197009

197009 197009

User Manual: 197009

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

Download
Open PDF In Browser	View PDF

September, 1970

t, - z:d EiBitXiiq

Vol. 19, No.9

ut e Sr!&

la~d automation

r£ HNOLOGY

JICALS SEC

1263385628

°UBL LIBRARY
. RLOS

"Cedar Grove"

0104

ST
*01270
95113

PICTURE PROCESSING AND PSYCHOPICTORICS
edited by BERNICE SACKS LIPKIN, Bethesda, Maryland, and AZRIEL ROSENFELD,
University of Maryland, College Park, Maryland
This book brings together 'papers on pictorial pattern perception and picture processing.
The contributors discuss the factors that influence the detection of objects in complex
images, and computl\r techniques for manipulating these factors. They also describe
programming systems and computer algorithms for image processing and on-line experimentation. The book covers three classes of psychophysical variables: contrast and
border; shape and geometry; and texture. It first reviews psychological data on the role
of each factor in perception, and then treats the computer analysis and manipulation of
images with respect to the factor. Third Quarter 1970.

COMPUTER TECHNIQUES IN IMAGE PROCESSING
by HARRY C. ANDREWS, Department of Electrical Engineering, University of Southern
California, Los Angeles, California
with contributions by WILLIAM K. PRATT, Department of Electrical Engineering, University of Southern California, Los Angeles, California, and KENNETH CASPARI, ITT Electro
Physics Laboratories, Hyattsville, Maryland
This treatise presents methods of digital computation for digital image processing. It
borrows techniques from classical Fourier optics, linear systems, communications, and
orthogonal transform theories. Some of the valuable features contained in the book are:
a description of optical techniques for image enhancement, emphasizing the earlier results
of diffraction and diffraction-limited imaging systems; three different implementation
techniques-one entirely optical, a second combining optical and digital techniques, and
one purely digital; studies of digital image coding for both digital storage and communication and of the theory of image coding. August 1970, 187 pp., $10.50.

METHODOLOGIES OF PATTERN RECOGNITION
THE PROCEEDINGS OF THE INTERNATIONAL CONFERENCE ON METHODOLOGIES OF PATTERN
RECOGNITION HELD AT HONOLULU, HAWAII, ON JANUARY 24-25-26, 1968.

edited by SATOSI WATANABE, University of Hawaii, Honolulu, Hawaii
Pattern recognition, once considered as being nothing more than a mystifying computer
stunt, is quickly becoming a respectable branch of scientific art. This book, which presents articles written by thirty outstanding authorities, evaluates the present and discusses
the future of pattern recognition. Each author emphasizes the "philosophy" of his approach rather than the mathematical derivations and experimental data, and provides
the reader with a self-contained survey of lasting value. 1969, 579 pp., about 150 figures
and illustrations, $16.00.

RECURSIVENESS
by SAMUEL ElLENBERG, Department of Mathematics, Columbia University, New York,
New York, and CALVIN C. ELGOT, IBM Thomas J. Watson Research Center, Yorktown
Heights, New York.
This monograph provides an algebraic development of elementary aspects of the theory
of recursive functions. Its algebraic approach will contribute greatly to the long range
goal of developing a theory for digital computer programs using recursive functions.
Readers familiar with finite automata theory or mathematical linguistics will note that
operations utilized in this monograph also play a central role in those studies. 1970, 89
pp:, $6.50.

ADVANCES IN COMPUTERS
Series Editors: FRANZ L. ALT, American Institute of Physics, and MORRIS RUBINOFF,
University of Pennsylvania and Pennsylvania Research Associates
VOLUME 10
Guest Editor: WALTER FREIBERGER, Center for Computer and Information Sciences, Brown
University, Providence, Rhode Island
CONTENTS: CHARLES DeCARLO: Humanism, Technology, and Language. PETER WEGNER:
Three Computer Cultures: Computer Technology, Computer Mathematics, and Computer Science
BRYAN THWAITES: Mathematics ill 1984-The Impact of Computers. E. E. DAVID, Jr.: Computing from the Communication Point of View. FREDERICK P. BROOKS, Jr.: Computer-Man
Communication: Using Computer Graphics in the Instructional Process. ANDRIES VAN DAM
and DAVID E. RICE: Computers and Publishing: Writing, Editing and Printing. ULF GRENANDER:
A Unified Approach to Pattern Analysis. ROBERT S. LEDLEY: Use of Computers in Biomedical
Pattern Recognition. WILLIAM PRAGER: Numerical Methods of Stress Analysis. J. H. AHLBERG:
Spline Approximation and Computer-Aided Design. DANIEL L. SLOTNICK: Logic Per Track
Devices. Author Index-Subject Index.
September 1970, about 290 pp., $14.50.

ACADEMIC
.
PRESS ' 'i~ ,~.
formation is in the box on the
i shaved with a blade. Only way
Nhiskers off. But for the post two
19 the Abercrombie 60-Second
ose, clean, comfortable shove, all
lng into my shirt 60 to 62 seconds
'hat's the best port.

Inside this box, a new

last Wednesday, Ronnie D
to us, tried the Abercrombi
Did it in 56

second~.

Another friend took 64 sec(
First time we tried the Aberc
cleaned our face smooth, c
And you're right - somebo
shoving is about the dullel
thing a man does. And we
thing that would keep yo\.
we did the next best thingwhiskers off your face - cl
a matter of seconds.

over with so fast.
,d Shove comes in a genuine leo.
case just by itself - makes a fine

---'----------.

~BIE' & FITCH

Iton, Iowa 52732
tpaid
Abercrombie 60~ 32.S0each.
[oney Order

Result! The Abercrombie
shave you closely and (
can return it, simple as th
Some friends in Europe I
Second Shave for us. Tt
heads - instead of the
difference you see. Flip
start purring - covers m
method of shoving we'vl
Each head is covered w
foil screen. A sort of sw

__________

~Zip,

________

haver to cover cost of handling
Jew York, New Jersey, Florida,
;olorado and California add ap-

112-299-7-11111610111

.. -_ .... _------,----,

Vol. 19, No.9
September, 1970
Editor

Edmund C. Berkeley

Associate Editor

Sharry langdale

Assistant Editors

linda Ladd Lovett
Neil D. Macdonald

Software Editor

Stewart B. Nelson

Advertising Director
Art Directors
Contributing Editors

CDI"I~uters
and automation
The magazine of the design, applications, and implications
of information processing systems.

Bernard lane
Ray W. Hass
Daniel T. Langdale
John Bennett
Moses M. Berlin
Andrew D. Booth
John W. Carr III
Ned Chapin
Alston S. Householder
Peter Kugel
Leslie Mezei
Rod E. Packer
Jean E. Sammet
Ted Schoeters
Richard E. Sprague

Advisory Committee

T. E. Cheatham, Jr.
James J. Cryan
Richard W. Hamming
Alston S. Householder
Victor Paschlc:is

Fulfillment Manager

William J. McMillan

Computers and Privacy
14

REGULATIONS FOR INFORMATION SYSTEMS
[A]
by Prof. Calvin C. Gotlieb, Univ. of Toronto
A classification of information systems is proposed, along with
an examination of the goals, methods, and costs of regulating
information systems.

The Computer Industry
18

PROBLEMS OF LIABILITY FOR THE EDP SERVICES INDUSTRY
[A]
by Milton R. Wessel, Attorney
The attorney for the Association for Data Processing Service
Organizations (ADAPSO) discusses the responsibilities in the
1970's of the EDP services industry to: the public, its customers, its shareholders and owners, its employees, and its
suppliers.

THE USER/MANUFACTURER INTERFACE
[A]
by David F. Stevens, Lawrence Radiation Laboratory
The current state of the uneasy alliance between manufacturers
of computer systems and their users - with some specific suggestions on how to improve the relationship between the users
and the manufacturers.

COMPUTERS, AUTOMATION, AND COMPUTER·ASSISTED ACTIV.
ITIES
[E]
by Edmund C. Berkeley, Editor, Computers and Automation
.
Implications of the change in emphasis from "automation" to
"computer-assisted".

RELIABILITY OF INFORMATION IN C&A by Norwood Jones and the Editor

Adt'el'lising Contact
THE PUBLISHER
Berkeley Enterprises, Inc.
815 Washington St., 617-332-5453
Newtonville, Mass. 02160

Editorial Offices
BERKELEY ENTERPRISES, INC.
815 WASHINGTON STREET,
NEWTONVIllE, MASS. 02160
CIRCULATION AUDITED BY
AUDIT BUREAU OF CIRCULATIONS

Computers and Automation is published monthly
(except two issues in September) at 815 Washington St., Newtonville, Mass. 02160, by
Berkeley Enterprises, Inc. Printed in U.S.A.
Subscription rates: United States, 11 monthly
issues and two issues in September (one of
which isa directory issue) - $18.00 for 1 year,
$36.00· for. 2 years; 12 monthly issues (without
directory issue in September) - $9.50 for 1
year; $18.00 for 2 years. Canada, add 50¢ a
year for postage; foreign, add $3.50 a year
for postage. Address all U.S. subscription mail
~to: Berkel,ey Enterprises, Inc., 815 Washing.
ton St., Newtonville, Mass. 02160. Second Class
(>ostage paid at Boston, Mass.

.' 10

SJCC 71 - CALL FOR PAPERS
by Dr. Nathaniel Macon

COMMENT

[F]

Computer Programs and Syste'ms
22

SYSTEMS TEST
[A]
by Dwight M. Frazier, Jr., IBM Corp. Systems Mftg. Div.
Why computer systems should be tested before they are installed - and specific guidelines for setting up such systems
tests.

COMPUTERS AND CHEMICAL NOTATIONS
by Charles E. Granito

Postmaster: Please send all torms 'J!i79 to Berkeley
Enterprises, Inc., B15 Washington St., Newtonville,
Mass. 02160. © Copyright 1970, by Berkeley Enterprises, Inc.
Change of address: If your address changes, please
send us both your new address and your old address
(as it appears on the magazine address Imprint), and
aI/ow three weeks for the change to be made.

[F]

Computer Applications
28

THE "LANGUAGE EXPERIENCEII APPROACH IN TE,ACHING READ·
ING - COMPUTERIZED
[A]
by Mrs. Florine L. Way
How the computer is being used to help teachers use the pupils'
own language to create the content of their reading lessons a prize-winning essay in C&A's Martin Luther King Memorial
Prize Contest.

Computers and Society
33

OUR TOP PRIORITY

[A]

by William B. Johnson, Pres. and Chrmn. of the Bd., IC Industries Inc.
Why the key element of decision-making - the art of choosing
priorities - must be applied on a national level in order to
.find objective, balanced solutions to problems and terminate
obsolete activity.

"WHAT WE MUST DOli - COMMENT
by John McLeod, Editor, Simulation

[F]

AUTOMATED POLICE STATE
Reprinted from Dataweek magazine

[F]

Computers, Science, and Assassinations
39

PATTERNS OF POLITICAL ASSASSIN.ATION: How Many Coinci ...
dences Make a Plot?
[A]
by Edmund C. Berkeley, Editor, Computers and Automation
How the science of probability and statistics can be used as an
instrument of decision to determine if a rare event is: (1) within
a reasonable range; (2) unusual
strange or suspicious; or
(3) the result of correlation or cause or conspiracy. [See detailed contents on page 39.]

This Month's Cover
The front cover drawing was
one of the entries in C&A's Eighth
Annual Computer Art Contest
(see the August issue). The artist
is Lloyd Sumner, Computer Creations, P.O. Box 1842, Charlottesville, Va. 22903. The drawing
was programmed in ALGOL and
produced with the aid of a B5500
computer and a CalComp 565
plotter.
Lloyd Sumner is one of the
first people to devote his full time
to the development of computer
art. He has exhibited his work
throughout the world, and has
lectured on computer art in Europe and the United States. He
holds a degree in Engineering
Science from the University of
Virginia.

COMPUTER·ASSISTED ANALYSIS OF EVIDENCE REGARDING THE
ASSASSINATION OF PRESIDENT JOHN F. KENNEDY - PROGRESS
REPORT
[A]
by Richard E. Sprague
A report on the nature and progress of a computer project
initiated at the National Committee to Investigate Assassinations.

Der)(~1'tments

Advertising Index

Across the Editor's Desk
Applications
Education News
Research Frontier

Calendar of Coming Events

Classified Ads

CORRECTION

[F]

Computers Abroad
32

REPORT FROM GREAT BRITAIN

[C]

by Ted Schoeters
A look at the development of data transmission networks in
the UK, including the new Post Office Corp. network, and the
facilities of the National Physical Laboratory.

Letters to the Editor

Monthly Computer Census

New Contracts

New Installations

NewProduds and Services

Punch lines •.•

Key

Computers and Fun
10

NUMBLES

[C]

by Neil Macdonald, Asst. Editor, Computers and Automation

PROBLEM CORNER
by Walter Penney, CDP

[C]

[A] -

Article

[F] -

Readers' Forum

[E] -

Editorial

[C] -

Monthly Column

EDITORIAL

Computers~

Automation, and
Computer-Assisted Activities
In "the old days" of the computer field, which is about
20 years ago, there was much talk about the "automatic
factory", the "automatic office", and "automation". There
was a continuing argument about who had composed the
word "automation", and there was another continuing
argument about whether automation really was anything
newer in 1950 than it had ever been before.
Ten years ago we published a definition of the word
"automation", in the 5th edition of the "Glossary of Terms
in Computers and Data Processing":

could do something all by themselves, nowadays there is
more emphasis on the contribution of the human being and more effort to develop the system of "man and
computer working together", each aiding the other.
Is this change permanent or temporary? Will machines
by themselves unaided by human beings take over almost
every productive activity? In the future, will we really have
large-scale automation, or will we have large-scale computer-assisted activities?

1. Process or result of rendering machines self-acting
or self-moving.
2. Theory or art or technique of making a device or a
machine or an industrial process more fully
automatic.
3. Making automatic the process of moving pieces of
work from one machine tool to the next.

I think the present fashion is a temporary turn in the
spiral of development. I think in years to come there will be
greater and greater strides towards full automation.
Here are a few of the reasons:

This last kind of automation was sometimes referred to
as "Detroit automation", because at one time there was a
famous automatic assembly line for automobile engines.
This production line had approximately 500 stages or
stations; it took in at the front end a raw casting for an
automobile engine, and put out at the far end a fully
finished and inspected engine for installation in an automobile being constructed.
Nowadays, the "in" word is "computer-assisted". There
are dozens, and perhaps even hundreds, of processes and
activities which are called "computer-assisted". A few of
these activities have almost become fields in themselves,
such as
computer-assisted instruction;
computer-assisted education;
computer-assisted medical diagnosis;
computer-aided design;
computer-assisted laboratory instrumentation; etc.
A definite change in viewpoint has occurred. Whereas
the old viewpoint implied that machines and computers

1. Computers will become far cheaper than they are
now;
2. More and more programming will include provisions for taking care of more and more exceptions;
3. More and more activities still performed by human
beings, such as visual observing and recognizing of
objects, will become mechanized, because engineers will find out how to mechanize them. In
fact, what seems to be today a permanent hard
core of difficult tasks to give to computers, will
yield to technical advances.
4. Human beings want to live more, play more, and
work less, especially work less in the sense of
having uninteresting monotonous tasks to do, like
picking grapes or sweeping subway platforms.
5. Repetition of work soon becomes monotonous for
any human being - but it is duck soup for a
machine which regularly has no emotions designed
into it; etc.
Prediction for twenty years from now: Less and less
"computer-assisted" activities - more and more full
automation.
What will human beings do? More and more of what
they like, less and less of what they must.

COMPUTERS and AUTOMATION for September, 1970

Letters To The Editor
"The Limitations of Computers" Comment
I found your magazine in the new
library on Michigan State University's
campus in East Lansing: I read your
editorial in the June, 1970 issue ["The
Limitations of Computers", page 6],
and I believe your three points at the
end of the editorial are very well said
and I enjoyed reading them.
Teachers, lawyers, doctors, dentists,
accountants and others all have to be
certified by the states in which they
practice. What is being done about the
computer operators and programmers?
I would be interested in hearing from
anyone who is working on the certification by a state board of computer
professionals.
THOMAS R. ZICK, SR.
613 Cherry St.
Stockbridge, Mich. 49285

Articles That Would Never Appear
Elsewhere
Presently I am engaged in research
involving the future use of mini-computers in large MIS configurations. The
finished product of my research may
be of interest to your readers, and I
will send you a copy for consideration
for publication.
Regardless of your decision regarding my paper, I intend to continue as
one of your subscribers. Perhaps it is
because you maintain a paid readership that you are able to publish articles that would never appear elsewhere.
STANLEY JAFFIN
211 N. Piedmont St.
Arlington, Va. 22203

Patents on Software
In the February issue of Computers
and Automation, Howard A. Si1ber
expressed his feeling that if a computer program were translated into another computer language, it would circumvent copyright protection. [See
"A Hypothetical Interview Between
the President of a Computer Software
- Company and a Patent Attorney Specializing in Protection of Computer
Programs", Feb. issue, p. 16.] This is
not true with international copyright,
which protects the "story line". Inter-

national copyright provides protection
only if the same function (book) is
performed by essentially the same
steps (story incidents, sentences). I
believe a significant court case could
be built by showing that nearly all
subroutines in the copy of a computer
program did like jobs, with like calling
sequences, as did the original program.
If such likeness cannot be shown, I
believe that even patent protection
would be futile for that program, since
one would be attempting to exclude
something derived by an apparently
independent mental process. I would
be interested in 'hearing comments
from others on this subject.
RICHARD MERRILL
34 George Rd.
Maynard, Mass. 01754

Railroads Seek Tougher Terminal
Railroads want anew, tougher terminal device to communicate data
from yards to computers and other
terminals. So the Management Systems
Dept. of the Association of American
Railroads has put together a verbal
picture of what's needed.
Terminals for the input and output
of all kinds of data - waybill information, train consists; payroll reports and
the like ~ are noW being used by
railroads in some 2,000 locations.
The kind of device needed would
be able to stand up under conditions
found in the typical railroad yard substantial vibration, much dirt, wide
variations in heat and humidity - with
a minimum of maintenance.
Another desirable feature would be
suitability for installation in the field
and modular features that would allow
a user to operate either a simple basic
device or to add compatible
components.
The AAR isn't seeking bids on the
deVice, but is sending the specifications to member roads and potential
manufacturers. A vendors' meeting
will be scheduled if sufficient interest
is shown.
Any of your readers that are interested are invited to write to me for
copies of our specifications.
R. A. PETRASH, Executive Director
Data Systems Div.
Association of American Railroads
Washington, D.C. 20036

XEROX

FicheFile
When We designed a total microforms system for a major farm
equipment manufacturer, we developed the FicheFile to meet their
microfiche storage needs. Now, for
the first time, 4 x 6" microfiche can
be instantly located, removed,
cleaned, lubricated, and refiled in
an attractive, rotating desktop unit.
SUPERB PROTECTION. Fiche are
sheltered in individual dust- and
scratch-proof compartments.
MAXIMUM FLEXIBILITY. Three
basic capacities to match all storage requirements. Choice of grey
or light tan.
SPILLpROOF. Fiche are gently but
firmly gripped to prevent spillage.

XEROX FicheFiles
100 capacity FicheFile .... $69.50
64 capacity FicheFile .... 59.95
32 capacity FicheFile .... 49.95
8 capacity expansion units.. 8.00
F. O. B. Ann Arbor, Michigan

Microforms
Systems
A Service of University Microfilms
Xerox Education Group
204 North Zeeb Road
Ann Arbor, Michigan 48106
313: 761-4700
"Xerox" is a registered trademark of Xerox corporation·

COMPUTERS and AUrOMATION for September, 1970

READERS' FORUM

"WHAT WE MUST 00" - COMMENT
I. To the Editor, from John McLeod, Editor, Simulation,
and Executive Director, SCi World Simulation, Simulation
Councils, Inc., P. O. Box 2228, La Jolla, Calif. 92037

I appreciate your republishing "What We Must Do" by
Dr. John Platt [June issue, p. 41]. I had some strong
reactions to that article, which I expressed in a letter to the
author [see below] .
Incidentally, when I read your very good editorial, "The
Limitations of Computers" [June issue, p. 6], I was
reminded that I had not complimented you on having the
guts, or whatever it takes, to publish the Sprague article in
your May . issue. That was a most remarkable - and
frightening - study. But, regardless of the proof, there's
one thing that I simply can't believe, and that is that 50
people could be involved in a conspiracy of that complexity
without somebody "blowing" it!
II. Letter to Dr. John Platt, Mental Health Research Institute, Univ. of Michigan, Ann Arbor, Mich. 48104, from
John McLeod.
I have just finished reading the Computers and Automation reprint of your November 28, 1969 Science article
(which I somehow missed).
I am impressed by two things: the similarity of our
thinking on the subjects which you discussed; and your
failure to recognize - or at least to emphasize - the fact
that computers, which you acknowledge are a part of the
technology that "did not create human conflicts and
inequities, but ... has made them unendurable", might be
the most powetful tool at our disposal to "devise new
mechanisms" and "better ways of heading off these multiple crises ... ".
I do not believe that "the application of the full
intelligence of our society is likely to be adequate" if that
intelligence is unaided. I firmly believe that it is beyond the
capability of the human mind to assimilate, correlate,
recall, and manipulate all the information required to make
the intelligent decisions that will be necessary in the crisis
situations which you describe. I firmly believe that, by and
large, our decision-makers would select the best (or at least
8

better) courses of action if they could thoroughly understand the situations. with which they are dealing, and the
consequences of their actions - but first there must be
understanding!
Now I will admit to a strong bias (born of 20 years of
experience in the synthesis and analysis of complex systems) in favor of computer modeling and simulation, the
latter being the use of the model for scientific experimentation. However, although you call for "inventive minds, who
can put together methods, organizational designs, or 'social
inventions' " you do not mention simulation by name. So,
instead of our looking at what I recommend as simulation,
let's look at it as a special kind of information storage and
retrieval system.
In this system we will enter all of the pertinent information relative to the problem under study, but we will also
enter information concerning how the individual pieces of
data are related to, and influence other pieces of data~ and
on what time scale. Then, although input data will be
available for retrieval if desired, the normal outputs of the
system will not be this raw data but the effect on data of
interest when the input data is modified. In other words,
the system will answer the question, "what if?". What if
some input data were in error?, or what if it were to be
modified by changing conditions? You will, of course,
recognize the foregoing system as a computer model, and
the "what if" exercise as a simulation.
Philosophy aside, the time for action has passed, and
very little action has been taken. I stated my feelings on the
subject a year ago when I proposed a simulation of the
world. Current progress, I think, is rather encouraging in
view of the fact that we have had no outside funding. Now,
however, circumstances preclude further financial support
from Simulation Councils; and how fast we can progress on
the basis of the voluntary efforts of the Trustees and others
who have expressed an interest remains to be seen. I have,
therefore, prepared a 'preliminary proposal in the hope of
obtaining a grant to carryon our work. I would welcome.
your constructive criticism.
I hope that you will find our efforts along the lines of
those for which you have so lucidly called, and that you
can be interested in helping us make this project "fly" - in
time!
0
COMPUTERS and AUTOMATION for September, 1970

AUTOMATED POLICE STATE
R~printed with permission from Dataweek, Business Publications Ltd., Mercury House, Waterloo Rd., London SE1,
England, Vol. 11, No. 10, June 3, 1970.

During May, four students were killed by National
Guardsmen at Kent State University in Ohio. Hard-hatted
construction workers demonstrating their support for President Nixon in New York City bashed peace demonstrators
with lead pipes. At Jackson, Mississippi, two black students
died and at least a dozen were wounded when ,police
opened fire on a crowd with deadly 00 buckshot, reaching
far beyond self-defense or crowd control. Augusta, Georgia,
suffered a Watts-type riot, and before it was under control
six blacks were dead, allegedly shot in the back.
Respected American visitors,' short haired, capitalist,
often Republican, use words like "revolution", "dictatorship", "polarization", in a way that is chilling.
In this context, the innocuous installation of computer
systems in Los Angeles and Sacramento to help California

law enforcement officers check out stolen' cars and su
pected criminals brings mixed emotions.
On the one hand, we worry that people will aSClll.J<': Lc
the computer great, menacing teeth, and a ferocious snarl
that is not at all characteristic of the dumb beast.
On the other hand, we worry that people will not
comprehend the potential for abuse of a near-perfect,
on-line network. It's 14 years yet to 1984, but events in the
past year on the social, technical, and ecological fronts
make them look like very short years indeed. The power of
the computerized information system, and the fear it
engenders, may be very strong weapons in the California
police arsenal.
The system is admirable, a feat of technology, and it
represents a successful merging of private industry and
government applications. But great care must be taken to
ensure that it works for the good of all.
The message is, don't be afraid of computers, people;
0
just be afraid of people.

RELIABILITY OF INFORMATION IN C&A - COMMENT
I. From Norwood Jones
Data Processing Officer
First National Bank
EI Dorado, Ark. 71730

As a user of Burrough's equipment, I have resolved each
time I have read your "Monthly Computer Census" figures
to write you. Until now the letter has never been written,
but here it is. The method used in listing Burrough's
equipment and the data in your "Computer. Directory and
Buyers' Guide" illustrate a lack of knowledge of their
products. I am referring to the models listed as B-IOO,
B-200, B-300, and B-500.
It would be easy for one unfamiliar with Burrough's
equipment to believe these were separate computers, when
in reality they are variations of the same machine. In fact,
the name as often refers to a contract agreement as to a set
of machine specifications. For example, I can change my
B-300 in to a B-500 by signing a new contract while keeping
all my present hardware. A similar situation would be to list
all the models of the IBM 360/20 series as separate
computers.
Burroughs refers to this series of machines as the
B-IOO/B-500 group. The CPU on all these machines is
basically the same. The basic instruction set and character
representation is identical, but some versions have a larger
instruction set and added capabilities. Cycle time is either
10 or 6 microseconds and core memory comes in three
sizes: 4.8K, 9.6K or 19.2K.
Your effort in presenting the true pictures of these
machines would assist in correcting a distorted image of an
excellent machine.
With reference to your list of characteristics of this
COMPUTERS and AUTOMATION for September, 1970

equipment in your Directory, the following changes need to
be made:
Bits per digit = 6.
Digits per Alphabetic = 1.
Word length = 12 digits.
Number of addressable positions in memory =
4.8K, 9 .6K, or 19.2K.
(e) Access Time = 6 u or 10 u.
(a)
(b)
(c)
(d)

The misinformation on equipment that I know about
caused me to distrust your information on the other
equipment in your Directory. This is one reason why I did
not order it this year.
II. From the Editor
Thank you very much for your most helpful letter.
The processes which we use for gathering information
for Computers and Automation for such items as the
Census and Characteristics of Digital Computers unhappily
do not produce the kind of accuracy and reliability we
would like. But because most companies treat this kind of
information as proprietary, we have to rely on responses to
surveys, and on people approving any information we can
gather from outside sources. We do try to indicate the
degree of accuracy of information for our readers - in the
Census, for example, we use various symbols to show the
sources of census figures, to show which figures are estimates, etc.
. In the 1970 Directory, we will be publishing the Digital
Computer C,haracteristics compiled by Keydata Corp. for
their "Computer Characteristics Review". We believe this
will be much more accurate and complete. But it is a
never-ending struggle to get good, reliable information to
0
publish - and we do appreciate your help.
9

SJCC 71 - CALL FOR PAPERS
Dr. Nathaniel Macon
'71 SJCC Technical Program Committee
P. O. Box 30130 .
Bethesda, Md. 20014
. The Spring Joint Computer Conference will be held in
Atlantic City, N.J., May 18-20, 1971. The purpose of the
technical program at the Conference will be to bring
together professional design specialists and users of computer technology, to communicate significant advances and
to discuss important issues. Papers are invited in areas of
interest such as:
Compu ters and Elections
Computers and the Media
Data Files Versus People
Education and Training
Historical Perspectives
Law Enforcement and Judicial Administration
Manpower Survey
Techniques and Practices of Embezzlement
Teleprocessing
The Shaping of our Policies
Transportation

All of these topics have hardware, software and systems
connotations. We seek an interplay among hardware people,
software people, and users.
Participants will be expected to deliver a talk at the
conference, provide an abstract for the technical program,
and pr~pare a paper for publication in the proceedings.
Deadline for initial submissions is October 9, 1970. Notices
of intention to contribute a paper, and inquiries, should be
sent to the address above.
0

NUMBLES
Neil Macdonald
Assistant Editor
Computers and Automation
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 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 709
ABU S E

I S

E R DEI 0

+ B U I T WD T I B K D

= WR ITT

KR=BU=WR

E NI NT 0

COMPUTERS AND CHEMICAL NOTATIONS
31862

Charles E. Granito, Manager
Information Services
Institute for Scientific Information
325 Chestnut St.
Philadelphia, Pa. 19106
Mr. Wiswesser's article on computers and chernical notations [April 1970, page 35] is a fine contribution, and
Computers and Automation is to be congratulated for
. bringing this notation to the attention of the computer
world.
The Institute for Scientific Information has for several
years provided subscribers to its Index Chemicus Registry
System (ICRS) with computer tapes containing information
appearing in Current Abstracts of Chemistry and Index
Chemicus. Over 500,000 new compounds in Wiswesser
Line Notation (WLN) form have now been covered by this
service. In addition, rather sophisticated, though simple to
use, software has been developed and tested.
The use of the WLN notation will be further enhanced
by the publication of lSI's computer produced and permuted WLN printouts, which will enable even those without computers to do many useful chemical substructure
searches. We invite your readers to write us at the address
above if they are interested in receiving copies of these
0
printouts.
10

36064

77365

87960

Solution to Numble 708
In Numble 708 in the August issue, the digits
9 are represented by letters as follows:

D;O
H=l
L=2
0=3
E,W=4

through

A,Y= 5
K=6
N=7
T=8
S=9

The message is: "They know as well that do not ask."
Our thanks to the following individuals for submitting
their solutions to Numble 707: M. Bestgen, Riverdale, N.Y.;
Roy J. Boeckler, Milwaukee, Wis.; A. Sanford Brown,
Dallas, Tex.; W. S. Butts, Virginia Beach, Va.; Murray A.
Chayet, Tucson, Ariz.; H. Royce Culpepper, Jr., Nashville,
Tenn.; T. Paul Finn, Indianapolis, Ind.; George Gluck,
Rockville, Md.; Hank Green, Durham, N.C.; William A.
Leonhardt, Cincinnati, Ohio; Lambert J. Simon, Irving,
Tex.; and Robert R. Weden, Edina, Minp. Numble 706:
G. P. Petersen, St. Petersburg, Fla.
COMPUTERS and AUTOMATION for September, 1970

