newell, Carnegie Mellon University
Max Palevsky, Scientific Data Systems
Prof. Anthony Ralston, State Univ. of N.Y., Buffalo
Prof. Norman R. Scott, University of Michigan
Eric A. Weiss, Springfield, Penn.
34

Inst. of Electrical and Electronics Engineers, Inc.
345 East 47 Street
New York, NY 10017

The Board of Directors of the Institute of Electrical and Electronics Engineers. has authorized the
reduction~ by 50%, of all dues and fees for those
members of the Institute currently unemployed. This
step was taken in recognition of the economic conditions currently prevailing.
IEEE members, who are currently unemployed
through involuntary termination and are actively
seeking re-employment, may continue their IEEE membership through December 1971, with all privileges,
publications and services covered by membership dues
(and by Group or Society fees, and subscription fees,
if any) by payment of one-half of the established
dues and fees. To take ~dvantage of this arrangement, please inquire of IEEE Headquarters regarding
details.
SOCIAL DATA PROCESSING CENTERS
IN THE FORM OF TELETERMINAL TIME
Joseph Auciel/o, Director
Computer Job Bank
10 Upland Road
Plympton, MA 02367

Perhaps I can make a suggestion that will eliminate some of the misunderstanding between certain
student groups and the data processing industry.
I suggest tnat "social data processing centers"
be set up to make computer power available to individuals or groups who are working on problems whose
solutions will benefit our society as a whole. The
data centers would be manned on a volunteer basis
by computer professionals. The computer time would
be given thru telecommunication links by computer
users; they would be free to give as much or as little CPU time as they choose to the social data center.
This proposal would allow some computer manufacturers and users a way of expressing (to varying degrees) their social concern. By computer use for
people and society and their problems~ the computer
perhaps will cease to be known by some as a mechanical monster responsible for our social ills, but
rather will be viewed more properly as an extension
of man's mind and intelligence.
1

COMPUTERS, SCIENCE, AND ASSASSINATIONS
Computers and Automation believes that the
possibility of conspiracies in the assassinations
of important American leaders in our times is of
the utmost interest and significance to every
American -- and especially to computer people,
because computers can be used: to handle large
amounts of information easily; to correlate the
information rapidly and accurately; to prove-or
disprove certain theories or possibilities of
conspiracy; etc. Therefore, computer people can
make a unique and important contribution to society in this area.
Accordingly, Computers and Automation is publishing from time to time articles and reports
on: _ investigations into assassinations; the major evidence; and the application of computers
to the evidence. Our purpose is to present important, useful, and authoritative information
objectively in order to find out the truth.

COMPUTERS and AUTOMATION for March, 1971

liThe Assassination of President
John 'F. Kennedy:
ANOTHER

VIEW

'The Application of Computers to
t,he Photographic Evidence

Benjamin L. Schwartz, Ph. D;
McLean, Va.

Comment
Outline

Note by Edmund C. Berkeley, Editor: In the May
1970 issue, "Computers and Automation" published a
32-page article by Richard E.; Sprague, computer professional, with the foregoing title.
This article contained: (1) eleven photographs
related to the assassination; (2) a two-page chart
of Dealey Plaza, Dallas, Texas, where President Kennedy was shot, showing (a) geographical details, (b)
the location (or probabl~ location) of many events
in regard to the assassination, and (c) the locations
of many of the 75 photographers; (3) a tabulation
made over many years by Sprague of some 510 photographs that were taken, and some details about each
one; (4) a bibliography of 22 references; and (5)
about ten pages of text. This article reported a
number of aspects of a study of this assassination
made by Sprague from 1963 to 1970. Part 3 of the
article discussed how computers could be applied to
the photographic evidence, consisting of over 25,000
frames plus 350 still pictures, a task seemingly beyond the capacity of any human being unaided by a
computer.
See the notice and advertisement about this article on page .44.
The following article by Dr. Benjamin L. Schwartz,
a mathematician and an operations research specialist, has been received, commenting on Sprague's article.
We are glad to publish Schwartz's article since
it is a thoughtful, important, and significant polemical attack on Sprague's article, and also on "Computers and Automation" for publishing Sprague's article.
Mr. Sprague has said that he does not wish to reply to Schwartz's article. But I do wish to, and
following Schwartz's article, I have put a response.
For readers who enjoy a donnybrook, the two articles are recommended.
I should like to add that in the course of an interesting (and friendly) conversation with Dr.
Schwartz, he said his basic position in regard to
an argument about a theorem is not whether the theorem is or is not true, but whether the theorem is
proved by the evidence offered. In other words, a
mathematical paper is acceptable and publishable
when the author demonstrates what he says he is setting out to demonstrate, and otherwise it is not acceptable and not publishable.
COMPUTERS and AUTOMATION for March, 1971

1.
2.
3.
4.
5.
6.
7.
8.
9.
10.
11.
12.
13.
14.
15.
16.
17.
18.
19.
20.

Introduction
Statement of Position I
Statement of Position II
Commentary I
Commentary II
Inconsistency I
Inconsi stency II
Computer Relevancy
Editorial License
Retreat I
An Impossibility
Retreat II -- An Invisibility
Insanity -- Retreat III
An Inaccuracy
An Implausibility
A Contradiction
Another Contradiction
An Irrelevancy
On Scientific Attitudes
Conclusion
Introduction

In the May 1970 issue of Computers and Automation,
Richard Sprague published a remarkable article enti tIed: "The Assa:ssination of President John F.
Kennedy: The Application of Computers to the Photographic Evidence". That article claimed to show a
number of spectacular things about the assassination. Among them were that the assassination stemmed from a conspiracy; that more than 50 persons
were involved in the conspiracy; that among these
people were members of the Dallas police force, the
CIA, anti-Castro Cuban exiles, and others; that Lee
Harvey Oswald was involved in the conspiracy, but
did not do any shooting; and that knowledge of all
the preceding items has systematically been denied
to the American people by a (second) conspiracy of
silence and/or distortions engaged in by the news
media and certain public figures. Unfortunately,
Mr. Sprague failed to prove his case, in my opinion
and apparently that of many other readers. In subsequent issues, he has withdrawn many of the claims
that he has provided conclusive proof. He still
holds to his contention that his version of the assassination is true, but he has admitted implicitly
that many of his arguments, as presented last May,
were faulty.
The present article brings together a number of
objections to the May article, including some that
Mr. Sprague has acceded to, and some other, new ones.
The reader should read (or reread) the original article in considering the material below.
You may feel that I am being unfair to Mr. Sprague
when I attack an argument of his which he has subsequently already abandoned. However, I feel it is
appropriate to include both the original claim and

the objections that forced him to withdraw it.
The reason is this: Mr. Sprague's own credibilityis part of the issue. In the article, he included many assertions of additional facts about
the assassination for which he gave the reader no
evidence. For example, he says (p. 50-51) "A .. .
conclusion ... drawn: from photographic evidence .. .
(is that) ... Oswald did not shoot Officer Tippit;
... two other men did." But Mr. Sprague does not
show us the photographic evidence that supports
this claim. We have to take his word for its existence. 'Hence, it is very much to the point to know
how good he is at drawing sound conclusions from evidence. When he is shown to be vulnerable on other
points, it must reflect in the credence we give to
such of his claims as the foregoing.
Statement of Position I

with my article; read additionally as you feel necessary; and then take sides.
Before going on, I would note one point On which
Mr. Sprague and I agree: the Warren Commission
report is seriously defective. But I will show you
that Mr. Sprague and the National Commission to Investigate Assassinations (NCTIA) are even further
away from meeting any objective scientific criteria
for accuracy, in either reporting or deductive
reasoning.
Mr. Sprague claims to show positively thot "There
was a conspiracy"; " ... at least three gunmen .. .
fired ... at President Kennedy"; ... "Oswald ... did
no shooting." (p. 30) He furthermore claims that
he will show thi s "to be true on the ba si s of substantial conclusive evidence", because" ... there
is conclusive evidence".

You, the reader, are certainly entitled to know
where each of us stands. For my own part, I certainly do not claim to be an expert on the events
of Dallas 1963. I am an interested layman, like
yourself, willing to consider new points of view
as they appear in the public media, but certainly
not so involved as to have made any original investigations.

It does not suffice, for Mr. Sprague to demolish
the Warren Commission: and indeed that had already
been done before him. Rather he must build his Own
case with "undeniable evidence" (p. 30) to survive
the same exacting scrutiny that was given the Warren Commission report.

On the basis of the evidence I have seen and the
arguments I have heard, my current belief about
Dallas 63 is the following: Oswald was the lone
assassin who shot and killed John Kennedy; there
was no conspiracy; there are many unexplained details, but all alternate explanations I have heard
are so far-fetched as to strain credibility beyond
reason and leave even more loose ends.

Mr.' Sprague's basic position is certainly a hard
one to attack. If evidence for the first conspiracy is fragmentary, uncoordinated and unsubstantial,
that merely "proves" how effective the second conspiracy has been. If you find any of the "strange"
events of the second conspiracy to be not so strange
in your eyes, -- well, Mr. Sprague has already admitted that each separate event has a reasonable
explanation.

These opinions are not strongly held, and all
are subject to change if new evidence comes to my
attention. But with the increased passage of time,
I think it less likely tha~ there will be change.
This expectation is reinforced by seeing Mr.
Sprague's miscellaneous collection of loosely COnnected tidbits offered under the claim of "incontestable proof".
Statement of Position II

If I read him correctly, Mr. Sprague has for his
main thesis the following: John F. Kennedy was
shot to death in Dallas by several members of a
large conspiracy, which Mr. Sprague calls the first
conspiracy.
There was also a second conspiracy, which came
into being spontaneously, after the killing. The
members of the second conspiracy were public figures and officials. The objective of the second
conspiracy was to suppress and conceal evidence of
the first. The evidence for the first conspiracy
was in large part successfully destroyed, distorted,
or hidden from the public by the second conspiracy.
The evidence for the second conspiracy is a number
of what Mr. Sprague calls "strange events". Each
of them separately, he admits, has a reasonable
explanation. The combination of all of them, however, is too unlikely, in his view, to be accidental.
Commentary I

For reasons explained in more detail in the final section, the writer thinks that it is important
for each reader to consider carefully the conflicting positions. Read Sprague's article: read here36

Commentary II

However, it seems to me that the relative invulnerability of Mr. Sprague's position is less a
proof of its validity than a tribute to his debating style~ He is ingeniously turning any apparent
damaging circumstance into an actual asset.
Inconsistency I

This ability to work both sides of the street
is manifest many times in the article. For example,
consider the question of people departing from the
murder scene. On page 36, he describes how several
men have been photographed shortly after the killing, running away. At least one of them leaves in
a conveniently parked, nearby car. This hasty departure is taken as evidence of guilt.
But a little later on (p. 50), he considers a
hypothetical conspiracy member called the umbrella
man. Does he make a fast getaway? Indeed notl
Mr. Sprague~escribes the many pictures taken later
on in the day that still show the umbrella man
hanging around aimlessly. This delay Sprague seems
to find also suspicious.
If both early and later departures can be interpreted as signs of guilt, I wonder what the innocents didl
(We shall return later for more about the umbrella man.)
I nconsistency II

Another example of Mro Sprague's unusual logic
is seen in his attitude toward the Warren Commission
and its members o He has bitter words (po 34) for
the unwillingness of Justice Warren and others to
COMPUTERS and AUTOMATION for March, 1971

give ear to and respond to the many criticisms of
their report. Their silence is construed as evidence confirming the NCTIA theory.l But when one
of the Commission (Sen. Richard Russell v p. 59)
does speak out along the lines Mr. Sprague seems to
desire 9 that too is taken as being in full accord
with Sprague's theories o Both the open mind and the
closed one work in his favor. He had devised chains
of argument under which any evidence 9 or no evidence
at al1 9 can be construed to help his case.

these has anything to do with the computer business.
But his other reason does. The article, he
claims "focuses on 0000 computer-assisted analysis
of 000. data 0.0. to solve a crime"o But as I have
noted v that certainly is not the "focus" of the
article: and the degree of justification obtainable
from this one. rather contrived, argument would not,
I am sure, win admission for any other subject as
remote from the main stream of interest of readers
of C&A.

Computer Relevancy
Retreat I

I confess to having originally plunged into the
article eagerly in hope of finding both new insight
into the John Kennedy assassination and new information about computer applications o Both hopes were
futile. I claim that the material is not computerrelated 9 and should not have appeared in C&Ao

Mr. Sprague has now admitted the correctness of
this objection. In the July 1970 issue of C&A he
saysv "The conclusions stated in the article were
arrived at by the author before the computer appli~
cations to photographic evidence were begun".

Some quantitative facts tell the story. The
article occupies 32 pages of the issue. Of this 9
about one third is narrative text: another third is
tables 9 notes 9 and similar material: and the remainder comprises photographs and drawings. Of these
32 pages 9 less than three (1-1/2 text; 1-1/3 tables)
are given to computer applications. In this brief
section v the material consists of vague generalities
about how computers might be used. There is no description of any actual use of computers. Neither
is the material sufficiently specific to permit an
analyst to reduce the suggestions to practice.

In correspondence with me 9 Mr. Berkeley also has
admitted to me that the computer relevance of the
article is marginal. But Mr. Berkeley has also
pointed out that C&A does publish non-computer related material v e.g., Kingman Brewster's article in
the February 1970 issue. However, according to the
announced editorial policy concerning such articles,
they are to appear in a special department ("The
House is on Fire") specifically created to carry
such papers. Mr. Sprague's article was not published
in that department o It tried to "make it" on computer relevance, and failed.

In short 9 the conclusion this writer reaches is
that the computer part of the article title v and the
computer related addendum (Section 3) of the article
were afterthoughts v appended to give a tenuous justification for this particular journal to publish
the article.
Editorial License

Granting that this is simply this writer 9 s inference. there is some additional evidence to support it. I suspect that the editor 9 Mro Berkeleyv
believes Mro Sprague's thesis of a conspiracy and
desires to publicize ito I also suspect he is
aware of the defects in Mro Sprague's presentation.
Mro BerkeleY9 in his editorial (po 6). refers to
the article as "one of the most important 0000 we
have ever published". Among the reasons he cites
for his decision to publish are the "opportunity
to help important truths become known"o Yet
even while believing that Mr. Sprague has an "important truth." Mro Berkeley admits "It is possible
that 0000 Sprague's article does not prove 000 nor
adequately support 0000" the assertions.

.. 0.

Among Mro Berkeley's other stated reasons for
publishing are a social conscience: the use of glossy
paper in C&A (permitting high quality reproduction
of the photographic evidence for the readers to examine): and his relative freedom from advertiser
pressure as a paid circulation magazine o None of
IMy own conjecture is that Justice Warren had foisted
off on him a difficult and unpleasant job for which
he was not qualified by training or temperament. After doing his best in the year of the Commission's
existence. he had no stomach for further acrimony
over a distasteful topic o
Of course. my making such a suggestion automatically makes me suspect as a member of the second
conspiracyo Indeed v I have been asked bluntly about
my connections with the CIAI Needless to sayv there
are none.
COMPUTERS and AUTOMATION for March, 1971

An Impossibility

On page 32, Mro Sprague describes an "interesting
event" concerning the discovery of Jim Hicks by New
Orleans District Attorney Jim Garrison.
"(Hicks) showed up voluntarily in Garrison's office o Garrison and 0000 Jones Harris suddenly realized that they had seen his picture before 0000 the
photograph Harris had seen is a picture of Hicks
(See Figure II)". (po 32)0 Figure 11, on page 33
is a somewhat blurred 7 x 9 enlargement. It shows
several men SEEN FROM THE REAR. The one identified
in the caption as Hicks has visible no part of his
face whatever. It is simply impossible that anyone
could "suddenly recognize" the subject of that picture.