PUNCH LINES ...
The very innovation and change of which we as scientists
are so proud, has many people running scared. Whereas the
world of science lives with change - in the form of new
phenomena and new ideas and new techniques as the steady
state - people who find their education outmoded and the
skills by which they earn their living superseded are not so
sure. When the pace and scale of innovation threatens to
outrun the adaptability of "the social system, there's bound
to be some strain. That is what we are feeling now on many
fronts. More engineers need to learn how to be persuasive in
public debate, how to gain mutual respect and confidence
for communications with political arms of government from lower levels on up. They need to help win understanding of the impact of new developments and to make science
an ally of society. When problems become sufficiently clear
to enough segments of society, the legislation and other
corrective measures that are needed can be passed and
enforced.
- Donald Hornig.
Vice Pres. and Science Advisor
Eastman Kodak Co.
343 State St.
Rochester. N. Y. 14650

Everyone should have a health data base that is kept as a
valuable paper. It should include not only information on
an individual's health history and that of his family, but on
other factors, such as economic and social, which may
affect his well being. Each individual should carry this
"problem oriented" record with him so that ifhe moves or
has an emergency the information is readily available. It
must be easy and cheap to establish this data base. The
collection of information could begin as early as the first
grade, when children could answer simple questions about
themselves and their families. We must have this informationwritten down, for the human memory is not good
enough to accomplish this task of remembering.
- Dr. Lawrence L. Weed
Prof of Medicine and Community Medicine
Univ. of Vermont
Burlington. Vt. 05401

Newspapers will look better and be produced more rapidly
and economically in the 1970's. Almost one-third of the
country's 1,758 daily newspapers are now using computers
to drive typesetting equipment. Many are exploring or have
adopted offset printing, photographic typesetters, highspeed data transmission, and mailroom automation. Because of this practical experience in changing traditional
procedures, the groundwork has been laid for newspapers
to utilize advanced communications technology.
- Richard B. Tullis. Pres.
Harris Intertype Corp.
55 Public Square
aeveland. Ohio 44113
COMPUTERS and AUTOMATION for September, 1970

The implication for today's violent dissenters is that technology has provided the establishment with a potential for
control far exceeding anything in history. Such items as
computer bugging, infra-red surveillance (already in use
at the White House), monitoring of telephone calls by
satellite, introduction of mind control or disorientation
drugs into the environment, and the like, already loom on
the immediate horizon. It is time to quit, for continuation
of violent protest will only hasten the takeover of technopoly; i.e., the monopolistic domination by technological
devices and systems over the thought, feelings, aspirations
and behavior of man, and. the resources at his disposal.
Perhaps at the community and professional level, channels
of analyses and discussion will emerge which will create the
constitutional solutions so necessary for the preservation of
a system of ordered liberty, in a world rushing headlong
toward technopolistic restraint.
- Prof Charles S. Padden
The John Marshall Law School
315 S. Plymouth Court
Chicago. Ill. 60604

While it's easy to see "the knowledge explosion" as a
supreme tool, we too seldom see it as a potential tyrant.
The hard reality, however, is that in attempting to gather,
process, absorb and disseminate information and knowledge
today, we find ourselves living more and more in the
confusion of tied-up telephones, computer printout, procedure manuals, stacked airplanes, unnecessary correspondence, meetings, mail, memoranda and aging files marked
"Must Read".
In one way, the basic problem is similar to that faced a.
generation ago by nuclear scientists. Given the inevitability
of a force that can be of enormous benefit to mankind, we
must still insure and maximize its benefit. We must impose
control on a potentially infinite and uncontrollable chain of
reactions.
- C. Peter McColough. Pres.
Xerox Corp.
Stamford. Conn. 06903

The strength of the computer revolution is now unmistakable, and within 10 to 15 years, 50 per cent of all
workers will be affected by or connected with the computer industry. Computers need people, thousands every
year. Employers are almost conducting auctions for the
services of experienced people - and people with no more
than 12 months within the industry are deemed as experienced.
'
- George Parkinson. Chairman
Data Processing Services Ltd.
Stourbridge. England
13

REGULATIONS FOR INFORMA TION SYSTEMS
Prof. Calvin C Gotlieb
Dept. of Computer Science
University of Toronto
Toronto 181, Canada

"It is possible that introducing regulation and licensing for information
systems might exact a price which is unnecessarily high at first, but I prefer
to see caution on the side of protecting rights. "
Information systems containing data about individuals
are needed increasingly in the public sector (census data,
tax records, medical statistics, police fields, etc.) and in the
private sector (records for banking, credit, insurance, etc.).
Governments need these records to carry out their responsibilities; planners and social scientists need them to understand our society and suggest measures to take it in the
directions considered desirable; business needs the records
for effective operations, service and management.

devices are leading to integrated information systems built
around computerized data banks. Availability of terminals
and ease of communications increase the danger that data
about a person may be distributed without restriction and
used for purposes detrimental to his interests. The dangers
are often aggravated by an unjustified confidence in data
coming from a computer.

The Trend Toward Integrated Systems

This chain of reasoning leads to proposals for regulating
data banks, for example, by licensing them. Opponents of
regulations argue that they are in general addressed to the
wrong problem and in any case unenforceable. Proponents
argue that it is essential to adopt firm measures before there
is an irretrievable loss of human rights.
It is accepted. in this paper that some regulation of
information systems is required. No new arguments for
regulation are advanced. Most people reading this have

There is a strong trend towards integrated systems - i.e.,
systems that contain heterogeneous information about
many people and which are accessible to different authorities for different purposes. The ability to gather more
accurate and consistent data, and the advantages of lower
costs and better coverage, make such systems irresistible.
But our laws offer poor protection against the misuse of
such systems. The concept of privacy is very poorly defined
in Canadian law. Rules governing which agency may collect
what information, how data should be verified and to
whom it may be disclosed, exist in only a few situations. As
a result the individual is not protected against the misuse of
information about him.
Technological advances in storage and communications
Based on a paper prepared for a conference on "Computers: Privacy
and Freedom of Information", held at Queen's University, Kingston, Ontario, Canada, May 21-24, 1970.
14

Controversy Over Regulation

"In my opinion the crucial question is: Can effective regulations be formulated and enforced?"
probably formed opinions on whether privacy is really
bdng eroded. In my opinion the crucial question is: Can
effective regulations be formulated and enforced?
Many of the arguments against regulation are based on
views that:
COMPUTERS and AUTOMATION for September, 1970

information systems are too thoroughly woven
into the fabric of our technology to be limited;
any regulation which would protect privacy would
exact unacceptable costs in loss of efficiency and
opportunities; and
some diminution of privacy is unavoidable.

stored on magnetic tapes, or on magnetic disks which may
be attached on-line to a computer, will undoubtedly continue for decades. But to make any regulation effective it
must be applicable both now and through the transition

This paper is an attempt to examine goals and to show that
adopting measures to achieve them is not impossibly
difficult.

"Regulations must be considered for any information system which contains data about individuals,
and not simply for computerized data banks. "

Goals
It is not difficult to list goals which are almost truisms.
For example:

The rights of an individual to privacy must be
protected as much as possible, consistent with the
legitimate needs of society.
Data concerning individuals must be as accurate as
possible.
The problems arise when it is necessary to interpret the
meanings of specific phrases such as "individual rights", and
"legitimate needs of society", in knowing when reasonable
steps to provide protection or ensure accuracy are being
proposed.
The first proposition above implies a middle position in
privacy vs. freedom of information. It suggests that a
person cannot expect to opt out of society by refusing to
recognize that some records must be kept about him - for
example, birth registration and social security number. But
it does assume that there are rights of privacy for an
individual. Since it is generally agreed that the legal concept
of privacy is poorly defined in Canadian law, if at all, it
follows that there must be legislation to define individual
privacy. This is an essential point. If I do not dwell on it
further in this paper, it is because I do not regard myself
qualified to discuss where in Canadian law such legislation
ought to reside - whether it should exist in a provincial law
or in federal, or both, and in which sections of the law it
would be most appropriate. I hope that very soon specific
proposals for legislation will emerge.

Relating a Legal Concept to Rules

period. Regulations must be considered for any information
system which contains data about individuals, and not
simply for computerized data banks. Since a system which
contains information about a person may take a wide
variety of forms, ranging from a telephone book to a
security file, it is necessary to identify those systems for
which regulation might be needed.

A Classification of Information Systems
It is proposed here to classify information systems
containing data about individuals according to three characteristics, each with two or three categories. The characteristics and categories are shown in Table 1.
Table 1
Category

Characteristic
Data Source

P - public record
S - supplied by individual
o
other

Distribu tion

I - internal
E - external

Inspection

A - automatic
R - upon request of individual
F - forbidden

Although the terms used for the categories convey a
general sense of their meaning, precise definitions have to
be given.

A legal concept of privacy, though necessary, is not
enough. It is also necessary to consider how the concept
should be interpreted when operating information systems.
In particular it is important to relate the concept to the
rules which those who are responsible for designing, assembling, and maintaining information systems, are to use in
handling data: At present these rules are ambiguous. It will
be taken as a general goal here that it is desirable to have an
explicit statement of how information is collected, how it is
verified and to whom and under what conditions it is
transmitted, for any information system, public or private.
This is desirable even for a police or security information
system. 1
,
In this paper the emphasis on the phrase information
system, rather than data bank, is deliberate. Although
computerized data banks are coming to the fore, most of
the data about individuals is still in the conventional form
of files or punched cards. The transition toward systems

complete print-out of the information about an individual

1 An exception must be admitted for those (hopefully rare) systems,
for which because of national security, the existence is not made
known.

is sent to him at specified points either periodically, or
whenever a change is entered. In the "request" category
there might conceivably be some small fee charged if a
person wishes to see his print-out, so as to discourage

COMPUTERS and AUTOMATION for September, 1970

Data Source
For Data Source it is clear enough when the data is
supplied by the individual himself, and since "other" is
defined by exclusion, the definition hinges on what is
meant ~y public record. This could be defined by listing
those sources which were acceptable - for example, public
service awards, vehicle registrations, records of criminal
convictions, voters' lists, etc. Careful thought would be
needed to choose the list and it would have to be reviewed
in the light of experience, but there does not seem to be
any inherent problem in arriving at a definition.

Inspection
For Inspection the "automatic" category means that a

nuisance requests, but there should be no other condition
imposed. In particular an individual must be allowed to see
the whole record (otherwise the category should be "forbidden"), and he should not be required to sign forms
which prohibit him from presenting claims for damages
arising out of improper operation of the information
system.

Distribution
The most difficult categories to define are "internal"
and "external", with respect to the Distributioncharacteristic. Generally internal is intended to mean that distribution
of information is restricted to the company or institution
which maintains the information system, unless there is
explicit permission of the individual about whom the data
pertains, in every individual case, to transmit it elsewhere.
However, in the case of government, federal or provincial,
the organization is so large that it would be necessary to be
much more precise than this, if the term "internal" were to
have any validity.
Perhaps internal should mean a single department or
office. For a company , a decision would have to be made
whether various subsidiaries were to be considered internal,
and even for a university the question arises whether
different faculties and schools are all to be considered as
internal to the one institution. If it turned out that it were
not possible to define internal distribution with enough
precision, it might be necessary to consider distribution for
specific items of information rather than for the whole
contents of the system. This possibility is discussed at
greater length below, but for the moment it is assumed that
the categories "internal" and "external" are meaningful.

Some Sample Oassifications
Using this system a number of common information
systems are shown in Table 2, along with their classification
according to Table 1. It will be observed that fourteen of
the eighteen possible types occur in this table. This is some
evidence that the classification into types is useful. Most of
these information systems have a long history of use, and
methods for operating them have evolved to minimize
problems of verification and accessibility.
In fact problems really .arise only for the types OEF and
SEF. Regulation of information systems could therefore
proceed by first identifying those of type OEF and SEF.
For all others, and this would include the overwhelming
majority of systems - company payroll files, who's who
lists, newspaper morgues, etc. - no regulations would
apply. This would in itself encourage those operating
systems to make their data open for inspection to the
individual concerned, and to restrict general disclosure if
possible, so that regulations would not apply.
Where it is judged essential to permit the transfer of
records (for example,between one law enforcement jurisdiction and another), or it is judged not desirable for a
personalwa)(s to have access to his complete record (as
might be the case with a doctor's report), the conditions for
allowing disclosure or preserving security would be spelled
out. I repeat that the essential purpose of the classification
is to allow attention to be focussed on any information
system where there are problems of security and disclosure,
and not just on computerized data banks.
16

Table 2
Classification of Some Information Systems
System

Type

Bank Account
Payroll File
Who's Who
Medical Report
Personnel File
Police File

OEA
OIR
SEA
OIF
OIF
OEF

Credit Record
Tax File
Telephone Book
Voters List
Sales Prospects' File
Sales Prospects' File for Sale

OER
OIF
PEA
PEA
PIF, OIF
PEF,OEF

Membership List
(Club, Professional Society)

SIA, SIR
SEA, SER
(if given to others)

Newspaper Morgue
Court Records
Welfare List
Census Record
Biographical File of Company

OER,OEF
PER
OER or OEF
SIF
OEA

Security Tags
Although this classification system could be adequate
for setting up regulations about information systems, in
actual operation it will probably be desirable to make much
finer distinctions about categories of data. The most difficu1tquestion will continue to be: who should have authority to receive specified items of data? It is impossible to take
a simplistic approach on this. Eventually the only satisfactory solu tion will be to attach security tags to every data
field, and use these tags to determine under what conditions the information may be disseminated.

Eventually the only satisfactory solution will be to
attach security tags to every data field, and use
these tags to determine under what conditions the
informa tion may be disseminated. "
In simple cases the user's authorization code may be
sufficient to determine which fields are available to him; in
more complicated situations it may be necessary to set up a
table which relates authorization codes to tags. 2 There are
definite overhead costs associated with security tags and
these are discussed further below. In my opinion it will
come to be recognized that these costs must be paid and a
security tag system will be a regular part of every information system. But the lack of experience with such systems,
and the fact that they are so far rare and would have to be
added to existing systems over a considerable period of
time, does not make it practical to suggest that proposed
regula tions on information systems make it necessary to
include security tags for data.
Three types of costs will be associated with the regulation of information systems: direct costs, overhead costs,
2The IBM manual "The Considerations of Data Security in a
Computer Environment" discusses, briefly, authorization techniques
and tables (p. 16). See also Ware et al.: Spring Joint Computer
Conference, 1967, pp. 279-303.
COMPUTERS and AUTOMATION for September, 1970

and inhibition costs arising from things which cannot be
done.

Direct Costs
It will obviously take funds to maintain regulatory and
licensing agencies. Other direct costs to be paid by the
purchaser of hardware will be for scramblers, and other
devices for protecting information, detecting possible taps
on the communication channels, etc. These devices would
certainly come into wider use if the operators of information systems were made to take on legal responsibilities for
safeguarding data.

Overhead Costs
Overhead costs would arise from the more complicated
software systems which would be needed. There would be
costs in making transcripts of data available to individuals
either automatically or on request, or simply in classifying
different data fields and assigning tags to them. Care might
have to be taken that this task was not expanded to the
point where it had a whole mystique attached to it, as is
said to be the case with data classified for military security.
There would be costs for storing tags and the time taken to
decode authorization numbers and match them against tags.
On this latter point it should be noted that there are
already information systems in which a deal of redundant
information is carried in the form of tags attached to data
fields. In the Marc II system now being adopted widely for
bibliographic information, perhaps 10% of the storage is
used for tags that identify data, facilitate access and
counting, etc. If this type of storage and processing overhead can be built into a system for handling bibliographic
information, it is not too much to expect that it will also be
built into systems for handling personal-data.

Inhibition Costs

which will have significant effects on the preservation of
individual privacy if adopted.
1) A legal concept of the invasion of privacy should
be introduced in Canadian law.
This should go beyond the present laws on nondisclosure
governing lawyers, physicians, bankers, employees, spouses.
It should be broad enough to be applicable to situations
involving wiretapping, credit bureaus, health information
systems, and other types of information systems.
2) Certain types of information systems should be.
licensed.
Legislation could follow the general line suggested in
Computers and Freedom 3 and Bill 182 4 , and Privacy and
Commercial Reporting Agencies. The licensing should be
for information systems and not merely data banks, and
should be based on a classification which categorizes systems according to their mode of operation. The transition
to computerized information systems should simplify the
application of controls.
3) Technical improvements which permit greater se·
curity and control over the transference of information should be encouraged.
Particularly needed are effective techniques for matching
data with authorized users, and inexpensive hardware for
maintaining security. The encouragement could take the
form of research grants on projects and incentives to
manufacturers and software companies to develop and
market systems, and publicize the methods and devices
already known. Systems analysts and designers should use
the tools presently available.
Man is learning that not all the effects of technology are
beneficial. Our concern over pollution is only one aspect of

"Man is learning that not all the effects of technology are beneficial. "

There would be inhibition costs because worthwhile
activities would be more costly or forbidden. It would be

"It would be more difficult or even impossible to
carry out certain types of planning studies and
experiments in the social sciences if access to
personal data became more restricted. "

more difficult or even impossible to carry out certain types
of planning studies and experiments in the social sciences if
access to personal data became more restricted. But we are
used to such inhibitions in medical and psychological
experimentation involving human beings, where there are
very careful legal and other regulations about what may be
done, and we accept the necessity for them. We will have to
accept similar inhibition costs when using personal
information.

Proposals
This paper concludes with three specific proposals
which, it is believed, are capable of implementation, and
COMPUTERS and AUTOMATION for September, 1970

the review which has to be undertaken about many side-effects of technology. Especially serious are those cases where
the processes are almost impossible to reverse. A polluted
lake is a much greater problem than a polluted river. I do
not feel that we are far along an irreversible process in the
way we are allowing information systems to operate now.
But it is noteworthy that every review of the tolerance for
pollution or radiation leads to a downward revision of the
permitted levels.
I feel that it is possible that introducing regulation and
licenSing for information systems might exact a price which
is unnecessarily high at first,. but I prefer to see caution on
the side of protecting rights. Experience has shown that
overprotection is in fact, very rare. In my opinion if the
problems regarding protection of individual privacy are
explained to the public, and to those responsible for
political and legislative action, and the alternatives are set
out, they will be willing. to pay the price of keeping our
social environment healthy. In fact they may well insist
that the price be paid.
0
3 NCP Old Queen Street Paper: 8, Conservative Research Department 1968
4 An Act to Provide for Data Surveillance, 2nd Session 28th
Legislature,. Ontario 1968-69

PROBLEMS OF LIABI,LITY FOR THE EDP SERVICES INDUSTRY
Milton R. Wessel
Kaye, Scholer, Fierman, Nays and Handler
425 Park Ave.
New York, N. Y. 10022

I believe that the decade of the nineteen seventies will
see a broad expansion of the liability of the EDP services
industry's to all whose lives and businesses it affects. This
liability will be founded upon new and far-reaching socioeconomic considerations. The companies which fail to
recognize this and take protective action may well be
among those which do not survive.
Fundamental Shifts in Society
Any discussion of this industry's liability problems
during the next decade must be predicated in large part
upon an evaluation of what is happening in the streets and
on the campuses. Protests regarding Southeast Asia and war
genenilly, ecology, civil rights and poverty are not simply
the passing fancies of a bunch of kids or hot-headed
radicals, but a reflection of fundamental shifts in our
society which are changing all the rules.
Already we have witnessed many of the incidents of
private property, until recently a sacrosanct legal cornerstone of democracy and its free enterprise system, give way
to a host of intrusions based upon social considerations. No
longer can one be secure that he will be permitted to do
what he wants, when he wants with what he owns. If you
don't agree, ask Consolidated Edison of New York, or
General Motors, or the owner of a restaurant in Macon,
Georgia, or a home in Chicago.
Similarly, the former economic touchstone of gross
national product is beginning to yield to considerations of
quality and etological preservation, and we already have a
new concept of "net" national product, deducting pollution and depletion, with economic and social objectives·
modified as a result.

Milton R. Wessel is a law partner in the New Yark firm of
Kaye, Scholer, Fierman, Hays and Handler. He is also
attorney for the Association of Data Processing Service
Organizations (ADAPSO), and he represents the American
Federation of Information Processing Societies (AFIPS).

And so it is all along the line. One has only to examine
the decisions of the United States Supreme Court in a
variety of areas since the nineteen fifties - civil rights,
criminal prosecutions, even t4e Association of Data Processing Service Organization's (ADAPSO's) bank case extending the citizen's right to call government officials to
<:1ccount - to see the legal consequences of these enormous
changes.
The law is in fact developing rapidly, to reflect this
economic, social and political revolution. And nowhere are
the changes in the law during the seventies likely to be
more dramatic than in the EDP services industry, for it is a
new one having major societal impact where these modern
developments do not meet the resistance of history and
tradition - what lawyers call stare decisis - to the same
extent as in older established industries.
Liability to the Public
Industry problems of liability during the 1970's can be
considered in terms of the two major groups to which
liability extends: (1) the public (including governments and
competitors); and (2) those with whom the industry has
contractual relationships - customers, owners, employees
and suppliers. A recent decision, the impact of which has
not yet been fully appreciated either in industry or legal
circles, may be helpful in furnishing an overview of the
current and anticipated broadening of liability to the first
of the groups to which industry liability extends, the
public.
.
A Kentucky citizen, John T. Swarens, purchased an
automobile on credit through the Ford Motor Credit
Company. Hp paid his monthly charges, but the inevitable
error occurred and collection agents visited. Swarens satisfied them by exhibiting cancelled checks. The visit was
followed by a second, and Swarens again showed his
cancelled checks. But on the third visit, Swarens exhibited
only his shotgun and chased the men away, unsatisfied.
Ford seized and sold the vehicle, and Swarens brought suit
for compensatory and punitive damages. A jury, sustained
all the way through, awarded Swarens his full damages plus
250% in punitive damages, a quasi-criminal type of punishment. The Kentucky Court of Appeals went out of its way
COMPUTERS and AUTOMATION for September, 1970

"In an effort to avoid liability, EDP services organizations should introduce
measures which will: (1) guard against the dissemination of erroneous
information from a data bank; and (2) permit members of the public zn
appropriate circumstances to obtain assurances concerning their files. "

in a final paragraph to make clear the obligations being
imposed upon EDP service organizations. It said:
Ford explains that this whole incident occurred
because of a mistake by a computer. Men feed data to
a computer and men interpret the answer the computer spews forth. In this computerized age, the law
must require that men in the use of computerized
data regard those with whom they are dealing as more
important than a perforation on a card. Trust in the
infallibility of a computer is hardly a defense, when
the opportunity to avoid the error is as apparent and
repeated as was here presented. 44 S.W.2d 53, 57
(Oct. 17, 1969)
It is significant to add that despite much talk, no
legislature has yet imposed a statutory obligation upon
computer users to interpose such human judgment between

"It is significant that despite much talk, no legislature has yet imposed a statutory obligation upon
computer users to interpose human judgment between computer and public. "

computer and public. Swarens is thus far more important in
its general teaching than in its specific ruling, for the court
imposed the obligation on its own, even authorizing sanctions criminal in character. By parity of reasoning, computer services companies which feel secure on the privacy
issue must take note, despite their success thus far in
blocking any statute giving individuals the right of access to
data banks or imposing additional statutory liability for
disseminating erroneous information. In a proper case, I
predict that a court will take this action on its own without
a statute, based upon the expanded social concepts which
are entering the law.
In an effort to avoid liability, EDP services organizations
should, individually and through groups, introduce measures which will: (1) guard against the dissemination of
COMPUTERS and AUTOMATION for September, 1970

erroneous information from a data bank; and (2) permit
members of the public in appropriate circumstances to
obtain assurances concerning their files. The ADAPSO
Board of Directors has the matter of privacy and related
problems under intensive consideration, but up to this time
neither it nor any EDP services organization has taken
action.

Impact on Licensed Professionals
As EDP services continue to expand - and despite the
gloom, the present financial crisis is temporary and profitable expansion will return - their impact upon the licensed
professions such as accounting, engineering and even medicine and law will increase. I believe that intensified regulatory activity will result during this decade, including injunctive and punitive action against services organizations for
unlawful practice. The wise services organization will exercise caution in these areas, as well as take preventive action
such as by appending the disclaimer recommended by
ADAPSO to its reports.} I regret having to report that only
a handful of companies have yet done so.

Concentration of Services
Industry economic forecasters are agreed that the present trends towards concentration of EDP services into
fewer but larger economic units and proliferation of communications applications, will accelerate during the seventies. As services organizations thereby become regional and
national rather than strictly local units, their activities are
bound to become subject to closer antitrust scrutiny, both
by public enforcement agencies and private parties seeking
the windfall of treble damages. Liability for violation of the
antitrust laws can be huge, not just in terms of damage
recoveries, but also because of the enormous direct and
indirect costs that litigation of this kind inevitably
generates.
Despite restrictions seemingly imposed by the United
States Supreme Court, I believe that the application of the
related· state laws of unfair competition 2 will probably
continue to be intensified in this decade as they have been
for some time now, with the result that the industry will
19

also experience increased liability to competitors for conduct such as alleged theft of trade secrets or interference
with advantageous commerical relations.
On the plus antitrust side, I predict that the RobinsonPatman Act provisions relating to price discrimination will
continue to enjoy the disrespect of most antitrust· authorities, and that it will not be extended either by statute or
judicial construction to apply to purely EDP services.

The answer to all of this? Contractual disclaimers and
the errors and omissions insurance coverage soon t,o be'
offered through ADAPSO will be helpful. But first and
foremost EDP services organizations must furnish high
quality service, so that there are as few errors as possible. In
addition, however, because errors are inevitable in a commercial environment where costs are relevant, caution also

Liability to Customers

"First and foremost EDP services organizations
must furnish high quality service, so that there are
as few errors as possible. "

The liability of manufacturers and suppliers generally to
their customers has been expanding for a half century and
more, and the decade of the 1970's will also see the EDP
services industry's liability to customers extended far beyond anything which presently exists. A Minneapolis decision, now on appeal, imposed an almost one-half million
dollar liability on a service center despite written contractual disclaimers, and without any proof of scienter or

"A Minneapolis decision, now on appeal, imposed
an almost one-half million dollar liability on a
service center despite written contractual disclaimers, and without any proof of specific intention to misrepresen t. "

suggests fair dealing with customers through such means as
the circularization of the ADAPSO position paper on
error-free servicing, designed specifically to guard against
this kind of liability. The EDP organization which sends the
ADAPSO paper out to customers, has a useful defense to
any claim of misrepresentation by one of its employees, for
the paper itself constitutes evidence that the customer
should not have relied on an obviously unauthorized representation. Here again only a few companies have followed
this procedure, in spite of the fact that the organization
which fails to send out the paper will have a far heavier
burden to satisfy when a client refers to a salesman's
promise, and then proves that the company failed to follow
industry suggestions to disseminate this paper.
Liability to Shareholders and Owners

specific intention to misrepresent. A Los Angeles decision,
subsequently withdrawn but still significant as suggestive of
the trend, sustained a customer's right to sue for one
million dollars punitive damages for improper service. These
reveal only the top of the iceberg, for underneath lies the
possibility of what lawyers call "secondary" or "consequential" damages. One large brokerage house, in folding,
blamed its problems in large part upon computer difficulties. The losses to the brokerage house alone were catastrophic; but in addition, the threat of claims by customers
who were injured or by the stock exchange which covered
some of the losses, has been posed and cannot be ignored.
The former legal rule of caveat emptor - "let the buyer
beware" - meaning that a seller could say pretty much
what he wanted without liability, is already well on its way
out, and is being replaced by a variety of bothjudicially-imposed and statutory sanctions protecting the consumer. I
believe that this trend favoring the consumer will also be
accelerated during the seventies, especially in the EDP
services industry, to be replaced by presumptions and rules
such as res ipsa loquitur; 3 placing a heaVy burden of
responsibility on the vendor of services. Last year a Minnesota legislator introduced a statute which provided:
Whoever is injured ... [by computer error] ...
shall recover therefor unless the person who caused
the injury establishes that he wa~ not negligent.
ADAPSO reacted immediately and positively as it should
have, and the statute was defeated. But the result was only
to gain time, not ultimate victory, for I predict that in a
proper case, a court without a statute will impose liability
on just this theory, in effect shifting the burden of
establishing proper conduct on to the members of this
industry.
20

The period since World War II has seen a broad expansion of the liability of directors to shareholders, partners
and other owners; although individual director liability is
not asserted, the EDP services industry already has a
three-quarter billion dollar derivative stockholder action
pending for failure to take action to protect the interests of
owners.4 I predict this area of liability will also be greatly
extended in the seventies, accompanied by an expansion of
enforcement activity by the Securities and Exchange Commission (SEC), as soon as that much harassed agency is able
to get some of its more current and urgent problems under
control.
Statutory Software Protection
The EDP services industry is characterized by huge
investments in software, on the values of which stockholders and owners rely. During the 1970's, there is likely
to be established a new statutory form of software protection unlike copyrights, patents and trade secrets, although
combining some portions of each. Reliable legal protection
for software under any circumstances, however, is still some
years off.

"Reliable legal protection for software is still some
years off. "

Today, decisions of the Supreme Court suggest that even
trade secret protection may be in danger. The recent
write-offs by industry giants of millions of dollars of
COMPUTERS and AUTOMATION for September, 1970

intangible software investments, because they proved
worthless, are small compared to the losses which would
result from invalidation of trade secret protection. There is
an obvious exposure to personal liability to stockholders
for failure to disclose the possibility of such write-offs or
losses based upon presently existing legal impediments
which should have been recognized. The answer? Fair
disclosure, and far greater caution in valuing intangibles.
Let me add parenthetically in connection with relationships with owners, that the present financial crunch has
made the raising of money extremely difficult and in many
cases impossible, with disastrous consequences. This industry can no longer expect the public funds so sorely needed
for expansion to be channeled to it whenever requested, on
the expectation of huge multiples of earnings. In addition
to a return to the fundamentals of sound economic growth
and true (as distinguished from purely accounting) profit,
and attention to all the common alternatives of financial
planning, it must pay attention during this decade to the
special financing techniques which have been used so
successfully in some other industries, and which have made
them far more attractive than this one to many sophisticated investors. s
Liability to Employees
During the seventies, as mass production techniques
grow - and they must and will if the industry is to prosper
- organized labor will enter the industry. It has already
done so in a few cases. The EDP manager must know how
to deal with it, or he will have added just one more problem
to those which he already faces. He should be especially
alert to the legal requirements in this field, and companies
should work together to counteract the union strategy of
"divide and conquer".
It seems fair also to anticipate that as population
increases, the economy expands, and concepts of freedom
mature further, employment restrictions will be more difficult to enforce. Covenants limiting an ~mployee's right to
approach customers or otherwise compete during the next
decade will be even more confined in space and time than
they are today.
Moreover, as the industry grows during this decade,
enforcement of the Fair Labor Standards Act and similar
wage, hour and working condition statutes will become
more intensive, perhaps on the basis of a single-industry
campaign. This has been the experience in other new or
developing industries; the liability for failure to adhere to
these laws can be substantial. The industry must also be
alert to the possibility of increased governmental and
private enforcement of the civil rights and equal opportunity laws, especially as even the smaller services organization becomes no longer a local enterprise.
Liability to Suppliers
I have kept discussion of liability to suppliers for last,
because it is the one area about which I am most optimistic.
The chances are that during the 1970's the increased
liability of suppliers to the EDP services industry will far
outpace any increase in the industry's liability to suppliers.
I believe· also that irrespective of their outcomes, the
pending antitrust litigations will result as a practical matter
in significant competitive benefits to this industry. Indeed,
some of these benefits may already have been experienced.
COMPUTERS and AUTOMATION for September, 1970

But still a word of caution, added with some trepidation
out of fear of being misunderstood because of the strong
feelings about IBM held by so many: Too many in this
industry blame IBM and SBC for their own commercial
failures, and file suit as proof - I count nine major suits
against IBM alone and threats of many others. The courts

"The EDP manager who regards any litigation as a
substitute for proper operation of a business does
so at his own peril. "

will ultimately determine the merits of these cases, and I
certainly am not perspicacious enough to predict their
outcome. But the EDP manager who regards any litigation
as a substitute for proper operation of a business does so at
his own peril.
Conclusion
What does all this add up to? The EDP services industry
will finally grow up during the seventies, in a legal environment in which the rules are not yet defined. However, it is
clear that fundamental principles of integrity, quality and
the like are going to become more and more dominant.
Professionalism and ethical practices are not just moral
commitments; they are sound as commercial judgment as
well.
I believe that in the seventies .industry liability will be
broadly expanded. It will be predicated in large part upon
morality and ethics, which are far better measures of legal
standards of conduct than some of the conceptual considerations of the past. Viewed from afar and very long range,
this can only be good, and the EDP company which attends
properly to its affairs and survives will be a better one for
0
it.

References
1. The disclaimer recommended by ADAPSO states: "The foregoing presentation has been processed electronically by computer
from data furnished to us, has not been analyzed by professional
[accountants] [engineers] [architects] [doctors] [lawyers] or
other persons purporting to have expert knowledge of [accounting]
[engineering] [architecture] [medicine] (law] and is furnished
without [accounting) [engineering] [architectural) [medical]
(legal] opinion of any kind.
2. A federal law of unfair competition may even be enacted.
The former Lindsay Bill is now being revised by a National
Coordinating Committee of lawyers from the American Bar Association, the United States Trademark Association, and other bar
associations, in consultation with the Antitrust Division of the
.
Department of Justice.
3. Meaning that in the absence of a contrary explanation,
certain conduct will be presumed to have been the result of
negligence. Literally, "the thing [conduct] speaks for itself".
4. Although this article is concerned only with substantive
liability problems, the industry's exposure will also undoubtedly
continue to be enlarged during the decade as the result of broadened
procedural remedies such as the class and citizen's actions.
ADAPSO's bank litigation, in which the United States Supreme
Court enlarged the standing to sue concept, is a good example.
5. e.g., the "tax shelter", which is particularly suitable to the
EDP services industry and has given it an as yet untapped, but very
real,. financial competitive advantage because of limitations imposed
on other industries by the Tax Reform Act of 1969.
21

SYSTEMS TEST
Dwight M. Frazier, Jr.
IBM Corp.

P.O. Box 12195
Research Triangle Park, N.C. 27709

An article in the Harvard Business Review, January-February, 1968, reads: "The most successful U.S. manufacturers today are making much greater and more sophisticated
use of the computer than they were just three years ago.
The day may not be far distant when those who analyze
business failures can add another category to their list of
causes - failure to exploit the computer."
To "exploit the computer," many modern corporations
are installing, or planning to install, very complex interplant, inter-divisional, computer-oriented data processing
systems which will serve management at all levels, from
manufacturing supervisors to corporate presidents.

Design and Development Costs
Thousands, even millions, of dollars are being invested in
the design and development of such systems. When installed, the output from these systems is used to make vital
decisions which can mean the success or failure of a
corporation. Management must be able to assume that the

Dwight M. Frazier, Jr. is an associate programmer at
IBM's Systems Manufacturing Div. facility. He was graduated
from East Carolina Univ. in 1963 with an A.B. degree in
mathematics and a minor in physics. Prior to joining IBM in
1968, he was a programmer-analyst with the Carolina Telephone and Telegraph Co. in Tarboro, North Carolina, for
three years.
22

information from these systems will always be available and
will always be correct.
After installation, what happens if the new system stops
working, even for a short period of time? Old systems have
been discontinued, and no decision-making data is available.
An even graver situation would be an undetected malfunction. Decisions based on incorrect data can be and likely
will be wrong, resulting in time, manpower, and dollars lost
or wasted.
Since today's fast-moving, fast-changing and highly-competitive environment demands the exploitation of the computer for the development and installation of modern data
processing systems, how can needed systems be installed
and the undesirable consequences such as those mentioned
above be avoided? I believe the answer is a step which is
often not formally included in the development plans for
computer-oriented data-processing systems. This step is
called "systems test."

What Is a Systems Test?
Systems test is a test of all the computer programs and
associate procedures which make up a computerized data
processing system. Its purpose is to insure that they fit
together and perform as specified in pre-programming plans
(User Manuals, Design Specifications, etc.).
What can a systems test do for you? To help answer this,
let's look at an experience which occurred in an IBM
manufacturing plant in the southeastern United States. This
plant's Systems and Programming department was assigned
responsibility to develop a data processing system which
could be installed in nine of the corporation's manufacturing plants. They were located in many sections of the
United States and one was in Canada. This system was to
control orders and bills for certain customized products
whose sales amounted to millions of dollars each month.
The orders were received at each plant from the corporation's sales division, whose offices were in every state, and
several foreign countries. The billing information was to be
returned to the sales division at a central location via
magnetic tape.

Approaching Perfection
Management concluded that, to install such an interplant, inter-divisional system at remote'locations, the system itself must very closely approach perfection. To guarCOMPUTERS and AUTOMATION for September, 1970

"It is quite possible that without a good system test, some projects may
incur installation problems and be scrapped unnecessarily because of misunderstandings, lack of education, and a natural resistance to change. "

•
antee this perfection, systems test was included in the
development phase for this system. It was to last three
weeks, and was to be performed at a location other than
the developnig or installing plants. Installing plants were to
send representatives (both user and programming personnel) to review the system, watch it operate, and examine
the test results. The installing plant's representatives were
to have the responsibility of determining if the system
could or could not be installed and made operational in
their plant's environment.
The developing plant assigned impartial personnel the
responsibility of preparing for this systems test. Plans were
developed and test data were prepared. Most of this test
data was run through the system at the developing plant to
help programmers unit-test (a test of an individual program)
their programs. By doing this, many "bugs" (errors) were
discovered, and corrections were made before the official
three-week systems test began.

Success of the Test
Once the systems test began, and logistics problems
involved in assembling and making lodging accommodations
for fifty or so people had been solved, things went quite
well. Every plant reviewed the structured test and added
additional test cases. Some even ran sample live data
through the system.
Program errors were discovered and fixed. Sign-offs were
received from all plants stating that the system could be
installed and made to operate in their environment. In fact,
all plants were running live with the new system within six
weeks after the completion of the systems test.
This example shows the importance of a systems test
and what role it can play in the development cycle of a
large, sophisticated system. However, this was only one
system and only one systems test. The following is a list of
benefits which can usually be derived if systems tests are
properly planned and executed:
1. More Reliable System - During the systems test,
many errors (hopefully all) will be discovered and
fixed. This can greatly reduce the parallel run time
necessary after installation. Also, some of these
errors will probably be the type that could go
undiscovered for years under normal production
conditions. Errors discovered and corrected beforehand will therefore not occur after installaCOMPUTERS and AUTOMATION for September, 1970

tion, when they may result in time, manpower and
dollars lost or wasted.
Education and Training - To prepare the test data
or examine the results, one must become thoroughly knowledgeable about the system. He is
forced to read and understand "User Manuals"
and "Design Specifications" to learn how to use
the system, how it is maintained, what it does
with the data, and how its output is to be
interpreted.
User Involvement - Because he is preparing test
data and examining results, the user becomes
acquainted with the system prior to its installation. Otherwise, the testing responsibility is usually left entirely up to Systems and Programming
personnel, who often dislike the task and often do
not do a thorough job. Now, instead of having to
use a system tested entirely by someone else, the
user is able to use a system he has helped test.
Increased Confidence - Because users have been
involved in the testing of the system and have
helped make it operate successfully, they will not
hesitate to use the system after it is installed.
Their confidence in the system is greatly
increased.
Diagnostic Test Deck - This can be condensed
from the system test to be used for a quick system
check-out when future program changes or modifications have to be made.
Common· Programs and Common Documentation
- If the system is to be installed at other locations, this becomes a very important benefit. At
the end of systems test, all programs and documentation must be updated to the latest level and
released. In this way, all installing locations know
what the latest level is and can be assured of
receiving it.

Systems Test Guidelines
Systems tests must be designed to fit the system that is
being tested. Since all systems vary in purposes and complexities, no two systems tests will be exactly alike. However, there are certain basic rules and guidelines which
should be followed to develop and execute a successful
systems test. They are as follows:
23

1. Organize a "test team" as soon as design specifications have been completed. It should consist of representatives from Systems and Programming and from the user
areas. This team should report directly to management
which is in control of systems and programming and user
areas. A chairman should be elected or appointed, and a
regular meeting schedule arranged. The following is a list of
recommendations for test team activities:
A. Study Systems Flow - What programs make up
the system, and in which sequence are they
executed?
B. Study System - What is put into system, and what
comes out of system? What does each program in
the system do to the data? Usually it is a good
idea to look at system output first, then to go
back through every program in the system to
determine what the original input must be.
C. Examine System Environment - What type of
hardware (computer) is the system going to be
run on? What type of input and output devices
will be utilized (cards, disks, tapes, video display,
etc.)? What types of software (programs which
control the computer) will be utilized?
D. Prepare Test Plans - What test cases and runs will
be necessary to properly test the system? on what
dates will these runs be made? Test plans should
be considered very carefully. The test cases themselves will actually determine how good the test
is, and the dates will be used as targets which
must be met as the test progresses. Any slippage
in these dates will also mean a slip in the system's
in stallation.
E. Design Forms - Forms will furnish input data,
and will document expected results. What happens to this data after it goes into the system?
Does it come out on a report or go into a master
file for on-line retrieval later? Over-lay forms
(foils), which will facilitate the reading of non-formatted computer outputs (data set dumps, etc.),
are often very useful time savers. As the test
progresses, errors will be discovered in programming and documentation. They must be controlled. This establishes a need for a vital form,
the "trouble report".
F. Establish Control Procedures - These should be
used to handle various trouble reports which will
arise during the testing of the system. The status
of these trouble reports must be known at all
times. When was it received? What program was
affected? What was the level of the program? Who
is working on the problem? Once the problem is
fixed, the trouble report must be returned to the
originator with an explanation of what corrective
action has been taken.
G. Establish Control Center - To control (l) flow of
data between computers and test.team, (2) system
level, (3) program levels within system and
(4) Trouble Reports. Without a centralized control center to handle these functions, the system
may get out of control and may never be brought
to successful completion. This control center
should be manned by members of the system test
team. For examples of Control Documents see
exhibits 1,2 and 3.
24

2. Centralize Final Phase of Test - Bring together at a
central location representatives of all installing or using
locations. With the developing department's systems analysts, programmers and test team to review systems test
results, installing and using representatives may consist of
both user and programming personnel. The customers
should now be able to add additional test data if it is
deemed necessary. However, be wary of attempts to foil the
system with _unrealistic data (outside the specifications),
which could prolong the test cycle without accomplishing
anything really valuable.
At this time the control center becomes a very critical
part of systems test. Its role is increased to control flow of
data between computer, test team, and installing or using
representatives.
3. Obtain Sign-Offs - At the end of the final systems
test phase, all installing locations or customers should be
required to sign a document stating that the system does, or
does not, operate according to specs and is, or is not,
installable. Even though the system may not do everything
that users now want it to do, it must be installed if possible.
Otherwise desired modifications and changes could keep
the system from being installed indefinitely. Desired modifications and changes. should be documented as possible
updates to the system after its installation. If the sign off
concludes that the system does not operate according to
specs and cannot be installed, the problems should be
documented in a written letter to the developing project
manager with a copy to all responsible management, including the corporate president.
An economic evaluation of the problems and possible
solutions should be made, and either a redevelopment cycle
scheduled or the system scrapped.

Economic Justification
The cost of a good systems test will vary from system to
system, depending upon variables such as amount of test
data necessary, manpower, etc. Actual dollar savings ,as a
result of the systems test may be impossible to calculate
accura tely. However, the savings resulting from a short
installation time, elimination of bad decision-making data,
elimination of down time and maintenance time after
installation, etc., will certainly repay the cost many times.
It is quite possible that without a good system test, some
projects may incur installation problems and, be scrapped
unnecessarily because of misunderstandings, lack of education, and a natural resistance to change. This results in a
one hundred percent loss of the development cost and still
leaves the corporation without the needed system.

Summary
To exploit the computer, corporations need to develop
and install computer-oriented data processing systems.
However, before installing a computer-oriented data processing system, the corporation must be reasonably certain
it will work correctly. To do this, every system should be
required to clear a final hurdle before installation - the
systems test. If this hurdle is cleared successfully, then the
system is declared sound and installable. If not, the system
cannot be installed and must be either reworked or
scrapped. Regardless of the results, the systems test has
0
worked toward the good of the corporation.
COMPUTERS and AUTOMATION for September, 1970

THE USER/MANUFACTURER INTERFACE
David F. Stevens
Lawrence Radiation Laboratory
Bldg. 50A, Rm. 1135
Univ. of California
Berkeley, Calif 94720

"The user must recognize that some change is inevitable; the manufacturer
must realize that not all change is desirable. "
The relationship between the manufacturer of a computing system and its users was, until recently, a reasonably
well-defined, if somewhat nonproductive, affair. Unbundling, however, has introduced a measure of uncertainty
into that relationship, and the resulting turbulence may
provide us with an opportunity to redefine it (the relationship) in a manner profitable to both parties. The purpose of
this article is to examine the current state of this uneasy
alliance and to suggest how it can be improved.

A Parable
That a certain amount of friction has always existed
between manufacturer and user is clear from the following
story told about the first installation of a large scale system.
Since it was, in fact, the first system, it was quite simple.
The manufacturer was rather tractable by today's standards: there was only one portion of the system he insisted
that the users not diddle - a particularly critical tree
structure. But this, of course, was too much for one of the
more subtle graduate students, and he convinced the junior
systems programmer (there were only two: it was a small
installation) that a small change would surely not be
disastrous. The manufacturer discovered the change and
closed down the installation, driving out the programmers;
also, being rather old-fashioned, he laid a curse on the
graduate student: " ... dust shalt thou eat all the days of
thy life" .... Now this was perhaps over-reaction on the
part of the manufacturer, but his attitude is illustrative of
current attitudes, and users' behaviour certainly hasn't
changed at all.
One result of the long cold-war which manufacturer and
user have continued to wage against each other has been a
self-reinforcing hardening of attitudes, in which each side
has created an uncomplimentary stereotyped image of the
other and then, to a certain extent, adopted, as a model for
its own behaviour, the stereotype created by the other side.
Our examination of the current state of the interface begins
with a consideration of these stereotypes.
COMPUTERS and AUTOMATION for September, 1970

The Manufacturer as Seen by the User
The manufacturer, as seen by the user, is possessive: the
system is his product, and his pride of creation is inordinate. Because it is his creation, he refuses to recognize its
shortcomings or failures. Discrepancies between specifications and product are often eliminated by changing the
specifications (the product, being created perfect, is beyond
change). Also, during those sensitive formative months
when the product is developing, it is jealously screened
from all contact with the world, lest reality should accidentally shape that development. Requests by prospective
users for any level of detailed information are promptly and
firmly denied.
To the user, the manufacturer is myopic: his vision is
limited not only in distance, but also in direction ( he has
tunnel vision). He fails not only to foresee the users' future
needs, but also to foresee the uses to which they will put
his current products.

"The user feels that the small additions which the
manufacturer does adopt from him are often accepted in the same spirit of amused toleration with
which one accepts a mud pie from a four-yearald."
The manufacturer is also condescending: he knows that
lie has nothing to learn about computing from the user, and
that therefore he is able to create, in a communications
vacuum, products of universal utility. What small expertise
the user has he acquired from the manufacturer, and it is
several years out of date. The user can create nothing of his
own, worthy of notice by the manufacturer, and those
small additions which the manufacturer does adopt are
often accepted in the same spirit of amused toleration with
which one accepts a mud pie from a four-year-old .... One
especially notices that the toleration becomes ever more
strained with each succeeding offering.
25

To the user, the manufacturer is overconfident: he
overestimates the performance of his products and he
underestimates the time it will take him to deliver them.
(There are some users who would state this particular case
more strongly; they see the manufacturer as misleading: he
overstates the performance of his products and he understates the time it will take him to deliver th((m. The
difference is one of intent. I take the more optimistic view
here because if the other is, in fact, the true state of affairs,
then there is little hope that any measure of cooperation
between user and manufacturer can ever be achieved.) He
assumes that they will work as advertised. He assumes that
he has solved the problems of the world, and that his
products will need little correction, less modification, and
no extension.
Finally, the manufacturer, to the user, is a radical: he
subscribes to the Detroit philosophy of planned obsolescence, change for the sake of change. Good programmers
are creative people, and he must allow his programmers to
express themselves. (That this frequently imposes a distasteful burden upon the users' programmers is unfortunate, of
course, but it can't be helped.) Furthermore, it is in his
interest to keep the users somewhat off-balance and hence
somewhat dependent upon him.

The User as Seen by the Manufacturer
The user, as seen by the manufacturer, is possessive: the
system is his by purchase or lease; possession is nine points
of the law. He is unwilling to relinquish one memory cell,'·
one storage cycle, no matter how great the benefit. It must
be made to work in accordance with his conception of how
a system should work regardless of whether his concepts
were designed into it or not. Since the system is his, its past
and its future must be his also; his curiosity about these
matters is insatiable.
To the manufacturer, the user is visionary: his estimates
of the cost (to the manufacturer) of a "minor" change in
direction are unrealistically low; his estimates of the benefits to mankind of the implementation of his pet project are
unrealistically high. He ignores side effects.

"To the manufacturer, the user's estz"mate of the
cost of a "minor" change in direction are unrealistz"cally low; and his estimates of the benef£ts to
mankind of the implementation of his pet project
are unrealistically high. "

in his efforts to secure the realization of his inferences. He
is no less skillful in the creation of iron-clad contractual
commitments out of preliminary, tentative, incomplete,
and inaccurate data.
Finally, the manufacturer sees the user as an archconservative: he resists progress at every level if it involves change.
("The old ways are the good ways.") Technological improvements are complications to be mistrusted: old, familiar, trusted tools are not to be abandoned, no matter how
much more potent their replacements are. The game is
never worth the candle.

Who Is Which?
It should be emphasized that these stereotypes have
broad applicability: the term "manufacturer" is not limited
to "hardware manufacturer" any more than the term
"user" is limited to "occasional Fortran or Basic user". In
fact, most people involved with computing alternate between these roles, sometimes with bewildering rapidity in
the course of a single conversation.
The lines are most firmly drawn, however, and the
stereotypes most firmly entrenched, at two levels (both of
which exist in almost every shop): the interface between
the outside vendor and the systems programmers, and the
interface between the systems and applications programmers. One result of this situation might be that systems
programmers, having a foot in each camp, serve as conciliatory agents. They might, when dealing with the outside
vendor, remember that they, too, have delivered late; that
their own products have been released with live bugs; that
their own documentation has been incomprehensible or
non-existent. Remembering these things, they might react
with sympathy and understanding. They might, when dealing with the applications programmers, remember that they,
too, have resisted change; that their own programs have
failed to run on new releases; that the documentation they
have received has been incomprehensible or non-existent.
Remembering these things, they might react with sympathy
and understanding.
But of course they do not. They remember, certainly,
but instead of acting as fIlters they act as amplifiers. The
abuse hurled at them by the applications programmers is
hurled with increased vigor at the vendor; the scorn heaped
upon them by the vendor is heaped even higher upon the
applications programmers. The process is iterative and with
each iteration the stereotypes become ever more firmly
entrenched and reality approximates the stereotypes ever
more closely.

Changing the Image
The user is also selfimportant: his problems are the
most important problems in the world, and should be
solved first; his solutions are the most promising, and
should be tried first. Whatever is in his interest is in the
interest of the whole universe of computing. Features he
does not use are barnacles on the ship of progress. His is the
one clear voice of truth amid the constant tumult and the
shouting.
The user, to the manufacturer, is implacable: he is
unforgiving, vindictive, and possessed of a very long memory. The slightest delay is intolerable, the tiniest imperfection is inexcusable, the least hint of change in specifications
is totally unacceptable. The user is an expert in the creative
interpretation of reference documentation, and unrelenting
26

It is clear from the above that the relationship between
user and manufacturer has become a combative one highly ritualized, it is true, but combative nonetheless. It
will continue to be combative until the pejorative stereotypes have been replaced by more cooperative models, for
the user as well as for the manufacturer. The substitution
need not be instantaneous, but it must be undertaken by

"The relationship between user and manufacturer
has become a combative one - highly ritualized, it
is true, but combative nonetheless. "
COMPUTERS and AUTOMATION for September, 1970

both parties or it will be ineffective, incomplete, and quite
temporary. The models suggested below may not be optimum, but they are steps in the right direction.
The attitude a manufacturer adopts towards his creations should be somewhat akin to parental pride, remembering that good parents recognize and treat illness and
injury. Good parents foster the development of their
children (and much development takes place outside the
home), and recognize that they have limitations beyond
which they should not be pushed. The users, on the other
hand, should recognize that children are often recalcitrant,
and usually require some training before they are capable of
behaving properly away from home.

Standards of Communication
User and manufacturer need to adopt standards of
communication not involving ritual attack and defense. The
manufacturer should seek the users' advice before he fixes
the specifications of a new product; he should recognize
that users have a legitimate need for advance information.
Users should remember that advance information is often
inaccurate or misleading, and should accept that risk when
they accept the information. Dialogue concerning bugs
should be undertaken with the object of improving the
product, not as exercises in invective (by the user) or
skillful evasion of responsibility (by the manufacturer).
Both manufacturer and user need to broaden their
perspectives: The manufacturer needs to recognize that the
users harbor a wealth of experience and expertise which
could be tapped, if the manufacturer would make the
effort. The users need to recognize that the manufacturer
must be responsive to other users with other priorities.
Both user and manufacturer should try to look at things
from the other point of view before criticizing.

Progress Exacts Its Toll
Finally, user and manufacturer should work together to
define and regulate the development of the product. The
user should recognize that improvement is rarely achieved
without some sacrifice, that progress exacts its toll. The
manufacturer should recognize that change must be justified, and that his estimate of the cost of the change should
include the cost incurred by the users in adopting it. Th~
user must recognize that some growth (change) is inevita~
bIe; the manufacturer must realize that not all progress
(change) is desirable. User and manufacturer should agree
upon which changes should be provided by the manufacturer and which should be left to the user. The manufacturer
must expect the user to make some changes; those areas
where changes are likely should be designed to simplify
their installation. The user who makes such changes must
be willing to accept the non-conformity he has thus
created.
In brief, both the user and the manufacturer need to
grow .up a little. The present struggle is fun, but it is
childish. There is much loose talk about the coming-of-age
of the computer (which is a far different thing from the
coming age-of-the-computer): The computer cannot achieve
maturity until those who direct it do.
0

PROBLEM CORNER
Walter Penney, CDP
Problem Editor
Computers and Automation

PROBLEM 709: SQUARES

OR SOMETHING

Harry was beaming as he punched the keys of his trusty
desk calculator. "I'll have to publish this," he said, "I've
never seen it in any mathematics book."
"What's that?", asked John, "Did you prove Fermaes
Last Theorem or something?"
"Something, all right. I've been calculating factorials and
subtracting each one from the next larger square. And what
do you know - every difference is a square. At least for N
greater than three."
"Are you sure that isn't just coincidence?"
"Coincidence? I'm up to 10 ! and everything's working
out right. 10 ! is 3,628,800; subtract that from 3,629,025
which is 1905 2 and you get 225 which is 15 2 • Likewise 9 !
is 362,880; subtract that from 603 2 . and you get 729 or
27 2 • And so on down the line."
"Yes, you've got something there, all right, but I'm not
sure what."
.
"All right," said Harry. "I'll work out one more case on
this cement mixer, and if it's. O.K., I'll write a program for
the first 100 factorials. Here goes 11 !"
"I'm still skeptical," said John.
Has Harry stumbled onto something big?

Solution to Problem 708:
A Coin Tossing Program
The average number of tosses for H H H is 14 and for
H THis 10. The number of ways of getting H H H on the
third, fourth, ... toss is 1, 1, 2, 4, 7, 13, ... where each
value is the sum of the three preceding. For H T H the
corresponding figures are 1,2,3,5,9,16 ... where un :::
2U n -l ~ un~2 + un 3·
w

Readers are invited to submit prOblems (and their solutions) jor
publication in this column to: Problem Editor, Computers and
Automation, 815 Washington St., Newtonville. Mass. 02160.

Research for this article was done under the auspices of the U.S.
Atomic Energy Commission.
COMPUTERS and AUTOMATION for September, 1970

THE "LANGUAGE EXPERIENCE" APPROACH IN
TEACHING READING - COMPUTERIZED
Mrs. Florine L. Way

P.O. Box 499
Monticello, Fla. 32344

"Both the content and the language of each pupil's reading lessons need to
be relevant to his own life experience and understanding. "
Learning to read can be inhibited by the use of material
that is irrelevant, too difficult, or just plain boring. A
pupil's self-image can reach rock-bottom when he fails to
learn to read - and when he can't learn without someone
beside him helping him with every word. But where are
there enough teachers - or even enough dedicated volunteer tutors - to sit beside each pupil who can't read and
help him in a one-to-one relationship?
There is no single, simple answer to the problem of
teaching. everybody to read.

relevant content with which to learn the skills of reading.
Regardless of the difficulty of the words and concepts he
uses, he can already recognize them aurally, understand
them, and use them orally, before he dictates them. Once I
had to keep a dictionary beside me to spell the technical
words that a seventh-grade non-reader used, when he
dictated his own experience stories for me to type to create
his reading lessons. Real and vicarious experiences can be
arranged for the pupils, 'so that their vocabulary and
conceptual understanding can be internalized, before dictating their lessons.

What is "Language Experience" Reading?

Disadvan tages

One of the most promising practices for teaching beginning and remedial reading was introduced in the early
1900's by a teacher named Miss Flora Cooke. She used the
pupils' own language to create the content of the reading
lessons. This way of teaching reading has earned a respected
place in education. It has been used to teach: beginning
readers, retarded readers, and adult illiterates. It has been
used to teach Persian peasants, prison inmates, Maori
children in New Zealand, delinquents who were nonreaders,
children with limited backgrounds. The most common
name by which this method is known is the "Language
Experience" approach.
In the Language Experience approach, the teacher literally brings together the pupil and his environment to create
his reading lessons.

The disadvantages of using the Language Experience
approach are also clear. Teachers already have too much to
do, especially teachers of beginning readers, retarded readers, and adult illiterates. How can they take on the added
task of typing or writing for each pupil his own dictated

Advantages.
The advantages of using each pupil's own language are
clear. Regardless of his age, sex, race, social status, or
cultural background, each pupil in a classroom can have

Mrs. Florine' L. Way is one of the two prizewinners in the
Second Annual Martin Luther King Memorial Prize Contest
sponsored by Computers and Automation. (See announcement in Aug. issue, page 9.) Mrs. Way has been a classroom
teacher for twenty-nine years. She has an A.B. from the Univ.
of Miami, and an M.S. from Florida State Univ.

28'

"How can the teacher take on the added task of
typing or writing for each pupil his own dictated
lessons every day?

lessons every day? How can they make time to sit beside
each pupil until he attains the sound-sight relationship
between what he has dictated and what is on the page?
How can the teacher have time to make a test for each
pupil on his separate lessons to see how well he is learning?
What kind of reporting can be done when each pupil dicates
. different lessons about different experiences? The teacher
needs to have the time to use her humanness and her
expertnes') to motivate pupils, to interact with them, to
create the environment in which they can learn.
In an effort to solve some of the problems that are
involved in making the Language Experience approach
practical to use in a classroom setting, a small grant from
the Florida State University Research Council was obtained, and I conducted an evaluation pilot study at the
COMPUTERS and AUTOMATION for September, 1970

Federal Correctional Institution in Tallahassee, Florida, in
1965. This pilot study was successful. The results and the
responses of the illiterate inmate pupils (e.g., "Man, you got
to do something to get this way of teaching reading so
everybody that needs it can use it") indicated that the work
should be continued.