0 •••

Mro Sprague has subsequently explained that the
identification by Jones Harris was based upon the
build v hair style. and clothing of the subjecto Since
these are the only visible features of the man in
Figure 11. there seems to be nothing new in these
comments. If a reader previously doubted that a
person could be identified from a rear view9 he will
still doubt ito (Another input supplied by Harris
is that Mrs. Hicks supposedly confirmed the identification v afte~ the facto Obviously, this has no
bearing on Harris's original ·sudden realization'.)
In fact v as far as I can see, Mro Sprague has added
nothing except that he has nothing to add o Neither
do I. I am quite willing to have the reader examine Figure 11 and decide whether he believes that
a person like Jones Harris. who did not know the
subject, could identify him from the picture.
Retreat II - An Invisibility

In the original article 9 Sprague claimed that the
picture showed that Hicks was carrying radio equipment in his hip pocket. and an antenna could be seen
trailing behind him. This claim has now been withdrawn. It is acknowledged that no antenna can be
seen. (C&A, July 1970, p.32)
37

However, Mr. Sprague still claims that the antenna is there, and he cites as proof that he has
another print of the same photograph in which this
object is visible. Mr. Sprague adds that he will
show this to anyone who wishes to see it and comes
to see him.
It is clearly impossible for more than a few of
among the thousands of subscribers to C&A (many from
abroad) to make such a visit. For the vast majority
of the reading audience, the argument has been changed. The original theme was: "Lookl Here it isl
See for yours~ifl" It has now become, "I, Richard
Sprague, have seen the evidence. Take my word for
itl Believe in mel"
There can be no doubt that Mr. Sprague is honest
in his conviction. He has investigated, and he is
deeply convinced of the truth of his claims. His
sincerity is not at issue, and never has been. His
credibility is.
For that reason, I cannot accede to his plea that
I should take his word for somethingi and I do not
think that other members of the general readership
should either. Mr. Sprague is well aware of the
extent to which an argument is emasculated when the
evidence in support of it, instead of being openly
published for all to see, is limited to a small
audience. This is one of his main complaints about
the Second Conspiracy. He can hardly expect that
we will exempt him from the same criticism when it
applies to the other side of the case.
Insanity - Retreat III

Sprague adds that Hicks is now confined to a mental institution. Perhaps we are to believe that this
is part of the national conspiracy to suppress the
truth. For myself, I find it perfectly reasonable
that Mr. Hicks may be deranged. Even casual readers
of criminology literature are well aware that every
sensational crime brings forth a large number of
guilt-seeking mental cases to "confess". Mr. Hicks
mayor may not be onei but if he is not, it is up to
Mr. Sprague to show this.

not be sure.) His mouth is hidden by the shoulder
of the passenger in front of him. His expression
is masked from us. If Mr. Sprague sees the face
"contorted" he is using a vivid imagination, not a
scientific detachment.
This point has no apparent bearing on the assertion of conspiracy. I mention it because Mr. Sprague
raised it in the first place. Perhaps he thinks it
relates to his case. It certainly does have a
significance in our evaluation of his competence in
photointerpretation.
An Implausibility

On page 34, Mr. Sprague discusses the 6th floor
window from which the shots were alleged to have
come. He gives us two photographs that show this
window, respectively" •••• exactly 5.7 seconds before •••• and exactly 3.5 seconds after" the shooting. I am impressed with the preclslon of the time
measurement: 100 millisecond accuracy I
But I must question how such precise determinais possible. Regarding a different photograph
(Flgure 10) he explains carefully how precise time
determination is accomplished. In the razor-sharp
print of photographer Altgen's picture, the auto
wheel position can be measured relative to the center stripe. A comparison with the Zapruder movies
permits exact time matching.
ti~n

But for Figure 6 and 7, such an argument will not
do. Figure 7 is sharp, but it shows only a building
wall, with no distinguishing feature, no moving
elements (except people), no way to relate to other
pictures. In Figure 6, the street can be seen, but
the print is so fuzzy that the vehicles cannot even
be identified as to make, much less exact position
on the road. So the precise time determination, if
done at all, must be accomplished by some other
mechanism. But without being told what it is, the
reader is asked to accept that these "two pictures
destroy the Warren Commission's findings". (p. 35.)
I don't think Mr. Sprague has proved his case.
A Contradiction

On the question of Mr. Hicks' mental competence,
Mr. Sprague has been eloquently silent, despite
several reader inquiries. He gives his source of
information on Hicks confinement as a newspaper
story (despite his mistrust of the news media). He
leaves us no reason to doubt that Mr. Hicks is actually a valid mental case. It leads one to wonder
how good the case for conspiracy can be when one of
the star witnesses is apparently psychotic o The
exact question was addressed in the Clay Shaw trial
in New Orleans, cited by Mr. Sprague (p. 31). The
decision made by the jury there is well known: No
conspiracy.

Let us remain with those two pictures of the fatal window. Mr. Sprague correctly says "the first
•••• shows the window with only a box appearing in
it. The second •••• show~ the window with only the
box in it." He is right. 2
But look again. THE BOX HAS MOVED! Although we
may now know exactly when the pictures were taken,
or even which came first, we are sure that somebody
was at the window between those two exposures. If
Mr. Sprague's times are accepted as correct, then
in the IS-second period covering the shooting, someone was in the room, at the window.

An Inaccu racy

By far the clearest of the pictures in the article
is Figure 10, understandably taken by a professional
photographer, James Altgens. Mro Sprague rightly
draws many conclusions from a careful examination of
that picture. But one of them is certainly wrong.
He says "Governor Connally had been hit .0 •• He can
be seen with his face contorted." (p. 47).
Well, hardlyl The person identified as Gov.
Connally is turned away from the camerai and we can
see only a 1/4 view of his face. His ear and cheek
are clearly visible. The tip of his nose shows in
profile. And a corner of the left eye. (I think
the eye is open, but even with a magnifier, I can38

Yet Mr. Sprague has flatly asserted (p. 30): "At
least three gunman •••. (none of whom were in the
sixth floor easternmost window of the Texas School
Depository Building) ...• fired •••• at President
Kennedy." (My emphasis). And again (p. 35) "No one
fired 2.!!Y shots from the window." (Emphasis in
original). How Mr. Sprague converts certain evidence
that somebody was present (the moved box) into the
conclusion that no shot was fired is a step that
evades me.
2Almost. A small corner of a second box shows also
in Figure 7. But that might be because of a different angle of view.
COMPUTERS and AUTOMATION for March, 1971

Another Contradiction

Mr. Sprague considers a hypothetical "umbrell a
man" (mentioned previously), (p. 50) who might have
been the on-the-scene commander. The umbrella man
is supposed to be seen in Figure 9 with his umbrella
"open and low over his head". Figure 9 is too fuzzy
to be sure of. But there is, in fact, in Figure 8,
a corner of something that might well be an umbrella.
But the man himself cannot be seen; he is hidden
by the shoulder in the foreground. The statement
that he is holding it "close over his head" comes
from the same vivid imagination that created Gov.
Connally's facial contortion.
More important is the umbrella man's position,
In the text (p. 50), Mr. Sprague places him "in front
of the Stemmons Freeway sign", which is confirmed by
the drawing on pages 48-49. Even the reason is
given: "So that people on the western side can't
readily see him and people in front of the sign are
facing away from him." (p. 50),
But he was not in front of the sign. If there was
an umbrella it was behind the sign. In Figure 8, the
left hand vertical standard of the sign can clearly
be seen in front of the presumed umbrella. Or perhaps Mr. Sprague wishes us to believe that the umbrella ~ is in front of the sign, but the umbrella
is behind.

An Irrelevancy

Mr. Sprague devotes four pages of pictures and
half a page of text to "the 'Tramps f and the tPhoney
I am at a loss to explain why. They
prove nothing about the supposed conspiracy. All
they show is that there are some loose ends in this
case. But that was known and acknowledged long ago.
Poli~emanf".

Mr. Sprague appears to find it" signifiean't"
(p. 36) that the tramps were questioned but no
record exists of their being booked. He seems to
feel this is unusual.

It is not. Every day, all ove the country, people are detained for questioning, 3 sometimes in the
station house, and then released without being
booked. It is commonplace. It has happened to me.
It will happen several hundred times on the day you
read this. For Mr. Sprague to appear to find something suspicious in it indicates either a lack of
candor or an incredible naivete about police procedures.
On Scientific Attitudes

Are Richard Sprague and the NCTIA prophets without honor, or are they cranks: It l! important to
decide. Martin Gardner i in his witty and scholarly
book on pseudo-science, Fads and Fallacies in the
Name of Science (Dover, 1957) includes a chilling
reminder that in the 1930's, a renaissance of German
quasi-science paralleled the rise of Hitler. He asks
pointedly "If the German people had been better
trained to distinguish good from bad science, would
they have swallowed so easily the insane racial
theories of the Nazi anthropologists?"
I share with Sprague, Berkeley, and Gardner the
concern that the people should be informed. In his
opening chapter, Gardner gives several clues that
often help identify the quack or quack group. Here
is that list.
COMPUTERS and AUTOMATION for March, 1971

I. They work alone, without interacting with
other organizations in the field (in this case,
criminal investigation).
2. They claim this isolated status is involuntary and due to the prejudice of the establishment.
3. They continually remind readers of earlier
historical suppression of truth.
4. They publish out$ide conventional channels.
5. They believe themselves to be geniuses,
and those who oppose them, idiots.
6. Everyone is out of step but them; and they
proclaim their dedication to truth, and their absolute assurance that they have found ito
7. They believe that any opposition is the
result of a "dastardly plot", a conspiracy, if you
will, of the establishment o
8. Their attacks are aimed at the highest
persons and the best-established theories.
9. And finally. they often invent their own
jargon.
Gardner makes it clear that no one crank is expected to fit all these characteristics. Certainly
Mr. Sprague does not. He does not claim genius
status for himself, nor invent any private jargon.
And his target, the Warren Commission report, is
not sacrosanct.
But he does appear to believe absolutely in the
"truth" he has discovered; and that there is a vast
conspiracy opposing him (points 6, 7). Not only
does the NCTIA not work with the FBI and other criminal investigation organizations. it classes them
with the conspiracy (points 1 and 2). Sprague believes that all the news media in the country except
one newspaper gave a "very distorted" report of Clay
Shaw's trial in New Orleans (po 30). All but the
one were out of step (point 6).
In addition, Mr. Sprague has not succeeded in
getting a hearing in the normal channels and has
aired his view in a journal whose mainstream interests are entirely foreign to his subject matter
(point 4). He blames this lack of access to public
eye on the conspiracy of the publishers. (point 2)~
rather than the quality of his manuscript. (On thls
point, the reader must be his own judge.) He also
inserts an entirely irrelevant comment (p. 30), reminding the readers that (over 60 years ago) the
press was reluctant to believe and publicize the
accomplishments of the Wright Brothers (point 3).
The reader may find other matching characteristics.

Conclusion

Are those enough reasons to dismiss Mr. Sprague
and the NCTIA as cranks (in the criminology field,
at least)? I think SOi especially when they are
combined with inaccurate photointerpreting. But
each reader must decide for himself. And I believe
Mr. Sprague will agree with me on this one final
point, if on nothing else. Each of you shoulQ decide for yourself.

3In common, though slightly inaccurate parlance i
this is called "arrest".

2. Proof of Proposition I

Response
Edmund C. Berkelev
Editor, Computers and Automation

Outl ine
1. Two Propositions
2. Proof of Proposition I: "Oswald was not the sole
assassin and there was a conspiracy."
3. Evidence for Proposition II: Sprague's alternate
hypothesis
4. The Dimensions of an Article in "Computers and
Automation"
5, The Computer-Relatedness of Sprague's article
6, The Case of Jim Hicks
7. The Decision of the New Orleans Jury in the First
Trial of Clay Shaw
8, The Chart of Times of Events
9. Some Other Specific Points
10. Clues for Identifying Quacks and Cranks
II, Mental Resistance to New Ideas
12. Defects in the Warren Commission Report
1. Two Propositions

There are two completely different propositions
(or groups of propositions) put forward in the May
1970 article by Sprague (see page 30 of the May
issue).
One is a very narrow proposition, which I shall
call here Proposition I:
The conclusion of the Warren Commission [that
Oswald was the sole assassin and that there
was no conspiracyJ cannot be considered true
by any person who carefully considers the
crucial evidence -- such as the physics of the
shooting. the timing of a number of events,
and other important and undeniable facts. In
other words. Oswald was not the sole assassin,
and there was a conspiracy [page 30J.
Only in regard to this Proposition I does the article assert:
This article will develop that thesis. prove
it to be true on the basis of substantial.
conclusive evidence, and in particular some
analysis of the photographic evidence [page
30J.
The other proposition (or group of propositions)
I shall call it here Proposition II -- relates
to a number of topics: (1) the number of persons
in the first conspiracy ("more than 50 persons were
involved in the conspiracy at the time of the firing
of the shots"); (2) the number of shots fired
(Sprague maintains that 6 shots were fired. while
another of the authorities, Josiah Thompson. a professor of philosophy at Haverford College, author
of "Six Seconds in Dallas", maintains that four
shots were fired); (3) the second conspiracy to cover
up the first one; (4) arrests, with no record of
persons arrested; etc.
In the May article by Sprague, evidence is offered for Proposition II. But no assertion is made
that Propo si tion II is demonstrated. Instead, in many
places there are queries and questions about wh?t
the details might be for the alternate hypotheSIS.
40

Proposition I is that:
Oswald was not the sole assassin, and there
was a conspiracy.
The proof is extremely easy, because there are
many undeniable pieces of evidence, anyone of which
if firmly established is sufficient. (In the same
way it is easy to prove the negative theorem that
not all odd numbers are exactly divisible by 7, by
exhibiting a single odd number (such as 11) which
is not exactly divisible by 7.)
Here is what seems to me the simplest proof that
I am aware of:
1. It was physically impossible for one man,
shooting during the time the target (President
Kennedy) was in line of fire from the 6th
floor easternmost window of the Texas School
Book Depository Building, to have fired all
the shots (at least four, and probably five or
six) for which there is overwhelming physical
evidence.
2. Therefore, since two or more men were firing in precisely the same interval of approximately 6 seconds. there was a conspiracy to
assassinate Kennedy, and Oswald (it is entirely conceivable that he might have done
some of the shooting) was however not the sole
assassin.
Of all the proof which I have seen or studied,
the most spectacular and convincing to me is the
explosion of President Kennedy's head, to the left
and backwards at the time of the fatal shot, shown
in the Zapruder film at frame 313 (see p. 35 of the
May article by Sprague). I studied physics in preparatory school and again in college, and I know
that when an object is hit hard from a given direc~
tion, that object has a tendency to move in the direction from which the hit comes, because of the law
of momentum. It is impossible for any shot from the
direction of the Texas School Book Depository Building to have driven the President's head backward and
to the left as revealed by the sequence of frames
Z312, Z313, Z314.
The Warren Commission saw the Zapruder film, and
looked at it at least several times. For them to
have accepted the conclusion that Oswald was the
sole assassin and that there was no conspiracy demonstrates (from my point of view) that they cheated.
It also demonstrates a second conspiracy.
For LIFE magazine to continue to refuse to show
the original, clear Zapruder film year after year
since 1963, to my mind, demonstrates the enormous,
continuing power of the second conspiracy
the
slogan of which is "national security."
, I watched a bootleg copy of the Zap ruder film on
May 1, 1970; I watched it played over and over; run
forwards, run backwards, and stopped, at various
frames for study. The copy, though not very g'ood,
showed plainly the explosion of Kennedy's head when
struck by the fatal bullet. "Computers and Automation" can make arrangem~nts to show a bootleg
copy of the Zapruder film to audiences almost ariywhere in the United States and Canada. If any
reader is interested, he should write Computers and
Automation, Dept. F, 815 Washington St o , Newtonville, Mass., 02160.
COMPUTERS and AUTOMATION for March, 1971

3. Evidence for Proposition II:
Sprague's Alternate Hypothesis

A large number of statements in the May article
are included in Proposition II. But Proposition II
is in the nature of an alternative hypothesis, advanced by Sprague.
It is evident that many of the details of any alternative hypothesis may and probably will differ
from one investigator to another, as each one reflects on the evidence he is familiar with and as he
attempts to reconstruct what he thinks probably
happened.
In regard to Proposition II, the May article does
not assert proof; the article expresses simply the
normal position of an author of any article: "Here
is my view of the situation; and here are some of
the reasons which support my view." And my function
as editor is essentially to make sure that the author states his view of the situation in a clear and
fairly reasonable way, and offers clear and fairly
reasonably supporting evidence for his view.
Among the supporting evidence are these facts:
that Sprague has over six years studied the photographic evidence and other evidence; has read many
of the references carefully; has put together from
primary evidence a number of strands of secondary
evidence; that he personally has made many trips to
Dallas and elsewhere to see the scenes and to interview witnesses and photographers; that he has in his
files copies of 200 of the photographs; that he has
written descriptions of 200 more photographs which
he did not secure copies of; that he has failed to
see only some 100 of the photographs -- or 20%; that
he attended many of the days of the New Orleans trial
of Clay Shaw; etc.
This background makes Sprague a well-qualified
author to put forward Proposition II; but that proposition in the May article is not asserted as "true
on the basis of substantial, conclusive evidence,"
as a careful reading of the article will plainly
show.
4. The Dimensions of an Article
in "Computers and Automation"

One of the charges made by Dr. Schwartz relates
to not giving evidence or supporting information for
some of the statements included in Proposition II -for example, a statement that Oswald did not shoot
Officer Tippet. The charge is that the statement
was made, and not supported by any details.
An article is not the same as a book or a treatise;
it is much shorter. Inevitably, an article will not
have room to contain for every controversial statement the data, analysis and reasoning which may constitute reasonable support of a statement. Furthermore, it is not necessary in an article to deal with
all the background of each statement made.
Four significant and important references are
on page 35 of the May article, and a commentary is given for each of these. In addition a
bibliography containing references to 18 more books
is given on page 60.
ci~ed

5. The Computer-Relatedness
of Sprague's Article

The charge is made that the material is not comCOMPUTERS and AUTOMATION for March, 1971

puter-related and should not have appeared in "Computers and Automation"; and that this non computerrelated article did not appear in the "House is on
Fire" department of "Computers and Automation"o
It is simply not true that the article is not
computer-related. Part 3 of the article (see pages
56 to 58) discusses at length a number of ways in
which computers should be and could be applied to
the massive problems of information handling that
are involved in dealing with over 25,000 pictorial
frames recording various aspects of an important
historical event.
The title of the article was:
The Assassination of President Kennedy: the
Application of Computers to the Photographic
Evidence
With the benefit of hindsight, I now believe that it
would have been better (in the sense that there
would have been less misquotation of the article) if
the title had actually been:.
The Assassination of President Kennedy: the
Prospective Application of Computers to the
Photographic Evidence
However, it is right, proper, and adequate to say
"the application of" when one means "the prospective
application of". For example, for many years articles on "the application of computers to the game of
chess necessarily had to mean "the prospective application". For only in the last few years have
computers had reasonably good computer programs for
playing chess.
As editor, if I had the same decision to make
again, I would again have included this article in
our magazine. I would have known now some of the
minor corrections which have been been published in
later issues of "Computers and Automation" -- of
which the most unfortunate perhaps is in the spatial
chart of Dealey Plaza; the numbers in the scale of
the chart, namely 10, 20, 30, 40, should be replaced
by 20, 40, 60, 80 respectively.
It is certainly not necessary that a computer related article be published in the C&A department
"The House is on Fire".
6. The Case of Jim Hicks

In regard to Jim Hicks, several charges are made:
"The S-shaped loop of the antenna is not visible in
the picture as published in 'Computers and Automation' in May." (See page 33). This is true. The
picture that went to the printer showed the antenna,
faintly; the picture as printed in the magazine did
not show it. This point is discussed in the July
1970 issue of "Computers and Automation" on page 32.
The charge is made that "It was impossible for
Jones Harris to recognize Jim Hicks from the back
when Harris was in District Attorney Jim Garrison's
office in New Orleans." This charge is ridiculous.
The event actually happened o I have talked to Jones
Harris in New York in recent weeks and he confirmed
that this did happen.
The charge is made: "If Mr. Hicks is now confined to a mental institution, it may well be that
Hicks is a mental case o " This supposition is undeniable. However, from 1963 to early 1969 when he
was in District Attorney Jim Garrison's office in
New Orleans, Hicks was not so confined, and it is
41

quite plausible that during that time he was sane o
It is even likely that he did what he said he did 9
establish a radio communications center in the Adelphi Hotel in Dallas so that radio communication among
the firing teams could take placeo
7. The Decision of the New Orleans Jury
in the First Trial of Clay Shaw

In the first trial of Clay Shaw in New Orleans in
February and March of 1969 the jury found Clay Shaw
"not guilty" of participation in the conspiracy to
assassinate President Kennedyo The May article (see
page 31) says:.
The record of the trial as published in The
Times Picayune contains many indications that:
10 Clay Shaw did know and meet with Lee
Harvey Oswald (dead)9 David Ferrie (dead)9
and Jack Ruby (dead)9 and exchange money
with them o Twelve witnesses saw them together in twos and threes at various times
and pI aces

der film (roughly ten seconds after the last
shot) up to the first of Murray's photos
(about 3 minutes) is covered by several movies
and TV sequences.
The charge is made that "The precise method of time
determination is not stated, and the reade~ is asked
to accept 5.7 seconds before and 3.5 seconds after".
Wel1 9 how detailed can you get in a magazine article of limited length?
If one has over 25,000 frames and 350 still pictures, and persons and cars moving through a scene
with 75 photographers; if one coordinates all that
one can find out, with the positions of the tires
of the Kennedy limousine in the motorcade, the Zapruder frames~ and the time scale figured out by the
Warren Commission and the FBI; etc. etc. etc o -one should be able to go quite a way in establishing
times. Actually, a detailed description of the
method of computation of the 3.5 and 5.7 second times
exists, and is available on request from Sprague.