An Experimental Class

An experimental class at the Institute followed the pilot
study. During that time, it was found that technology could
be used to perform many of the chores for making the
Language Experience approach practical to use in a classroom. From experience as a classroom teacher and from the
reactions of the illiterate prison-inmate pupils, I designed
and wrote the specifications for a project called DOVACK
(Differential, Oral, Visual, Aural, Computerized, Kinesthetic). The Florida State University Computing Center whose
director is Dr. E. P. Miles, Jr., developed computer support
for this project.
In the DOVACK project of implementing the Language
Experience approach with a computer, the essential attributes are: (1) to give each pupil a chance to dictate his own
reading lessons to be processed by the computer for
independent study with his dictaphone; (2) to give each
pupil a chance to use multi-sensory techniques in his
independent study; (3) to give each pupil a chance to be
tested on his own vocabulary; and (4) to provide the
teacher with evaluations and reports at regular intervals to
be used in finding, and planning for, the needs of each
pupil.
It was necessary to obtain federal funds in order to
field-test and perfect the project. ESEA Title III granted
the necessary financial support to the Jefferson County
School Board, for me to direct a DOVACK project for
three years (1968 to 1971) in Monticello, Florida.
The following is a brief description of the project in
Monticello. There are two classrooms, two black teachers
and four black aides. This year, they serve one hundred
pupils a day. (A larger pupil-to-aide and to-teacher ratio is
possible under different circumstances. Presently there are
many visitors, many other demands on staff time, requirements for field-testing teaching strategies and techniques,
etc.) At present, all of the pupils are black. The classes
range in size from four beginning readers in kindergarten to
fifteen severely retarded readers in the sixth grade. Other
classes are for pupils from the first, second, fourth, and
fifth grades. It is anticipated that this fall the minimum
class size will be thirteen, and that a larger total number of
pupils will be served, although the staff will remain the
same.

Technological Support
In this project, technology supports learning, but does
not intrude. There is a dictating· machine on each desk,
which is used by approximately four pupils at different
periods during the day. Teacher aides use these machines
for transcribing the dictations of pupils, and the teacher
uses them for voice-taping the models of their lessons.
Sound-resistant material protects the classrooms from
noise. Aides punch cards on key punches to submit to the
computer, or use a Teletype for direct communication with
the computer at the Florida State University Computing
COMPUTERS and AUTOMATION for September, 1970

Center, which is about thirty miles from the project center.
The aides have learned on the job how to operate the
machines and use the necessary precision in their records.

The Teacher
It is not necessary for the teacher to understand computer programming or the technical aspects of the project. In
order to use a computerized Language Experience method
of teaching successfully, the teacher's attitudes, teaching
philosophy, and strategies for creating the learning environment, are crucial - just as they are in all teaching.

"It is not necessary for a teacher to understand
computer programming in order to successfully use
a computerized Language Experience method of
teaching. "

The teacher helps the pupils expand their experiences
and broaden their understanding, so that they can internalize their expanded vocabulary and concepts. The teacher
supplements the pupils' real experiences with vicarious
experiences by reading to them, shQwing films and film
strips, etc. All of the techniques and strategies that are used
with other Language Experience teaching can be used
effectively when the method is computerized.
The project capitalizes on the experience, folk humor,
and prior comprehension of each pupil. When each pupil
dictates his own lessons from his own experience and
understanding onto the dictating machine, he uses the
language patterns to which he is accustomed to speak,
listen, and understand. Both the content and the language
of each pupil's reading lessons need to be relevant to his
own life experience and understanding, so that each pupil's
own self-image and self-confidence are raised by assuring
him of positive support for the way in which he communicates with his family and his peers.
In addition to the strategies and techniques ordinarily
used in the Language Experience approach to teaching
reading, there are strategies and techniques peculiar to
DOVACK which enable the teacher, supported by technology, to fit to the individual three essential elements or
phases: teaching, testing, and reporting. [The documentation for computer support for these three phases, which
comprise the DOVACK FORTRAN (and COMPASS) package, is available on request.]

Objectives
Inherent in all of the educational objectives for this
project is the goal of enhancing each pupil's own self-image.
The teacher, supported by technological help, works
with each pupil to help him achieve the following objectives:
- develop skill in manipulating his own equipment
and materials;
- become independent and self-pacing in his study
habits;
develop favorable attitudes toward reading; and
develop proficiency in word recognition, word
attack skills, and general reading achievement.
29

Agenda
Each pupil has an Agenda on which various items are
listed. He chooses the several activities that he is going to
engage in the following day. Then he numbers them on his
Agenda in the order in which he plans to do them. Each
day, when he comes into the room, he gets his box
containing his materials and proceeds independently to
follow his Agenda for that day. After he completes a task,
he records the time beside it. The Agenda includes such
items as: Dictate a story. Read my story to the class. Study
WORD ATTACKER, which is a simple device for teaching
the sound-sight relationships between the names and sounds
of the letters and the consonant blends.)

enunciation and standard pronunciation. The pupil keeps it
in his box so he can use it for review any time he needs it.
As he listens, he follows the text with his eyes. He listens as
many times as he feels is necessary and then studies his
word list. Then he puts aside his earphones and studies
without earphones. When he studies his word list, if he does
not know a word, he looks at the number beside the word
which shows the line on which it first appears; he goes back
and reads that sentence again and recognizes the word in
context. With the WORD ATTACKER and other phonics
devices, the pupils are taught various other skills that they
can use in attacking unrecognized words. Finally, when the
pupil finishes studying yesterday's story, he proceeds to the
next item on his Agenda.

Procedure
For example, let's observe a boy in a class, during the
week of Martin Luther King's birthday. The teacher has
read stories to the pupils about Dr. King and has shown
them film strips and pictures. They have discussed his life.
Then each pupil dictates into the microphone of his
dictating machine what he understands from the lessons.
Here is the last part of the dictation of one boy in the fifth
grade, severely retarded in reading:
. . . He was a good man. They had to walk fifty
miles; Boy, they must have been tired. When one had
to use the bathroom, they just waved their hand and
a truck come along. There was some colored people
and some white people walking .... He was a leader
of the Negro. He liked to be good to others. He just
wanted them to be good to him. All he wanted was
peace and white persons and colored to go together.
He was a good man. He wanted peace. That is the end
of my story.
. The pupil finishes dictating his story, removes his dictating record, and takes it to the basket for the aide. The aide
transcribes the story on punch cards. The aide spells the
words correctly as she transcribes, but she does not change
the language usage, nor the grammatical structure of the
pupil's dictation. This transcription is included with others,
to be submitted to the computer each day for batch
,
processing.
When the pupil returns to his desk, he proceeds to the
next item on his Agenda. This might be: study his yesterday's story; join a peer for a mutual sharing of stories; etc.

Self-Teaching
The next day, this pupil receives the story he dictated
the day before, prepared by the computer for his independent study with his dictating machine. Each line of his text
is numbered. Each new word in the story is alphabetized
with a number beside it showing the line on which the word
first appears in the text of the story. He puts on his
earphones and listens to the verbatim voice tape the teacher
has made of his story. This voice-taping by the teacher of
each story for each_pupil serves as a model for clear

"Each day, a pupil receives the story he dictated
the day before" prepared by the compu.ter for his
independent study with a dictating machine. "
30

Testing
Every six days this pupil, along with the other pupils in
each class, receives a Random Sample Vocabulary Recognition (RSVR) Test. The words in his latest story are
compared by the computer with all the words in his stored
word bank; all new words in his story are added to the
bank. At the time for each RSVR Test, the computer
generates a random number and derives each pupil's test
from his own stored list of new words that he has used in
his dictations for that six-day period. The computer uses a
sample size designed to yield a standard deviation equal to
5 per cent of the estimate.
Each pupil administers his test to himself at his dictating
machine. This is a test of word recognition only. He says
the number beside the word (the words are numbered by
the computer) and then pronounces the word. Then he
takes the dictated record to the test basket. At the same
time the computer prints out all of the tests, it punches a
card for each pupil with a record of his work for that
six-day period. After checking, the tests, the aide completes
the record on each pupil's card. The completed deck is
returned to the computer for calculating and compiling
reports.,

Reports
After each six-day RSVR Test, reports. are printed out.
The estimated number of new words learned, and the
estimated rate of new words. learned per class period
present, are based on results. of the Te,st.
There is a ten-item pupil progress. report for each pupil
and a summary report of these ten items for each class.
Three dig,!ts of each pupiPs identification' number are used
to plot his position on the graphs., One histogram shows for
each pupil the total number of new words. dictated during
that six-day cycle; another shows the: estimated number of
new words learned during that cycle.
One coordinate, graph shows a plot of type-token ratios
against RSVR Test scores. for each pupil in each grade. (The
type-token ratio is the ratio of the: number of new words
used to the total number of words used.) Another coordinate graph shows for each pupil in each grade a plot of the
percentage, of new words learned against the number of new
words dictated. For the latter coordinate graph, to supplement the teacher's judgment" a transparent overlay is
furnished . The midpoint of the overlay is placed over the
computer-located coordinate point of the medians. In each
quadrant, a paragraph on the overlay suggests the kind of
help that the pupils who fall in that quadrant need.
COMPUTERS and AUTOMATION for September, 1970

Where
there's
magnetic
tape
you usually find this. .
Reason: MS-200 Magnetic Tape Head Cleaner is recommended
by leading manufacturers, prescribed by a major TV network and
has proved itself at hundreds of broadcast and computer installations. MS-200 is the safe, thorough, convenient way to get rid of
oxide dust. You can even apply it while tape is running.
Price: $2.90/ can in cartons of 12 16-oz. cans.
Trial order: 4 cans @ $3.75/can.
Price~ f.o.b; Los A~geles, Chicago or Danbury, Conn.

U.S. and
foreign patents
pending.

miller.stephenson chemical co.,lnc.
ROUTE 7, DANBURY, CONNECTICUT 06810

Alternate Forms of Communication

Following the Reports

The teacher and the aides write composite stories for
each pupil, using his own vocabulary. These composite
stories serve several purposes. They give each pupil a chance
to recognize his own vocabulary in a different context.
They give him a story to read about himself, and, since they
are frequently shared, about his peers. In the composite
stories, there is a subtle introduction to standard usage, but
the content is still based on each pupil's own vocabulary.

During the class periods following the reports, the
teacher attempts to meet the needs of the individual pupils
as indicated in the reports. Some pupils need to extend
their experiences, expand their speaking vocabularies, and
use a greater number of new words in their dictations.
Some need to work more on using the Language Experience
technique for finding and recognizing words in context.
Others need to interact more with peers and to exchange
.
stories with them_

Criteria of Success
"Here" is one kindergartener's own story: 'The
rooster is to the mama hen. The house is made of
wood. The one what drinking water is the sister
and the one what ain't is the brother. The hen is
looking at the rooster . .. ' "

The pupils understand that both ways of communicating
are good - the non-standard and the standard. (For
example, here is one exuberant kindergartener's own story:
"The rooster is to the mama hen. The house is made of
wood. The one what drinking water is the sister and the one
what ain't is the brother. The hen is looking at the
rooster .... ")
COMPUTERS and AUTOMATION for September, 1970

Three criteria are being used to judge the success of the
field-tests of the project: (1) adaptability to the populations for whom it was deSigned; (2) effectiveness in meeting
the specified learning objectives; and (3) economic
feasibility.
Hard data are being collected for one part of the
evaluation. A formal report will be made on this phase of
the evaluation when the data are analyzed. Especially
interesting should be the analysis of the correlation between each pupil's own rate of learning his own vocabulary
and his standardized test scores in mental maturity and in
reading achievement.
A teleprocessing version of the DOVACK Model is being
field-tested this year, ·using paper tape for input and output
via telephone line. It is hoped that eventually it can be used
on a shared-time, shared-cost basis.
D
31

REPORT FROM GREAT BRITAIN
While the common carriers in America are under heavy
pressure from computer users and manufacturers to provide
satisfactory solutions to the network problem, there are
signs in Britain that things are on the move.
Several significant events in the past month add up to
the conclusion that a decision may be forthcoming soon on
the form which a UK network (possibly connected to the
European continent) could take.

The Network of the Post Office Corp.
On the side of the monopoly, the Post Office Corporation, a 48 kilobit transmission link has been set up connecting London with Birmingham, and has functioned since
July 1 as the fastest data service in Europe. Its main
function for the moment is to enable users to familiarize
themselves with problems of using such a rapid means of
transferring data in bulk. Equally important, if the service is
to be a commercial success, is for manufacturers of terminal
equipment to use the new switched network to develop 48
K units compatible with Post Office installations - very
little standard interface equipment in this group has so far
been built.
.
Almost simultaneously with the 48 K switched network
announcement, the Post Office disclosed that the number
of terminals in the UK is now 14,000, and is doubling each
year. This suggests much higher installation figures than the
conservative 50,000 by 1973 and 1/2 million by 1983 that
the monopoly common carrier is using for planning purposes.

The National Physical Laboratory's Facilities
While all this was going on, the "ginger group" at the
National Physical Laboratory threw open its doors to the
computer press to show what it had achieved in the
construction and operation of a digital network. It is
providing flexible communications between all parts of the
very extensive laboratory, with a spur to the NPL's ship test
facili ties.
I call it a ginger group because for several years it ·has
been prodding the elephantine Post Office in the path it
should tread to provide a satisfactory network according to
NPL's lights. This would be based on a trunk system with
high-speed digital characteristics having small control computers at its nodes. These would function as the nerve
centres to which local computers would connect, serving
either groups of peripherals or subscribers' own computers.
Nodal design would give a response time of ten milliseconds between the five furthest points in Britain. Information is broken up into "packets" of 1,024 bits and the
message switching computers at the nodes would have
capacity to handle 2,000 of these packets a second.
The NPL network now in use applies these principles which have already been adopted by at least one American
organization - and has a small computer to which all
32

peripheral equipment connects directly (or through multiplexers if the units are slow in operation). The "node"
computer switches packets of information to give intercom·
munication between any of the laboratory's devices. These
can be simple printers, displays, experiments, and laboratory computers. There will also be a large computercontrolled magnetic file available to all users.
The main objective in establishing the design has been to
set up a system which will cope with any sudden demands
for extra services and which is independent of any particular technology.

The Plessey Terminal·
A significant feature is the use of a data terminal built
by Plessey Telecommunications Research. It operates at
1.152 bits/second or fast enough to transmit all Shakespeare's plays in under two minutes. Plessey has also been
the recipient of an important slice of repayable government
aid - $10 million to be precise - to which it will add a
similar sum. The whole is being applied to the construction
of a stored-program control system for electronic telephone
exchanges.

Traditionalists vs. Innovators
This development program, aimed at the second half of
the decade, should go a long way to solving the increasingly
serious problems of telephone communications in Britain.
But will it help in the development of a full-scale switched
data network? Many observers think not.
This is because there appears to be a fight between the
traditionalist telephone engineers and the innovators who
believe their biggest customer in the second half of the
decade will be the computer. The stored program controller
is about as far as the traditionalist will go. Any mention of'
a separate data system sets off an immediate panic reaction.
This is because - as in America - there is an immense
amount of money sunk in local connecu"ons from subscribers to exchanges. In Britain the estimate is for about
$1 billion worth of equipment which, the Post Office says,
cannot be discarded and must work out its useful life.
Nevertheless, we are seeing the emergence of the first,
fast, dedicated system linking three major cities now. And,
who knows, perhaps it will link all major centres in five
years from now.

Ted Schoeters
Stanmore, Middlesex
England

COMPUTERS and AUTOMATION for September, 1970

)

liThe House is on Fire" - THE PROFESSION OF INFORMATION ENGINEER

Computers and Automation believes that the profession of
information engineer includes not only competcm;e in handling information using computers and other means, but also
a broad responsibility, in a professional and engineering
sense, for:

Accordingly, this department of Computers and Automa·
tion will publish from time to time articles and other
information related to socially useful input and output of
data systems in a broad sense. To this end we shall seek to
publish here what is unsettling, disturbing, critical - but
productive of thought and an improved and safer "house" for
all humanity, an earth in which our children and later
generations may have a future, instead of facing extinction.
The professional information engineer needs to relate his
engineering to the most important and most serious problems
in the world today: war, nuclear weapons, pollution. the
population explosion, and many more.

-- The reliability and social significance of pertinent input
data;
- The social value of the output results.
In the same way, a bridge engineer takes a professional
responsibility for the reliability and significance of the data
he uses, and the safety and efficiency of the bridge he builds.
for human beings to risk their lives on.

OUR TOP PRIORITY
William B. Johnson
President and Chairman of the Board
IC Industries, Inc.
135 E. 11 th Place
Chicago, Ill. 60605

"The blunt truth is that there could be ample warning of almost every
emergency and agonizing dilemma before it hits us between the eyes provided we organize to generate the forecasts."
Propriety impels me to warn the sqeamish that they are
about to see a man walk, quite deliberately, into a vast field
of quicksand. I propose to speak of an area where a fellow
must be fast on his feet and keep in constant motion to
avoid sinking into the morass without a trace. I'm not sure I
can avoid that fate, but in my view someone should give it a
try.
My proposition simply is that an effort must be made to
set up on firm ground a system of national priorities for
attaining essential, long-range social and economic objectives. Presently, we are lurching erratically from one crisis
to the next. We react frantically to current public expressions, which are as volatile as quicksand and involve hidden
pitfalls that are just as treacherous.
Public opinion is, of course, the decisive voice that
influences national policy. Unfortunately, much of the
public is often poorly informed and their views are too
frequently shaped by emotional forces that distort the
rational judgment essential for reaching sound decisions.
Further, public attitudes are often conflicting and they
change continually, sometimes capriciously. In the absence
of priorities, stopgap solutions to problems of current
public concern are contrived hurriedly even though the
problems should have been anticipated a decade earlier and
resolved primarily by logical analysis.

This article is based on an address which Mr. Johnson delivered to
the Town Hall of California, an organization of 6000 businessmen,
in Los Angeles on June 2, 1970.
COMPUTERS and AUTOMATION for September, 1970

To underscore the constant shifts in public opinion, let
me take you back two short years. At this time in 1968, the
Presidential primary campaigns were.reaching a climax here
in California. The most controversial issues then were the
war in Vietnam, racial conflicts and urban deterioration.

The Priorities: Political and Social
~

What are the major political and social issues today? I
think you'l1 agree that they are inflation, the war, environmental pollution and the rising incidences of crime and
narcotics addiction. None of the difficult problems that
commanded the public's attention in 1_968 has been resolved. The problems are just as urgent now as they were
two years ago, but with the exception of the war, the
nation's demands for decisive action are focused on an
entirely new set of priorities.
There generally is substantial agreement among architects of our society - economists, sociologists, ecologists,
scientists and politicians - on' what should be done to
improve the quality of life and promote the nation's
welfare. But there invariably is chaos when it comes to
deciding which problem should now receive the main thrust
of our resources and our coordinated planning and
implementation.
The normal political reaction is to downgrade the priority issue, to please as many as possible by doing a little bit
of everything. The inevitable result, of course, is that
nothing is done properly. The band-aids we slap on unsightly sores do not remedy the source of spreading infection.
33

What I'm proposing is practiced by every business
executive as a matter of necessity. The component parts of
a business enterprise are not different from the various
departments of government. The requests for capital expenditures always exceed the money a company can afford,
even in boom times. Even when every request can be
justified, some projects must be deferred and others abandoned because more urgent plans take precedence in timing
or importance. There are many fancy definitions of decision-making, but they mostly boil down to one key element
- the art of choosing priorities.

Benefits of Evaluating Priorities
The country will derive three immediate benefits if
public priorities -are evaluated in perspective, by a rational
and orderly process.
First, problems will be analyzed and approached objectively instead of by quick response to a series of heated
emotional emanations. All of us are acutely aware that·
shrill propaganda by special pleaders now exerts undue
influence on national policy. The squeaky wheel gets the
most grease. Congress must at least be exposed to expert
analysis, free of political pressures, or there will be no
judiciOUS allocation of public funds and public effort.
The second benefit is that priorities will help restore a
sense of balance and continuity to the federal government.
!n a less tur~ulent era, the interrelationship of important
Issues was more visible and pointed the way to national
goals. Although the thorny issues have increased greatly,
they can - and must - be integrated into an overall
pattern. That can be done only by weighing alternatives and
sticking to the judgments reached. But we are continuing to
fling ourselves onto white horses and charge off madly in all
directions, pursuing multiple, inconsistent alternatives. We
need the benefit of choice~

"A broad-gauged ordering of federal programs will
expose an incredible mishmash of bureaucrats who
are impeding progress to current goals, by maneuvering to preserve little empires set up to achieve
objectives that have become obsolete. "

The third benefit brought by a system of priorities will
be reflected immediately in the budget. A broad-gauged
ordering of federal programs will expose an incredible
mishmash of bureaucrats who are impeding progress to
current goals by maneuvering to preserve little empires set
up to achieve objectives that have become obsolete. Those
of us with corporate experience know how insidiously these
empires spring up and hang on. We clean them out periodically - although not often enough - but in Washington
they keep on growing with every new administration.

A Remedy: Known but Unused
Twenty-one years ago a commission headed by former
President Herbert Hoover submitted a landmark report on
the reorganization of the federal government. Its Number
34

One recommendation was the elimination of two-thirds of
the agencies reporting to the President. Since then, there
has been such a vast proliferation of agencies and commissions and task forces that it's safe to say 75 percent of them
can be dropped with a corresponding increase in efficiency
- and no loss in terms of achieving "what the world needs
now."
A typical example of confusion and cross-purpose in
Washington is the current study of weather modification.
The Department of the Interior looks at it from the
standpoint of increasing rainfall. The Department of Agriculture is interested mainly in reducing crop losses. The
Federal Aviation Administration is pushing for measures to
combat fog. None of the three agencies is concerned
primarily, or even significantly, with total systems effects
or possible side effects.
Sometimes it seems that painful experience has taught us
nothing. Just last month a Presidential study group submitted a report to the White House identifying continuing
leadership in science and technology as a vital national goal.
I quote from the report: "Our national progress will
become even more critically dependent upon the excellence
of our science and technology. A vigorous, high-quality
program aimed at advancing our scientific and technological
capabilities - including the social, economic and behavioral
components - is vital to all national goals and purposes."
Here we go again. The study group threw practically
everything into the pot. It is foolhardy, and terribly
wasteful, to blaze away indiscriminately with a battery of
howitzers instead of appointing a squad of expert riflemen
to aim at the most imperative targets.
The crux of the s~ory, as reported by The New York
Times, came in the closing paragraphs. The study group,
tacitly conceding that maybe we cannot dominate the
entire scientific spectrum, recommended that the Office of
Science and Technology, which is headed by the President's
chief science adviser, be designated to establish priorities in
the field. It is rather typical that the one realistic note was
added as an afterthought.

Underaction and Overreaction
Look at our overreaction to environmental pollution,
which suddenly has assumed the gravity of a full-scale
national emergency. Please believe I am not trying to
dismiss the problem, nor am I implying that we have been
sold an unduly alarming bill of goods. But I do know that
we have been bombarded with many wildly conflicting
predictions and a great deal of emotionalism and frenzy.
As a concerned citizen and businessman, I would like to
know whether the dangers of pollution have a more serious
physical and psychological impact on American life than
the housing shortage, which has been plaguing us for a
generation. Before we embark on a giant crash program to
stamp out pollution, at any cost, I'd like to know something more about that cost. For example, will a convulsive
effort to correct our abuses against nature delay - or
derail - the drive to correct other historic abuses perpetrated against our fellowmen. The social climate also is a
vital part of our environment.
I don't mean to belabor the pollution issue, but it is a
conspicuous example of the tendency of government on
federal, state and local levels to react too late, and then too
violently, to questions that were swept under the rug years
COMPUTERS and AUTOMATION for September, 1970

ago. Industry suddenly has become the arch-villain in the
agitation against pollution. The crusaders blithely ignore
the fact that there has been a drastic and sudden change in
the rules of the game. For more than a century, all that the
public asked of technology was that it turn out a succession
of cheap new products and services that made life a little
easier and more convenient.

"Industry suddenly has b~come the arch-villain in
the agitation against pollution. Yet for more than a
century, all that the public asked of technology
was that it turn out a succession of cheap new
products and services that made hie a little easier
and more convenient. "

Since cheap, efficient consumer goods was the public's
top priority, industry responded by raising the standard of
living in America to unrivaled heights. Now, suddenly,
critics are rejecting the benefits of mass production and are
applying new criteria to the technology that makes it work.

Industry Efforts to Curb Pollution
Industry concedes that pollution must be curbed. Some
20 percent of the capital expenditures of steel companies is
directed to environmental control, with a resultant drop in
productivity per million dollars of investment. Detroit is
pouring the equivalent of a fair-sized South American
country's annual budget into research to reduce air pollution by automobiles. Manufacturers of bottles, cans and
paper cartons are earmarking millions of dollars for the
development of self-disintegrating containers. Airlines are
experimenting with noise-abatement devices.
A lot more will be done as soon as Washington sets up
guidelines for questions that surely cannot be ducked
indefinitely. But there are also the questions of who is to
pay for environmental control and what it will cost the
people. What is the top priority - a pollution-free environment or low cost products and services? Suppose the
former makes the latter impossible - and surely we can't
have it both ways. If Washington yields to popular pressures
and makes industry, and thus the product users, absorb the
burden of curbing pollution as a cost of doing business,
we'll see an unknown degree of violence done to another
facet of the environment - the daily cost of living. That
may indeed be the best answer, but it should be done
knowingly, if at all, and there should be no later surprise at
the consequences.

The Whipping Boys of Change
The growth and stability of small businesses is a highly
desirable element of the free enterprise system, but some of
us have become aware that the campaign against pollution
is beginning to take a heavy toll of small or marginal
operations. In states where stringent pollution laws' have
been adopted, many small foundries that cannot afford to
COMPUTERS and AUTOMATION for September, 1970

comply with the regulations have shut down or are for sale
at a sacrifice. It is a predictable cinch that similar difficulties will cause widespread havoc in other sectors of the
economy.

The Public Utilities
What about public utilities? They have been made
whipping boys for air and water pollution with complete
disregard for the facts. Until a few years ago, regulatory
agencies were slow to approve the use of cleaner fuel and
the installation of purification equipment in public utility
plants. Public Service Commissions, concerned mainly with
holding down rates, generally withheld the necessary permission.
A knock-down fight is brewing in Chicago, my hometown, over power rates. Chicago always generated its power
by burning cheap coal from nearby Illinois mines, but this
coal is relatively high in sulphur. In response to today's
public agitation, and apparently without considering technological decontaminants, Chicago has simply outlawed the
coal of its own state and now is bringing low-sulphur coal
from Wyoming. It hardly is necessary to hold a degree from
the Harvard School of Business to understand that rates
must be increased to meet the cost of hauling more
expensive coal halfway across the continent, but the utility's application for the rate increase is bitterly opposed by
those whose agitation has contributed to its necessity.
This is the public-be-damned attitude with a reverse
twist. If the rate increase is not granted, the investors in the
utility bonds and stocks will be thrown to the wolves for
the benefit of pressure groups who demand higher-cost
electricity at the old low rates.
It has become a cliche to say that critical events erupt
with bewildering rapidity in the pressure-cooker we call our
world. It is comforting for our government officials - and
businessmen - to take refuge in that bromide to excuse
copping out on their responsibilities to take effective action
when they are confronted by difficult decisions. The blunt
truth is that there could be ample warning of almost every
emergency and agonizing dilemma before it hits us between
the eyes - provided we organize to generate the forecasts.

The Auto: Boon or Bad?
Nothing illustrates so clearly the fallacy of identifying
national goals by emotion, without consideration of eventual consequences, as the evolution of the automobile.
Although the motor vehicle has contributed heavily to the
serious financial plight of the railroad industry, of which
my business is a part, I must confess that it has had a more
profound impact on our society than any invention of the
twentieth century.
It is impossible to envision the United States as it is
today without automobiles and trucks. The majority of the
population would be living in abject poverty in rural areas.
Industry would be concentrated in huge, dreary complexes
that would make the factory towns of Victorian England
look like garden spots. The amenities of suburban life,
leisure and cultural interests would be confined - as they
always had been - to the very rich. Yes, the motor vehicle
opened up alluring vistas of freedom, mobility and flexible
freight transportation. Who could have foreseen that Henry
35

Ford's marvelous Model T would lead to poisoned air and
congestion that is strangling our cities?
Yet, innumerable gentlemen made precisely those forecasts when cars began to burgeon on the streets and road~

"There were fewer than 10 million motor vehicles
registered in 1920, compared with more than 100
million today. But even then, traffic engineers were
drawing up projections of monumental congestz"on
in cities and intolerable driving conditions on
highways. "

of America after World War One. There were fewer than 10
million motor vehicles registered in 1920, compared with
more than 100 million today. But even then, traffic
engineers were drawing up projections of monumental
congestion in cities and intolerable driving conditions on
highways in the event that private cars supplanted public
transit systems for commuting to work and engaging in
social activities.

40 Years of Congestion
So the federal government and states proceeded to pour
tens of billions of dollars into highways that compounded
congestion. Even back when I was going to college in the
1930's, it was, true that a boy on a bicycle could cross
. midtown Manhattan faster than an automobile could. Traffic in downtown Boston, Chicago, Philadelphia and other'
major cities was a worsening mess. It seemed to be obligatory for every comedian doing a radio or TV show from
Hollywood to pull gags about the jams and accidents on the'
freeways around Los Angeles - although I doubt that the
conditions were a joke to people who had to endure them.
Getting to work or going down to see a movie at night
became an increasingly harrowing ordeal, and still the top
priority in transportation was ignored. In the last 15 years,
$200 billion of taxpayer funds have been spent on highways compared with $61 million on railroads, mostly since
1966 in the northeast corridor from Washington to Boston.
In 1966, after 20 years of earnest petitions, the federal
government finally created a Department of Transportation, but still there is no balanced transportation system or
fund allocation to achieve one. Last year $16 billion were
spent on highways and only $30 million on railroads.
The shape of things to come was seen on last April 22,
Earth Day. New York City banned all vehicles from a major
traffic artery and closed Fifth Avenue for two hours at
noon, turning it into a pedestrian mall. Inevitably, all
private cars and trucks will ultimately be banned or heavily
taxed during the day in the business sections of all cities, as
vehicle registrations keep pace with the population increase.