20 There were at least three gunmen in
Dealey Plaza firing at President Kennedy on
November 229 1963~ from at least two directions v and therefore there was a conspiracy
What Garrisoi failed to prove to the satisfaction of the jury was that Clay Shaw was involved in the conspiracyo
The charge is made that "The decision made by the
jury in New Orleans was no conspiracy." This is not
the decision that was made by the jurY9 and the
charge is false.
The jury was not asked nor empowered to give any
verdict on the question of a conspiracy to shoot
President Kennedyo The jury simply decided, on the
basis of evidence put before them~ that it was not
proved that Clay Shaw participated in such a conspiracY9 whether or not such a conspiracy existed.
8. The Chart of Times of Events

The May article contains on page 51 Chart 2, a
schematic timing chart of photographs 9 movies, and
events in Dealey Plaza around 12:30 pm on November
22, 1963. On page 57 appears the following text.
The FBI and the Warren Commission staff made a
careful analysis of the timing of the frame s
in the Zapruder filmv and correlated these
times with the positions of the presidential
limousine and the other cars in the motorcade.
(See Chart I). Consequently, there exists an
accepted time scale at 1/18 second intervals
which applies to the events in the photographs: the scale consists of eighteenths of
a second (named with the frame numbers of the
Zapruder film) during the crucial 6.8 seconds
of the fatal interval. I was able to extend
this _~eference technique by using four other
movies stretched end to end to cover the period
from the moment Kennedy rounded the corner of
Houston and Main Sts. up to the beginning of
the Zapruder film.
A preliminary correlation of the times in the
various movie sequences and still photographs
is shown in Chart 2 [not "Table 1" -- the correction was published in the June issue of
"Computers and Automation", page 22J for the
period of Kennedyts travel through Dealey Plaza ••••• The period from the end of the Zapru42

I think the charge is an example of the fallacy
of pettyfogging.
9. Some Other Specific Points

The charge is made that "the second photo of the
6th floor easternmost window of the TSDB shows that
'THE BOX HAS MOVED'."
Actually, the second photo is taken from a different angle and a different distance; the box has
not moved, as may be shown by more photos besides.
The charge is made that "the four and a half
pages of pictures of the tramps and the phoney
policemen prove nothing about the supposed conspiracy."
Actually they do prove something significant:
the presence at the scene of persons who have not
been accounted for, and who therefore might have
been involved in the conspiracy. In regard to the
"phoney policeman", the May article says (page 39)
"There is no record of this man being a member of
the Dallas police force. He is the only Dallas
policeman of those appearing in the hundreds of
photographs taken Nov. 22 who is wearing an earpiece or radio communication device in his right
ear (see Figure 2). Why?"
I consider it remarkable that the one man in a
Dallas police uniform who cannot be identified as a
Dallas policeman is the one who is equipped with a
device which could be a communicating device. (How
many policemen on active duty as police have to
wear hearing aids?)
The charge is made that it not unreasonable to
detain persons, question them, and then release them
without making any record whatever -- even on an
occasion when the President of the United States has
been assassinated.
Well, that is not the way it seems to me. It
seems to me remarkably careless not to make a record
of the names and other information of such persons
on such an occasion -- so careless that there might
well have been some intention not to.
The charge is made that "the umbrella man is
hypothetical". This is perhaps a friendly word for
"imaginary," or "fictitious," or "nonexistent".
Actually, there are at least half a dozen picCOMPUTERS and AUTOMATION for March, 1971

•

tures showing the umbrella man (see the account on
page 50 of the May article). And there is a good
picture of all of him (with his umbrella folded) on
page 228 of "Six Seconds in Dallas" by Josiah Thompson.
10. Clues for Identifying Quacks and Cranks

Nine clues listed by Martin Gardner in his book
"Fads and Fallacies in the Name of Science" are
present~d by Schwartz. The charge is made that
Sprague and his article, (and the act of "Computers
and Automation" in publishing it) are to be classified as the behavior of quacks and cranks, by reason
of satisfying (assertedly) 7 of the 9 clues.
This is an argument from a generalization (not a
universal statement) and is. not logically sound for
every case. Over and over again a generalization,
however suggestive it may be, fails because it does
not conform with all the cases.

Not one word by Dr. Schwartz admits even that the
subject is important and worth investigating -whereas any scientist must always admit that an
important subject in which evidence is offered for
judgment is worth investigation.
Sprague's article is like an iceberg with a
little tip above water. Not once does Dr. Schwartz
say "I'd like to look at more of your evidence."
Actually, McLean, Va., where Dr. Schwartz lives, is
within a few miles of the openly accessible, large
central collection of the evidence in the case, at
the headquarters of the National Committee to investigate Assassinations, 927 15 St., N.W., Washington,
D.C. 20005.
12. Defects in the Warren Commission Report

Dr. Schwartz says "I agree that the Warren Commission Report is seriously defective". But Dr.
Schwartz does not say how in detail.

For example, one of the strongest possible generalizations (on the basis of our experience with
all the days that men have lived) is that "the sun
will rise tomorrow". Yet in the far distant future
(some ten billion years from now) it is predicted
by astronomers and physicists that the sun will explode, having exhausted its supply of hydrogen for
its nuclear furnace, and when that happens the entire earth will be vaporized in the solar nuclear
explosion. Then there will no longer be a solid
earth to turn on its axis, and therefore there will
be no more tomorrows.

Actually, the Warren Commission Report is so defective that it is flatly and fatally wrong:

11. Mental Resistance to New Ideas

The clear and obvious reason is that they did not
want to get to the bottom. And the clear and obvious reason for that lack of desire is that there
were indications that the Central Intelligence Agency was involved in Kennedy's assassination and probably the FBI was also involved -- and so "national
security" was at stake. Furthermore, the Commissioners did not know what the people of the country
would do if they were officially and soon told "the
Central Intelligence Agency had a hand in killing
President Kennedy" and "The FBI knew about the plot
ahead of time and issued no warnings". And so the
Commissioners drew back from that chasm that yawned
in front of them.

Facts are stubborn things. Once a person accepts
the demonstrated fact that "Oswald was not the sole
assassin; there was more than one person firing; and
therefore there was a conspiracyv" a foundation stone
is loosened, and the theory that the Warren Commission sought to establish comes tumbling down, like a
collapsing building.
A great deal of human progress occurs because of
stubborn facts; and often in human history persecution and death has been the reward for those who
insisted on them.
To balance the quotation from Martin Gardner, I
would like to quote from "The Art of Scientific Investigation" by W. 1. B. Beveridge published by
Mercury Books, London, 1950:
The mental resistance to new ideas is partly
due to the fact that they have to displace
established ideas. New facts are not usually
accepted unless they can be correlated with
the existing body of knowledge; it is often
not sufficient that they can be demonstrated
on independent evidence. Therefore premature
discoveries are usually neglected and lost.
An unreasoning, instinctive mental resistance
to novelty is the real basis of excessive
scepticism and conservatism.
In my opinion, Dr. Schwartz displays (1) excessive skepticism, (2) extraordinary resistance to
new knowledge, (3) a profound ignorance, and (4)
an extraordinary cockiness since he is aware of
his ignorance.
Not one word in his polemical attack on the May
article by Sprague acknowledges the obviously extensive research and investigation and the accumulation of evidence which Sprague (and various other
authors and researchers) have amassed.
COMPUTERS and AUTOMATION for March, 1971

- Oswald was not the sole assassin;
- there was more than one gunman firing;
- there was a conspiracy.
Furthermore, the Commissioners had the money,
the resources, and the subpoena power to get to the
bottom of the plot. They clearly did not choose to
get to the bottom (at least a majority did not so
choose).

I do not blame the Commissioners for not wanting
to get to the bottom of the plot -- they may well
have saved their own lives that way. However, their
report was a betrayal of the people of the United
States.
I have not tried in this response to deal with
every single one of the points raised by Dr. Schwartz
in his thorough and well-organized,{but most one-sided
attack) on the article: "The Assassination of President Kennedy: The Application of Computers to the
Photographic Evidence" by Richard E. Sprague publIshed in our May, 1970 issue. (Footnote: I believe
I could deal with all of his points!)
However, I do agree with Dr. Schwartz: The best
procedure is to recommend to each reader who is interested that he read Sprague's article and consider
it for himself.
If you are interested and cannot find an easily
accessible copy of the May, 1970, issue, please order
it from us (see the notice on page 44). Then read
the article, return the issue to us in 7 days in
salable condition, and you will owe us nothing. In
this way, you can judge for yourself.

THE MA Y ARTICLE
THE ASSASSINATION OF PRES IDENT JOHN F. KENNEDY:
THE APPLICATION OF COMPUTERS
TO THE PHOTOGRAPHIC EVIDENCE
Computers and Automation, published in its May
issue a 32-page feature article, "The Assassination
of President Kennedy: the Application of Computers
to the Photographic Evidence"o
In this article, Richard E. Sprague, President,
Personal Data Services, Hartsdale, N.Y., states
that analysis of the evidence proves:
•

that the Warren Commission conclusions
(that Lee Harvey Oswald was the sole assassin, and that there was no conspiracy)
are false;

•

that there were at least four gunmen firing
from four locations, none of whom was
Oswald;
that the conspiracy to kill Kennedy involved over 50 persons (of whom several are
identified in the article) including members of the Dallas police, and elements of
the Central Intelligence Agency of the
United States; etc.

The evidence published in this article includes
eleven important photographs. One of them shows
Jim Hicks, who admitted he was the radio communicator among the firing teams at Dealey Plaza, with
his radio transmitter in his back left pocket. The
article includes a tabulation of over 500 photographs (counting a movie sequence as one photo)
taken in and around Dealey Plaza, Dallas, Texas,
Nov. 22, 1963, at the time of President Kennedy's
assassination and shortly thereafter. Both a spatial chart and a timing chart of the events and the
photographs are included in this article.
Sprague, a computer professional for over 24
years, has as an avocation, studied the old and the
new evidence for over 6 years, and has analyzed
over 400 of the 500 photographs.
The work in computerized analysis of over 300
still photos and over 25,000 frames of movie sequences has been started.
To obtain your copy of this extraordinary report,
please complete and mail the following order:
- -(may be copied on any piece of paper)To:

Computers and Automation, Dept. P
815 Washington St.
Newtonville, Mass. 02160

Please send me ( ) copy(ies) of the May 1970
issue containing the article on President
Kennedy's assassination. I enclose $4.00
( ,) check ( ) money order for each copy.
(Please do not send cash.) If not satisfactory, returnable in 7 days for full refund (if
in salable condition),
My name and address are attached.
Discount:

10%, 5 to 9 copies;
20%, 10 or more copies
Special price: $1.00 for students (attach evidence);
$1.00 for non-profit organizations

An excerpt from the Nay article

Part 2. The Photographic Evidence

The assassination of President John F. Kennedy
was the most photographed murder in history. Approximately 75 photographers took a total of approximately 510 photographs, either before or du~ing or
within an hour after the events in Dealey Plaza, and
either there or nearby or related to those events.
The word "photograph" in this context includes both
still photos and movie sequences. The number of
frames in a movie sequence ranges from about 10 to
about 500; and in the count of 510 photographs given
above, the 10 to 500 frames of a single movie sequence are counted just as one photograph. The total number of frames is over 25,000.
The Warren Commission examined 26 photographs,
about 5 percent of the 510. The FBI examined about
50 photographs, or about 10 percent. The most famous of all the photographs is the Zapruder film,
which had over 480 frames.
Many of the photographs were taken by professional photographers. About 30 of the photographers
were professionals who worked for newspapers, television networks, and photographic agencies.
The Warren Commission did not interview a single
one of the professional photographers, nor did the
Warren Commission see any of their photographs.
Fifteen of these professionals were actually in
the Kennedy motorcade, no further than 6 car lengths
behind the Kennedy car. Five of these photographers were television network cameramen. The Warren
Commission looked at none of their photographs.
Two of the photographers were from the White
House. One of these men (Thomas Atkins) was the
regular photographer for the White House. He made
a special film for Lyndon B. Johnson. Atkins used
his own film plus some footage obtained from the
television photographers. Johnson looked at the
film and then put it away. This film is now stored
with the Kennedy Memorial Library materials in a
warehouse in Washington, D.C.; it is stated to be
"unavailable" to researchers. The Commission did
not see this film, nor did they interview Atkins.
Because the professionals used movie cameras of
professional quality, their films are exceedingly
revealing and valuable as primary evidence. The
Warren Commission looked at none of these films.
Chart 2 of this article shows the times of about
50 of the photographs taken in Dealey Plaza during
Kennedy's passage through it.
Table 3 of this article lists over 510 photographs so far identified and known to exist or to
have existed -- with possibly a few borderline cases.

COMPUTERS and AUTOMATION for March, 1971

District Attorney Jim Garrison on
The Assassination of President Kennedy:
A Review of Heritage of Stone

Neil Macdonald
Assistant Editor
Computers and Automation

This book is important and interesting.
The
words used, the narrative thread, reveals an
entirely sane but deeply angry man, and a remarkably courageous one, who is appalled at the assassination of political leaders in the United States.
Whether or not you believe that District Attorney
Jim Garrison of New Orleans, La., is a publicity
seeking charl?tan and fool as the Establishment
press has portrayed him, or an honest and brave
investigator as the press has NOT portrayed him,
this book is very significant, well and clearly
written, and filled with facts,
But there is no
substitute for reading the book for yourself
which this reviewer highly recommends.
Interestingly enough, there is no mention in the
book per se of the defendant Clay Shaw nor mention
of his trial by Garrison in New Orleans in February
and March, 1969 -- except that three footnotes do
mention ,Clay Shaw. Presumably this is because the
second trial of Clay Shaw (for perjury in denying
that he knew certain persons) which began in January 1971 would be prejudiced by the inclusion of
such material in Garrison's book. The book mainly
discusses persons such as David Ferrie (dead) who
knew Lee Harvey Oswald, and other persons who knew
Oswald, while he lived in New Orleans.
Beginning of Garrison's Interest

On one occasion, in the autumn of 1966 Garrison
was in New York talking to Senator Russell Long of
Louisiana, and the talk turned to the assassination
of Presiden_~ Kennedy.
Garri son reports he was astonished to;hearLong say that he felt there was a
question about the Warren Commission's inquiry.
Up to then Garrison had assumed that the question
had been "fully and honestly" looked into by the
Warren Commission.
After Long's remark, when Garrison went back to
New Orleans, he began to read the 26 volumes of the
Warren'Commission'sHearings and Evidence, and also
the Commission's conclusions.
It became clear to Garrison that the official
conclusion that Kennedy had been killed by a single
man with a rifle shooting at him from behind was
totally impossible.
It even began to seem to
Garrison that Lee Oswald quite possibly had not
fired any shots, and had been a mere scapegoat.
So what was the meaning of these 26 volumes of
evidence?
Why was the government lying to the
people?
Who had killed President Kennedy?
and
why?
These questions took Garrison down a long road,
and involved him in a great collision with what he
called an "enormous domestic intelligence organization which would seek to discredit and destroy anyone who dared to challeng~ its authority."
COMPUTERS and AUTOMATION for March, 1971

Heritage of Stolle / Jim Garrison / G. P. Putnam's Sons,
200 Madison Ave., New York, N.Y. 10016/ 1970,
hardbound, 253 pp, $6.95

He inventories a long list of efforts and successes by many parts of the federal government,
both official and unofficial, to impede, prevent,
and discredit his investigation, deny him subpoenaed evidence, etc.
Some Evidence that the Warren Commission
Conclusions are False

Referring to the end of the day of President
Kennedy's assassination, November 22, 1963,
Garrison writes: "The leaders of our government
knew that President Kennedy was not killed by Lee
Oswald."
Garrison cites as evidence:
- the testimony of the overwhelming majority
of the witnesses at the scene;
- what the doctors in Parkland Hospital in
Dallas observed and reported;
- what the color movie film taken by Abraham
Zapruder showed.
This evidence, says Garrison, makes it "perfectly
clear" that the shot that killed Kennedy and'blew
his brains out came from the front, and not from
the rear where supposedly Oswald was firing from
the 6th floor of the Texas School Book Depository
Building.
In another place Garrison points out that Dr.
Pierre Finck (one of the doctors) testified that
more lead was removed from Texas Governor John B.
Connolly's wrist than was missing from the bullet
that the Warren Commission asserted had gone
through the President and the Governor.
Garrison
says that this testimony "virtually destroys the
official conclusions of the Warren Commission".
This is the same John B. Connolly, Jr., who has
just been appointed by President Nixon to be
Secretary of the Treasury of the United States
a man who knows from personal experience and other
information that the Warren Commission conclusions
are false; a man who has never publicly stated
"President Kennedy was assassinated by a conspiracy."
Why Was Kennedy Assassinated?