Alternatives - and Costs
New York has a serviceable, although deteriorating,
subway system that keeps the city viable.. But what will Los
Angeles do when it becomes necessary to bar the private
cars of workers, shoppers and visitors? Los Angeles is the
36

only city of its size in the Western world that does not have
a single commuter railroad line. Its municipal transit system
is reput~dly something less than adequate. What is Los
Angeles planning to do when the freeways no longer are
permitted to disgorge cars into business sections? Has
serious consideration been given to constructive plans to
cope with the situation?
Deciding on the top priority in any broad area that
affects the public interest involves something much more
profound than comparing the costs of alternatives. It
embraces a philosophy of government - and that is the
crux of the transportation problem.

"Deciding on the top priority in any broad area
that affects the public interest involves something
much more profound than comparing the costs of
alternatives. "

Permit me to pose a hypothetical situation that is not as
. far-fetched as it may seem. President Nixon recently called
for a crash program cutting emissions of pollutants by
automobiles up to 93 percent by 1980, regardless of the
cost or the effect on the performance of cars. In March,
The Wall Street Journal carried a front-page story reporting
that many top engineers in Detroit doubted that the
internal combustion engine can be redesigned to meet those
requirements. Other authorities declared that a cleaner
engine will boost the price of cars by several hundred
dollars, increase maintenance bills and add to operating
costs by yielding from 10 to 16 percent less mileage on
unleaded gasoline. Apparently, the President will soon
propose a substantial tax on the more efficient leaded
gasoline.

An Example: Fuel Cells for Automobiles
Now, as many of you know, scientists have been experimenting for more than 50 years with the fuel cell, a battery
that runs on air by converting it into its chemical components. The fuel cell emits absolutely no pollutants and it is
more than a visionary dream. It is the source of electrical
power in the Apollo space vehicle and Allis-Chalmers
Company has converted it to experimental use in tractors.
There are two problems that have barred the adaptation
of fuel cells for automobiles. First, the battery must be
recharged frequently and there are no facilities to perform
that service. That's not a serious obstacle; stations would
spring up overnight if they were needed. The more imposing complication, however, is the weight and the cost or
the fuel cell which now rule it out as a substitute for the
internal combustion engine in cars.
No one who is even vaguely familiar with the small
miracles wrought by technology in this generation doubts
for an instant that the fuel cell eventually will be reduced in
weight and cost to make it practical for automobiles. You
may recall that when computers were introduced 20 years
ago, they had more than 1 ,000 vacuum tubes and were so
expensive and bulky that their commercial applications
were extremely limited. Today, computers contain no
vacuum tubes and compact desk models are on the market.
COMPUTERS and AUTOMATION for September, 1970

Similar advances in the fuel cell unquestionably will be
made.
Let us assume, as an intellectual exercise, that General
Motors is on the verge of the big breakthrough. Officials of
the company meet with Congressional committees and tell
them: "We can completely eliminate pollutants from cars in
three years by spending $200 million on research and
development of fuel cells. We can reduce pollutants 90
percent in five years by spending $200 million on redesigning internal combustion engines. We can't underwrite both
programs. On which one do you want us to concentrate?"

Effects on the Oil Industry
Your impulse is to select the fuel cell. It will achieve
better results faster. But hold on a minute.
The giant oil industry has invested billions of dollars in
equipment and installations in this country and abroad.
Cutting its market in half will throw a monkey wrench into
the nation's economy, perhaps touch off another recession.
What about the millions of employees and stockholders of
oil companies? Oil still will be vital to the nation's defense.
Should the federal government subsidize the industry to
ensure ample reserves, and what will the cost be? There are
222,200 gas stations in the United States. They support
two million people, counting the families of owners and
attendants. What happens to them and the operators'
investments, as business dwindles with the gradual disappearance of gasoline-powered vehicles? Incidentally, the
100 million cars and trucks on the roads today would not
ordinarily be phased out entirely for some 10 years. Whose
obligation will it be to maintain enough gas and service
stations to keep them running?

nological developments before they reach widespread use.
He proposed to set up an Assessment Board to advise
Congress - which votes the money for more than half the
nation's research programs - on emerging technologies that
are likely to be beneficial to society. By the same token,
the Assessment Board would spell out for Congress those
developments that might lead to unwanted or unintended
results and, therefore, should be restricted.
The board would solicit the advice of authorities in
appraising a wide spectrum of activities affecting the public
interest, ranging from human organ transplants to the
ecological implications of building dams, from the impact
of automation to genetic engineering. In short, Representative Daddario's bill provided the machinery for a system of
priorities after considering the benefits and drawbacks of
technological trends, much like the procedure of the Food
and Drug Administration in studying new medical products
for damaging side effects. The bill was shelved in committee.
o

The Priorities: Sifting and Weighing
The solution to the original problem seemed clear at
first, but on further thought it opens up a barrel of snakes.
There are so many ramifications that Congress is reluctant
to choose between the internal combustion engine and the
fuel cell. Is a 10 percent reduction of air pollution below a
minimum tolerable level, achieved two years sooner than it
otherwise might be, worth the economic dislocation it
entails? Congress seeks a Presidential recommendation, or
leaves the decision wholly to him.
35/e" x 5" x 'l'e"

"Is a 10 percent reduction of air pollution below a
minimum tolerable level, achieved two years sooner than it otherwise might be, worth the economic
dislocation it entails?"
R-100-6

Other holders, trays and accessories available.
Free sample and information sent upon request.

All right, Mr. President, what's your decision? The
President is not, of course, personally qualified to make
one. He must have the counsel of competent experts who
are qualified to reach a balanced judgment on the basis of
evidence presented on both sides of the question. Most
important, it must be a nonpartisan panel.
In 1967 Representative Emilio Daddario of Connecticut
introduced a bill to establish an early-warning system to
identify the probable consequences, good or bad, of techCOMPUTERS and AUTOMATION for September, 1970

BEEMAK PLASTICS
7424 Santa Monica Blvd. Lo. Angele., Calif. 90046

Phone:.213-87&-1770

The Need for an Assessment Board
I'm all in favor of something along the lines of Mr.
Daddario's proposal - with certain modifications I'll dis~
cuss in a moment. Let me first explain, howeve~, the need
for a board to identify national priorities. Friends and
associates, who have heard me sound off on this idea,
understandably throw up their hands and groan, "Another
federal agency! We already have too many bureaus, boards
and committees in our hair, driving us up the wall with red
tape and regulations."
There are, indeed, too many agencies and it is for that
precise reason that a master board is needed to coordinate
and eliminate hundreds of bureaus that run around aimlessly putting out brush fires. Their activities are fragmented, diffused and often biased because there are no
clear guidelines to what national policy should be.
Identifying national priorities is an enormously difficult
job bristling with political booby-traps, but it is a responsibility the federal government has already assumed, and I
merely suggest that it be done in a more sensible way. It
may sound new, ,but I believe there is nothing heretical in
asserting that a prime function of a republican form of
government is to look ahead, recognize trends and practices
that hold the seeds of social problems, then chart the
courses that best serve the interests and ambitions of the
greatest number of citizens. Businessmen have identical
responsibilities to their customers, stockholders and communities. Progressive executives meet that obligation by
setting up profit centers that draw, upon all available
scientific methods to solve their problems.
There is no earthly reason why a national priorities
board cannot follow a similar procedure of analysis and
evaluation, then submit its recommendation - with a
minority report where appropriate - to the President.

Public Endorsement
Public opinion must express itself through duly elected
representatives' and under our Constitution the ultimate
authority must lie with the Congress, not the exe'cutive of
an Assessment Board. This will temper the unwarranted but understandable - objections o"f the people to a dictatorship of eggheads on the board. Franklin D. Roosevelt's
Brain Trusters aroused just such antipathy a generation ago,
and the influx of Ivy Leaguers who accompanied John F.
Kennedy to Washington met similar opposition. National
policy must have the strong endorsement of the people to
be effective, a lesson several Presidents have learned the
hard way.

The Political Hazards of Priority-Picking
Mr.' Daddario's proposal would set up a 13-man board
composed of a director appointed for a six-year term, two
Senators, Two Representatives, the Controller General and
seven members appointed by the President. I don't want to
get bogged down here in the specifics of the board's
composition, but surely I would not include members of
Congress. The country needs definitions of goals that are
not distorted by partisan political considerations. That is
one of our major problems now.
Another important revision should be made in Mr.
Daddario's bill. He would have' the director of the board '
report to the Controller General, who then will send the
38

recommendations to Congress. You know what will happen. The recommendations will be lost in a maze of reports
by ad hoc committees, winding up in musty pigeonholes
crammed with old blueprints for a brave, new world.

"The country needs definitions of goals that are
not distorted by partisan political considerations.
That is one of our major problems now. "

I believe it essential for such ,a board to report directly
to the President, just as does the Council of Economic
Advisers. The prestige and high visibility of his office must
be used to help gain acceptance of policies opposed by
special interests.

An Obscured View of the Future
We must hope that the public will support the right
decisions if it possesses a reasoned evaluation of all the facts
pertaining to major issues. At least we would have a better
chance than we have today. Lord knows we are overwhelmed by facts, but we need broad-gauged analysis of
information to put them in proper perspective. A songwriter assures us that on a ,clear day we can see forever. It is
comforting to embrace that romantic notion, but on most
issues it is a harsh reality that currently we aren't looking
much beyond today's crisis - or the next election. And it is
not air pollution alone that is obscuring our view of what
the future can be.
We have no clear concept of our role in the international
arena, of our ultimate social goals, of the proper relationship, of capital and labor to government, or of foreign
policy versus domestic objectives. Science has given us the
tools and techniques for evaluating alternatives and then
reaching sound, balanced judgments. A priorities board,
chosen for competence rather than political considerations,
can help forge policies that will point the way to the
attainment of logically reasoned objectives.
Formulating a philosophy of government is a task for
people with far more imposing credentials than I possess.
I'm just a pragmatic businessman with a legal background one who knows the government is dissipating talents and
money on opportunistic and bureaucratic wheel-spinning.
The main thrust of our attention is now devoted to laws
and regulations that are merely ad hoc expedients to meet
given situations. But the squeaky wheel should not always
get the grease. And although the Daddario proposal may
not be the best approach to the problem of defining
national priOrities on a reasoned and orderly basis, there
should be some approach, so that we do get the three
benefits of objectivity, balance and termination of obsolete
activity. We really can't do everything that everybody

wishes; and decibel count is no basis for making the hard
choices. The basic policy problems confronting us are
10ng-hau1. Nothing short of a deliberate and rational process
of evaluation will guide us safely through the series of crises
that will continue to beset us in our own time.
0
COMPUTERS and AUTOMATION for September, 1970

COMPUTERS,

SCIENCE,

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.
Already a computerized analysis of information regarding the assassination of President John F.
Kennedy is underway in Washington.

AND

ASSASSINATIONS

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. Since this subject is not
receiving adequate and comprehensive coverage anywhere else that we know of, Computers and Automation
has taken the responsibility to publish.
No scientist, no honest man, ever refuses to consider new evidence or to correct errors. If corrections are needed or new evidence appears, Computers
and Automation will publish both.

PATTERN.S OF POLITICAL ASSASSINATI'ON:
How Many Coincidences Make a Plot?

Edmund C. Berkeley, Editor
Computers and Automation

CONTENTS
Are certain killings so extraordinary that the
theory "it is just a coincidence" must be abandoned and replaced by the theory "these killings
are correlated and there exists a correlation~
cause~ or conspiracy of some kind"?

Edmund C. Berkeley concentrated in mathematics while attending Harvard College, and graduated in 1930 with an A.B. summa cum laude.
From 1930 to 1948, except for 3-172 years active duty in the U.S. Navy. he did actuarial
work, first in the Mutual Life Insurance Co.
of New York, and then in the Prudential Insurance Company of America. He passed 12 professional actuarial examinations from 1931 to
1941 and became a Fellow of the Society of
Actuaries in 1941. He is the author of 12
books, including "Probability and Statistics:
An Introduction through Experiments", 121
pages, published in 1961, which has sold over
15,000 copies. This book also accompanies a
scientific kit bearing the same name, which
is now published by Math-Master, Big Spring,
Texas. He has been editor of Computers and
Automation since 1951.

COMPUTERS and AUTOMATION for September, 1970

The Case of the Fifteen Russian Generals
The Definition of Conspiracy
The Varieties of Conspiracy
The Conspiracy of Silence: The Concert of
Ideas or Attitudes
The Argument from Authority
The Argument from Tell-Tale Facts
The Argument from Statistical Reasoning:
Statistical Interlude
Calculating the Expected Number of Events
Measuring Spread or Scattering
Deciding Between Pure Coincidence and
Definite Correlation
Conditions for the Statistical Distribution
of Rare Events to Apply
The Instrument for Decision: End of the
Statistical Interlude
Applying the Instrument for Decision to the
Case of the Russian Generals
The Case of Political Assassinations in
Germany 1918 to 1932
The Case of the Black Panthers
The Case of The Assassination of Two
Kennedys and Martin Luther King, Jr.

40
40

41
41
41
42
42
42
43
43
43
44

44
45
46
47

Exhibits and Tables
Exhibit 1: Three Dispatches from
The New York Times
Table 1: Categories of Possible Deviations
Table 2: Instrument for Decision About
Rare Events:
Reasonable Variation;
Unusual or Strange or Suspicious
Correlation or Cause or Conspiracy
Exhibit 2: Black Panthers Meeting
Violent Death
Exhibit 3: Inquest on Dead Panthers
Opens Up Many Questions

40
43
44

46
47

The Death of Fifteen Russian Generals
In The New York Times during May 1969, there appeared three short reports about deaths of Russian
generals. These are shown verbatim in Exhibit 1.
In summary, according to these reports, 15 Soviet
generals died in the period April 19 to May 20,
1969. ,
This raises a significant question:
Are there enough generals in the Soviet army
so that 15 of them can die from natural causes
in 30 days? or can we confidently conclude
from a scientific calculation, that "coincidence" is extremely unlikely and therefore a
common element, perhaps a conspiracy, was responsible for their deaths?
In order to answer this question a naive person
might think of inquiring of the government of the
Soviet Union, for it might seem reasonable to ask
for more information. But the reticence of the
Soviet government to give out information is well
known. For example, in the summer of 1966 there
was a bad earthquake in Tashkent; I happened to be
in Moscow on a vacation at the time, and asked my
Intourist guide, "How many persons were killed?"
I was told, "In this country we do not make public
figures like that." Another example: during at
least 40 years, Aeroflot (the Soviet government
airline) has been flying scheduled air flights in
the Soviet Union;' it seems virtually impossible
Exhibit 1
THREE DISPATCHES FROM
THE NEW YORK TIMES, MAY, 1969

May 12, 1969:

A SOVIET GENERAL IS DEAD; 12th REPORTED IN 3 WEEKS

Moscow, May 11 (UPI) -- The Defense Ministry newspaper Krasnaya Zvezda today reported the death of
Maj. Gen. Avgust A. Nemme, the 12th Soviet general
to have died since April 19.
The death was announced in a notice signed by
"a group of comrades in arms". Nothing in the recent obituaries of Soviet generals has suggested
that there is any connection between them.
General Nemme was identified as a retired veteran of the armed forces. His age was not given, but
he was described as a veteran of the 1918-1921 civil
war and of World War II. No other details were
given.

May 18, 1969:

ANOTHER SOVIET GENERAL DIES

Moscow, May 17 (AP) -- The death of another general was announced today, bringing to 14 the number
of Soviet generals whose deaths have been disclosed
in the last three weeks. Krasnaya Zvezda, the official newspaper of the Soviet Defense Ministry,
reported the death of Lieut. Gen. G.K. Volkov at
the age of 70.

May 22, 1969:

SOVIET GENERAL DIES

Moscow, May 21 (UPI) -- The military newspaper
Krasnaya Zvezda reported today the death after a
brief illness of Lieut. Gen. Aleksandr G. Chernyakov.

that that airline should have had no accidents in
the Soviet Union, resulting in the deaths of passengers. But though I have searched for any announcement by Aeroflot of plane crashes in the Soviet
Union, and of the resulting number of deaths, I
have never found any reports.
There exists however another way to answer the
question about the deaths of 15 Russian generals in
30 days -- by a scientific calculation of the probability:
1. We calculate the number of expected deaths.
This turns out on conservative assumptions
to be 4.
2. We know the reported number of actual
deaths: 15.
3. We look in an appropriate probability table
to determine what is the chance that a
deviation as big as 15 or bigger could
occur without correlation, cause, or conspiracy.
4. We determine the answer: the chance is
less than 6 out of 100,000.
On the basis of this calculation we can come to a
very confident conclusion:
There definitely exists a ·strong correlation,
an extreme departure from normal happenings,
in the "coincidental" deaths of 15 Russian
generals within one month.
One possibility, of course, is that the 15 Russian generals were together in a plane, and the
plane crashed. Or that the 15 generals were in a
conference room, and the conference room was bombed.
Or something else. But the most reasonable explanation of 15 deaths occurring from April 19 to May 20
"after a short illness", appears to be that a secret
action of the Soviet government or the Soviet Communist Party produced the deaths of at least 10 or
11 of the 15 generals. The basic reason for this
supposition is that the Soviet Union has had a history of purges and liquidations, notoriously while
under the rule of Stalin; and even currently, persons who picket, protest, or dissent are treated
extremely harshly. The supposition is supported by
a scientific calculation which is described later.
The scientific calculation of probabilities can
also be applied to some questions of tremendous importance to the people of the United States:
1. Are the political assassinations of liberal
American leaders during the 1960's truly
the work of lone individuals in each case?
2. Does a thorough study of the evidence concerning the assassinations of President
John F. Kennedy, Senator Robert F. Kennedy,
and Martin Luther King, indicate conspiracies in their deaths? and perhaps even
a common conspiracy?
3. What is it that is happening in the United
States that has produced the deaths by
assassination of three outstanding leaders
in five years?
It is important for Americans to determine answers
to these questions. If we decide that these three
deaths are pure coincidence, we take one course of
action -- mainly, sit back and hope. If we decide
that these three deaths demonstrate correlation or
cause or conspiracy, we take other courses of action
-- mainly, we organize to put a stop to the assassination of liberal American leaders. The interpretation of events necessarily guides the behavior of
citizens.
The purpose of this article is to look into some
of the facts and some of the science that bears on
these questions.
The Definition of Conspiracy
What is the meaning of the word "conspiracy"?
One unabridged dictionary gives this definition:
COMPUTERS and AUTOMATION for September, 1970

conspiracy: 1 a: an illegal, treasonable, or
treacherous plan to destroy another person,
group, or entity; e.g., the conspiracy to murder Caesar; e.g., his theory of the trade
union movement as a conspiracy against the
unorganized worker -- L. A. Fiedler.
b: an agreement manifesting 'itself in words or
deeds and made by two or more persons confederating to do an unlawful act, or use unlawful means to do an act which is lawful: confederacy. 2: a combination of persons banded
secretly together and resolved to accomplish
an evil or unlawful end; e.g., a conspiracy
made up of storm troopers and disgruntled
aristocrats. 3: a striking concurrence of
tendencies, circumstances, or phenomena as
though in planned accord; e.g., the portentous conspiracy of night and solitude and
silence -- Ambrose Bierce
conspiracy of silence: a secret agreement to
keep silent about an 6ccurrence, situation or
subject, esp. to promote or protect selfish
interests, e.g., local manufacturers were accused of a conspiracy of silence on the childlabor situation
Webster's Third New International Dictionary, unabridged, published by G. &
C. Merriam Co., Springfield, Mass., 1961,
2062 pp.
The Varieties of Conspiracy

There are many varieties of conspiracy.
Over 5000 lynchings of Negroes have occurred in
the United States over more than 100 years. Was
this a conspiracy by Southern whites? To a large
extent, yes, by the Ku Klux Klan at least, and
other organizations; but in a wider sense, no. For
many of the lynchings were locally organized and
sprang from local attitudes of race hatred; akin to
the hatred that produces genocide on a very large
scale.
During the civil rights drive in the 1960's,
Medger Evers, a Negro leader working on voter registration, was shot in Mississippi. A white man
was arrested and tried; he was tried twice. The
evidence was conclusive that he had shot Medger
Evers, but in each trial the jury refused to find
him guilty. Was this a conspiracy by each of the
juries? In one sense, yes. In another sense, no.
Certainly, the jurors represented the strong feeling of a great many whites in Mississippi -- that
it was trivial for a white man to shoot and kill a
black man.
The Conspiracy of Silence: The Concert of Ideas or Attitudes

In a conspiracy of silence. the members of a certain class of persons, because of their point of
view, or the way they tend to behave, or the kind
of education which they have received, or their
economic interest, etc., stay silent about matters
that almost cry to high heaven. No spoken agreement is necessary for this kind of conspiracy to
exist: simply the common interests of a group of
persons. The principle may be expressed quite simply: "Don't rock the boat".
One example is the conspiracy of "respectable"
people about the misdoings of their own kind. Tobacco firms, knowing that they will have to cut
down their advertising of cigarettes soon, step up
their current advertising so as to "hook" more people before they are required to stop; and there is
no protest. Drug company XYZ charges as a price
ten times the cost of antibiotics, until the Federal government compels it to cut th~ price down to
COMPUTERS and AUTOMATION for September, 1970

three times the cost -- but no other drug companies
talk about it. The air in Los Angeles is sometimes
so dangerous, due to pollution by automobile engines, that school children are forbade to physically exert themselves during recreation time. Yet
the automobile and oil industries are certainly not
breaking their backs to quickly eliminate the gasoline-driven car.
Various important things happen as a result of
the conspiracy of silence. The damage that a bad
condition produces continues much longer than it
otherwise would. The profit which a lawbreaker
makes by breaking the law grows much greater, and
continues much longer than it otherwise would. The
spirit of the young people in a society tends to
br,eak, because the older people set a bad example
in tolerating and sometimes defending evil conditions that the older people ought to be able to
change and ought to try to change.
The Argument from Authority

In the case of political assassinations, many
people take the position that they cannot know
whether or not some conspiracy occurred untii accepted authorities have investigated and announced
a conclusion. They wait for authorities to speak.
For example, the American people by and large
suspended judgment after the assassination of President John F. Kennedy in Dallas, Texas, on November
22, 1963. Ten months later, the Warren Commission
announced that Lee Harvey Oswald was the sole assassin. Because of the Warren Commission's official prestige, many Americans discarded their previous doubts and accepted the findings. This acceptance allowed them to continue their ordinary
way of living without a nagging suspicion of a conspiracy that had been covered up.
In 1968, two more political assassinations occurred: Reverend Martin Luther King in Memphis,
Tennessee, on April 4, 1968; and Senator Robert F.
.Kennedy in Los Angeles on June 5, 1968. Again,
the majority of the American people suspended judgment and awaited the conclusions of official investigations. Again, authoritative sources reported
that a sole assassin was responsible: James Earl
Ray, who pleaded guilty to the murder of Martin
Luther King; and Sirhan B. Sirhan, who pleaded not
guilty in the shooting of Senator Kennedy. Both
were convicted in trials and sentenced.
The authorities, the standard newspapers, and the
establishment now consider these cases "closed".
And it seems as if they have often done their best
to stop these cases from being reopened.
It is of course sensible to study the findings
of official groups in regard to political assassinations. In such study the basic questions to be
asked include the following:
Has the official group looked into all the
important questions, including cui bono
(to whose advantage?)?
-- ,Have the findings reported on all the evidence systematically and impartially?
Has the group honestly and fairly evaluated the evidence that clashes with their
own conclusions?
It is frequently true that the "authoritative"
sources are not disinterested experts. For example, after Senator Kennedy's death the Los Angeles Police Dept. certainly did not want to seem to
appear as incompetent as the Dallas Police appeared
after President Kennedy's death. Perhaps this is
one of the reasons why their report, as expressed
by Robert A. Houghton, Chief of Detectives, in his
book, Special Unit Senator, deliberately suppressed
very important and disturbing evidence of a conspiracy.
41

Likewise, government officials certainly had "a
lot at stake" in the report of the Warren Commission.
What if there was a conspiracy? How would it affect the nation and the government? The principle
of "a lot at stake" also applies to the unofficial
investigators who seek the fame that may accompany
startling conclusions. Therefore, it is always important to consider the interests of the person or
group providing explanations.
People should not be misled by the "cult of the
expert": that only experts know the truth, and
that ordinary people should believe experts, to the
degree that the expert has authority. "Truth is
not shaped so that it can fit into the hand of a~y
one person", says an old maxim.
TWo of the most important arguments for dealing
with the political assassinations, reports about
them, and the question "How many coincidences make
a plot?" - are the argument from tell-tale facts,
and the argument from statistical reasoning.
The Argument from Tell-Tale Facts

Although it is sensible (and necessary) to study
the findings of official groups in regard to political assassination, it is sometimes difficult or even
impossible to consider their conclusions to be true.
There are too many "tell-tale" facts, that provide
a "dead give-away" of important information.
In the case of the political assassinations of
two Kennedys and Martin Luther King, important questions appeared immediately after major events in
connection with the assassinations. A great many
of these questions remain either without answers,
or with extremely improbable answers, because of
tell-tale facts. Here is only one of the many such
questions regarding each of the three assassinations:
1) How could Jack Ruby walk into the Dallas
police station on November 25, 1963, with
a gun, without being stopped, at just the
right time to shoot and kill Lee Harvey
Oswald - who had claimed he was a "patsy"?
2) Where and how did James Earl Ray acquire
the large sums of money he used while
hiding in the United States, Canada, England; Belgium, etc., until his arrest at
London airport?
3) Who was the "girl in the polka-dot dress"
who was reported to have run down a
stairs and said, "We've shot him. We've
shot him;~ and who was seen with Sirhan
B. Sirhan prior to Senator Kennedy's assassination by at least five witnesses?
(Only two of these witnesses are mentioned
in the book SpeCial Unit Senator by Robert
L. Houghton.)
A reasonable estimate is that there are dozens
of tell-tale facts of this kind, that upset the official or authoritative conclusions, for both the
assassinations of Senator Robert F. Kennedy and
Martin Luther King - and hundreds of such facts in
regard to the assassination of President John F.
Kennedy.
After all, there is a single tell-tale fact which
if it existed could completely disprove the Warren
Commission's conclusion that Lee Harvey Oswald was
the sole assassin of President Kennedy. This would
be a photograph of the 6th floor easternmost window
of the Texas School Book Depository Building between
the first shot and the last shot at President Kennedy, showing no one at all in that window. There
is reason to believe that such a photograph exists
or did exist, taken by a man named N. Similas; that
photograph was acquired by the Federal Bureau of
Investigation, and is now unavailable. (See the
article "The Assassination of President John F.
42

Kenne9Y: The Application of Computers to the Photographic Evidence" by Richard E. Sprague, in the May
1970 issue of Computers and Automation, Table 2,
p. 50, and Table 3, p. 56.)
But whether or not many tell-tale facts are available, there is often available another very strong
and powerful argument: the argument from statistical reasoning.
The Argument from Statistical Reasoning: Statistical Interlude

Over and over again in ordinary, everyday life,
we apply arguments from statistical reasoning, based
on the laws of probability. Much of the time we
are not even conscious of doing so. Often we use
these arguments rather unscientifically, and sometimes incorrectly because of prejudice or habit.
But all of us have a practical knowledge of many
basic concepts of probability and statistics. This
is revealed by words and phrases that we use correc tly over and over: "probably, likely, often,
seldom, almost always, almost never, the chance is
good that •••• , maybe and maybe not", etc. All of
these terms refer to facts studied in the science
of probability and statistics.
The arguments from statistical reasoning take a
variety of forms. Also, they require observations,
and counting and classifying of observations.
When we are examining rare events, however, events that are reasonably' expected to be unusual,
it happens to be easy to apply powerful statistical
reasonIng. We then come out with important conclusions, leading to decisions that have an extremely
good chance of being relJable.
For example, the whole business of insurance is
built around the rarity of events insured against.
Suppose your house is worth $20,00~ and you insure it against fire for one year. Suppose the annual premium the insurance company asks you to pay
is $80. The insurance company has basically two
costs: expenses for operating; and the cost of
paying claims. Suppose $30 of the premium goes for
expenses of operating. Then $50 is available from
your premium (and a similar amount from a great
many other people's premiums) for paying claims.
If the chance of your house burning in one year is
1/500, then the average claim cost for insuring
your house (and many similar houses) is $40, and
the insurance company has $10 per customer which
is left over for profits and contingencies.
It is a good bargain on both' sides. You have
the important safety of protection against fire;
and the insurance company can pay its claims and
its expenses and stay in business. In the long
run, the risks will all average out. The more contracts the insurance company enters into, the smaller will be the fluctuations, the scattering, the
spread. Such fluctuations can wipe out a small
company; but a big company can weather them.
Calculating the Expected Number of Events

First, we have to explain how the expected number of events can be calculated. How is the calculation made?
In the science of probability and statistics,
the expected number of events is always computed
from the following rule:
(the population out of which the events occur)
TI~ffiS (the probability of the event occurring)
EQUALS (the expected number of events)
For example, if you toss a coin 100 times, and
the probability of getting heads is 1/2 (in other
words a 50-50 chance), then the expected number of
heads is 50. Of course, common everyday experience
will tell you that fluctuation is likely. It would
COMPUTERS and AUTOMATION for September, 1970

be unusual if you should obtain exactly 50 heads;
the result you would obtain might be 47 heads, or
51 heads, etc.
But the ratio of number of heads to total number of tosses will come closer and closer to 1/2,
as you toss more and more times. And if that should
not happen, you would be certain that some bias was
present. Perhaps you are tossing in a biased way,
or perhaps the coin is badly made (perhaps bulgy
on one side and flat on the other), etc. Human beings have proved this law of probability literally
billions of times; it is not possible for the laws
suddenly to apply no longer.
Measuring Spread or Scattering

•

Having estimated the expected number of events,
and knowing the actual number of events, we now
come to a different question:
What is the probable spread or scattering or
deviation of the number of events? For example, suppose 4 is ·the expected number of
events and 15 is the actual number observed?
Is 15 a reaso~ably probable fluctuation?
This question has a very definite answer, determined by the laws of probability and statistics.
The applicability of the laws in this case depend
on several conditions which are here true; these
conditions will be described more completely further on. The particular law of probability applying here is called the "Poisson Distribution"; this
law will also be ~escribed more completely further
on. The laws of probability and statistics (including the Poisson Distribution law) have been
demonstrated by experience to be valid in literally
billions of instances.
Drawing on this knowledge, one can calculate
with complete correctness that a given wide deviation is extremely rare as a chance coincidence, but
is instead almost certainly due to definite correlation or cause or conspiracy. To make this calculation, first,we compute a measure of spread or
deviation called the standard deviation. Let us
call it S (the statisticians use a small Greek letter S and call it "sigma"). Second, (in this case,
the Poissondistribution case), we apply the scientific formula for computing S, which is:
(The standard deviation S)
EQUALS THE SQUARE ROOT OF
(the mean M)
Since M is equal to 4, S is equal to ~.
The standard deviation is a useful measure of
~egree or extent of deviations.
As soon as we
measure a deviation (degree of spread) in units
equal to the standard deviation, then we can immediately use a probability law to say how common or
how rare such a deviation is. This is explained in
the next section.
Deciding Between Pure Coincidence and Definite Correlation
Suppose we consider all the deviations that we
may find as actual observations; we can classify
them according .to the chances that they will occur.
(See Table 1, "Categories of Possible Deviations.")
Thjs table says the following:
• Deviations that are within two standard deviations away from the mean are in a
REASONABLE RANGE; the chance of getting
such a deviation is 19 out of 20 •
• Deviations that are further out than two
standard deviations from the mean but not
so far out as four standard deviations
from the mean are ·UNUSUAL or STRANGE or
SUSPICIOUS; and the chances of getting
such a deviation are less than lout of
COMPUTERS and AUTOMATION for September, 1970

Table 1
CATEGORIES OF POSSIBLE DEVIATIONS
Possible
Deviations

Categor:y:

Chances
of Observing

(2)

(3)

Two standard deviations from the
mean or closer

Reasonable
range (R)

19 out of 20

Further than two
standard deviations
from the mean but
nearer than four

Unusual,
strange or
suspicious
range (U)

Less than 1
out of 20
but more than
6 out of
100,000

Four standard
deviations from
the mean or still
further out

Correlation,
cause, or
conspiracy
range (C)

Less than 6
out of
100,000

20 but more than 6 out of 100,000.
• Deviations that are four standard deviations away from the mean or are further
out still are almost certainly due to
CORRELATION, or CAUSE, or CONSPIRACY:
the chance of such a deviation is less
than 6 out of 100,000.