Garrison writes vigorously and well, even if
not scientifically.
He discusses the reasons why
Kennedy was assassinated.
He advances the statement: "A man who cares too much for the human race
may find himself living in a hostile environment.
His humanity may not be regarded as dangerous so
long as his voice cannot be heard by too many
people but if he is eloquent, or if he is in a position to affect the affairs of the nation, then his
45

humanity will be regarded by some men as a great
threat".
He comments that: "All superstates engaged in
efforts to gain power must maintain extensive
intelligence efforts at home.
They must seek to
maintain control of individuals and ideas lest
their international war adventures lose the support
of the populace at home •... The issue is power,
immense power
"After the United States ascended to the position
of the most powerful military nation in history, in
the midst of its accumulation of the most effective
death machinery of all time, there occurred the accident of the election of a President who regarded
the entire human race with compassion.
By the
time this happened, the cold war had become our major
industry, and the Central Intelligence Agency had
become the clandestine arm of our military-industrial
complex, and, in the process, the most effective
assassination machine in the world."
Unfortunately, Garrison nowhere in his book
raises the question of what the Central Intelligence
Agency may be doing outside of the United States,
nor the question of the vast (and wicked even if
thoughtless) assent given by the American people
and by the United States Congr-ess to the assignment
of the CIA to commit assassinations and carry out
other violent, corrupt, hidden, etc., interference
with other nations and other governments outside of
the United States
all actions financed by an
authorized budget of billions of dollars a year.
The United States of course has no moral right whatever to engage in such practices in the rest of the
world.
Diligent vs. Careless Investigation

Garrison comments: "When an assassination of a
national leader is not supported by elements of the
government, it is predictable that the government
investigation will be effective and relentless.
... All information contributing to the discovery
of the whole truth will be welcome .•. When the
criminals are caught, the machinery of justice will
be firm and uncompromising.
" •.. However it is another matter when an assassination is supported by powerful forces within the
government.
The ••. protective guard of the
President suddenly will have become curiously impotent, for its operation will be known intimately by
the assassins.
The assassination apparatus will
be extraordinarily effective.
Federal investigative agents ... will move like sleepwalkers.
High
officials reviewing the affair will diligently examine many irrelevant items" such as Lee Oswald's
record of a smallpox vaccination in 1951 - "but
will casually overlook the most pertinent evidence
relevant to the assassination."
The Cover Story

Garrison says: "In a country with advanced
technology for news distribution, the removal of a
nation's leader will never be attempted unless
those sponsoring the murder feel assured that they
will have an effective degree of control over the
dissemination of the news.
Government control
must be at a high enough level to guarantee the
subsequent distribution of official news releases
encouraging the belief, that however tragic the
accident, it was essentially meaningless and all

is well •... Creation of a believable cover for
an assassination is routine for an intelligence
agency of a major government.
The cover story
which is initially distributed by the press release
creates a degree of acceptance virtually impossible
to dislodge.
This is the case especially when
the official fiction is supported by the prearranged activities of a decoy pointing in the
direction of a false sponsor of the assassination.
The actual events of the assassination become irrelevant.
All that remains relevant is the
cover story issued to the press and the power to
control the investigation and conceal the evidence.
Understanding of the Forces

"We must begin to recognize history as it is
happening to us", says Garrison.
"We can no longer
toy with illusions.
Our war adventures in Asia
are not related to national security in any rational
sense .•.• "
To understand the forces involved
[in the murder of Jack Kennedy] and their motivation is to understand all of the once-mysterious
assassinations of the 1960's, which in each instance
achieved the elimination of a public figure who
opposed our massive military expedition into Asia:
If

The Link of Assassination to War

Garrison remarks that: "A new political instrument
has been created.
It provides for the permanent
removal of men whose philosophies do not coincide
with that of the dominant power structure of the
United States. .••
Justice is not so blind that
it pursues the most powerful forces in the country.
Nor is the press so. committed to truth that it
wants the burden of knowledge of what is happening.
Sooner or later the relationship of assassination at home and war abroad must come to be understood. ..•
I have written this book so that the
truth about the murder of John Kennedy finally may
be brought out for every American to see.
• .• I
have sought to show what has been done to our
country by men who believe in solving problems by
the use of force •••• I wrote [this book] in the
hope that it might illuminate the peril which
surrounds us. Welcome to the fight.
Nature and Quality of the Book

The contents consists of: acknowledgments; Foreword; One Friday in November; Part 1, Illusion, with
five chapters, The Execution, Ornaments, Power,
the Quarry, Justice; Part 2, Reality, with five
chapters, The Craft of Deception, Traces of Intrigue,
The Ides of November, Nightfall, The War Machine;
'an Appendix, John Kennedy and Nuclear Militarism;
Notes; and Index.
This is a scholarly book: over 300 notes are
given in pages 233 to 244; and over 1500 entries appear in the index pages 245 to 253.
Thus a great
deal of evidence is carefully cited for many details.
But it is not a scientific book, in the sense that
substantial evidence is offered for hypotheses of
great generality.
Errata noticed by this reviewer: p. 95, line 6:
replace "Xaxier" by Xavier"; p. 104, line 16: replace
"Accessibility" by "Access".
This book is worth reading, rereading, and studying, and the knowledge in it should be applied by
every American.

COMPUTERS and AUTOMATION for March, 1971

ACROSS THE EDITOR'S DESK
APPLICATIONS
SEARCH FOR UNDERGROUND OIL
AIDED BY COMPUTER-STORED
HISTORY OF OIL DRILLING
ACTIVITY

Hi stories of 700,000 oil well s
drilled in the United States since
the 1890s are stored on 2,600 reel s
of magnetic tape in an unusual library at Petroleum Information,
Inc., Denver, Colo.
The firm is
using an IBM computing system to
keep track of almost every important well ever drilled in thi s
country, as an aid to geologi sts
faced with increasingly difficul t
drilling decisions.
Bill Goodin, executive vice president of the firm (which gathers
and distributes information to the
petroleum industry), said that 2.5
million oil well s have been drilled
in the United States.
Many wells
have been drilled in each of the
major petroleum basins. It is estimated that 60% of the area under
thi s country has sedimentary deposits of rock and sand that could
c~ntain oil.
Geologists seeking
oil can receive reports on an area's
hi story to aid in determining where
to drill.
The oil well hi story file was
started on an IBM System/360 Model
40 and now contains information on
700,000 well s. Each day new reports
are added to the file (each year
32,000 newwells are drilled). The
company soon will transfer the information to an IBM System/370 Model
155.
Information for the file, as
well as for the company's other
reports and publications, is gathered through the cooperation of
maj or petroleum producers, individual smaller operators, and state
and federal governmental agencies.

UNITED AIRLINES IS USING
COMPUTER SYSTEM TO CURB
STOLEN TICKET USE

Uni ted Air Lines has installed
a special purpose computer system,
developed by TRW Data Systems, Torrance, Calif., to check for stolen
and counterfeit airline tickets at
Los Angeles International Airport.
Capt. Frank Beeson of the Los Angeles Pol ice Department,
described
Los Angeles as the nation's bl ack
market capitol for peddling stolen
airline tickets.
He said United
Air Lines is the first major carrier
to take a substanti al positive step
COMPUTERS and AUTOMATION for March, 1971

in helping to combat the serious
and increasing use of stolen airline tickets.
The TRW system utilizes 23 keyboard terminals, 17 located at
boarding gates and 6 at the ticket
counters of United's two airport
satellite buildings. As passengers
check in, the ticket number is put
into the computer through the touchtone terminal by United counter or
check in personnel. An instantaneous reply from the computer with
a green light indicates the ticket
is valid; a red light indicates
that the ticket may be stolen.
Numbers of tickets stolen anywhere
in the country can be entered into
the system.
It is estimated that there is
more than $90 million outstanding
in stolen airline tickets in the
Los Angeles area. Los Angeles International Airport has the highest
rate in the country in the use of
stolen tickets.
COMPUTER-AIDED TECHNIQUE
TECHNIQUE FOR GROWING
SINGLE-CRYSTAL SILICON INGOTS

Computers have been harnessed
wi th crystal growing furnaces to
produce low dislocation - density
single-crystal sil icon ingot s, such
as the one shown at the left in the
picture.
Such ingots, 2~ inches
and larger in diameter, are sliced

and pol i shed to form the sil icon
wafers which are then processed in
a series of complex operations to
produce monolithic logic and memory
chips. These chips are then packaged, as shown, to form the circuitry for IBM computers.
The technique utilizes an IBM
1800 Data Acqui si tion and Control
System to monitor and control cri tical process parameters such as
silicon melt termperature, crystal
lift and rotation speeds, and crucible 1 ift and rotation speeds.
The Teal-Little modification of
the Czochralski process is used to
produce the ingots. Computer-:aided
techniques minimize thermal disturbances during the crystal grow-

ing process, which can result from
comparatively gross adjustments by
human operators. Consequently, the
computer enables a more consistent
production of low dislocation-density single crystals.
The technique was described by
Kenneth E. Domey (of the IBM Components Division East Fishkill Facility, Hopewell Junction, New York)
at the Semiconductor/IC Processing
and Production Conference and Exhibi tion which was hel d 1 ast month
in Anaheim, Calif., in a presentation titled "Computer Controlled
Growth of Single-Crystal Ingots."
MODERN MARINERS AIDED
BY OCEAN-GOING MINICOMPUTERS
AND ORBITING SATELLITES

A small ocean-going computer is
the heart of a satell ite navigation
shipboard system which can pinpoint
the position of a ship anywhere in
the world, regardless of the weather,
time of day or distance from land.
The modern mariner no longer has
to climb on the deck and shoot stars
wi th hi s sextant to determine hi s
posi tion; he can now reI ax in a
heated cabin and watch a teletype
automatically print out the latitude
and longi tude to wi thin 150 feet of
his actual position.
The Transi t Navigation System
(developed originally for Navy submarine use, but since made available to non-mili tary users) couples
a Hewlett-Packard minicomputer wi th
a speci al ized communication recei ver.
The system automatically digests information transmitted from
orbiting satellites and prints out
a ship's longitude and latitude on
a teletype.
More than 60 shipboard navigation systems buil t and programmed
by Magnavox are now in use; most
are controlled by HP minicomputers.
Because of their compactness, entire
systems are frequently transferred
between ships. About half the systems are aboard oceanographic research or support ships. The other
half are'used by navigators of oil
exploration ships, commercial transports, passenger ships, cable laying ships and Navy and Coast Guard
ships.
COMPUTER HELPS PRO FOOTBALL
TEAMS MAKE DRAFT SELECTIONS

A direct communications link to
a 1 arge-scare computer helped 8
National Football League teams pick
among the top college prospects
during this year's draft on Janu-

ary 28-29. The computer, an 1108
located at Univac's Midwestern Computer Center, supplied immediate
ratings on players to BLESTO VIII,
an organization of the Chicago
Bears, Detroit Lions, Philadelphia
Eagles, Pittsburgh Steelers, Minnesota Vikings, Baltimore Colts,
Miami Dolphins, and Buffalo Bills.
(BLESTO VIII is a wholly owned subsidiary of the eight member teams.)
The 8 teams were able to request
and receive information on a UNIVAC
DCT-500 Data Communication Terminal
installed in the Philadelphia Eagles'
office and linked to the UNIVAC 1108
in Oak Brook, Ill.
Using the information, each team advi sed its
representative at the draft in New
York on which names to choose. (The
computer contains files on 957
college players.)
The ratings of the top draft
pro spec t s were compil ed from numerous reports of scouts. These reports cover ten general traits, 12
position traits, and the individuaI's size, speed, school, and injury hi story.
Members of BLESTO
VIII pay a fixed fee to support.
eight area scouts and five scout
supe~visors
throughout the U.S.
This is believed the first time a
direct link to a computer has supplied immediate data on progress of
the draft to professional football
teams. Other scouting organizations
al so use computers extensively, but
not for immediate information on
the draft itself.
FLORIDA FIRM USING
COMPUTE R TO I DENTI FY
NEW PRODUCTS, PROCESSES
AVAILABLE FOR LICENSING

A Florida-based firm is using an
IBM System/360 Model 20 to collect
and disseminate information on new
·products and processes that can be
licensed to other companies.
On
January 4, 1971, Dr. Dvorkovitz &
Associates began issuing continuous
reports on the newest available offerings from around the world to
U.S. and foreign fi rms in the pharmaceutical and chemical process industries, and for patented electronic and mechanical devices.
Dvorkovi tz representatives scout
all parts of the world for products
and processes which may be of interest to firms in other countries.
Dr. Vladimir Dvorkovitz, founder
and president of the firm, sends
clients in many countries computerproduced reports on such unusual
items as the formula for a plastic
that self-destructs in sunlight and
a real butter that needs no refrigeration and can ,be produced in ei ther
1 iquid or sol id form. Several hundred items are reported every week.
48

"Our computer-based system, which
was under development for more than
two years, enables us to provide
each client with a highly individualized report that will always be
up to date", said Dr. Dvorkovitz.
The computer automatically indexes
each offering by its basic process
or product line, and print-outs
incl ude such information as a basic
description of each item; worldwide patent status on a country-bycountry basi s i degree of commercialization; and identification of geographic areas where 1 icensing is
possible.
The name, address and
responsible party as licensor are
1 i sted and direct contact may be
made immediately.
Every licensable offering is required to meet
rigid quality standards before it
is accepted for listing.
There
are over 1800 items currently housed
in the computer's memory bank.

EDUCATION NEWS
COMPUTER TEACHES TEACHERS
IN EARLY DETECTION OF
HANDICAPPED CHILDREN

Elementary school eacers are
learning to recognize characteristics of handicapped children with
the aid of a new computer-assisted
instruction program recently developed by Penn State University,
University Park, Pa.
The program
is funded by U.S. Office of Education's Bureau of Educational Personnel Development and Bureau for
the Education of the Handicapped.
An IBM 1500 system, including
di spl ay terminal s, image proj ectors
and synchronized audio equipment,
is housed in a van which will travel
throughout Pennsylvania for on-site
presentations to teachers.
Some
1000 teachers per year throughout
the State of Pennsylvania are expected to receive instructions on
the IBM system.
Teachers, seated at computerdriven terminals, receive instructional materi al on spotting such
tell tale signs of brain damage or
other handicaps as garbled speech
patterns and resistance to discipline procedures.
The teacherstudent responds to graphically
di spl ayed material by touching a
1 ight probe to the screen or by
typing answers into the system
over the unit's keyboard.
Dr. Harold E. Mitzel, director
of the Penn State proj ect, said
that nearly one out of every six
children in the primary grades
suffers from some form of handicap
that requires special recognition
in an educational program.