Conditions for the Statistical Distribution
of Rare· Events to Apply
Above we referred to certain conditions which
enable the probability law called the Poisson Distribution to apply.
This law was named after a French mathematician
named Poisson; he was the first person to use it,
to study the distribution of rare events. The
particular historical case which he first studied
was the distribution of deaths of soldiers from
kicks by horses in Prussian regiments over the
years I
From this unusual beginning, the applicability
of this probability law has spread far and wide.
In the past the law has been called the Law of
Small Numbers or the Law of Rare Events -- but
these poor names for it have been abandoned.
The conditions under which this distribution applies are:
1. The events are independent; in other words,
the occurrence of one event (a death) is
~ associated with, has no likely influence upon, the occurrence of any other
event in the class of events being consiaered. (Clearly, if fifteen people
were together in a plane, and the plane
crashed, their deaths would not be independent. )
2. The ratio of the number of expected events
(M) to the number of the "population" out
of which the events occur (N) must be
"small". "Small" is regularly taken by
statisticians to mean less than 1/30. If
the proportion of M to N is greater than
1/30, then the formula for the standard
deviation changes to a more complicated
formula. And the Poisson Distribution
changes to another distribution called
the Binomial Distribution. If we make
these changes, then the formula for defining the standard deviation changes.
Of course, we can deal mathematically
with this case also, if we desire to.
Rut for rare events we do not need to.
43

The Instrument for Decision: tnd of the Statistical Interlude
For cases where the Poisson Distribution applies,
we can construct a powerful ttInstrument of Decision tt
that anyone can use.
This is a precalculated table of ranges of fluctuations. We have chosen three ranges of variations or deviations of fluctuations:
1) a range of fluctuations that are reasonable and to be expected;
2) a range of fluctuations that 'are unusual
or strange or suspicious;
,
3) a range that almost certainly (the chance
is of .99994 out of 100,000) indicates
correlation or conspiracy or common cause.
See Table 2.
This table applies when: (1) the number of the
population (out of which the events occur) is 30 or
more times the expected number of events; and (2)
the events are mutually independent -- in other
words, the occurrence of one event does not affect
or influence the occurrences of any other events.
For example, if the expected number of events is
4, it is reasonable that the actual observed number
be from 0 to 8 -- the chance is 19 out of 20 that
an observation will be in this range. If the actual
number of events is 9 or 10 or 11, this is unusual
or strange or suspicious -- the chance is less than

Table 2
INSTRUMENT FOR DECISION ABOUT RARE EVENTS:
-

REASONABLE VARIATION (R) i

UNUSUAL OR STRANGE OR SUSPICIOUS (U) i

CORRELATION OR CAUSE OR CONSPIRACY (C).

1!:

Category:

Chance,
19 out
of 20

Unusual, Strange,
or Suspicious
Range: Chance,
less than 1/20
but more than
6/100 1 000

Range of Correlation or Cause or
Conspiracy:
Chance,
6 out of 100,000
or less

to 2

3 or 4

5 or more

to 5

6 or 7

8 or more

to 7

8 or 9

10 or more

to 8

9 to 11

12 or more

to 10

11 to 14

14 or more

1 to 11

0, 12 to 15

16 or more

2 to 14

0, 1, 15 to 19

20 or more

3 to 17

to 2, 18 to 22

23 or more

7 to 23

to 6, 24 to 30

31 or more

10 to 30

3 to 9, 31 to 37

to 2,
38 or more

15 to 35

6 to 14,
36 to 44

to 5,
45 or more

18 to 42

9 to 17,
43 to 51

o to

27 to 53

15 to 26,
54 to 65

to 14,
66 or more

35 to 65

22 to 34,
16 to 78

to 21,
79 or more

Reasonable
Range:
Expected
Number of
Events

8,
52 or more

Note: This statistical table applies when the "population" (out of
which the "events" occur) is 30 or more times the expected number of
events, and the events are mutually independent.

lout of 20 but more than 6 out of 100,000. If
the actual number observed is 12 or more, then correlation or conspiracy or common cause is almost
certain; the chance of such an observation is less
than 6 out of 100,000.
The method which we have just described is a
powerful method for making decisions regardless of
authority. Just as anybody is able to decide correctly and prove to his satisfaction that 2 plus 3
equals 5, so anybody can decide that a figure is
far outside of a probable range, so far outside
that the figure indicates correlation or conspiracy
or common cause.
This power to practically decide that some occurrences are not a coincidence, but that there is
correlation or conspiracy or common cause, is truly
a great power. It increases your independence of
authority. It stops the wool from being pulled
over your eyes. It enables you to free yourself
from some of the misinformation which is offered
for you to believe. It helps you distinguish truth
from falsehood. Especially in the ttcredibility
gaptt between the government and the people of the
United States, this method is valuable for distinguishing what is probably true from what is definitely to be classified as untrue.
Applying the Instrument for Decision to the Case
of the Fifteen Russian Generals
Let us return now to the case of the 15 Russian
generals. How shall we apply the argument from
statistical reasoning, the instrument of decision,
to the deaths of 15 Russian generals in one month
from April 19 to May 20, 1969?
The procedure is:
1. Estimate the number of persons in the population.
2. Choose a typical or average age.
3. Look in an appropriate table of observed
annual death rates to determine the death
rate that applies to that age.
4. Multiply Item 1 by Item 3, obtaining the
expected number of deaths over a year.
5. Find the period of time for which these
persons were exposed to death (in years
and frac tions) •
6. Multiply Item 4 by Item 5 to obtain the
answer, i.e., the expected number of
deaths in the period observed.
We have very little information about the size
of the Soviet army and the number of Soviet generals. So, suppose we make two estimates, one which
is likely to produce the highest reasonable number
of expected deaths and another which is likely to
produce the lowest reasonable number of expected
deaths:
First Estimate:
1. Suppose we estimate the average age of
Soviet generals as 60. (Probably high.)
2. The annual death rate from all causes for
men aged 60 is 20 per thousand. (This
rate is the actual death rate shown in a
widely accepted Mortality Table called
the Commissioners 1958 Standard Ordinary
Mortality Table. This rate could reasonably apply to another 'advanced i_ndustrialized country like the Soviet Union.)
3. Suppose we estimate that the size of the
Soviet Army counting both men on active
duty and in the reserves, including veterans, is 12 million soldiers. (Almost
certainly overstated. Some of the mentions of Soviet generals in the Army
newspaper Red Star imply that a general
who is retired from the Army remains
classified as a general.
The size of
COMPUTERS and AUTOMATION for September, 1970

•

the active U.S. army is about 3 million
men; the size of the reserves perhaps 4 mi Ilion. But the Soviet population is greater
than the U.S. population, and the area of
the Soviet Union far exceeds that of the U.S.).
4. Suppose that there is one general for' every
5000 soldiers. (This is also probably
high. In the American army there is reportedly one general for every division
of 20,000 men; it is possible that more
than just "full" generals might be classified as generals by Red Star.)
With these suppositions, we can calculate how many
deaths we would expect:
1. There would be 2400 Soviet generals all together (12,000,000 I 5000).
2. There would be 48 deaths expected in a
year's time (2400 x 20 i 1000).
3. In the course of 30 days (or 1/12 of a
year), there would be 4 deaihs from natural causes (48 / 12).
4. We look in Table'2 in the line for the expected number of deaths equal to 4. We
find that 15 deaths imply correlation or
cause or conspiracy.
Second Estimate:
1. Suppose we ~stimate the average age of
Soviet generals as 55.
2. The corresponding death rate f!om the same
mortality table is 13 per thousand.
3. Suppose we estimate the size of the Soviet
Army as 4 million men.
4. Suppose we estimate tl!el)umber of generals
as 1 per 20,000 soldiers.
With these suppositions we again calculate how many
deaths we would expect:"
1. There would be 200 Soviet generals all, together (4,000,000 I 20,000).
2. There would be 2~6 deaths expected in a
year's time (200 x 13 /1000).
3. In one month there would be 0.2 (2.6 I 12)
deaths on the average from natural causes.
4. Suppose arbitrarily we call this 1 death.
We look in Table 2 in the line for the expected number of events equal to 1. Again
we find 15 deaths to imply correlation or
cause or conspiracy.
Other possibilities of course exist. For example, suppose that due to errors of reporting by
Red Star, a number of deaths of generals spread
over several months all happened to be reported in
one single month. However, if the information published in Red Star is to be believed, reporting 15
deaths of Russian generals in one month, the probability is overwhelming that a correlation or cause
or conspiracy affected the ~5 Russian generals who
"died after a short illness",
For a final check, let us determine what the effect of 15 deaths of generals per month would~mply
per year and per five years: 12 times 15, or 180
dead generals per year; and 180 times 5 or 900 dead
generals per five years. This rate of mortality
among generals seems on the 'order of magnitude of
Stalin's distrust of Soviet generals and their consequent widespread death.
The Case of Political Assassinations in Germany, 1918 to 1932

There is important confirmation of a pattern of
political assassinations in a number of historical
situations. One of these is the ~olitical assassinations that occurred in Germany "after she was defeated in World War I, and before the coming to power of Hitler as Chancellor of the German Reich in
1932. The following account is by E. J. Gumbel
who was a professor of statistics at the University
of Heidelberg, 1923 to 1932:
COMPUTERS and AUTOMATION for September

1970

••. The illegal military groups included an
array of fanatic terroristic organizations, small
in size, but important for their work of political
assassination in eliminating first the leaders of
the rWeimerJ Revolution, then prominent Republicans,
and finally-the enemies of the illegal rearmament •.• _
.•. Within the German population, especially
during the inflation period, substantial popular
backing, or acquiescence, was given to the evasion
of the Versailles Treaty. For the nationalists of
all shades these efforts were necessary for the restoration of national honor ••••
Terror as an Instrument for Enforcing Support
for Illegal Armaments. During the period of illegal German rearmament, terrorist methods were
wielded against the opponents of rearmament.
These included acts of personal violence carried
out by the terroristic nationalist groups supported by manipulation of the legal system.
Altogether, there were about four hundred political assassinations of the nationalists' foes. A
considerable literature was published in Germany
which detailed the charges •.••
The nationalist terrorists who enforced acquiescence in the rearmament of the Reich included
many men who later became Hitler's trusted adjutants, for overseeing the mass extermination program which the Nazis carried out during the Second
World War •••
The Role of the Law Courts. The political assassinations committed by the members of the former
Imperial and the secret armies put a heavy burden
on the administration of justice. The murderers
had to be acquitted and the victims .had to be shown
as guilty. This task was fulfilled by the employment of military courts which sided with the military men when they were accused of murder, by the
slowness of the justice enforcing agencies, by inability to find the guilty, by issuance of false
papers of identity by the police, etc.
Another procedure consisted in accepting at
face value the claim of the accused murderer that
the victim had tried to escape ••.•
To insure sentences members of the Reichswehr
responsible for the secret rearmament were called
as witnesses for the prosecution. In order to terrorize the public, many more trials were started
than could ever be completed .•.•
From "Disarmament and Clandestine Rearmament Under the Weimar Republic': (pp 203219) by E. J. Gumbel, in Inspection for
Disarmament by Seymour Melman and 19
authors, published 1958 by Columbia University Press, New York, 291 pp.
With a number as large as 400 deaths, it is
hardly necessary to invoke the instrument of statistical decision. In this case, . there is ample
significant and accepted evidence, which is reasonable arid overwhelming proof at first glance of
foul plai and not "coincidence".
Footnote:
E.~. Gumbel was one of a small group of pacifists
who exposed the illegal rearmament and the terrorisms connected with it. For this activity he was
three times' charged with high treason. In 1932 he
was dismissed from the University, and expatriated
by the Nazi government on its first list. He eventuall1~~~~ ~~ the United States and taught at Columbia University. During the summers 1953 to 1957
he was Visiting Professor at the Free University of
Berlin; and in 1956 his dismissal from the University of Heidelberg was declared void.

Exhibit 2

The Case of the Black Panthers

Let us now consider the application of the arguments from tell-tale facts and statistical reasoning
to the deaths of Black Panthers o
The Black Panthers are one of the newer groups of
young black men who are challenging the treatment
of black people in the United States. Their platform was adopted in October 1966; it contains a
number of planks, some progressive, some revolutionary, some expressing condemnation of the existing
power structure in strong and derogatory terms.
Furthermore, many of them arm themselves, and they
refer to policemen as "pigs".
In Exhibit 2 is a list of 19 Black Panthers meeting violent death, in the period April 1968 to December 1969. This list is derived from a report in
an article in Life magazine, Feb. 6, 1970. Eleven
of these 19 men-were shot by the police; and when
the inquest was held, the decision in every case
was "j ustifiable homicide" 0
The Black Panthers claim that there is a conspiracy to kill them off, and that 28 of their group
have been killed in accordance with this conspiracy.
Whether or not there is a conspiracy depends
partly on the meaning to be assigned to the word
"conspiracy", and partly on the interpretation of
the events that have happened.
Among the tell-tale facts to be considered are
these two:
• On December 4, 1969, in the predawn neriod,
the Chicago police raided the apartment in
Chicago where two Black Panther leaders
lived. There is str.ong evidence that almost all the shots were fired by the police
from outside the room, and that the Black
Panthers inside -- some of them sleeping
until the police raid -- fired only a few'
shots at the police outside. Fred Hampton,
Illinois chairman of the Panthers, and his
colleague Mark Clark were both killed.
See the newspaper account in Exhibit 3.
• Wesley C. Uhlman, Mayor of Seattle, Washington, stated in February, 1970, that he had
been asked by the federal government (or a
part of it) to organize a midnight raid on
Black Panther headquarters in Seattle, and
he had refused to do so.
A question we need to consider with our statistical "instrument of decision" is the significance
to be attached to the 28 deaths. The locations of
19 of the deaths are Chicago, Long Beach (Calif.),
Los Angeles, New Haven, OaKland, and San Diego.
The number of members of the Black Panthers is reported to be about 1200.
The number of natural deaths to be expected from
a group of 1200 Negro men, average age 25.. say, is
about 4 per thousand. Over 1 and 3/4 years the expected ~umber (M) of deaths among such a group of
1200 persons is 8 (1200 x 4 / 1000 x 3/4). If we
look in Table 2, on the line"for 8 events, we find
that 28 actual deaths would imply correlation,
cause, or conspiracy.
In this case, a life insurance company would not
apply a rate of death from natural causes to this
group of persons, but instead a higher rate, because
of additional hazards. Consequently, the argument
from statistical reasoning is not as sound as it
might appear to be on its face. However, the telltale facts together with the statistical argument
would seem to be strong evidence of a conspiracy by
elements of federal and city governments in the U.S.
against the Black Panthers.

BLACK PANTHERS MEETING VIOLENT DEATH
(Information chiefly from "Black Panthers: the
Hard Edge of Confrontation" in Life magazine, pp.
22-27, Feb. 6, 1970, Vol. 68, No.4)
1. Spurgeon Winters, 19, shot by police, Chicago,
November 1969. Inquest ruled it "j ustifiable."
2. John Huggins, 23. shot to death, UCLA lunchroom,
January,lY69; rival black militants convicted.
3. Walter Pope, 20. shot by police, Los Angeles,
October 1969: Inquest: "j ustifiable homicide" •
4. Alprentice Carter, 26, shot, UCLA lunchroom,
January 1969; rival black militants convicted.
5. Welton Armstead, 17, shot by police, Seattle,
October, 1969. Inquest ruled it "j ustifiable
homicide" •
6. Sidney Miller, 21, shot by storekeeper, Seattle,
November, 1969; storekeeper was not charged.
7. Fred Hampton, 21, Illinois chairman of the Panthers, shot Dec. 4, 1969, by police, Chicago,
in a predawn raid on his apartment. Inquest
termed it" justifiable homicide" •
8. Mark Clark, 22, shot by police, Chicago, December 1969. Inquest called it "justifiable
homicide" •
9. Tommy Lewis, 18, shot by police, Los Angeles,
August 1968. Inquest ruled it "j ustifiable
homicide" •
10. Sylvester Bell, 34, beaten and shot, San Diego,
August,1969; rival black militants have been
accusedo
11. Bobby Hutton, 17, shot by police, Oakland, Calif.,
April 1968. Grand jury ruled it "justifiable
homicide" •
12. Steve Bartholemew, 21, shot by police, Los
Angeles, August 1968. Inquest ruled it "justifiable" •
13. John Savage, 21, shot on San Diego street in
Mai 1969; rival black militant is awaiting
trial.
14. Frank Diggs, 40, found shot to death in Long
Beach, Calif., D~cember 1968. Assailant unknown.
~5.

Nathaniel Clark, 19, shot by wife, who Panthers
say is a police agent, in Los Angeles, September, 1969.

16. Larry Robertson, 20, shot by Chicago police,
July, 1969, died in September. Judged "justifiable" •
17. Robert Lawrence, 22, shot by police, Los Angeles,
August 1968. Inquest ruled it "j ustifiable".
18. Arthur Morris, 28, shot in 1968, Los Angeles gun
fight, not involving police. No one was charged.
190 Alex Rackley, 24, tortured, shot. New Haven,
Conn., May 1969. Panthers are charged with
murder.
COMPUTERS and AUTOMATION for September, 1970

Exhibit 3

The Case of the Assassination of Two Kennedys
and Martin Luther King

From The Gazette & Daily, York, Pa.,
January 26, 1970
INQUEST ON DEAD PANTHERS OPENS UP MANY QUESTIONS
Chicago -- The finding last Wednesday of a special coroner's jury that the deaths of two Black
Panthers in a predawn police raid Dec. 4 were "justified" were based, the verdict noted, "solely and
exclusively" on testimony presented by the police
and the coroner's office.
But that evi: !nce itself appeared to raise a
number of questions about the incident, which the
police have described as a gun battle lasting at
least 10 minutes, touched off by shot-gun blasts
from the occupants of the small West Side apartment.
Members of the Black Panthers declined to testify
because of pending criminal trials.
Among the unexplained points raised by the testimony were the following:
-- A police firearms expert testified ammunition
recovered from the apartment were the shell casings
of five .32-caliber bullets. He testified that
these did not match any of the alleged Panther weapons the police say they found and that the list of
police weapons carried on the raid did not include
any of .32 caliber.
-- The same'police expert, John M. Sadunas, testified that none of the 130-odd expended shell casings and slugs recovered from the apartment
matched 17 of the 19 weapons the police say they
seized.
-- Sadunas testified that three shot-gun cartridge casings the police said they found in the
apartment matched the remaining two shotguns. Yet
no slugs or shotgun pellets were recovered from the
area where the shotguns were supposed to have been
fired.
-- Although the police raiders repeatedly testified that there had been a number of shotgun
blasts directly out of the doorways of the two bedrooms, no pellets or slugs were recovered from the
opposite walls, and visitors to the apartment saw
no marks on the walls.
-- Police Sgt. Daniel Groth, who led the raid,
testified that there was no tear gas "available,"
although the raiders had drawn three shotguns and
a submachine gun from a special weapons arsenal.
-- Sergeant Groth also testified that no fingerprints were taken from any of the weapons the police said they seized.
-- All 14 policemen who took part in the raid
testified thaty they had virtually no prior knowledge of the ~lack Panthers, although they are assigned to the Special Prosecutions unit of the
State's Attorney's office, which was set up last
spring with a specific mission of keeping an eye
on black youth groups and street gangs.
-- Despite the heavy volume of fire described by
the police, the only injuries they received came
when one policeman cut his hand breaking a window
with his pistol and another was grazed by gunfire
coming through a wall, apparently fired by the
police.
Testimony about the three shots fired into
Hampton, who was the Illinois chairman of the
thers, raised a number of questions about the
of the bullets that were fired and the nature
the wounds.

Fred
Panpath
of

COMPUTERS and AUTOMATION for September, 1970

For most Americans, by far the most important
and most disturbing instances of political assassination in recent years in the United States are the
assassination of two Kennedys (in 1963 and 1968),
and of Martin Luther King·(in 1968).
President John F. Kennedy was shot to death about
12:30 pm while riding in an open car on November 22,
1963, in a motorcade in Dallas, Texas. Lee Harvey
Oswald was arrested in a movie theatre early that
afternoon, and charged with the crime, which he
steadily denied; then he was shot to death two days
later in the basement of the Dallas jail by Jack
Ruby, ,while being transferred from one jail to another. Ruby is now dead also.
Martin Luther King was shot to death while standing on the second floor balcony of a motor hotel in
Memphis, Tennessee, on April 4, 1968. He was shot
by a rifle bullet from the window of a rooming house
a short distance away. Some months later an escaped
convict, James Earl Ray, was arrested in London Airport, and was brought to trial. He pleaded guilty
on the advice of his lawyer at the time, was sentenced to 70 years in jail, and is now in jail in
Tennessee.
Senator Robert F. Kennedy was shot to death about
12:15 am, June 5, 1968, in the kitchen area of the
Ambassador Hotel in Los Angeles, just after he had
won the California Democratic primary election on
June 4. Sirhan B. Sirhan, a Jordanian who became a
naturalized American, was arrested in the kitchen,
with a gun in his hand, after he had fired eight
bullets. Sirhan was tried, pleaded not guilty, and
was sentenced to death. He is now in jail in California.
A question which confronts a great many people
in the United States is: Are these killings so extraordinary that the theory "it is just a coincidence" must be abandoned and replaced by the theory
"these killings are correlated, and there exists a
cor'relation, cause, or conspiracy of some kind"?
At the present time, the answer derived from applying the laws of probability and statistics is
inconclusive. It is easy to show that if these were
three normal people running ordinary risks, the expected number of deaths out of this group of three
young men (two Kennedys and Martin Luther Kin.g")
would be very close to zero and that even one death
occurring would be in the range of unusual, strange,
suspicious.
But there is the important question, what kind
of death rate should apply to these persons? If
the life of such a person was to be insured, a life
insurance company would charge an extra premium because this kind of person runs extra risk, extra
hazard.
Even so, the argument from statistical reasoning
declares there is very likely a correlation or cause
or conspiracy (or concert of ideas and attitudes),
and this is confirmed by the abundance of tell-tale
facts about the assassinations.
Probably the best summary at present is the following:
1. In regard to the assassination of President
John F. Kennedy, there was a conspiracy
in Dallas.
2. As to the killing of Martin Luther King,
James Ea.rl Ray appears to be ei ther a hired
gunman or a "patsy", either of which means
a conspiracy.
3. As to the killing of Senator Robert F.
Kennedy, there is evidence which points
toward conspiracy.
4. As to a possible inter-connection between
the three slayings, there are a number of
47

COMPUTER-ASSISTED ANALYSIS OF EVIDENCE
REGARDING THE ASSASSINATION OF PRESIDENT
JOHN F. KENNEDY - PROGRESS REPORT
Richard E. Sprague
Personal Data Services
Hartsdale, N. Y.

A computer project has been initiated at the
headquarters office of the National Committee To Investigate Assassinations (NCTIA), 927 15th St., N.W.,
Washington, D.C. The project was started in June,
1970, by the Committee's Executive Director, Bernard
Fensterwald, with the assistance of two board members.
Twelve volunteer workers have contributed to the
project during the summer months. Three of them
have designed a coding structure for feeding evidence into a computer system and storing iti they
have also coded an appreciable amount of evidence
to date. Six workers are contributing to system
design and programming work. Three others have offered computer time.
Specific Objectives

The obj ec tives of the proj ec t are fourfold.
First, to organize and store in computer-based form
all significant factual evidence pertaining to the
John F. Kennedy assassination.
Second, to make easily and efficiently available
to researchers: (1) .evidence having interest, and
(2) answers to queriei peitaining to the evidence.
Third, to provide an index to voluminous textual,
photographic, and physical material on the assassination, presently located in many different locations
around the United States.
Fourth, to provide an analytic capability comprised of a combination of information retrieval
and computing techniques to augment human efforts
in examining the evidence to solve the John F. Kennedy assassination.
The fourth objective is of rather longer range
than the first three. A pilot analysis program is
contemplated in which a specific set of dates,
places, persons,· organizations, and events will be
analyzed to see whether a proposed solution to part
of the planning for the assassination can be verified.
Coding Structure

The coding structure developed so far provides a
general framework for numerically coding raw data
elements and relationships among them in the following categories: People, organizations, places,
events, times, and objects. A coding form permits

parallels in. the cases which might indicate a common modus operandi and which
normally would be subject to official,
governmental scrutiny.
5. The absence of such scrutiny is extraordinary, and implies collusion of powerful
parts of the government and of the establishment.
Nowhere in the United States can concerned Americans apply to have their questions about political
assassinations answered reasonably. Only a few
agencies, such as courts and Congressional committees, have the power to issue subpoenas, compel the
appearance of witnesses, and ask direct questions.
The rights of the people to know are defective. 0
48

several relationships and data elements to be represented in one grouping. For example, two people
met at a certain time in a specific place. Or one
person is related to a second person and is also a
member of an organization.
Utilizing this coding structure, the coding volunteers have ploughed their way th·rough several important sources of evidence and information, producing several thousand completed coding forms for
eventual entry into a computer system. The sources
include: The Warren Report; NCTIA files of evidence; the book Accessories After the Fact by Sylvia
Meagher; the book Oswald in New Orleans by Harold
Weisberg; and others.
Systems Design

Two systems design approaches are under study.
The first is a batch-processing approach using
punched cards and tapes with standard printed reports. The output in this case would be primarily
organized indexes, to be placed in manually available files and scanned by researchers.
The second system is a random-access interactive
approach using disc files for storing all the evidence in coded form, and an English language translator to produce readable output. The programming
language under consideration for this system is Integrated Data Store (IDS). The researcher would be
able to formulate a series of questions and enter
them through an interactive terminal, receiving online answers at each step of his exploration.
Taking the pilot program as an example, the researcher will be able to formulate a problem as follows. He wishes to determine the likelihood that
meetings among a group of specified people took
place in a specific period of time in a particular
place or places in a specific city or geographic
area.
He can explore via his terminal all of the evidence pertaining to the whereabouts of the group of
people during the period of time in question. He
can also explore the events which took place in the
locations in question during the time period.
Thus in the long run he will be able to gather
evidence for and against his proposed hypothesis.
If any readers of Computers and Automation are
able to contribute assistance with key punching,
programming, machine time, or money to help continue
this research,it is suggested they write to the
National Committee To Investigate Assassinations,
927 15th St. N.W., Washington, D.C. 20005.