OSU STUDENTS LEARN WITH
AID OF COMPUTER-CONTROLLED
TAPE MACHINES

A bank of magnetic tape machines
controlled by a computer is helping
Ohio State University students learn
everything from basic French to advanced chemistry. The system, recently implemented as a supplement
to classroom instruction, consists
of approximately 440 student li stening devices or stations, linked to
the central control unit from varied campus locations by telephone
lines. Equipped with headsets and
telephone dials, the learning stations are located in libraries,
dormitories, schools within the
uni veri sty and even some fraternities and sororities.
Seated at one of the stations,
the student di al s a three-digit
figure representing the self-study
or lecture which interests him. An
IBM 1800 data acquisition and control system, located at OSU's Listening Center, accept s the student's
call and connects him with the desired course. If the tape is currently being used by others, the
student monitors' the recording to
completion and the computer automatically recycles it for his benefit.
This is done until all students on line to any particular
course of instruction disconnect.
The system is currently handling
approximately 4,000 student call s
a day.
HIGH SCHOOL DATA PROCESSING
NETWORK OFFERS SPECIALIZED
COMPUTER TRAINING TO
TEXAS STUDENTS

Built around two RCA Spectra 70
computers in Houston's (Texas) Region IV Education Service Center,
a new educational data processing
network will support 200,000 Texas
high school students by 1972 in
34 of the region's 56 districts.
Using classroom teletypewrtier
terminal s and the simplified computer 1 anguage known as BASIC, student s sol ve mathematic al probl ems
associated with chemistr~ physics,
physical science, computer science
and general mathematics courses.
These terminals operate in a timesharing mode with the Spectra 70/46
and Spectra 70/45. Student training in computer operation and program preparation, offered in vocation-technical and computer science
courses, is handled through card
reader-printer terminals, thus giving students access to the computer
systems through the day. The network also will be used by teachers
and school officials for classroom
scheduli ng , grade and attendance

COMPUTERS and AUTOMATION for March, 1971

reporting, test scoring, tax accounting and other administrative
tasks.
TWENTY-NINE SCHOOL
DISTRICTS IN MINNESOTA
JOIN IN 'COMPUTER CO-OP'

A "computer co-op" is providing
a quarter million students and
teachers in 29 independent school
districts in Minnesota with complete low -cost compu t~r services.
Services range from demonstrations
of a computer's problem-solving
abili ty and classroom compu ter-aided
instruction to unrestricted use of
a time-sh~red computer plus the
more usual administrative handling
capabilities of payroll and scheduling, and computer analysis of
data for research projects and experiments.
The Minnesota School
Districts Data Processing Joint
Board, which operates as TIES (Total
Information for Educational Systems)
has a staff of approxima te ly 30
professionals who administer the
educational programs and operate
the million dollar computer center.
TIES ~tilizes two computer systems. The first is a recently acquired Hewlett-Packard 2000A Time
Shared Basic system with telephone
links to 16 remote terminals. Seventy terminals will be used by the
schools in the TIES project during
the current school yeaL This system supplies demonstrations of a
computer's abilities and computeraided instruc tional programs.
A
professional staff, many of whom
are trained both in teaching and
computer technology, instruct and
assist the students from kindergarten through the twelfth grade.
The other system, which primarily handles the administrative
duties, is a Burroughs 3500 system
wi th 34 data input and display
terminals ins taIled in the districts. Ten high-speed line printers, for lis ting of long reports,
are to be ins taIled at centrally
located schools
The study of the potential use
of computers in elementary and secondary education began in the St.
Paul-Minneapolis area as a voluntary proj ec t in 1963.
The 19691970 school year marked the first
year of services to member school
districts in seven counties. The
member districts contributed approximately $3.50 per student during the pas t two years to cover
partial expenses of the program.
This took up the s lack as a federal
grant ran out in 1970.
Costs for
the coming school year (1970-1971)
to cover complete operating expenses
are expected to be around $5.00
per student. Previous to TIES, inCOMPUTERS and AUTOMATION for March, 1971

di vidual schools were purchas ing
time from commercial systems at $10
to $12 per hour.
Costs for the
same service now are approximately
$1.50/hour. WWhere once there was
a demand for more time than schools
could afford, they now use all the
time they need and cos ts are included in the $5.00 per student
yearly charge.

RESEARCH FRONTIER
LASER BEAM SENDS DIGITAL
DATA OVER LENGTHY
ATMOSPHERIC PATH

Lockheed Palo Al to Research Laboratory (Calif.) has announced it
is using a laser beam to send digi tal data at a high rate over a
1. 2-mile path across the nearby
hi lIs.
The new overland link consisting of a brilliant green
beam shot from a laboratory peephole to a hillside receiver - has
a transmission speed of 300 million
data "bits" per second.
Dr. Romayne F. Whitmer, head of
the firm's Electro-Optical Systems
Laboratory, said, "We believe this
is the highest rate ever achieved
over a long path through the atmosphere. Such a link could have large
commercial and military uses." He
said with advanced terminal equipment, the link could transmit the
entire 23 volumes of Encyclopedia
Britannica in five seconds.
Dense fog, heavy snow and other
extremely poor weather condi tions
probably would hamper the new laser
system just as they would hamper
other optical links
al though
foul-weather tests have yet to be
run, according to DL WhitmeL
But
he said the sys tern's proven superiority in normal weather is still a
maj or step forward. The link alleviates the problem of atmospheric
distortion because it uses a "frequency modulated microwave subcarrier" to impress the information
on the laser beam.
The result is
an FM system which, like FM radio,
is far less vulnerable to distrubances than other forms of transmission.
One application may be in the
field of cable television.
Laser
beams could carry 30 or 40 TV channels each - simultaneously - from
a city or regional distribution
center out to smaller centers, which
then would send the signals by cable
into viewers' homes. He said this
could bring immense savings in
cable-system construction costs.

A second use would be to link
computers, scattered around a city
or region, so they could pass data
back and forth automatically. Many
large compan ies and governmen t agencies require such communica tion between computers. This now is done
primarily wi th more cos tly cable
networks.
Some early use of such a laser
communication system is expected
between satellites in space, which
is free of the disturbances found
in the earth's envelope of air.
Bu t the new link acros s the Palo
Al to foothills proves there are also
many earthbound possibilities.

MISCELLANY
COMPUTER CUBISM?

No, this is a well known face
that has been precisely blurred by
a computer. It is part of an experiment by Leon D. Harmon at Bell
Laboratories, Murray Hill, N. J . ,
to learn the least amount of visual information a picture may contain and still be recognizable.

The picture is divided in to about
200 squares, with each square rendered in an even tone from one of
16 intensities of gray.
If you
still don't recognize theportrait,
try looking at it from 15 feet or
more, or while it's in motion, or
while you're squinting, or wi th
your eyeglasses removed.
Studies
of the "information content" of a
picture may be useful for designing future PICTUREPHONE® systems,
and for devising techniques for
computer storage of pictures.
(UTOOUn weqe~Q\f JO sllle~uod

aql)

NEW PRODUCTS AND SERVICES
NAM.E/ MODEL.

'Digital
-

NO.

Cogar System 4

GRI-909 computer

Honeywell Series 200
computers

UT-l system

FOR MORE INFORMATION

DESCRIPTION

A stand-alone minicomputer designed for word handling
as well as numerical computation / data system includes
fully p~ogrammable, random-access, monolithic memory
(2K expandable to 16K bytes), keyboard, two tape decks
and a CRT display
Four new models fully software compatible wi thtlie original GRI-909 / Model 10 is designed for system control
applications; basic memory is lK x 16 read-write, expandable /Model 20, similar to Model 10 wi th operator's
console for interaction with computer / Model 30 has
4K x 16 memor
ex andable
Model 40 series lar est
New models, 115 2, 1015 and 2015 have elements of Model
115 / Series family spans small-to-medium price-performance range / main memory sizes range from 16,384
to 262,144 characters; cycle times from 2.75 usec per
character on the 115 to 1.3 usec for two characters on
the 2015
Functions both as off-line computing sys tem and as
programmable remote batch data communications terminal
for time sharing applications / configured around the
Data General NOVA 1200 computer, system offers wide
range of peripherals

Cogar Corp.
Information Systems Div.
Cosby Manor Rd.
Schuyler, N.Y. 13502
Attn: Stephen Robbins
GRI Computer Corp.
230 Needham St.
Newton, Mass. 02164
Attn: Irwin M. Stone
Honeywell Information Sys tems
60 Walnut St.
Wellesley Hills, Mass. 02181

UNITECH, Inc.
1005 E. St. Elmo Rd.
Austin, Texas 78745
Attn: Frank Milstead

Memories
Ampex Model RGM Core
Memory

Braided U-Core ReadOnly Memory (ROM)
Cogar 70

DO 4314-1 Dl'sk Drive

ExpandaCore 620
MEGAMEMORY 1000

Micromemory 4000

Series 720 Controllers

For military and other severe-environment applications/
400 nsec access time; full cycle time, 1 msec / four
basic word-size configurations; capacities of 4,096 by
18 and by 36 bits, 8,192 words by 18 bits and 16,384
words by 18 bi ts
Model MP 256/36A for airborne and field service use /
a 256 x 36 bit (9K) ROM / 500 nsec cycle time; access
time is 250 nsec / physical size is 1.25" x 4.5" x
9.0" for a packing densi ty of over 180 bi ts per cu. inch
Monoli thic add-on memory for 360 computers /262K bytes
expandable to a one megabyte system in plug-in increments of 262K bytes / 425 access time; cycle time of
750 nsec / a complete. self-contained system
Replaces IBM 2314 and 2319 Drives; up to 9 drives may
be connected to matching control unit, DC 5314 / less
than 30 msec average access times / transfer rate is
321,000 bytes per second or 624,000 digits per second
with packed decimal data
Plug-compatible expansion memory for Varian 620/i 4K
to a maximum of 32K words / a 3D, 3-wire core memory /
1.8 msec full cycle timej access time is 750 nsec
Flexible, 2-wire, 2VzD core system for main frame memory extension or as peripheral memory / 850 nsec access
time with a 1.5 usec cycle time / storage capacities -32,768 words, 32-160 bits per word up to 524,288 words
of from 8-14 bits per word
2-wire, 2VzD organization / basic module capacity is
32,768 words of 18 bits per word / 800 nsec access time,
1.5 nsec cycle-time / key features are price, 1-0 flexibility. packaging density
Provides minicomputer users with upward compatible plugin data storage system capability / allows user to expand his storage from 0.6 million bits to over 70 million bi ts depending on which of 8 memory systems are used

Ampex Corporation
9937 West Jefferson Blvd.
Culver City, Calif. 90230
Attn: Chris Hoppin
Datapac, Inc.
3839 South Main St.
Santa Ana, Calif. 92702
Attn: Tom Lee/Ben Garrett
Cogar Corp.
Technology Div.
All Angeles Rd.
Wappingers Falls, N.Y. 12590
Potter Instrument Co., Inc.
532 Broad Hollow Rd.
Melville, N.Y. 11746
Attn: R. T. Grunenwals
Cambridge Memories, Inc.
285 Newtonville Ave.
Newton, Mass. 02160
Electronic Memories
12621' Chadron Ave.
Hawthorne, Calif. 90250
Attn: LeNore Plotkin

Bryant Computer Products
850 Ladd Rd.
Walled Lake, Mich. 48088

Software
DOS ASAP (Automatic
~ooling-with Asynchronous frocessing)
50

For users of IBM 360 under the Disk Operating System /
alleviates dependence of CPU on relatively slow unit
record devices / provides print, punch and card read
operations at maximum device speed

Universal Software, Inc.
12 Horseshoe Drive
Danbury, Conn. 06810
Attn: David W. Kearns

COMPUTERS and AUTOMATION for March, 1971

DESCRIPTION

NAME/MODEL NO.

FOR MORE INFORMATION

(Software, continued)
EASY PLOT

General Ledger System

A non-programmer application language for engineers and
other non-programmer professions to generate plotter
ou~put / need only a keyhoard terminal (such as a Teletype) and a plotter (works with most standard plotters)
Time-sharing system for public accounting firms / reduces accountant's write-up tIme by 50-85% / accessed
by remote terminal/no programming knowledge required /
applicable to all types of business

COBOL preprocessor / provides programmer with unlimited
abbreviation capability throughout all COBOL divisions
by both supplied and user defined abbreviations / macro
facility, syntax checking. output formatting, also supplied
PROFACTS (PROduct For- Provides current information for management review and
mulation~ccount
production purposes / written in COBOL for an IBM System/360 Model 30 or larger, using DOS / minimum of 64K,
ing, and ~osl ~s
tern)
4 tape drives and 2 disc drives
QWIK-TRIEVE""
A generalized interactive data storage and retrieval
system for time-shared computers / stores data in direct access files / handles both formatted data and
free data / responds to English-like query statements
SIRE (Symbolic InforIntegrated information storage, file management main~
mation Retrieval)
tenance and retrieval system / provides capability to
write applications in any language (English, French,
German, Japanese, etc.) / no technical programming
background required / available for UNIVAC 1106 and
1108, IBM 360/50 "and larger (minimum core,270K bytes)/
developed as in-house system by Boeing Company; Script
has exclusive world-wide marketing license agreement
Calculates federal, state and local withholding taxes
TAXCAL
handling calculations in one pass / modularized /
written in COBOL / requires 2500 to 18,000 bytes of
core
Unit Inventory TechFor small department stores, specialty shops and other
niques for System/3
retailers / produces time-saving inventory data reports
using IBM's System/3 Model 10 / written in English-like
RPG II programming language
MAGIC-Shorthand

Tymshare
525 Uni vers i ty Ave., Sui te 220
Palo Alto, Calif. 94301
Attn: Jim Harrison
Honeywell Information Services Operations
Mail Station Gl124
2701 Fourth Avenue South
Minneapolis, Minn. 55408
Information Management, Inc.
447 Battery St.
San Francisco. Calif. 94111
Fortex Data Corp.
230 North Michigan Ave.
Chicago, Ill. 60601
Westinghouse Tele-Computer
Systems Corp.
2040 Ardmore Blvd.
Pittsburgh, Pa. 15221
Script Associates, Inc.
225 108 NE
Bellevue, Wash. 98004

Datasonics, Inc.
663 Fifth Ave.
New York, N.Y. 10022
Attn: Martin Burack
IBM Corp., Data Processing Div.
112 East Post Rd.
White Plains, N.Y. 10601

Peripheral Equipment

Automatic Calling Unit, For interfacing business machines and computers with
Sola Basic Industries, G-V
ACU-1801
with tone or pulse dialing telephone systems / unit acControls Division
cepts stored telephone numbers in binary code, converts 101 Okner Pkwy.
to digital code, and automatically dials those numbers
Livingston, N.J. 07039
over any standard telephone line / four available models
Card Reader, Mod 250
Specifically for use with minicomputers / both handle
UniComp, Inc.
Card Punch, Mod 260
80 column cards / Reader has vacuum pick finge~ handles 18219 Parthenia St.
300 cards per minute; is 14"W, 18"D, 18"H, 47 pounds /
Northridge, Calif. 91324
Punch operates 120 cards per minute; is 22Yz"H x 19"W
DigiNet® 160 Series
Private line data multiplex systems equipped with "inGeneral Electric Co., Telecommunications Products Dept.
ternal diagnostics" / moves information over leased or
Section P, P.O. Box 4197
private four-wire voice-grade circuits / connects diLynchburg, Va. 24502
rectly to telephone line, no data set reauired / has
channel speed intermix capabil i ty / modul arl y con structed
For time-sharing and communications applications / has
Model 420 Portable
Data Access Systems, Inc.
Printer
built-in acoustic coupler / operates at switched speeds 503 Route 10
of 10, 15, and 30 cps using non-impact thermal print
Dover, N.J. 07801
head i prints 80 characters per line ! teletype compatible Attn: Robert T. Coppoletta
Model GT50/10Graphics High speed 10-bit device for converting hand drawn data Computek, Il,lc.
Tablet
to digi tal form / wri ting surface is 11~" x ll~tt / may
143 Albany St.
be used with graphic display terminals or as a standCambridge, Mass. 02139
alone unit
Attn: Dou las Drane
Punched-tape optical
Operates continuously
designed specifically for
Decitek
batch-type, continuous tape runs / tape transported on
reader
15 Sagamore Rd.
dual-sprocket, capacitor-start/capacitor-run AC motor
Worcester, Mass. 01605
drive at 700 characters per second
Attn: I. E. Spalding
Regiscan Reader
Hand-held automatic price tag reader for retailing / a
American Regitel Corp.
4 oz., trigger-operated device using light-emitting and 870 Industrial Rd.
photosensing diodes! automatically senses price and
San Carlos, Calif. 94070
merchandising information from Kimball or Dennison
Attn: Jerome Turk
tickets I reading time is less than one millisecond
':'QWIK-TRIEVE is a service mark of the Westinghouse Tele-Computer Systems Corporation
COMPUTERS and AUTOMATION for March, 1971

NAME/MODEL NO.