CORRECTION

In the May issue, in the article "The Assassination of President Kennedy: the Application of Computers to the Photographic Evidence", in the "Spatial Chart of Northern
Half of Dealey Plaza" pages 48 and 49, in the
lower center of the chart, in the "scale of
feet", please replace the numbers 10, 20, 30,
40 by the numbers 20, 40. 60, 80, respectively:

COMPUTERS and AUTOMATION for September, 1970

ACROSS THE EDITOR'S DESK
APPLICATIONS
REVENUE AGENTS TRACK DOWN
CIGARETTE BOOTLEGGING WITH
AID OF COMPUTER

Georgia revenue agents have put
their computer to work to help stop
illegal cigarette shipments. Cigarette manufacturers supply all
states with monthly computer tape
records of cigarette shipments
coming into a state. "By comparing
computer reports every month we can
spot sudden changes in volume that
might indicate bootlegging activity.
For example, if a city showed a
sudden decrease in per capita tax
paid cigarette sales, we would suspect it as a point of bootl egging
activi ty and send investigators,"
said Georgi a Tobacco Licensing Chief
Jerry Wynn, who
spearheaded the
program. Al though other states are
using the tapes, Georgia is the
only state to use them in thi sway.
Tapes are fed into the IBM System/360 Model 40.
A printout of
the information by wholesaler or
distributor gives the number of
cigarettes, packs, cartons, their
destination, their shipping point
in the state and computes the tax
as well. Manual auditing systems
formerly used by the state required
as much as a month to complete the
audit of each wholesaler - with
the computer, time has been cut to
one day.
Since the system is very new,
no accurate figures on collection
increases resulting from it are yet
available. Georgia's Revenue Commissioner, John A. Blackmon, said,
however, that cigarette bootlegging
is a national problem that results
in an estimated annual revenue loss
of nearly $100 million. " ••• Eventually we hope to share information
between states, which will help us
further in stopping the flow of
bootlegging activity from low tax
states to high tax states ," he said.
POST OFFICE TESTS COMPUTERIZED POSTAGE SYSTEM

Two companies working with the
Post Office are using computers to
determine postage for different
size packages and to automatically
print the amount
on pre-printed
labels. Parcel post rates are determined by weight
and di stance
being mailed.
Loa Corporation in
Omaha, a catalog mail order house,
stores rates from Omaha in their
System/360 Model 20. Sarah Coven-

try in Newark, N. Y., a maj or di stributor of costume jewel ry, stores
rates from Newark
in the firm's
System/360 Model 40.
The computer-aided mailing process bypasses multiple handling of
packages by the mailer and by the
United States Post Office.
As
orders arrive, clerks enter a code
number designating a specific item
of merchandise, its weight and to
whom it is being sent. The computer
matches thi s information with the
proper weight and destination rate i
prints the correct postage amount
on a pre-printed 1 abel, and, al so
prints the name and address of the
cus tomer in ZIP Code sequence. The
boxes are put in mail bags for delivery to the Post Office from
where they are immediately shipped.
The mail bags are not even opened
at the Post Office.
Billing is on a weekly basis.
Using computer listings of all
mailings, the Post Office charges
for total postage as figured by the
computer. Periodically, a postal
official verifies package weights
to ascertain computer-figured postage rates are correct. The computer
al so verifies i t8' calculations under
direction of a stati stical sampling
program. Official s at both firms
have reported reduced costs with
the computerized mailing process.
Postmaster General Winton M. Blount
added that by presorting their
packages according to ZIP Code,
del i very is speeded by 1 to H~ days.

bi rds. The system al so keeps track
of the egg s, by fl ock, from the day
they are laid until they hatch.
Thus researchers can determine i f
variations in feed and other factors
affect the number and quality of
the eggs. Additionally, the company plans feed formulations on the
computer and figures cost per pound
on birds as they are being readied
for market.
The company was a pioneer in the
development of the "MINI", an ~" to
12-pound bird specifically bred
for white meat and small size.
Williams also raises the traditional
1 arge turkey. Both are produced for
national and international markets.
COMPUTER PROVIDES INSIGHT TO
CHANGES IN GREAT LAKES

Researchers at the Great Lakes
Fi shery Laboratory at Ann Arbor,
Mich., are using a computer to interpret the effect changing lake
condi tions are having on the fi sh
popul ation in the Great Lakes. Dr.
George Y. Harry, director, said
laboratory researchers rely on the
IBM 1130 computer to interpret the
data being gathered both in the lab
and on the lakes.

CALIFORNIA COMPANY BREEDS
TURKEYS BY COMPUTER

A computer is keeping records
on individual birds and entire
flocks, allowing experts at Williams
Turkey Breeding Farms, Inc. (Oakdale, Calif.) to conduc t research
that results in improved breed
lines and rigid quality control.
Using an IBM 1130 system, Williams
researchers are able to select parent turkeys that will produce poul ts
with specific desirable characteristics. These attributes, including superior body conformation,
fast weight gain and high meat
yield, meet the specific requirements of the market in North America,
South America and Europe.
Computer records, noting characteristics, are kept on every
breeder bird dating back five
and six generations. When a turkey
with certain characteristics is
needed, the computer can give the
wing band 'numbers of the tom and
hen most likely to produce the"se

COMPUTERS and AUTOMATION for September, 1970

Researcher Tom Edsall, shown
above, examines a computer printout
of data gathered on Lake Trout. The
fish are shown in a tunnel-type
respirometer, which enables researchers to con trol wa ter pres sure,
temperature flow rate, dissolve
gasses and vary soluble toxicants
and observe fish reactions.
The
Fishery Lab (operated by the U.S.
Department of Interior) operates
four research vessels on the lakes,
recording data on fish population
by geographical area and type, and
noting biological, chemical and
physical changes taking place in
the water. "Researchers combine
these da ta wi th "catch" information
and other historical records provided by both commercial fishermen
and state agencies for computer
processing. The computer correlates
the d~ta into charts for conservationists.
49

"By unders tanding the changes
that are taking place in the lakes,"
Dr. Harry said,
"we can suggest
corrective measures to head off
developing problems.
Changes in
species composi tion also give clues
about changes taking place in lake
water."
According to Dr. Harry, measures
already have been taken that are
aimed at re-storing a biological
balance in the lakes.
HOSPITAL PROJECT USING
COMPUTER SYSTEM TO AID
SURVIVAL OF FETUSES

At the Department of Obstetrics
and Gynecology, Univ. of Chicago
Lying-In Hospital (Illinois~ medical researchers are developing a
small
computer-based moni toring
system for obstetric patients in
the active stages of labor.
The
system is designed to recognize
those conditions crucial in the care
of patients which could" mean the
life or death of the fetus. The
resiliency and reserve of the human
body are such that subtle pathophysiologic changes can occur wi thout
being fully assimilated by the obstetrician. All too frequently, it
is only when the cumulative effects
of many deviations from normal labor
become evident that the doctor is
aware of a problem.
Wi th the system's ability to
continuously record the fetal heart
rate and the uterine contraction
pattern, diagnosis of the patient
and fetus can be more accurate as
events occur.
A PDP-8/I (Digital
Equipment Corp.) used in the sys tern,
gives the flexibility to alter the
mode of data analysis and display
as it is necessary.
Software in
the computer also can be changed as
experimentation dictates.

COMPUTER LOOKS THROUGH
MICROSCOPE

The National Bureau of Standards
(U.S. Department of Commerce), in
a collaborative effort wi th the
National Institutes of Health, has
adapted a microscope for use wi th
a digital computer to direct and to
record the presence of obj ects on
the slide. The scanning system was
designed by Philip Stein (of the
Bureau's Center for Computer Sciences and Technology), Dr. Lewis
Lipkin (of the National Insti tute
of Neurological Diseases and Stroke)
and Dr. Howard Shapiro (of the National Cancer Institute). It will
be of use in neuropathological
studies, in autoradiograph counting,
and as a tool in the des ign of systems .for automating image processing.
The system consists of a microscope, an image plane scanner, a
motor-driven stage, a computer, and
the necessary interface hardware.
It not only scans automatically in
the .2S and y... directions on the microscope slide, but can do so repeatedly at sequential ~ positions
to obtain three-dimensional data.
Computer control allows the operator to manually posi tion the slide
to locate features of interest and
to note their locations in the computer memor~ When later instructed,
the stage can return to within one
step in each axis of every posi tion
previously noted - the system automatically performs a scan at each.
The scanning microscope was buil t
on an existing special-purpose microscope frame. It offers a choice
of ei ther wide-spectrum or monochromatic light, the illumination
desired being switched by a mirror.
A fixed portion of the light is
passed through a special lens system to a photomul tiplier tube to
provide a reference for specimen
illumination.

Presently, the "front-end" or
data acquisi tion section of the
system is an intrauterine catheter
and strain gauge for moni toring
uterine contractions which feeds
into a polygraph. The fetal heart
rate is obtained from a skin-clip
elec trode and a cardiotachometer,
which also is fed into a polygraph.
One aim of the development plan is
to simplify this section so that
physicians, nurses, or obstetric
assistants can easily use it.
The system also includes a visual display via closed-circui t television - in the patient's room.
The display is generated by the
DEC KV81 I graphics portion of the
computer system and a small storage
scope, which is slaved off of the
main display scope. The small scope
is photographed with a closed-circuit tefevislon camera.

NIH.
The scan is manipulated by
pressing buttons on the operator's
console, which interrupts computer
control. The system computer controls the scan but processing the
resulting data requires large computers.
At present a high speed
telephone line is the link with an
image processing computer.

EDUCATION NEWS
UNIV. OF PITTSBURGH
INAUGURATES NEW PROGRAM
IN INFORMATION SCIENCE

A new Interdisciplinary Doctoral
Program in Information Science at
the University of Pittsburgh (Pa.)
is headed by Professor Allen Kent,
Director of Communications Programs
at the University, and Dr. Anthony
Debons, professor of information
science.
It is supported in part
by a contract from the Commonweal th
of Pennsylvania Science and Engineering Foundation.
The program
accepts students wi th undergraduate
maj ors in almost any discipline,
but when they complete their studies
in information science they will
have well-rounded backgrounds in
the behavioral, systems, computing,
engineering and basic sciences.
The program is designed to provide the student wi th a deep appreciation of the role of man in his
use of informatIon and communication
technology in day-to-day pursuits.
How the human will use computer information is an important concept
in information science, Professor
Kent points out. The program is one
of the few information science programs in the country wi th an interdisciplinary emphasis. Most other
academic programs in information
science have stressed the technological aspect of the process.
POSTAL CLERKS IN ENGLAND
TAUGHT BY COMPUTER

The scanner is connected in a
closed-loop configuration with a
small general-purpose computer that
controls the scan.
The picture
shows NBS computer engineer Philip
Stein viewing a slide through a
computer-controlled microscope that
he developed in collaboration with

Great Britain's Post Office System plans to integrate a computerized "teaching machine" into its
operations that will enable up to
20 operators to' be simultaneously
taught letter-coding techniques.
(The electronic coding and sorting
process will be central to the flow
of mail through the new pos t office
system.) Using television-like keyboard devices, the Honeywell Model
516 comput"er will be able to display up to 4,000 different lessons
on the screen by selecting specific
informa tion stored in its memory.
The system also can incorporate
remedial lessons to correct continuous errors on the part of the
operator.

COMPUTERS and AUTOMATION for September, 1970

Operators first will be taught
to touch type; then to code groups
of letters; and third. read and
code two addresses simultaneously.
Stations will be moni tored by a
postal supervisor who can display
his comments to trainees via the
screen during instruction.
The
computer itself presents an explanatory text before each lesson, then
performs a cri tique based on a "par"
for the course of instruction.

"TYPING CLASS" AT BOWLING
GREEN (OHIO) IS COMPUTER LAB
Students preparing for careers
in computer science or bus ines s
"experiment" in an unusual class at
Bowling Green State University.
Using typewri ter-like communications
'terminals (IBM 2741s) linked to a
central computer, undergraduates
wri te their own computer programs
and solve complex problems in quanti tative terms as part of two maj or
fields of study at Bowling Green.
Students maj oring in information
systems or computer science are
provided with "hands-on" experience.
Fifteen terminals are linked to
an IBM System/360 Model 50, which
serves all terminal users concurrently, while enabling each to proceed through computer-assisted exercises at his own pace. The computer oriented maj ors at Bowling
Green are offered by the College of
Business Administration and the
College of Arts and Sciences.

IBM MANUFACTURING CAl
NETWORK HELPS EMPLOYEES
LEARN MORE THAN 50 SUBJECTS
Manufacturing employees at IBM's
plant at Poughkeepsie, N.Y., take
manufacturing-related courses at

are available to IBM employees
using this system. An IBM System/360
Model 50 computer located in' ,Poughkeepsie serves terminals at 11 IBM
manufacturing plants from coast to
coast and in Canada.
In addition'
to Poughkeepsie, the computer-assisted instruction system embraces
IBM manufacturing locations at:
Kingston, East Fishkill and Endicott, N.Y.; Rochester, Minn.; Boca
Raton, Fla.; Boulder, Colo.; Burlington, Vt.;
Raleigh, N.C.; Sari
Jose, Calif.; and Toronto, Canada.
Other locations are in the process
of linking to the system.

COMPUTER SCIENCE OFFERED
AS ADULT EDUCATION COURSE
BY NEW JERSEY HIGH SCHOOL
Res idents of this Philadelphia
area community (Cherry Hill, N.J.)
will have an opportuni ty to learn
computer science through an adult
education course being offered by
the school department this fall.
The course will be called Computer
Concepts.
The teacher will be
Sidney Rubenstein, a member of the
Mathematics Department of Cherry
Hill West High School.
A recently acquired PDP-8/L computer (Digital Equipment Corporation) will be used as a demonstration device., Students also will
get "hands on" experience with the
computer in both the programming
and computer operations segments
of the courses. " ••• The adult education course will offer the people
of our communi ty the chance to
increase their knowledge of a force
becoming more and more influential
in the business world at a price
they can easily afford", remarked
Mr. Rubenstein.

RESEARCH FRONTIER
UNIVERSITY OF UTAH SCIENTISTS
RESTORE CARUSO RECORDINGS
USING DIGITAL COMPUTERS
At the Uni vers i ty of Utah, a
computer scientist is devising a
method of "resurrecting" the voices
of recording artists from the muted,
hollow-sounding, antique records of
the dis tant past.
The technique
uses digital computers and already
has been dramatically demonstrated
on the voice of Enrico Caruso, the
great Ital ian operatic tenor who
died in 1921.

computer terminals in the computerassisted instruction system. Over
50 courses, ranging from chemical
safety to data processing principles

Directing the research is Dr.
Thomas G. Stockham, Jr., professor
of electrical engineering and a
leading au thori ty on "compu ter filtering proces ses:'
According to

COMPUTERS and AUTOMATION for September, 1970

Dr. Stockham. sound waves from recordings can be fed into a computer.
converted into numbers and then
re-recorded on numerical tapes.
"By using a complex, delicate and
precise program in the computer,
we can manipulate these numbers
wi th the hope of reversing the
orig inal recording proces s and coming back out with sounds that have
objectionable resonances removed",
the computer scientist said.
The
process is called "numerical dereverberation".
Dr. Stockham believes it holds promise for a wide
range of future advances in a variety of scientific areas, as well
as'in the recording industry.
The Caruso experiment actually
developed as a sideline to the primary research Dr. Stockham and his
colleagues are conducting on contract
wi th the Advanced Research Proj ec ts
Agency of the Department of Defense.
They are investigating the use of
digital computers to process pictures
and sound, the bas ic elements of
human communication.. Using computers, the scientists can restore
photographic images that are blurred,
enhance details in pictures, eliminate reverberation in sound, and
may be able to improve high frequency radio transmission.
Work on the antique recordings
began one night when, while talking
about filtering with some friends,
one chanced to remark, "Wouldn't it
be nice if computers could be used
to res tore the real voice of Caruso. "
Dr. Stockham began working with the
mechanical problems involved in such
a restoration, and was intrigued by
the similarity between these problems and others wi th which he had
been previously working.
Finally
he devised the initial computer
program from which Caruso's voice
was "resurrected."
While Caruso had a voice powerful enough that it came through the
crude equipment of his day, the low
and high frequencies were missing.
Distortions were so gross there is
reasonable doubt today about the
most salient characteristics of
his voice.
Dr. Stockham believes
the computer-restored Caruso voice
gives a more accurate sound on which
to judge its character and qual i ty.
He also is working on the elimination of surface noise and other related problems.
Experiments such as the Caruso
one, Dr. Stockham says, greatly
enhance scientific knowledge about
manipulation of pictures and sound.
"They are illus trative of how the
computer will supplant conventional
elec tronics for the future handling of communications information. II
he says.

NEW PRODUCTS AND SERVICES
NAME/ MODEL NO.

DESCRIPTION

FOR MORE INFORMATION

Digital
Control Data 3170 System

PDP-ll/15

For multiprogramming users / total hardware and software compatibility with CDC's 3300 and 3500 systems /
expandable core memory, 49,000 to 131,000 24-bit
words / 1.75 usec memory cycle time; 1 usec access
time / may be field-upgraded
General-purpose computer designed for original equipment manufacturer (OEM) applications / full compatible
with, and has same central processor capabilities as,
PDP-ll/20 / core memory, 4096 16-bi t words (expandable to 32,768); 1.2 usec cycle time; 500 nsec access time

Control Data Corp.
8100 34th Ave. So.
Minneapolis, Minn. 55420
Attn: Kent R. Nichols

For use supervIsIng operation of several small dedicated control computers, as second-level computer in
hierarchical computer system, and as data gathering
system in management reporting network / includes
PDP-IS processor, 16,384 words core memory and RSX-15
real-time multiprogramming, executive software moni tor
For nuclear research and analytical applications /
includes a GEOS single or dual parameter 4096-channel
analyzer interfaced to DEC's 16-bit PDP-ll/20 computer / system provided with application-oriented
software package
Card oriented phototypesetting system / allows access
to any point in composed material prior to setting of
type / chief components are a typesetting keyboard
which is connected to a keypunch, and a card reader
which activates a phototypesetter

Digital Equipment Corp.
146 Main St.
Maynard, Mass. 01754
Attn: Edgar E. Geithner

Random access system / composed of proprietary COMFILE
Data Storage System (storage to 72,000 characters for
any mini- to medium-sized computer), a COMFILE plug-in
interface; and a COMFILE utility program / average access of 350 msec
Modular system can be altered to produce 60 different
configurations / 4096-16,000 word capacities available,
lengths in 8, 12 and 16-bit formats / 1.6 usec full
cycle time with 650 nsec access time for read/regenerate and clear/write operations
For Century 200 computers / direct access files with full
track disc capaci ty of 29.8 or 59.6 million bytes / average access time including latenc~ 72.5 msec; maximum
data'transfer rate, 315,000 bytes per second / the 657-101
is a single-spindle disc unit; 657-102, dual spindles
Capacities to 10,240 bits; word lengths to 80 bits per
word /90 nsec access time; 190 nsec cycle time / has nonv()latile form of storage that is mechanically al terable;
individual bits, words, or entire memory contents can
be modified
A planar core memory specifically for greater storage
capabili ty in N/C systems / for use with UMAC 6 controls,
it accepts data in same manne~ thus no operator or programmer retraining is required / reported as more economical means for applications requiring over 32 offsets
High speed (7200 rpm) provides 4.4 msec access time /
stores 5 million to 20 million bits, densities to 1800
bpi / available with 256, 512 or 1024 tracks, each with
own read/write flying head / option - controller logic
to interface with user's central processor

Compat Corp.
177 Cantiague Rock Rd.
Westbury, N.Y. 11590
Attn: W. Craig Meyer

Digital Equipment Corp.
146 Main St.
Maynard, Mass. 01754
Attn: Dennis C. Goss

Special Purpose Systems
PDP-15/35

QUANTA~

Analyzer/Computer System

RACE (Random Access Composition Equipment)

Digital Equipment Corp.
146 Main St.
Maynard, Mass. 01754
Attn: Howard Steiner
Warlock Computer Corp.
Route 7
Georgetown, Conn. 06829
Attn: Allan Kumble

Memories
COMFILED Memory System

ICM-161 Core Memory
System

NCR 657 Disc Units

NANOROM 90 READ-Only
Memory System

UMACORE Memory

SA 8000 Series Drum

Honeywell Computer Control
Old Connecticut Path
Framingham, Mass. 01701
National Cash Register Co.
Main & K Sts.
Dayton, Ohio 45409
Memory Technology Inc.
83 Boston Post Road
Sudbury, Mass. 01776
Attn: Paul Rosenbaum
Vickers UMAC Division of
Sperry Rand Corp.
2500 Trans Canada Hwy.
Point Claire 730
Montreal, Quebec, Canada
Systematics/Magne-Head Div.
General Instrument Corp.
13040 S. Cerise Ave.
Hawthorne, Calif. 90250

COMPUTERS and AUTOMATION for September, 1970

DESCRIPTION

NAME/ MODEL NO.

FOR MORE INFORMATION

(Memories, continued)
Varian 620 Disc Drives

Four low-cost disc drives, Models 38A, 38B, 38C and 39,
for 620 computer product line / storage capacities from
30,000 to 585,000 words / Model 39 (585,000 words) can
be expanded with addition of slave for total capacity
of 1,170,000 words

Varian Data Machines
2722 Michelson Drive
Irvine, Calif. 92664

Proprietary installation management package to support
IBM System/370 Models 155 and 165 / permits rational
cost accounting of multiprogrammed system usage / designed to interface with IBM OS/MVT Control Program /
offered on a fully-maintained prepaid lease basis
Assembly language subroutine / enables COBOL programmer
to test and set bits / requires less than lK of core and
relies on register usage for maximum processing speed /
distributed on lease basis for one time fee of $595
For churches and synagogues / Contribution Record Reporting System (CRRS) is designed to provide more frequent and in-depth report to individual giving units /
Membership Master Profile (MMP) includes complete demographic, interest, activity and membership information
about each individual member of congregation / systems
are compatible and provide flexibility to fit requirements in congregations of any denomination and size I
wri tten in COBOL / available as a complete software
system or as a processing service
Automatic test data generator for DOS and OS COBOL programs / generates data by actually reading the COBOL
Source Program / runs on IBM/360 with 24K of program
core available
A FORTRAN IV symbolic debugging system / written in
AED-O language for operation in an interactive mode
on the IBM 360/67 under the CP-67/CMS operating system / uses symbolic notation of source program during
debugging
Data Set Dump, Update, and Map utility programs for
OS/360 users / eliminates most control card preparation / available for a free 21 day trial evaluation

Systems Dimensions Ltd.
770 Brookfield Road
Ottawa 8, Ontario, Canada
Attn: N. L. Williams

Software

ACCOUNTPAK

B2BCONV ("Bi ts .to Bytes
Conversion")
CHURCH CRRS/MMP

DATAGEN

FDS

MASTER SERIES

PIPTO

For the trucking industry / a profit improvement program for terminal operations / designed to reduce labor
costs and improve productivi~y

TOTAL

An integrated Data Base Management System / manages
virtually unlimited number of data sets and allows for
entry and association of each of data sets with any
other data set in the data base / system is selfoptimizing_/ minimum design level normally 32K

TriValent Enterprises, Ltd.
P. O. Box 654
Greenwich, Conn. 06830
Attn: Norman M. Kittredge
Tri-Data Systems
7301 Washing Ave. So.
Edina, Minn. 55435

Programming Sciences Corp.
6 Eas t 43rd St.
New York, N.Y. 10017
Attn: Howard Linzer
SofTech, Inc.
391 Totten Pond Rd.
Waltham, Mass. 02154
Attn: Alder M. Jenkins
Infodata Systems Inc.
1901 No. Fort Myer Drive
Arlington, Va. 22209
Attn: Donald H. Stromberg
Walter Frederick Friedman
and Co., Inc.
280 Madison Ave.
New York, N.Y. 10016
Cincom Systems, Inc.,
2181 Victory Pkwy.
CinCinnati, Ohio, 45206
Attn: Tom Nies
or
Tom Richley

Peripheral Equipment

Addo-X Model 25-653-32
Data Input Machine
DGC-301 Time-Sharing/Instrument Data Transfer
System

531 Series Interface
Unit

Produces machine readable tape that can be transferred
directly to disc or magnetic tape through reader / has
optical character recognition capability, programming
features, and optional check digit verification
Designed for researchers, production test, medical and
nuclear analysts, and process control personnel /
couples various instruments to remote computer via a
time-sharing terminal/accepts BCD data, stores data
in memory, decodes data and presents it to terminal /
directly compatible with all standard BCD, integrated
circui t instruments and most older transistor instruments
Ties mini- or midi-computer into System/360 / connects
IBM 360 Interface Channel-to-Control unit to the minicomputer's standard I/O / minicomputer/531 combination
is exact plug-to-plug replacement for IBM 2803 Controller, he~ce requires no special 360 programming

COMPUTERS and AUTOMATION for September, 1970

Addo-X Inc.
437 Madison Ave.
New York, N.Y. 10022
Attn: Tom Kibbe
Data Graphics Corp.
8402 Speedway Drive
San Antonio, Texas 78230
Attn: John Thompson

Datawest Corp.
7503 E. Osborn Rd.
Scottsdale, Ariz.

DESCRIPTION

NAME/ MODEL NO.

FOR MORE INFORMATION

(Peripheral Equipment, continued)
KI-680 and KI-880 DATASCRIBE~/Concentrator

Magnafax 860 facsimile
telecommunications
unit
Model 9600 Card Reader

Optical Mark Reader
(OMR 8421)

SEACO Model 401 COM Recorder

Typeliner Model III Remote Terminal Printer

For off-line data acquisition and/or remote batch appli- Vanguard Data Systems, Inc.
1642 Kaiser Ave.
cations / communicates with Teletype or IBM 2741 data
terminal compatible devices / records in computer comIrvine, Calif. 92664
patible format on 2400 ft. reels of 7 or 9-track magnetic
Attn: David A. Bowman
tape / includes an unattended answering feature
A two-speed model which offers option of 3 minute or 6
Magnavox Systems, Inc.
minute transmission for full-page documents through a
80 Park Plaza East
coupling to the telephone / can send to or receive from
Garden State Pkwy.
majority of facsimile units utilizing the Direct Dial'
Saddle Brook, N.J. 07662
Tele hone Network / skilled 0 era tor not re uired
Attn: Jim O'Malle
For IBM System 3
96-column photoelectric device reads
Bridge Data Products, Inc.
cards at rates to 600 cpm / comes complete with trans738 South 42nd St.
port, power supply, electronics, and skins / desk top
Philadelphia, Pa. 19104
sized, weighing less than 50 Ibs. / optional field conversion kit gives multiple card capability
For daily inventory control applications in department
Automata Corp.
and chain stores / desk top unit reads pencil marked,3~
1305 Mansfield Ave.
inch wide, fan-fold strip of any length / using inteRichland, Wash. 99352
data coupler, transmits via dial network to central
receiver at effective rate of 400 characters per second
'Converts computer output (off-line or on-line) to micro- SEACO Computer-Display, Inc.
2826 West Kingsley Rd.
film at rate of 36,000 characters per second / characters
and symbols selected from 70-plus set and printed on
Garland, Texas 75040
pages up to 140 characters per line, 64 lines per page /
formatting under complete control of operator / lease
and service contracts available
For use with CRT terminals with plug-to-plug compatibil- Data Computing, Inc.
2219 W. Shangri La Rd.
ity and with any modem / 80 colums upper and lower case
alphabetic and standard ASCII 64 character set / 1001pm Phoenix, Ariz. 85029
Attn: Donald E. Oglesby
print speed; printout on pinfeed, fanfold paper in multiple sets up to six and width of 9-7/8"

Components
Analog Circuit Modules

LSls for 16-digit desk
calculators

Fifteen additions to analog module line includes multiplexers, dual amplifiers, sample and hold modules, D-A
and A-D units / all have 12-bit accuracy, are compatible
with DTL and TTL circuitry / used mostly for interfacing
analog devices to a computer
Large scale integrated circuits (LSls) available in 8
types: 2 memory units; five calculation control units;
one display control unit / each measures 132" square /
cuts overall size, weight to one-half of current machines

Digital Equipment Corp.
146 Main St.
Maynard, Mass. 01754
Attn: Dennis C. Goss
Hitachi America, Ltd.
437 Madison Avenue
New York, N.Y. 10022

Data Processing Accessories
COM Support Equipment

IBM System/3 Accessories

Magnetic Ink Character
Tester, Model MCT

Line of viewer stations and cassette storage equipment
for Computer Output Microfilm Systems / includes cantilevered aluminum-frame stations for viewe~ and viewerprinter equipment
Coordinated accessory equipment specifically for IBM
System/3 / includes small tray to hold 2250 96-column
cards; modular cabinets hold to 30 trays and allow
front to back or cross access filing, tub file, etc.
Examines, evaluates signal level of printed characters
to be scanned on a reader/sorter / tells whether documents are properly encoded according to American Banking
Assoc. standards / for printers, banks, utilities, etc.

Tab Products Co.
2690 Hanover St.
Palo Alto, Calif. 94304
Attn: Vernon Craig
Wright Line
A Div. of Barry Wright Corp.
160 Gold Star Blvd.
Worcester, Mass. 01606
Kidder Press Co., Inc.
279 Locust St.
Dover, N.H. 03820

Service for libraries and informa tion centers / complete
listings and ordering information for Institute of Electrical and Electronics Engineers publications / bi-monthly
Technical evaluation of IBM's new computer line / emphasizes hardware and software facilities; discusses its
degree of System/360 compatibility / $25.00 per copy

IEEE
345 East 47th St.
New York, N.Y. 10017
Datapro Research Corp.
2204 Walnut St.
Philadelphia, Pa. 19103

New Literature

IEEE Publications Bulletin
The IBM System/370: an
Independent Appraisal

COMPUTERS and AUTOMATION for September, 1970

NEW CONTRACTS
Sperry Rand Corp., Univac
Defense Systems Division,
St. Paul, Minn.

Lockheed-California Co.

Collins Radio Co., Dallas,
Texas

Litton Industries, Data Systems Division, Van Nuys, Calif.

Control Data Corp., Minneapolis, Minn.

The National Center for Atmospheric Research (NCAR), Boulder, Colo.

Litton Industries, Data Systems Div., Beverly Hills,
Calif •

Naval Electronics Systems
Command

Burroughs Corp., Detroit,
Mich.

U.S. Post Office Department

Syner-Data, Inc., Beverly,
Mass.

Database, Ltd., London,
England

Computer Sciences Corp., Los
Angeles, Calif.

National Aeronautics & Space
Administration (NASA)

Sperry Rand Corp., Univac
Division, Philadelphia, Pa.

Bundeslanderversicherung,
Vienna, Austria

URS Data Sciences Co., Falls
Church, Va.

U.S. Army

IBM Corporation, Los Angeles,'
Calif •

State of California, Dept.
of Health Care Services

Logicon, Inc., San Pedro,
Calif.

U.S. Navy

Philco-Ford Corp., Communications and Technical Services
Div., Ft. Washington, Pa.
Stanwick Corp., Washington,
D.C.

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

Peripheral Equipment Corp.,
Chatsworth, Calif.
Logicon, Inc., San Pedro,
Calif •
Hazeltine Corp., Little Neck,
N.Y.

Computer Machinery Corp.,
Los Angeles, Calif.
U.S. Air Force Space & Missile Organization
Jet Propulsion Laboratory,
Pasadena, Calif.

Conrac Corp., New York, N.Y.

Dallas Cowboys Football Club,
Inc. , Texas
Department of Manpower and Immigration, Ottawa, Canada
U.S. Department of Health, Education and Welfare

Information Science Industries
(Canada), Ltd., Ottawa
Computer Sciences Corp., Los
Angeles, Calif.

U.S. Navy

System Development Corp. (SOC),
Santa Monica, Calif.

County of Los Angeles, Calif.

Public Safety Systems Inc., a
subsidiary of General Research
Corp., Santa Barbara, Calif.
American Regitel Corp. San
Carlos, Calif.