DESCRIPTION

FOR MORE INFORMATION

(Peripheral Equipment, continued)
T-66 Credit Control
Terminals

230 Dataplottetv

VertaTape, Paper Tape
to Magnetic Tape
Converter

A credit card reading terminal; automatically communicates with a computer, obtains credit authorizations in
seconds / operates over public telephone network I can
be programmed to access anyon-line computer in continental United States
Designed for time-share users I self-contained, desktop
device, compatibly interfaced to keyboard terminals and
acoustic couplers / operates at maximum speed in all directions'l includes easy-to-use FORTRAN plot ting subroutines
Makes conversion all off line I converts coded information from 5, 6, 7 or 8 channel paper tape to 64 character magnetic tape code set and edited for computer input I plug-in patch-board programmed for each customer's particular application

Audac Corp.
175 Bedford St.
ourlington, Mass. 01803
Attn: Peter Bryan
Electronic Associates, Inc.
185 Monmouth Pkwy.
West Long Branch, N.J. 07764
Attn: Ed Sharpe
Datascan, Inc.
1111 Paulison Ave.
Clifton, N.J. 07013
Attn: Doug Schwartz

Data Processing Accessories
Magnetic Tape
Cartridge

Completely compatible with IBM Magnetic Tape Selectric
Typewriters I 120-foot / exacting computer tape quality standards I backed by a one-year warranty

TM/33 Teletype

An automatic tape handling system
may be used to
simultaneously feed and collect tape for reading or
simply collect tape after perforating I mounts rigidly
to the Teletype, no tools or modifications required

Tape Handler

Memorex Corp.
1180 Shulman Ave.
Santa Clara, Calif. 95050
Attn: Jerome M. Kell
Data Specialties, Inc.
1548 Old Skokie Rd.
Highland Park, Ill. 60035

Computer. Related Services

Computer Assisted
Diagnosis for Internal Medicine

Computer Law
Dictionary

Moonlighters' JobMatching Service

Gives physician ready access to diagnostic information
screened from medical literature I computer is called
by way of teletypewriter installed in doctor's office,
hospital or other central point; findings are entered
into computer via numbers corresponding to those listed
beside particular findings in operating manual; computer compares these with memory bank; responds with a
printed listing of those diseases or syndromes found to
be supported by the findings / computer also provides
a literature reference and additional findings physician should look for in determining diagnosis I program encompasses diagnoses in general area of adult
medicine
For use by legal profession / contains over 50,000 words
and terminologies used by the profession -- in essence,
the language of the lawyer I Omnitec 800 portable data
terminal is link between the 'dictionary' and computer
stored data relevant to United States Court decisions /
when queried computer searches memory, finds proper response in form of legal citations, transmits response
via telephone and terminal automatically types it out /
service will be operational initially in and for Pennsylvania; dictionary soon will serve all of state and
federal computer libraries
A computerized "job bank" and referral system that
matches skills of moonlighters (temporary and part time
employees) with specialized requirements of employers /
encompasses virtually all skills required by industry
and the professions I free to employers; available to
moonlighters for a one-time registration fee of $10

Mead Johnson Medical Services
Evansville, Ind. 47721
Attn: Rolland M. Eckels

The Computer Searching Service Corp.
11 Erie St.
Garfield, N.J. 07026

Moonlighters, Inc.
9060 Santa Monica Blvd.
Los Angeles, Calif. 90069
Attn: Robert L. Rod

New Literature

Magnetic Disk Memories

A 72-page report on the industry covering 51 manufacturers / study encompasses four main aspects of the
industry: markets, technology, products, and companies I designed to permit rapid evaluation of competitive products, annual shipments, price changes,
technological developments, competitive new products
and growth opportunities for major and independent
manufacturers plus breakout for minicomputer disks

High Technology West
1060 Crenshaw Blvd.
Los Angeles, Calif. 90019
Attn: Eugene Kurchak

COMPUTERS and AUTOMATION for March, 1971

NEW CONTRACTS
Computer Sciences Corp., Los
Angeles, Calif.

RCA Corporation

Computer Technology Inc.,
Dallas, Texas

Barnett First National Bank
of Jacksonville, Fla.

Philco-Ford Corp., Willow
Grove, Pa.

U.S. Army Electronics
Command, Ft. Monmouth, N.J.

The Medicus Corp., Dallas,
Texas
Electronic Associates, Inc.,
West Long Branch, N.J.

Baptist Medical Centers of
Birmingham, Ala.
American Electric Power
System

Greatermans Group, South
Africa

The National Cash Register
Co., Dayton, Ohio

Burroughs Corp., Detroit,
Mich.
Univac Division of Sperry
Rand Ltd., London, England

U.S. Postal Service
British European Airways

Entrex, Inc., Lexington,
Mass.
Computer Sciences Corp., Los
Angeles, Calif.

Redifon Data Systems
United Kingdom
U.S. Naval Electron Systems
Command, Special Communi6ations Project Office

Peripheral Equipment Corp.,
Calif.

Consolidated Computer

Princeton Electronic
Products, Inc., North
Brunswich, N.J.
Ampex Corp., Culver City,
Calif.

Smith, Kline & French, Inc.,
Smith, Kline Instruments

Burroughs Corp., Detroit,
Mich.

U.S. Air Force

Sierra Research Corp.,
Computer Products Group,
Burlington, Mass.
The Center for Research
Libraries, Chicago, Ill.

Babcock and Wilcox Co.,
Naval Nuclear Fuel Div.,
Lynchburg, Va.
Ford Foundation

Chatswo~th,

Sigmatics, Newport Beach,
Cal i f.
Carnegie-Mellon Univ.,
Graduate School of Industrial Administration,
PittSburgh, Pa.
National Archives

Texas A & M University

Defense Atomic Systems
Agency Field Command,
New Mexico
The Richard King Mellon
Charitable Trusts
Ford Foundation

M.I.T., Dept. of Architecture
Architecture Machine Gp.,
Cambridge, Mass.
RCA, Cherry Hill, N.J.

Graham Foundation for
Advanced Studies in the
Fine Arts
U.S. Army, Pacific

Comma Corp.,

Data Processing Financial &
General Corp., Diebold Computer Leasing, Inc., Randolph
Computer Corp., and Talcott
Computer Leasing

COMPUTERS and AUTOMATION for March, 1971

A five-year contract to develop elements
of software for U.S. Navy Aegis defensive
missile system including radar processing,
command & control & training
A three-year data processing facilities
contract to handle design and operation
of bank's computer applications; also for
BankAmericard service center
Expansion of Defense Department's Overseas Automatic Digital Network; for receiving, storing & routing messages
Installation of a medical data processing
center in the two-hospital system
A computer-controlled simulator of a
power plant control room for fossilfueled power plant training in the
electric industry
Retail data-capturing equipment, initially
300 NCR 5 control registers for point of
sale data collection; additional 200
registers to follow
Design, development and installation of
a Code Controlled Letter Sorting Machine
Two hundred UNISCOPE 100 visual display
units; will enable 22 European cities to
be visually linked to BEA's BEACON seat
reservation system
Ten System 480's, computer based data
entry equipment, to be called "SEE-CHECK"
Providing technical support and management assistance in communications; will
provide studies and analyses in support
of naval communications projects
A contract including 250 PEC 6000 Series
tape transports for new KEY-EDIT data
preparation system
Electronic display equipment to be utilized in medical data processing systems
Model ECM-65 core memory systems and TM1624 magnetic tape units to be linked to
an IBM 360/65 system
Development of an advanced aerospace multiprocessing computer to be used in future
high performance aircraft and space
vehicles
Installation of an SDA-770 System for
factory data collection, planning and
control
A grant for support of a program to
provide widest access to machine readable forms of 1970 Census Summary Data
and public use samples on magnetic tape;
also to reduce cost of using data for
research purposes
Provision of programming support for an
automated financial management system
Development of a management information
control system to upgrade efficiency of
hospital administration; study of ambulatory care facilities
A two-year grant for indexing the papers
of the Coritinental Congress
Support of a fellowship program for
advanced study in computer-aided
architecture
Seven Spectra 70 systems, having a sales
value of $11.7 million, to provide automated logistic support for U.S. Troops
in Sotheast Asia; contract includes an
option for additonal 14 Spectras
Computer maintenance agreements; authorizes Comma to sign computer maintenance
contracts with any of leasing companies'
customers, should they so desire; approximately $600 million of IBM/36O equipment
is owned and leased by the four firms

$11 million

$5+ million

$2,685,000
$2.2 million
$2+ million

$1.5+ million

$1.3 million
$1.2 million

$1+ million
$1 million

$750,000
$670,000
$475,000+
$469,000

$300,000+
$225,000

$219,380
$200,000

$150,000
$40,000

NEW INSTALLAliONS

Control Data 3170 system

North Carolina Agricultural and
.Technical State University,
Greensboro, N.C.

Control Data 3300 system

Massachusetts Registry of Motor
Vehicles, Boston, Mass.

Control Data 7600 system

Westinghouse Electric, Westinghouse Nuclear Center, Monroeville, Pa.
J. C. Penney Company, Inc., New
York, N.Y.
Blue Cross of Southern California,
Los Angeles, Calif.

Hewlett-Packard 2000B system
Honeywell Model 4200 system
Honeywell 8200 system

The American Mutual Liability
Insurance Co., Wakefield, Mass.

IBM System/3

Auburn Hosiery Mills, Inc.,
Auburn, Ky.
The Fate-Root-Health Co.,
Plymouth, Ohio

IBM System/3 Model 10
.IBM System/360 Model 20
IBM System/370 Model 155

George R. hlein News Co.,
Cleveland, Ohio
Blue Cross and Blue Shield of
Alabama, Birmingham, Ala.

IBM 1130 system

Samuel Gary Oil Producer, Denver,
Colo.

IBM 1800 sys tem

The Budd Company, Automotive Division, Detroit, Mich.
Gemini Chit Fund Corp., Singapore
Guaranty Bond State Bank, Mt.
Pleasant, Texas
Revillon, Paris, France

NCR Century 100 system

NCR Century 200 system

UNIVAC 118-11 and 1106 systems

UNIVAC 1106 system

Alliance Machine Co., Alliance,
Ohio
U.S. Ceramic Tile Co., Canton,
Ohio
Swedish Government Office of Organization and Management (Statskontoret), Stockholm, Sweden
(4 systems: two 1106s and tw0418-IIs)
Mankato State Colleg~ Mankat~ Minn.
New Holland Division of Sperry Rand
Corp., New Holland, Pa.
Ontario Government, Computer Services Center, Toronto, Canada

UNIVAC 9200 system
UNIVAC 9400 system

Shrewsbury Borough Council,
England
The Roman Catholic Archdiocese of
New York, Data Systems Center

XDS Sigma 3 system

Goodyear Tire & Rubber Co.,
Bakersfield, Calif.
Link-Miles Ltd., London, England

XDS Sigma 5 system

American Council on Education

XDS Sigma 6 system

University of Tulsa, Kendall
Campus, Tulsa, Okla.

XDS CE16 systems

Administrative, training and scientific data processing applications; also plans to offer computer
time to local small colleges
(system valued at $600,000i
Building and maintaining automobile registration
files; providing remote inquiry arid retrieval system on driver record and related information; generating accident and statistical reports; preparing
excise tax bills for cities and towns; issuing
driver license renewals; other related applications
Use in designing reactor systems for nuclear power
plants
An in-house time-sharing system dedicated entirely
to business applications
The ini tial step in a maj or program to speed its
data processing operations; second H4200 will be
delivered in the spring
Replacing three present systems; H8200 will be used
for writing and rating policies, collection, claims
processing and administrative functions
Providing customers with in-depth data and sameday order filling
Labor and payroll reports, orders, invoices~ later
applications will include job order costing and
production control
Daily sales analysis of newspaper and magazines;
helps 'keep stockroom empty'
Speeding benefit payments to over 1 million Alabama
menmers; facilitating the full claims service for
enrollees in Medicaid and in parts A and B of Medicare
Helping monitor oilfields of independent oil producer and handling complex division-of-interest accounting for producer's investors; also bill processing, cost accounting, monthly reports for state
and federal governments
Providing production reports; monitoring production
of 250 machines which make automobile body components
Processing some 22,000 depositor accounts
The heart of a new Central ·Information File
A variety of data processing tasks including billing, stock and statistical work; will later include the entire banking application
InventoTY control, handling accounts payable; also
processing accounts receivable and payroll
Preparation of payrolls and general accounting
Use of the Centrala Bilregistret (Stockholm), the
central automobile registration agency for Sweden;
systems also be used for census work
(systems valued at $4~6 million)
Administrative and. academic work; also for use by
3 other colleges and later some high schools
Heart of multi-million dollar world-wide management
information center
(system valued at $2 million)
Various departments of the provincial government requiring data processing including Treasury and Economics, Revenue, Labor, Lands and Forests, etc.
(system valued at about $1.25 million)
Local government applications including payroll,
mortgage and investment loans and taxes
Use in administrative offices, education office
and in student/teacher training; includes e.g., fund
raising reports, pension plans for both layment and'
pri~sts; grade reporting, student scheduling, etc.
Incorporation into process control systems to automate automobile seat cushion production facilities
Use in flight simulator being built for BOACjwill
be used to train pilots and flight engineers
Assisting in studies of nation's higher education
problems (over 400 colleges are participating)j main
use as massive data storage and computing system for
researchers
Student instruction; administrative processing;
research activities (both for graduate students
and faculty); and information storage and retrieval
services for the petroleum industry - the North
Campus will have a remote job entry terminal
COMPUTERS and AUTOMATION for March, 1971

MONTHLY COMPUTER CENSUS
Neil Macdonald
Survey Editor
Q)MPUTERS AND AUTOMATION
The following is a summary made by Q)MPUTERS AND AUTOMATION of reports and estimates of the number of general purpose electronic digital computers manufactured and installed, or to be manufactured and on
order. These figures are mailed to individual computer manufacturers
from time to time for their information and review, and for any updating or comments they may care to provide. P lease note the variation
in dates and reliability of the information. Several important manufacturers refuse to give out, confirm, or comment on any figures.
Our census seeks to include all digital computers manufactured anywhere. We invite all manufacturers located anywhere to submit information for this census. We invite all our readers to submit information that would help make these figures as accurat~ and complete as
possible.

The following abbreviations apply:
(A) -- authori tati ve figures, derived es senti ally from information
sent by the manufacturer directly to COMPUTERS AND
AUTOMATION '
C
figure is combined in a total
(D)
acknowledgment is given to DP FOCUS, Marlboro, Mass., for
their help in estimating many of these figures
E
figure estimated by COMPUTERS AND AUTOMATION
(N) -- manufacturer refuses to give any figures on number of installations or of orders, and refuses to comment in any
way on those numbers stated here
(R)
figures derived all or in part from information released
indirectly by the manufacturer, or from reports by other
sources likely tO'be informed
(S)
sale only, and sale (not rental) price is stated
X -- no longer in production
information not obtained at press time

Part I of the Monthly Computer Census contains reports for United
States manufacturers. Part II contains reports for manufacturers
outside of the united States. The two parts are published in alternate months.
SUMMARY AS OF FEBRUARY 15, 1971

NAME OF
MANUFACTURER
Part 1. United States Manufacturers
Autonetics
Anaheim, Calif.
(R) (1/69)
Bailey Meter Co.
Wickliffe, Ohio (A) (2/71)

Bunker-Ramo Corp.
Westlake Village, Calif.
(A)

(12/70)

Burroughs
Detroit, Mich.
(N)

(1/69-5/69)

Computer Automation, Inc.
Newport Beach, Calif. (12/70)
Control Data Corp.
Minneapolis, Minn.

(A)

(R)

(9/70)

NAME OF
COMPUTER
RECOMP II
RECOMP III
Bailey 750
Bailey 755
Bailey 756
Bailey 855
BR-130
BR-133
BR-230
BR-300
BR-330
BR-340
BR-1018
205
220
BIOO/B500
B2500
B3500
B5500
B6500
B7500
B8S00
208/808
216/816
GIS
G20
LGP-2l
LGP-30
RPC4000
636/136/046 Series
160/8090 Series
924/924-A
l604/A/B
1700
3100/3150
3200
3300
3400
3500
3600
3800
6400/6500
6600
6800
7600

DATE OF
FIRST
INSTALLATION
11/58
6/60
11/61
2/65
4/68
10/61
5/64
8/63
3/59
12/60
12/63
6/71
1/54
10/58
7/65
2/67
5/67
3/63
2/68
4/69
8/67
6/68
3/69
7/55
4/61
12/62
9/56
1/61
5/60
8/61
1/60
5/66
5/64
5/64
9/65
11/64
8/68
6/23
2/66
8/64
8/64
6/67
12/68

AVERAGE OR RANGE

OF MONTHLY RENTAL
$(000)
2.5
40-250
200-600
60-400
100-1000
2.0
2.4
2.7
3.0
4.0
7.0
23.0
4.6
14.0
2.8-9.0
5.0
14.0
23.5
33.0
44.0
200.0
5.0
8.0
1.6
15.5
0.7
1.3
1.9

(S)
(S)
(S)
(S)

NUMBER OF INSTALLATIONS
Outside
In
In
World
U.S.A.
U.S.A.

NUMBER OF
UNFILLED
ORDERS

32
6
13

35
6
18

o
o

160
79
15
18
19
19

X
X
X

(S)
25-38
28-31

27-40
30-33

X
X

117
190

64-69
62
72-81
4

1
143
157

52-57
44
65-74
4

(S)
(S)

12
18
7

2.1-14.0
11.0
45.0
3.8
10-16
13.0
20-38
18.0
25:0
52.0
53.0
58.0
115.0
130.0
235;0

60
13

150
170
295
20
165
322
75
29
610
29
59
106-180
83-ll0
55-60
200
20
15
39
20
85
85
1
1

130
215
X
X
X
X
X
X
X
X

o
C
C
C
C
C
C
C
C
C
C
C

Total:
160 E
Data General Corp.
Southboro, Mass. (A)

(2/71)

Datacraft Corp.
Ft. Lauderdale, Fla. (A) (10/70) DC
Digiac Corp.
Plainview, N.Y. (A) (12/70)
Digital-Computer Controls, Inc.
Fairfield, N.J. (A) (2/71)
Digi tal Equipment Corp.
Maynard, Mass.
(A)

(2/71)

COMPUTERS and AUTOMATION for March, 1971

8.0
9.6
5.4

(S)
(S)

6.9

(S)
(S)
(S)

NOVA
SUPERNOVA
NOVA 1200
NOVA 800
SUPERNOVA SC
6024/1
6024/3
Digiac 3060
Digiac 3080
Digiac 3080C
D 112

2/69
5/70
12/70
4/71
6/71
5/69
2/70
1/70
12/64
10/67
8/70

11.9
54-200
33-200
9.0
19.5
25.0
10.0

PDP-l
PDP-4
PDP-5
PDP-6
PDP-7
PDP-8

ll/60
8/62
9/63
10/64
ll/64
4/65

3.4
1.7
0.9
10.0
1.3
0.5

813
102
8

(S)

(S)
(S)

(S)
(S)
(S)

9
21
25
16

9
21

48
40
90

C
C
C

4
45

C
C

10
50
45
100
23
160
1440

1
200
X
X
X
X
X
C

NAME OF
MANUFACTURER
Digital Equipment Corp. (Cont'd)

Electronic Associates Inc.
LongBranch, N.J. (A) (2/71)
EMR Computer
Minneapolis, Minn.
(N)

(12/70)

NAME OF
COMPUTER
PDP-8/1
PDP-8/S
PDP-8/L
PDP-9
PDP-9L
PDP-I0
PDP-11
PDP-12
PDP-IS
LINC-8

DATE OF
FIRST
INSTALLATION
3/68
9/66
11/68
12/66
11/68
12/67
3/70
9/69
2/16
9/66

640
8400
EMR 6020
EMR 6040
EMR 6050
EMR 6070
EMR 6130
EMR 6135
EMR 6155

4/67
7/67
4/65
7/65
2/66
10/66
8/67

~VERAGE

OR RANGE
OF MONTHLY RENTAL
$(000)
0.4
0.3
1.1
8.0
10.5
17.0

1.2
12.0
5.4

6.6
9.0
15.0
5.0
2.6

(S)

NUMBER OF INSTALLATIONS
Outside
In
In
U.S.A.
World
U.S.A.
C
3698
C
C
C
1024
3902
C
C
C
436
C
C
C
48
C
145
C
546
C
C
475
C
C
6
C
15
C
142
C

5
19

Total:
103 E
General Automation, Inc.
Anaheim, Calif.
(R) (6/70)
Hewlett Packard
Cupertino, Calif.
(A) (2/71)
Honeywell Information Systems
Wellesley Hills, Mass.
(A) (2/71)

SPC-12
SPC-16
System 18/30
2114A, 2114B
2115A
2116A, 2116B, 2116C
G58
GI05A
GI05B
GI05RTS
G1l5
G120
G130
G205
G210
G215
G225
G235
G245
G255 T/S
G265 T/S
G275 T/S
G405
G410 T/S
G415
G420 T/S
G425
G430 T/S
G435
G440 T/S
G615
G625
G635
G655
H-110
H-115
H-120
H-125
H-200
H-400
H-800
H-1200
H-1250
H-1400
H-1800
H-2200
H-3200
H-4200
H-8200
DDP-24
DDP-1l6
DDP-124
DDP-224
DDP-316
DDP-416
DDP-516
H112
H632
G3010
G4010
G4020
G4040
G4050
G4060
H1602
H1642
H1644
H1646
H1648
H1648A

1/68
5/70
7/69
10/68
11/67
11/66
5/70
6/69
6/69
7/69
4/66
3/69
12/68
6/64
7/60
9/63
4/61
4/64
11/68
10/67
10/65
11/68
2/68
11/69
5/64
6/67
6/64
6/69
9/65
7/69
3/68
4/65
5/65
12/70
8/68
6/70
1/66
12/67
3/64
12/61
12/60
2/66
7/68
1/64
1/64
1/66
2/70
8/68
12/68
5/63
4/65
3/66
3/65
6/69
9/66
10/69
12/68

0.25
0.41
0.6
1.0
1.3
1.4
1.2
2.2
2.9
4.5
2.9
16.0
6.0
8.0
12.0
13.0
17.0
20.0
23.0
6.8
11.0
7.3
23.0
9.6
17.0
14.0
25.0
32.0
43.0
47.0
80.0
2.7
3.5
4.8
7.0
7.5
10.5
30.0
9.8
12.0
14.0
50.0
18.0
24.0
32.5
50.0
2.65
0.9
2.2
3.5
0.6

2/67
8/64
12/66
6/65

11/68

12.0

C
C
C
C
C
C
C
C
C

Total:
1350 E
6

o
C
C
C
C
C

Total:
15 E

1075
663
1252

200-400

420-680

o
o

35
15
145
40-60
3
15-20
45-60

15-30

10-40

170-300
50-100
20

1
15
17

70-100
20-30

23
20-40
180
30
800
150
800
46
58
230
130

4
15
125
20
18
10

155
45
23
18

620-1080

11
35
16
160
57-77
3
15-20
60-90
10
15-45
240-400
70-130
26

26
23-43
75
160
220
275
40
15
90

55
6
5
60
2
2

1.2
3.2
2.0
6.0
6.0
3.0
7.0
2.0

40
25

C
C
C
C
C

NUMBER OF
UNFILLED
ORDERS
C

45
20
1

255
30
960
370
1075
86
73
325
185
10
20
185
22
20
14
90
250
250
60
450
350
900
75
12

200
65
24
20

60
X
X
X

x
X

X
X

COMPUTERS and AUTOMATION for March, 1971

NAME OF
MANUFACTURER
IBM
White Plains, N.Y.
(N)

(D)

(1/69-5/69)

Interdata
Oceanport, N.J.
(A)

(2/71)

NCR
Dayton, Ohio
(R)

(2/71)

Phi1co
Willow Grove, Pa.
(N) (1/69)
RCA
Cherry Hill, N.J.
(N)

(5/69)

Raytheon
Santa Ana, Calif.
(A)

(2/71)

Scientific Control
Dallas, Tex.

C~rp.

(A)

(6/70)

Standard Computer Corp.
Los Angeles, Calif.
(N) (12/70)
.
SystemS Engineering Laboratories
Ft. Lauderdale, Fla.
(A)]

(6/70)

UNIVAC J;)iv. _of Sperry Rand
New York, N.Y.
(A)

(2/71)

COMPUTERS and AUTOMATION for March, 1971

DATE OF
FIRST
NAME OF
INSTALLATION
COMPUTER
System 3
1/70
305
12/57
650
10/67
1130
2/66
1401
9/60
1401-G
5/64
1401-H
6/67
1410
11/61
1440
4/63
1460
10/63
1620 1, 11
9/60
1800
1/66
7010
10/63
7030
5/61
704
12/55
7040
6/63
7044
6/63
705
11/55
7020,
3/60
7074
3/60
7080
8/61
7090
11/59
7094-1
9/62
7094-11
4/64
360/20
12/65
360/25
1/68
360/30
5/65
4/65
360/40
360/44
7/66
360/50
8/65
360/65
11/65
360/67
10/66
360/75
2/66
360/85
12/69
360/90
11/67
360/195
Model 1
12/70
Model
5/67
Model 4
8/68
ModelS
11/70
Model 15
1/69
304
1/60
310
5/61
315
5/62
315 RMC
9/65
390
5/61
500
10/65
Century 100
9/68
Century 200
6/69
1000
6/63
200-210,211
10/58
2000-212
1/63
301
2/61
501
6/59
601
11/62
3301
7/64
Spectra 70/15
9/65
Spectra 70/25
9/65
Spectra 70/35
1/67
Spectra 70/45
11/65
Spectra 70/46
Spectra 70/55
11/66
250
12/60
440
3/64
520
10/65
703
10/67
704
3/70
706
5/69
650
5/66
655
10/66
660
10/65
670
5/66
4700
4/69
DCT-132
5/69
IC 4000
12/68
IC 6000
5/67
IC 7000
8/70
810
9/65
810A
8/66
810B
9/68
840
11/65
840A
8/66
840MP
1/68
Systems 86
I & II
3/51 & 11/57
8/62
III
8/56
File Computers
Solid-State 80 I,ll,
90, I, II, & Step
8/58
418
6/63
490 Series
12/61
1004
2/63

AVERAGE OR RANGE
OF MONTHLY RENTAL
$ (000)
1.1
3.6
4.8
1.5
5.4
2.3
1.3
17.0
4.1
10.0
4.1
5.1
26.0
160.0
32.0
25.0
36.5
38.0
27.0
35.0
60.0
63.5
75.0
83.0
2.7
5.1
10.3
19.3
11.8
29.1
57.2
133.8
66.9
150.3
(S)

232.0
11.0
20.0
15.0
25.0
35.0
14.0
2.5
8.7
12.0
1.9
1.5
2.7
7.5
7.0
40.0
52.0
7.0
14.0-18.0
14.0-35.0
17.0-35.0
4.3
6.6
9.2
22.5
33.5
34.0
1.2
3.6
3.2
12.8
(S)
9.8
(S)
19.0
(S)
0.5
2.1
2.1
2.7
1.8
0.9
9.0
16.0
17.0
1.1
0.9
1.2
1.5
1.5
2.0
10.0
25.0
21.0
15.0
8.0
11.0
30.0
1.9

NUMBER OF INSTALLATIONS
In
In
Outside
U.S.A.
U.S.A.
World

40
50
2580
2210
420
180
156
1690
194
285
415
67
4
12
35
28
18
10

15
18
1227
1836
450
140
116
1174
63
186
148
14

55
68
3807
4046
870
320
272
2864
257

1
27
13

13
4
10
6
4690

o
5075
1260
65
480
175
9
14

o
5

45
N/A
280
9
N/A
15
8
400
150
950
1100
1100
350
16
16
12
140-290
22-50

2
4
4
3276
4
3144
498
13
109
31
4
3

NUMBER OF
UNFILLED
ORDERS

471

563
81

41
21
13
70
15
6
14
10
7966
4
8219
1758
78
589
206
13
17

o
o

20
260
300
40
64
17

70
X
90
30
13

300
45
500
1800
300
90

700
195
1450
2900
1400
440

X
X
X

100-130
1

240-420
23-51

o
24-60
90-110
68-70
65-100
84-180
1
11
115
20
26
171
14
54
23
137
41
1
19
45
8
9
4
24
111
75
3
36
31

o
23
25

1-5
35-60
18-25
20-50
21-55

o
1
20
1

31
78
14

o
o
o
o
o
o
o
o
o
o
5

25-65
125-170
86-95
85-150
105-235
1
12
175
20
27
202
49
68
23
137
41

3
11

X
0

0
X

4
23

9
4
24

216
76

32
26
X
X

38
31

o
X

X
X

36
11
628

112
86

35 E

2130

20 E

210
76
75
1502

X
X

19
45

o
o

X
20 E

NAME OF
MANUFACTURER
UNIVAC (Cont I d.)

Varian Data Machines
Newport Beach, Calif.
(A)

(2/71)

Xerox Data Systems
E1 Segundo, Calif.
(R)
(2/71)

DATE OF
NAME OF
FIRST
INSTALLATION
COMPUTER
1005
4/66
1050
9/63
1100 Series (except
1107,1108)
12/50
1107
10/62
1108
9/65
9200
6/67
9300
9/67
9400
5/69
LARC
5/60
620
11/65
620i
6/67
R-620i
4/69
520i
10/68
520/DC
12/69
620/f
11/70
XDS-92
4/65
XDS-910
8/62
XDS-920
9/62
XDS-925
12/64
XDS-930
6/64
XDS-940
4/66
XDS-9300
11/64
Sigma 2
12/66
Sigma 3
12/69
Sigma 5
8/67
Sigma 6
6/70
Sigma 7
12/66
Sigma 9

AVERAGE OR RANGE
OF MONTHLY RENTAL
$ (000)
2.4
8.5
35.0
57.0
68.0
1.5
3.4
7.0
135.0
0.9
0.5
0.4
1.6
0.5
1.5
2.0
2.9
3.0
3.4
14.0
8.5
1.8
2.0
6.0
12.0
12.0
35.0

NUMBER OF INSTALLATIONS
In
outside
In
U.S.A.
U.S.A.
World
299
637
936
62
138
200
9
8
87
1051
387
8

10-60
150-170
93-120
20
159
28-35
21-25
60-110
10
15-40

7-10
5-12
1
14
0
1
10-15
0
6-18

9
11
56
175
144
3
2
75
1300
50
150
25
7
12-62
157-180
98-132
21
173
28-35
22-26
70-125
10
21-58

24-35

5-9

29-44

3
114
822
49
0
0

NUMBER OF
UNFILLED
ORDERS
90 E
10 E
X
X
75
850
550
60

E
E
E
E

X
400
30
330
25
125

Westin - Continued from page 7

peripheral equipment and expensive consulting services to
recommend, design, and install information systems. This
may, represent a typical instance of early development costs
of a powerful tool before its true potential is realized.
Furthermore, in a capitalist system, with its basic principle
of caveat emptor, the political. answer may be that the
consumer (even the government consumer) has to learn
through some costly experiences just what the new mousetrap can and cannot do. There is at any rate nothing to
suggest that socialist or communist nations have leaped over
these early experimental stages, or that they have conducted them at lesser cost per advance.
Be all that as it may, our interviews show that a new
tough-minded attitude has been developing among government professionals, agency officials, department executives,
and key legislators 'who oversee data processing expenditures. This suggests that the first era of bandwagon effects
and easy money may be over, and that systems in the
1970's will be under much more severe requirements to pay
their way. Nevertheless, there may still remain a need - in
the interests of protecting democratic government - for
some new processes and institutions (or at least for even
greater awareness by existing agencies) to review the creation and use of computerized information systems, not
only to secure citizen rights to privacy and due process, but
also to bring the political assumptions and policy preferences embedded in data bank systems under far better
and more open public scrutiny than was true in the 1960's.
The detailed findings of our study deal with the early
effects of data bank developments on the internal structures and role relationships of the government agencies that
adopted them, with the changing relationships of those
agencies to other government agencies, with some new
patterns of intergovernmental relationships, and with the
relationship of data bank-owning agencies to legislative and
interest group processes. For these findings, and for a
general analysis of possible trends in the next five years,
readers will be able to consult our finished report.
60

CORRECTIONS
In the February 1971 i.ssue of Computers and Automation, the followi ng corrections should be made:
Page 18, Figure 5e: In W903, line 2, replace
"SUBRAPUBIC" by "SUPRAPUBIC".
Page 42, "The Golden Trumpet":
In the la st
paragraph, line 3, replace "Gold" by"Quin".
Page 49, col. 1: Replace "spectographic" by
"spectrographic" in six places - the subheading at the top of the page, and in lines
10, 20, 30, 44, and 46.
Page 63, Advertising Index, col. 2, line 1:
replace "Page 3" by "Page 2".
(The proofreader for certain parts of our February issue qualifies as Rip Van Winkle.)
TIME-SHARING PHONE BOOK
Eugene C. Gaines, Jr., Pres.
Time-Sharing 'Enterprises, Inc.
3401 Market St.
Philadelphia, PA 19104
Thank you for your letter requesting a review
copy of our Time-Sharing Phone Book.
A copy is enclosed.
I hope things are going well for you and for your
magazine.
A note of explanation -- the TIME-SHARING PHONE
BOOK is intended to attract new users to time-sharing, by making them aware of the great wealth of
time-sharing services which are as close as a local
telephone call.
The time-sharing field needs help -- and new customers -- now. I will much appreciate your mentioning "helps you get started in choosing a time-sharing service" in anything you might wish to print
about the Time Sharing Phone Book.
COMPUTERS and AUTOMATION for March, 1971

Siklossy - Continued from page 15

CALENDAR OF
COMING EVENTS
Mar. 1-3, 1971: Data Processing Supplies Association, Spring Membership Meeting, The Doral Hotel & Country Club, Miami, Fla. /
contact: Data Processing Supplies Association, 1116 Summer St.,
Stamford, Conn. 06905
Mar. 1-3, 1971: First International Symposium on Fault-Tolerant Computing, Huntington-Sheraton Hotel, Pasadena, Calif. / contact: Dr.
Francis P. Mathur, Sec'y, IEEE Technical Comm. on Fault-Toleranl
Computing, Jet Propulsion laboratory, Calif. Institute of Tech.,
4800 Oak Grove Dr., Pasadena, Calif. 91 i03
Mar. 9-13, 1971: INfL 71, the 5th International Exhibition of Industrial
Electronics, Basel, Switzerland I contact: Sekretariat INEL 71,
CH-4000, Basel 21, Switzerland

WRON_ CC

15 THE FATHER OF CE

Figure 3. A quiz on another tree proposed by the student.

trees, since it draws them so prettily. But it knows more
than has been apparent so far. The tutor can ask questions
about the tree that the student has made it draw. Figure 3
shows the tutor's comment on an answer from the student.
The student was asked to select a node from the tree and
node CE was selected. The student was asked which node
was the father of CE, and answered UFO.
I mpossible Problems

The diagnostic capabilities represent a major asset of our
teaching system. They free the designer from having to
think out all possible answers a student might be expected
to give. Maybe more important still, they are very helpful in
making use of "impossible" problems for tutorial purposes.
For example, in Figure 3, the question: "Which node is the
father of BC?" does not make sense, since BC is not the
label, but the contents of a node. Some other examples of
impossible problems are:
Giveme a set with -2 elements, (there are no such sets);
Is the set A a subset of the set (B A C), (A is not a set). We
feel that impossible problems have excellent pedagogical
value, since we wish our students to learn not only how to,
but also when to and when not to.
Conclusion

Computers know, or can be made to know, many of the
subject areas that we wish to teach. We have given examples
of two computer teaching systems that have, at their hearts,
programs that understand, in an operative sense, what they
are trying to teach. The two subject areas that we have
described are elementary set theory and computer information structures. In the latter area, the competence of the
computer tutor is enhanced by its superior display abilities.
Using a CRT, past and present events can be exhibited
concurrently, and dynamic motions can be explicated.
0
Reference
Sikl6ssy, L. Computer Tutors that Know what they Teach,
Proceedings FlCC, 1970, p. 251-255.
2 The information structures teacher is programmed in FORTRAN
on the CDC 6600 computer and uses the CDC 252 display system.
The figures are photographic prints of microfilm pictures of the
CRT.

COMPUTERS and AUTOMATION for March, 1971

March 10, 1971: Fourth Annual Symposium on .Automatic Data
Processing (sponsored by Federal Executive Board of Federal ADP
Council of New England), Sheraton-Boston Hotel, Prudential Center, Boston, Mass. / contact: Thomas T. Donovan, Air Force
Computer Operations Div. (MCCO), l. G. Hanscom Field, Bedford,
Mass. 01730
March 17-18, 1971: Spring Conference of The Association for Systems Management, Royal York Hotel, Toronto, Ontario, Canada /
contact: Donald T. laughton, Chmn., Special Conference, North
American life Assurance Co., 105 Adelaide St. West, Toronto 1,
Ontario, Canada
Mar. 22-24, 1971: Ninth Annual Symposium on Biomathematics and
Computer Science in the Life Sciences, Univ. of Texas Graduate
School of Biomedical Sciences / contact: Office of the Dean, Univ.
of Texas Graduate School of BiomediCal Sciences at Houston,
Div. of Continuing Education, P.O. Box 20367, Houston, Tex. 77025
Mar. 22-24, 1971: Numerical Control Society'S Eighth Annual Meeting
and Technical Conference, Disneyland Hotel, Anaheim, Calif. /
contact: William H. White, Numerical Control Society, 44 Nassau
St., Princeton, N. J. 08540
Mar. 22-25, 1971: IEEE International Convention & Exhibition, Coliseum
& N.Y. Hilton, New York, N.Y. I contact: IEEE Headquarters,
345 E. 47th St., New York, N.Y. 10017
Mar. 23-26, 1971: Third National Meeting of the Information Industry
Assoc., Host Farm Resort, Lancaster, Pa. / contact: Paul G. Zurkowski, IIA Washington, 1025 Fifteenth St., N.W., Washington, D.C.
20005
Mar. 29-Apr. 2, 1971: Datafair '71 Conference, Nottingham Univ.,
Nottingham, England I contact: Datafair '71 Conference Office, The
British Computer Society, 21 Lamb's Conduit St., london, W.C.1,
England
Apr. 1-2, 1971: ACM Symposium on Information Storage and Retrieval, Univ. of Maryland, College Park, Md. / contact: Dr. Jack
Minker, Computer Science Center, Univ. of Maryland, College Park,
Md. 20742
Apr. 1-2, 1971: Virginia Computer Users' Conference, Virginia Polytechnic Institute and State Univ., Blacksburg, Va. / contact: Prof.
Bruce Klein or Joe Collins, Computer Science Dept., VPI & SU,
Blacksburg, Va. 24061
Apr. 5-8, 1971: The First National Educational Technology Conference, American Hotel, New York, N.Y. / contact: Conference
Manager, Educational Technology, Englewood Cliffs, N.J. 07632
Apr. 13-16, 1971: Ninth Annual Convention of the Association for
Educational Data Systems, Royal York Hotel, Toronto, Ontario, Canada I contact: AEDS Convention, P.O. Box 426, Don Mills, Ontario,
Canada
May 3-5, 1971: Data Processing Supplies Association, Affiliate Membership Meeting, Copenhagen, Denmark / contact: Data Processing
Supplies Association, 1116 Summer St., Stamford, Conn. 06905
May 11-13, 1971: IEEE (Institute of Electrical and Electronic Engineers) 1971 Region Six Conference, Wood lake Inn, Sacramento,
Calif. / contact: Dr. D. H. Gillot, Co-Chmn, IEEE Region 6 Conference, Sacramento State College, Dept. Of Electrical Engineering,
6000 Jay St., Sacramento, Calif. 95819; or, Dr. R. F. Soohoo, Program Chmn., IEEE Region 6 Conference, Univ. of California at
Davis, Dept. of Electrical Engineering, Davis, Calif. 95616
61

Classified Advertisements

PROBLEM CORNER

360's BOUGHT AND SOLD
ALSO 360 LEASES
-

360's - 7074's - 1401's - Other Mfgs. Complete Systems or Components
GEORGE S. HcLAUGHLIN ASSOCIATES, INC.
785 Springfield Avenue
Summit, New Jersey 07901
(201) 273-5464

1316 DISK PACKS (IBM) $95-$125
Guaranteed
24 packs used until January 1971.
2-5 years old. Buy any number.
F.O.B. Walla Walla.
Contact: H.R. Magnuson,
Controller
Whitman College
Walla Walla, Wash. 99362

Walter Penney, CDP
Pro blem Editor
Computers and Automation
PROBLEM 713: A FAMI L Y PROBLEM

"What's that you're drawing - a family tree?", asked
Joe coming upon Pete deep in thought.
'''Well, in a way, yes. You might say it's a family tree of
three-bit numbers", said Pete, pointing to the diagram he
was studying.
1

IBM ENGINEER RECEIVES $50,000 AWARD
Michael J. Ambrosio
IBM Systems Development Division
Poughkeepsie, NY 12602
IBM Senior Engineer William F. Beausoleil has received a $50,000 award for an outstanding contribution. A check for the award amount was presented
to Mr. Beausoleil on Jan. 26 in Poughkeepsie, N. Y.,
for his work in devising a low-cost technique that
results in significant increases in usable monolithic circuit memory modules for computers.
During his 14 years in IBM, Mr. Beausoleil has
received nine other awards. He also holds five U.S.
and eight foreign patents.

Announcement

100

101

110

III

"It's something that came up in our Data Structure course."
"Nice, but except for showing that each number has two
descendants in the next generation I don't see much to it."
"The idea is that if we have a tree like this with all the
binary numbers from 1 to 2 n - 1 and scanned it from left
to right we'd like to know what would be the kth one we
hit", Pete explained.
"What do you mean by scanning from left to right?",
asked Joe.
"It's what's called postorder. In this example we'd have
4,2,5, 1,6,3, 7."
"Well, it shouldn't be too diffIcult to figure out the
position of any number or the number in any position."
What number is in position k?

COMPUTER DIRECTORY AND BUYERS' GUIDE, 1971
Solution to Problem 712: A New Gematria?
The 17th annual COMPUTER DIRECTORY AND BUYERS'
GUIDE issue of Compu ters and Automation, a special
directory issue, will be published in June 1971.
It wi 11 contain more than twenty kinds of valuable
reference information, including an alphabetical
Roster of Organizations in the field of computers
and data processing and a Buyers' Guide of Products and Services in the computing field.
All
listings in the DIRECTORY issue are FREE.
If your organization has recently entered this
field - or if you are not sure that we have your
organization's name on our mailing list to receive
an entry form forthis year's directory - please
use the entry fQrm appearing in the 1970 directory
issue (or a copy of it) - or wri te us at once
asking for an entry form:

Directory Editor
Computers and Automation
815 Washington Street
Newtonville, Mass. 02160
Entry forms will be mailed shortly. The closi ng date for recei pt of entry forms for thi s
year's directory is April 15, Thursday, in our
office.

The three cases which involved all six letters are CB X
ED = ABEF, DB X ED = CABF and EB X DD = CADF. Of
these only the last (corresponding to 235 X 221 = 51935)
has one letter occurring three times and the other letters
once each.
Readers are invited to submit problems (and their solutions) for
publication in this column to: Problem Editor, Computers and
Automation, 815 Washington St., Newtonville, Mass. 02160.

ADVERTISING INDEX
Following is the index of advertisements. Each item
contains: name and address of the advertiser / page
number where the advertisement appears / name of
agency, if any
ACADEMIC PRESS, III Fifth Ave., New York, N.Y. 10003
/ Page 2 / Flamm Advertising
FROST & SULLIVAN, INC., 106 Fulton St., New York,
N.Y. 10038/ Page 3/ Austin Kelley Advertising
NEW YORK TIMES Book & Education Div., 229 West 43
St., New York, N.Y. 10036 / Page 64 / Kingen
Feleppa O'Dell
PROFESSIONAL & TECHNICAL ·PROGRAMS, INC., 866 Third
Ave., New York, N.Y. 10022 / Page 27 / Henderson
& Roll, Inc.
WM. C. BROWN COMPANY PUBLISHERS, 135 S. Locust St.,
Dubuque, Ia. 52001 / Page 63
I

COMPUTERS and AUTOMATION for March, 1971

.......

__ ............_ . _ - - - - - - - - - - - - - -

****************************

NUMBLES
NUMBER PUZZLES FOR NIMBLE MINDS
-AND COMPUTERS
Neil Macdonald
Assistant Editor
Computers and Automation

Los Angeles Trade - Technical College
and V.C.L.A. Extension

A "numble" is an arithmetical problem in which: digits
have been replaced by capital letters; and there are two
messages, one which can be read right away and a second
one in the digit cipher. The problem is to solve for the
digits.
Each capital letter in the arithmetical problem stands for
just one digit 0 to 9. A digit may be represented by more
than one letter. The second message, which is expressed in
numerical digits, is to be translated (using the same key)
into letters so that it may be read; but the spelling uses
puns or is otherwise irregular, to discourage cryptanalytic
methods of deciphering.
We invite our readers to send us solutions, together with
human programs or computer programs which will produce
the solutions. This month's Numble was contributed by:
Stuart Freudberg
Newton High School
Newton, Mass.
NUMBLE 713
MAN I S

H CRT H H
CFL

= STM = CTN

S 0 AHCS
R ADD I R D H

Prob. 600 pages-7V4" x 9Va"-cloth-1971-Prob. $9.95

INTRODUCTION TO DATA PROCESSING emphasizes the
fundamental principles, the importance of data processing
and the necessary guides for successful data processing
appl ications.
From the introductory explanations the reader is ;·ven a
concise, clear and basic understanding of data proc ssing,
and the important aspects of the field. To facilitate co plete
coverage of the area, subjects such as data processing history, the role of punched card, data representation in the
computer, computer components and programming, modern
third generation programming languages (BAL, COBOL, RPG,
FORTRAN, PL1), are covered.
Emphasis throughout is on business data processing applications. The final section deals with the management function of a computer system.

Of Special Note ... Workbook, Teacher's Manual Available

•

• No previous knowledge of data processing is required.
Comprehensive presentation of important programming
languages.
Inclusion of IBM/3, a most recent added dimension to
Punched Card Data Processing.
Coverage of BASIC programming language.

**
**

**
**
**
**
**
I

480 pages-8V2" x 11"-paper-1969-$8.75

THE

AF E I S L

*t
UJcb brings you
t three books by •••
*t Carl Feingold, CPA, CDP
*** ~~ INTRODUCTION TO
I** '4IlII DATA PROCESSING
*
**
**
*
***
***
* ..
** .
** FUNDAMENTALS
OF PUNCHED
CARD
DATA
PROCESSING
***
**
#
*

4278

0963

5595

Solution to Numble 712

In Numble 712 in the February issue, the digits 0
through 9 are represented by letters as follows:
S=O
F,V=5
R=1
N=6
H=2
E=7
T=3
0=8
D=4
U=9
The message is: The fed hound never hunts.
Our thanks to the following individuals for submitting
their solutions - to Numble 711: C. L. Agrawal, Claymont,
Del.; Marijoe Bestgen, Riverdale, N.Y.; Debra Bruno, Cliffside Park, NJ.; T. P. Finn, Indianapolis, Ind.; John H.
MacMullen, Eden Prairie, Minn.; L. J. Mathiason, Chillicothe, Ohio; G. P. Petersen, st. Petersburg, Fla.; Vincent
K. Roach, New York, N.Y.; Harold L. Smith, Thomson,
Ga.; and Robert R. Weden, Edina, Minn. - to Numble
7012: Marijoe Bestgen, Riverdale, N.Y. and G. P. Petersen,
St. Petersburg, Fla. - to Numble 7011: Bill Call, Saginaw,
Mich. and Lee Olson, Delaware, Ohio - to Numble 7010:
SSG Raymond L. Cowen, Gunter AFB, Ala. and Krishna
Moorthy, Kanpur, India.
COMPUTERS and AUTOMATION for March, 1971

I
I

**
**
**
**
**
**
**
**
**
***
**#

FUNDAMENT ALS OF PUNCHED CARD DATA PROCESSING
presents the basic operations of punched card equipment.
Initially, the book explains the various punched card machines as to operations as well as the wiring of necessary
control panels. Then it considers the relationships of punched
card equipment and the computer, emphasizing such items
as flow charting principles. Each chapter contains the operational and panel wiring. A Teacher's Manual is available.

Of Special Note ...

• Composite catalogue of basic machine operations which
lists entire operation of each machine.
• Presents wiring of necessary control panels with operational and panel wiring in each chapter.
• Shows relationship and function of machines.
• Considers relationship of punched card operations to computer processing.
Teacher's Manual contains: Key to questions after each
chapter in text, answers to problems at end of text. Semester project and overall solution to it, and teaching suggestions.

FUNDAMENTALS OF
COBOL PROGRAMMING
272 pages-8V2" x 11"-palJer-1969-$6.95
FUNDAMENTALS OF COBOL PROGRAMMING is written in
simple outline form and describes the characteristics of
COBOL and provides rules for writing in COBOL. Illustrative
programs and examples of COBOL statements are included.
Geared to the 360 Computer, this text contains detailed
explanations accompanied by examples with step-by-step
presentation of rules for writing in COBOL.

Of Special Note ...
• Includes latest COBOL specifications and all new improvements of COBOL.
• Describes characteristics of COBOL in detail.

For examination copies write Mrs. Carolyn Gantz

VVM. C. BROVVN
COMPANY PUBLISHERS
135 South Locust Street,
Dubuque. Iowa 52001

***
**
**

**
**

**
**
**
**
**

I
I

**************************

The New York Times and
Computers and Automation
announce a practical guide to
the most elusive components in
computers and data processing ...

PEOPLE

Who they are ...
What they do ...
Where they do it ...

I
\

.. ,

•

•i

~ 15
~

fnH

!§H

~I !~

~ ~;

~
~

l.l<.

f§

~ =~
I
~

~gij
1~.

BH
g~c
aD

!1lI

tilz
~c

:1
~~

~~
~ifj

I.~

»
0
t

~
~

ffi

Until now, it has been well-nigh impossible to keep track of the thousands of highly skilled professionals
engaged in the world's fastest growing profession.
The painstaking task required to inventory the qualifications and backgrounds of the 15,000 most necessary professionals in every branch of
the computer field has now been accomplished. The oldest magazine in
the field, Computers and Automation, and the information retrieval
services of The New York Times
have pooled their resources to produce the Fifth Edition of

WHO'S WHO IN
COMPUTERS AND
DATA PROCESSING
This is the most extensive register of
computer professionals ever published - the first of its kind in seven
years. It is arranged for your convenience in three volumes:
1. Systems Analysts and Programmers
2. Data Processing Managers and
Directors
3. Other Computer Professionals
(from professors of computer science to attorneys versed in the
computer field)
Each of the more than 15,000 specialists is covered by a separate capsule biography detailing: Birth Date
... Education . .. Year Entered Computer Field ... Title ... Honors ...
Memberships . .. Special Skills (from
applications to logic to sales) ...

PLUS both home and business addresses. For example:
CHAPIN, Ned / consultant / born: 1927 /
educ: PhD, lIT; MBA, Univ of Chicago /
entered computer field: 1954 / main interests: applications, business, logic, management, programming, systems, data
structures / title: data processing consultant / organization: InfoSci Inc, Box 464,
Menlo Park, CA 94025 / publications,
honors: 3 books, over 50 papers; member,
over 12 associations; CDP; lecturer for
ACM / home address: 1190 Bellair Way,
Menlo Park, CA 94025

This unique reference is particularly
useful not only for personnel managers, employers and recruiting organizations but also for libraries,
conference planners, directors of
computer installations . . . anyone
who needs to keep up with the peopIe who count in the field.

Reduced Pre-publication Price.
h
.
bill b
T e reservatIOn form e ow wi ring
you the 3-volume set at the reduced
price of $60 ... an immediate saving of $15 from the regular price of
$75. There is no risk involved. 10Day Free Examination.
WHO'S WHO IN COMPUTERS
AND DATA PROCESSING
Edited by Edmund C. Berkeley
3 volumes in durable hard-cover
bindings

Retail price $75 the set

Pre-Publication Price $60
- --- - - - - - - - - - - - - - - - --- - - - - - - Who's Who in Computers and Data Processing,

-",

815 Washington St., Newtonville, Mass. 02160
Please send 3-volume set WHO'S WHO IN COMPUTERS AND DATA
PROCESSING at pre-publication price of $60 (regularly $75). You will bill
me after delivery and I may return the set within 10 days if not fully satisfied.

l
t

NAME

TITLE
ORGANIZATION
ADDRESS
CITY
YOUR SIGNATURE

STATE & ZIP
P.o. # (IF COMPANY ORDER)

o Check here if you also wish to receive
COMPUTER DIRECTORY AND BUYER'S GUIDE Comprehensive geographical listing of: Computer and data-processing services, commercial time-shared services, courses
and seminars, consulting services, leased equipment, software, campus computer centers,
computer associations, users groups. PLUS: More than 2,000 applications of computer
and data processing equipment, a world computer census - and more! Order now at $20
L~~~~icatio~25~ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _~

I
)

Source Exif Data:

File Type                       : PDF
File Type Extension             : pdf
MIME Type                       : application/pdf
PDF Version                     : 1.3
Linearized                      : No
XMP Toolkit                     : Adobe XMP Core 4.2.1-c043 52.372728, 2009/01/18-15:56:37
Producer                        : Adobe Acrobat 9.1 Paper Capture Plug-in
Modify Date                     : 2009:03:24 21:45:23-07:00
Create Date                     : 2009:03:24 21:45:23-07:00
Metadata Date                   : 2009:03:24 21:45:23-07:00
Format                          : application/pdf
Document ID                     : uuid:7d43d3e2-61fd-490e-b3e2-0f62a10f1f02
Instance ID                     : uuid:4363ff25-29db-47c7-a46a-69e31fed8d1e
Page Layout                     : SinglePage
Page Mode                       : UseNone
Page Count                      : 60

EXIF Metadata provided by EXIF.tools

197103

197103 197103

Navigation menu

Versions of this User Manual:

Views

Navigation