City of Anaheim, Calif.
American Industries, Salt Lake
City, Utah

COMPUTERS and AUTOMATION for September, 1970

Design and development of nine UNIVAC 1832
Multiprocessor computers for the U.S. Navy's
S-3A anti-submarine aircraft, and provide
all systems software
Design, build, test and deliver automatically-controlled communications equipment
for the newly designated class of U.S. Navy
ships known as LHA, general purpose amphibious assault ships
A CDC 7600 computer system to expand current research activities in such areas as
mathematical modeling of global atmospheric
circulation patterns and convective storm
systems, and global studies of long term
climatic changes '
Development and production of automated
Direct Air Support Centrals (DASC) for
the U.S. Marine Corps. DASC will automate
control of fighter-bombers and helicopters
50 letter sorter machines used in high
speed processing of mail; ninth in the
series of contracts to the Company since
1958 for letter sorter equipment, total
value of contracts to date is $46.8 million
700 computer printing terminals: contract
gives Database exclusive rights to sell
and distribute Syner-Data's Alpha line
printer and Beta serial printer in the
United Kingdom, western Europe, Australia,
southeast Asia, and Union of South Africa
Operational support to the Simulation
Laboratory at NASA's Ames Research Center
near Mountain View, Calif.
A UNIVAC 1106 computer system: use will include retrieval of data concerning insurance contracts, contract processing and
control, various mathematical tasks
Continued URS support to the U.S. Army
Computer Systems Command (USACSC), Fort
Belvoir, Va.
Design, develop and install a prototype
data based, communications system that will
improve processing of claims under the
State's Medi-Cal piogram
Continuation of work on the TACDEW System
(a large digital simulation of the Naval
"at-sea" environment) which is used to
train Naval personnel in combat direction
and electronic warfare
A modern telecommunications network expanding communications between Luzon and Cebu
islands in the Philippines
Continuation of data processing services,
by Stanwick's Causality Data Company, related to Navy ship scheduling, budgeting,
inventory management. etc.
Model 6000 Series tape transports to be
used in the CMC9 KeyProcessing Systems
Missile guidance and control system performance evaluation studies
Develop a Digital-Video Display System for
use in support of planetary missions; will
use displays and prints in analyzing mission performance
Initial phase of a $1.5+ million scoreboard,
scoring and information system
A computer software system: lSI will provide over-all computerized services
A computer-based information system to ass,ist the Rehabilitation Services Administration to increase scope, quality of services, while cutting administrative costs
Preparing the design specifications for the
County's computerized Regional Justice Information System (RJIS)
A computerized traffic records system for
use by Police Dept., City Traffic Engineer,
count~ state,national safety organizations
100 80-column teleprinters lior use in medical data system) for doctor's offices, clinics

$401- million

$30 million
(approximate)

$8.6 million

$7.2 million

$6.1 million

$5.6 million

$3.3 million
$3 million
(approximate)
$2.4+ million
$2.2 million

$1. 6 million

$1.5+ million
$1.5 million

$1.4 million
$1 million
$849,000

$500,0001$500,0001$350,000

$308,000
$91,000

NEW INSTALLATIONS

Burroughs B340 system

First National Bank, Indiana, Pa.

Burroughs B500 system

City of Columbia, S.C.

First National Bank and Trust Co.,
Ardmore, Okla.
Burroughs B3500 system

Oak Park Trust & Savings Bank of
Illinois
Order of the Sisters of St. Joseph,
Orange, Cal if.

Control Data 3500 system

Hoogovens (the Royal Dutch Blast
Furnaces and Steel ~orks)

Control Data 6400 system

IABG, Munich, Germany

Honeywell Model 110 system

ALESCO (The American Library and
Educational Service Co.), Div. of
Paulist Press, Paramus, N.J.
Cambrian College, North Bay,
Ontario, Canada
Weber County, Ogden, Utah

Honeywell Model 115 system

IBM System 3

Andover Institute of Business,
Andover, Mass.
Sterl.ing Compu ter Sys tems, Los
Angel es , Cal if .
Dearborn Public Schools, Dearborn,
Mich.
Hayes International Corp., Air Force
Publications Div., Middle Rive~Md.
Refuge Assurance, Manchester,
En land
Marvin Windows, Warroad, Minn.

IBM System/360 Model 25

Gym-Dandy, Inc., Bossier City, La.

Honeywell Model 1200 system

Honeywell Model 3200 system

NCR Century 100 system

First National Bank, St. Joseph,
Mo.
Garden City Cooperative Equity
Exchange, Garden City, Kans.
Medicenters of America, Inc.,
Memphis, Tenn.
Northland Milk, Minneap·olis, Minn.

NCR Century 200 system

RCA Spectra 70/35
RCA Spectra 70/45
Univac 9000 Series computers

Brockton Public Markets, Brockton,
Mass.
Hertfordshire County Council,
Hertford, England
State Revenue Department, Alabama
American Home Foods, a div. of
American Home Products Corp.,
Mil ton, Pa.
Westinghouse Tele-Computer Systems Corp., Pittsburgh, Pa.
(30 systems)

Univac 9300 system

Anastasi Bros., Inc., Philadelphia,
Pa.
B. C. Hospitals Association,
Vancouver, B.C.
Vancouver Stock Exchange Ltd.,
Vancouver, B.C.

Univac 9400 system

Escaut Insurance Company, Antwerp,
Belgium

Univac 9200 system

Checking accounts, proof and transit, Golden Passbook,stockholder accounting,account reconciliation
(system valued at over $180,000)
Water Department billing, collection, analysis;
police court docket; traffic division, ticket enforcement, statistical analysis; City data processing, taxes, etc.; automotive accounting, inventory, analysis, of emergency medical services
(system valued at over $430,000)
General banking applications as well as muniCipal
water billing, safe deposit boxes, amortization
scheduling; bank is service bureau oriented
(system valued at over $500,000)
Demand deposit accounts, installment loan, proof/
transit, trust, bond portfolio, etc.
(system valued at about $600,000)
Processing and storing business and scientific data
collected from nine hospitals operated by the Order
(system valued at over $413,000)
Central data base and information systems development; also expansion of its computerized production
control system. presently on dual 3300 installation
Expansion of firm's scientific and technical services to German aerospace industry and government.
agencies
(system valued at $2.5 million)
Inventory management, order entry, invoicing and
sales information applications
Instruction in data processing and for administrative applications
Property tax accounting, payroll, budget accounting,
voter registration applications
Computer education applications
Applications including inventory control, accounting
services and statistical report preparation
Business and school services for the school district, which includes a community college
Inventory control and distribution of U.S. Air
Force forms and publications
All areas of life aSsurance
Producing a variety of financial statements and
reports
Provides production control, parts scheduling, sales
monitoring, sales forecasting
Personal trust accounting, certificates of deposit,
payrOll , etc.
Blending feed for cattle, billing, grain inventory;
also used several hours daily by Garden City
National Bank
A patient accounting system for nationwide complex
of over 40 recuperative care facilities in 24 states
Maintaining route settlement accounts, ice-cream
inventory, accounts payable
Order billing, inventory control, automatic reorder, sales analysis, and payroll for 580 employees
A variety of applications such as costing, stores
control, highway design and payrOll for 35,000
Processing state income tax returns, motor vehicle
registrations, state sales tax data
Chef Boy-Ar-Dee plant customer billing for some 150
warehouses, and general accounting chores
(system valued at $1 million)
Serving as remote terminals (at various Westinghouse
installations throughout U.S.) to a large-scale computer in Tele-Computer Center in Pittsburgh
(systems valued at about $2 million)
Applications including cost estimating, job cost
distribution, payroll and general accounting
Expediting operations of 37 hospitals with 16,000
employees
Providing instantaneous information on volumes and
prices; also prepares statistics and performs
accounting operations
Correlating accident statistics with premiums
charged; also developing improved techniques to
provide better and faster service to policy holders

COMPUTERS and AUTOMATION for September, 1970

MONTHLY COMPUTER CENSUS
Neil Macdonald
Survey Editor
COMPUTERS AND AUTOMATION
The following is a summary made by COMPUTERS 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. Please note the variation
in dates and reliability of the information. Several important manufacturers refuse to give out, confi rm, 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 accurate and complete as
possible.

The following abbreviations apply:
(A)

authoritative figures, derived essentially from informat(on
sent by the manufacturer directly to COMPUTERS AND
AUTO/IATION
figure is combined in a total
acknowledgment is given to DP Focus, ~larlboro, Mass., for
their help in estimating many of these figures
figure estimated by CO/IPUTERS AND AUTO/IATION
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
figures derived all or in part from information released
indirectly by the manufacturer, or from reports by other
sources I ikely to be informed
sale only, and sale (not rental) price is stated
no longer in production
information not obtained at press time

C
(D)
E
(N)
(R)

(S)
Part I of the Monthly Computer Census contains reports for United
X
States manufacturers. Part II contains reports for manufacturers
outside of the United States. The two parts are published in alternate months.
SUMMARY AS OF AUGUST 15, 1970
DATE OF

NAME OF
MANUFACTURER
Part I. 'Unit~dStat~s'MaM~fact~~~rs
Autonet i cs
Anaheim, Calif. (R) (1/69)
Bailey Meter Co.
Wickl iffe, Ohio (R) (6/70)
Bunker-Ramo Corp.
Canoga Park, Calif.
(A)

( 10/69)
Burroughs
Detroit, Mich.
(N)

( 1/69-5/69)

Control Data Corp.
Minneapol is, Minn.
(R)

(9/70)

NAME OF
COMPUTER
RECOMP II
RECOMP III
Bai ley 750
Bai 1ey 756
Bai ley 855
BR-130
BR-133
BR-230
BR-300
BR-330
BR-340
205
220
Bl00/B500
B2500
B3500
B5500
B6500
B7500
B8500
G15
G20
LGP-21
LGP-30
RPc4000
636/136/046 Series
160/8090 Series
924/924A
1604/A/B
1700
3100/3150
3200
3300
3400
3500
3600
3800
6400/6500
6600
6800
7600

FI RST
INSTALLATION
II/58
6/61
6760
2/65
4/68
10/61
5/64
8/63
3/59
12/60
12/63
1754
10/58
7/65
2/67
5/67
3/63
2/68
4/69
8/67

7155

4/61
12/62
9/56
1/61

AVERAGE OR RANGE
OF MONTHLY ·RENTAL
$ (000)
2.5
1.5
40-250 (S)
60-400 (s)
100-1000(S)
2.0
2.4
2.7
3.0
4.0
7.0
4.6
14.0
2.8-9.0
5.0
14.0
23.5
33.0
44.0
200.0

NUMBER OF INSTALLAtioNS
In
Outs i de
In
World
U.S.A.
U.S.A.
30
6
32
13
5
160
79
15
18
19
19
25-38
28-31

0
0
3
5
0

52-57
44
65-74

12
18
7

o
I

1.6

2.1-14.0
11.0
45.0
3.8
10-16
13.0
20-28
18.0
25.0
52.0
53.0
58.0
115.0
130.0
235.0

2/69
5/70
5/69
2/70
12764
10/67
11/60
8/62
9/63
10/64.
11/64
4/65
3/68
9/66
11/68
12/66
11/68
12/67
3/70
9/69
2/16
9/66

8.0
9.6
54-200
33-200
19.5
25.0
3.4
1.7
0.9
10.0
1.3
0.5
0.4.
0.3

o
I

20
X
X

2
2

o
o
o

27-40
30-33
64-69
'62
72-81
4

o
I

295
20
165
322
75
29
610
29
59
106-180
83-110
55-60
200
20
15
39
20
85
85

15.5
0.7
1.3
1.9

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

30
6
35
18
5

NUMBER OF
UNFI LLED
ORDERS

I
I

X
X
X
X
X
X

117
190
8
60
13
5
X
X

X
X
X
X
X

X
C
C
C
C

C
C
C
C
C
C
C
C

Total:
160
Data General Corp.
Southboro, Mass. (A) (6/70)
Datacraft Corp.
Ft. Lauderdale, Fla. (A) (9/70)Digiac Corp.
PI ainview, N.Y. (A) (2/70)
Digital Equipment Corp.
Marnard, Mass.
(A
(6170)

NOVA
SUPERNOVA
602471
6024/3
Dlgiac 3080
Di~iac 3080c
PD -I
PDP-4
PDP-5
PDP-6
PDP-7
PDP-8
PDP-81l
PDP-8/s
PDP-8/L
PDP-9
PDP-9L
PDP-IO
PDP-II
PDP-12
PDP-15
LI NC-8

COMPUTERS and AUTOMATION for September, 1970

(s)
(S)
(S1
(S)
(5)

(S)

1.1
8.0
10.5
17.0

9
17
14
5
SO
40
90.
C
C

(S)

C
C
C
C
C
C
C
C
6

650
22
9
17

5
10
C
C
C
C
C
C
C
C
C
C
C
C
C

45
100
23
160
1450

X
X
C

H57

1020
2350

C
C

4~5

41
Ilt4
27
275
15
142'

C
C
C

C
C
C
C
Total:
1350

NAME OF
MANUFACTURER
Electronic .Associates Inc.
Long Branch! N.J. (A~ (6/70)
EMR Computer
Minneapolis, Minn.
(N)
(6/70)

NAME OF
COMPUTER
640
8400
'EAR 6020
EMR 6040
EMR 6050
EMR 6070
EMR 6120
EMR 6130
EMR 6135

DATE OF
FIRST
INSTALLATION
4/67
7/65
4765
7/65
2/66
10/66
8/67

AVERAGE OR RANGE
OF MONTHL V RENTAL
$(000)
1.2
12.0
5.4
6.6
9.0
15.0
0.8
5.0
2.6

NUMBER OF INSTALLATIONS
In
Outside
In
Worl d
U.S.A.
U.S.A.
85
19
C
C
C
C

35
6

120
25

Total:
90 E
General Electric
Phoen i x, Ariz.
(N)
(9/70)

Process Cont ro 1 Computers:
(A)
(6170)

Hewlett Packard
Cupertino, Cal if.
(A) (9/70)
Honeywell
Computer Control Div.
Framingham, Mass.
(R)
(9/70)

Honeywe 11
EDP Di v.
We 11 es 1ey Hill s , Mass.
(R)
(6/70)

IBA
White Plains, N.V.
(N) (D)
(1/69-5/69)

58
105A
105B
10$RTS
115
120
130
205
210
215
225
235
245
255 T/S
265 T/S
275 T/S
405
410 T/S
415
420 T/S
425
430 T/S
435
440 T/S
615
625
635
655
4020
4040
4050
4060
2114A
2115A
2116A~

2116B
DDP-2
DDP-116
DDP-124
DDP-224
DDP-316
DDP-416
DDP-516
H112
H632
H1648
H-1l0
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
System 3
305
650
1130
140 I
1401-G
1401-H
1410
1440
1460
1620 1, 11
1800
7010
7030
704
7040
7044
705
7070, 2
7074
7080
7090
7094-1

5170

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
2/67
8/64
12/66
6/65
10/6B
11/67
11/66
5/63
4/65
3/66
3/65
6/69
9/66
10/69
12/68
11/68
BJ6B
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
1170
12/57
10/67
2/66
9/60
5/64
6/67
11/61
4/63
10/63
9/60
1/66
10/63
5/61
12/55
6/63
6/63
11/55
3/60
3/60
8/61
11/59
9/62

\.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
6.0
3.0
7.0
2.0
0.25
0.41
0.6
2.65
0.9
2.2
3.5
0.6

420-680

620-1080

11
35
15
145
40-60
3
15-20
45-60

0
0
1
15
17

11
35
16
160
57-77
3
15-20
60-90
10
15-45

15-30

10-40
170-300

70-100

240-400

50-100

20-30

70-130

23
20-40

3
3

26
23-43

46
20
1
2

1.2
3.2
12.0
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

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

180
30
800
150
800
46
58
230
130
4
15
125
20
18
10
0
40
50
2580
2210
420
180
156
1690
194
285
415
67
4
12
35
28
18
10
44
13
4
10

15
2
C
C
C
C

Total:
30 E

200-400

148
45
22
18

NUMBER OF
UNFI LLED
ORDERS

75
160
220
275
40
15
90
55
6
5
60
2
2
3
0
15
18
1227
1836
450
140
116
1174
63
186
148
14
1
1
27
13
3
3
26
2
2
4

194
65
23
20
915
663
1156
90
250
175
60
300
225
700
45
10
15
255
30
960
370
1075
86
73
325
185
10
20
185
22
20
14

X
X

55
68
3807
4046
870
320
272
2864
257
471
563
81
5
13
2
41
21
13
70
15
6
14

COMPUTERS and AUTOMATION for September, 1970

NAME OF
MANUFACTURER
IBM (Cont'd.)

NAME OF
COMPUTER
7094-11
360/20
360/25
360/30
360/40
360/44
360/50
360/65
360/67

360175

I nte rdata
Oceanport, N.J.
(A) (9/70)
NCR
Dayton, Ohio

(R)
(9/70)

Phi 1 co
Wi Ilow Grove, Pa.
(N) (1/69)
RCA
Cherry Hi 11, N.J.
(N)

(5/69)

Raytheon
Santa Ana, Cal if.
(A)
(9/70)
Scientific Control Corp.
Dallas, Tex.
(A)
(6/70)

Standard Computer Corp.
Los Angeles, Calif.
(N) (6/70)
Systems Engi neeri ng Laboratories
Ft. Lauderdale, Fla.
(A)
(6/70)

UNIVAC (Div. of Sperry Rand)
New York, N.Y.

(R)
( 1/69- 5/69)

Varian Data Machines
Newport Beach, Cal if.

(A) (9170)

Xerox Data Systems
El Segundo, Calif.
(N)

(4170)

DATE OF
FI RST
INSTALLATION
4/64
12/65
1/68
5/65
4/65
7/66
8/65
11/65
10/66
2/66

360/85
360/90
11/67
360/195
Model 2
7768
Model 3
3/67
Model 4
8/68
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
2000-210,211
10/58
2000-212
1/63
301
2161
501
6/59
601
11/62
3301
7/64
Spectra 70/15
9/65
Spectra 70/25
9/65
Spect ra 70/35
1/67
Spectra 70/45
11/65
Spectra 70/46
Spect ra 70/55
11/66
250
12/60
440
3/64
520
10/65
703
10/67
704
3/70
706
5/69
50
5
10/66
655
660
10/65
670
5/66
4700
4169
DCT-132
5/69
I C 4000
12168
IC 6000
5/67
I C 7000
6/69
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
III
8/62
Fi Ie Computers
8/56
Solid-State 80 I, II,
90 , I , I I, & Step
8/58
418
6/63
490 Series
12/61
1004
2/63
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
620 i
6/67
R-620 i
4/69
520i
10/68
11170
620
520/DC
12/69
XDS-92
4/65
XDS-910
8/62
XDS-920
9/62
XDS-925
12/64
XDS-930
6/64
XDS-940
4/66
11/64
XDS-9300
Sigma 2
12/66
Sigma 3
12/69
Sigma 5
8/67
Sigma 7
12/66

COMPUTERS and AUTOMATION for September, 1970

AVERAGE OR RANGE
OF MONTHLY RENTAL
$ (000)
83.0
2.7
5.1
10.3
19.3
11.8
29. I
57.2
133.8
66.9
150.3
(S)
232.0
0.25
0.4
0.6
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
0.7
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

NUMBER OF INSTALLATIONS
Outside
In
In
Worl d
U.S.A.
U.S.A.
4
6
10
4690
3276
7966

5075
1260
65
480
175

3144
498
13
109
31

8219
1758
78
589
206
13
17

9
14

o
5

4
3

o
o

400
125
350
1100
600
150

300
45
600
1550
200
50

16
12
140-290
22-50

o
5
70
95

X
X
X

24-60
90-11 0
68-70
65-100
84-180

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

25-65
125-170
86-95
85-150
105-235

100-130

1
11

o
1

155

175

1
20

20
27
181
18

6
4

44
23
137
41

9
3

o
o
o
o
o
o
o
o
o

24
211
75

5
1

216
76

38
31

16
40

3
36
31

23
25
13

o
o
6

6
9
3

10 E
X
32
26
X
X
2
2

X
X
X

9
8

o
3

9
11

38
127
106

18
48
38

56
175
144

11
628
299
62

o
o

112
86
2130
936
200

2
7-10
5-12

1
14

o
1

10-15

6-18
5-9

X
20
35
20
90
10

E
E

X
X
75
850
550
60

E
E
E
E

75
1200
30
125
10-60
150-170
93-120
20
159
28-35
21-25
60-110
10
15":40
24-35

o
o
X

35.0
57.0
68.0
1.5
3.4
7.0
135.0
0.9
0.5
0.4
0.5
1.6
1.5
2.0
2.9
3.0
3.4
14.0
8.5
1.8
2.0
6.0
12.0

1
12
10
X

16
40

210
76
75
1502
637
138

X
X
X

8.0
11.0
30.0
1.9
2.4
8.5

700
170
950
2650
800
200

240-420
23-51

18
230
175

20
26
161
12
40
23
137
41

NUMBER OF
UNF I LLED
ORDERS

18
12-62
157-180
98-132
21
173
28-35
22-26
70-125
10
21-58
29-44

X
400
30
330
125
25

E
E

CALENDAR OF COMING ,EVENTS
Sept. 1·3, 1970: 25th National Conference, Association for Computing
Machinery, New York Hilton, New York, N.Y.! contact: Sam Matsa,
ACM '70 General Chairman, IBM Corp., 410 E. 62nd St., New York,
N.Y. 10021
Sept. 2·4, 1970: The Institution of Electrical Engineers (lEE) Conference
on Man·Computer Interaction, UK National Physical laboratory, Ted·
dington, Middlesex, England! contact: Roger Dence, lEE Press Office,
Savoy Place, london WC2, England
Sept. 14.15, 1970: Society for Management .Information Systems
(SMIS) 1970 Conference, The Shoreham Hotel, Washington, D.C. !
contact: The Society for Management Systems (SMIS), c/o Prof. John
F. McCarthy, Jr., P.O. Box 16, Benjamin Franklin Station, Washington, D.C. 20044
Sept. 14·16, 1970: First Canadian Computer Show and Conference,
Montreal, Canada / contact: Jack McCaugherty, James Lovick Ltd.,
Vancouver, British Columbia, Canada
Sept. 14·17, 1970: 15th Annual Technical Symposium of The Society of
Photo·optical Instrumentation Engineers (SPIE), Anaheim, Calif. /
contact: SPIE, Attn: Symposia Vice President, P.O. Box 288, Redondo Beach, Calif. 90277

Sept. 15·17, 1970: 16th Annual Seminar of the American Society for
Industrial Security, Sheraton Boston Hotel, Boston, Mass. / contact:
Edward G. Goulart, Public Relations Chmn., M.I.T. lincoln laboratory, 244 Wood St., lexington, Mass. 02173
Sept. 16-18, 1970: Digital Equipment Computer Users Society (DECUS)
European Branch, Kunsterhaus, Munich, Germany / contact: Martha
Ries, DECUS European Secretary, c/o DECUS International Office, 81
Route de l' Aire, 1227 Carouge, Geneva, Switzerland
Sept. 17·18, 1970: Computer Science and Statistics Symposium, sponsored by the Los Angeles Chapter of the ACM, University of California, Irvine, Calif./ contact: Dr. Mitchell O. locks, C-E-I-R Professional Services Div., Control Data Corp., 6060 W. Manchester, los
Angeles, Calif; 90045; or Dr. Michael E. Tarter, Assoc. Prof., Dept.
of Mathematics and Dept. of Medicine, University of California,
Irvine, Calif. 92664
Sept. 22·24, 1970: The Computers and Communications Conference
(IEEE), The Beeches, Rome, N.Y.! contact: Jerold T. McClure, Conference Chairman, P.O. Box 182, Rome, N.Y. 13440
Sept. 22·24, 1970: Univac Users Association Fall Conference, Roosevelt
Hotel, New Orleans, lao / contact: User Group Relations, Univac
Division, Sperry Rand Corp., P.O. Box 500, Blue Bell, Pa. 19422

Sept. 14·24, 1970: 1970 FID (International Federation for Documenta·
tion) Conference and International Congress on Scientific Informa·
tion, Buenos Aires, Argentina! contact: U.S. National Committee for
FlO, National Academy of Sciences, 2101 Constitution Ave., Washington, D.C. 20418

Sept. 28·30, 1970: 6th Annual Meeting of the Association of American
Railroads Data Systems Div., Annual Equipment Show,· Regency Hyatt House, Atlanta,' Ga. / contact: Trade Associates, Inc., 5151 Wisconsin Ave., N.W., Washington, D.C. 20016

Sept. 15·16, 1970: 4th Annual Instrumentation Fair, Washington Hilton
Hotel, Washington, D.C. / contact: Norm Ward, AD-TECH, P.O. Box
475, Mclean, VA 22101

Sept. 29, 1970: 'Symposium on "Present and Future Uses of Computers
in the Chemical Industry", sponsored by the Soci'tlty of Consulting
Chemists and Chemical Engineers, White Plains Hotel, White Plains,

CLASSIFIED
ADVERTISEMENTS

ADVERTISING
INDEX

GEORGE S. McLAUGHLIN
ASSOCIATES, INC.
will buy or sell your used System/
360, 1400, or 700 Series

Special Type Heads
Prepared for IBM
Selectric Equipment
Camwil supplies custom modi·
fications of existing IBM and
Camwil type heads for Selectric
typewriters, Composers, accounting machines, and computer terminals.
We also mold new type heads
to your specified formats or you
may choose from our stock of
molded type heads which includes computer and teletype
codes, foreign languages, chemical and library formats.

CAMWIL, INC.
835 Keeaumoku Street
Honolulu, Hawaii 96814

201-273-5464
785 Springfield Avenue
Summit, New Jersey 07901
PDP9 LEASE FOR SALE
STANDARD PAPERTAPE READER
;PUNCH & TTY
KG09H Ext Mem Control
• MM09A (9, 129) Mem Extension
• CR03B Card Reader & Control
· KF09A Auto Priority Interrupt
KP09A Pow Fail & Auto Restart
· DB97A PDP9-PDP7 Interprocess
Buffer
· BULK STORAGE INTERFACE
Control Logic for IBM 2321 Data
Cell
(800 x 106 bytes; 175-600 ms)
· Data Line Interface for Twenty
Serial Lines
$2762/Mo.
C. Miller, Dir; Research Data Facility
2200 Webster St., San Francisco, Cal.
415-346-5135

Following is the index of advertisements. Each item contains:
Name and address of the advertiser / page number. where the
advertisement appears / name
of the agency, if any
Abercrombie & Fitch, P. O. Box
2991, Clinton, IA 52732 / Page 3
/ Media Selection Corp.
Academic Press, Inc., 111 Fifth
Ave., New York, NY 10003 / Page
2 / Flamm Advertising
Beemak Plastics, 7424 Santa Monica
Blvd., Los Angeles, CA 90046. /
Page 37 / Camwil, Inc., 835 Keeamoku St. ,
Honolulu, HI 96814 / Page 62 /
Richard T. Clarke Co.
Computers and Automation, 815
Washington St., Newtonville, MA
02160 / Page 64 / Miller-stephenson Chemical Co. ,
Inc., Route 7, Danbury, CT 06810
/ Page 31 / Michel-Cather, Inc.
University Microfilms, Xerox Education Group, 204 North Zeeb Rd. ,
Ann Arbor, MI 48106 / Page 7 /
Brian Connelly Advertising, Inc.

COMPUTERS and AUTOMATION for September, 1970

N.Y. I contact: Association of Consulting Chemists and Chemical
Engineers, Inc., 50 East 41st St., New York, N.Y. 10017

York, N.Y. / contact: Data Processing Supplies A~sociation, 1116
Summer St., P.O. Box 1333, Stamford, Conn. 06904

Oct. 5-7, 1970: 1970 Symposium on Feature Extraction and Selection
in Pattern Recognition, Argonne National Laboratory, Argonne, III. I
contact: David Jacobsohn, Argonne National Laboratory, Argonne,
IL 60440

Oct. 26·28, 1970: Forum of Control Data Users (FOCUS) Regional Conference, Statler Hilton Hotel, Washington, D.C. / contact: William
I. Rabkin, FOCUS Exec. Sec., c/o Itek Corp., 10 Maguire Rd., Lexington, Mass. 02173

Oct. 5-9, 1970: Computer 7G--lnternational Computer Exhibition,
Olympia, London, England I contact: M. F. Webster, Leedex Limited,
100 Whitechapel Road, London, E.l., England

Oct. 26-29, 1970: 25th Annual iSA Conference & Exhibit, Civic Center,
Ph1ladelphia, Pa. / contact: K. F. Fitch, Meetings Coordinator, Instrument Society of America, 530 William Penn Place, Pittsburgh, Pa.
15219

Oct. 7-9, 1970: American Production and Inventory Control Society
13th Annual International Conference, Ohio Convention Exposition
Center, Cincinnati, Ohio I contact: APICS National Office, Suite 504,
Watergate Bldg., 2600 Virginia Ave., N.W., Washington, D.C. 20037
Oct. 11-14, 1970: 33rd Annual Meeting of the American Society for
Information Science (ASIS), Sheraton Hotel, Philadelphia, Pa. I contact: ASIS 1970 Convention Chairman, Dr. Eugene Garfield, Institute
for Scientific Information, 325 Chestnut St., Philadelphia, Pa. 19106
Oct. 12-13, 1970: Sixth National Data Processing Conference of the
Information Processing Association of Israel, Tel Aviv Hilton Hotel,
Tel Aviv, Israel I contact: S. Shalish, Chmn., Information Processing
Association of Israel, P.O.B. 3009, Jerusalem, Israel
Oct. 12-14, 1970: SIAM (Society for Industrial and Applied Mathematics') 1970 Fall Meeting, Hotel Lenox, Boston, Mass. I contact: Genera I Chairman, SIAM 1970 Fall Meeting, 33 South 17th St., Philadelphia, Pa. 19103
Oct. 12-16, 1970: USE Fall Conference, Sheraton-Biltmore Hotel,
Atlanta, Ga. / contact: User Group Relations, Univac Division,
Sperry Rand Corp., P.O. Box 500, Blue Bell, Pa. 19422
Oct. 13, 1970: Second Annual TDCC (Transportation Data Coordinating
Committee) Seminar on the Computerization of Transportation Data
and Information Systems, Presidential Ballroom, Statler Hilton,
Wushington, D.C. / contact: Transportation Data Coordinating Committee, 1101 Seventeenth St. NW, Washington, D.C. 20036
Oct. 14-16, 1970: ADAPSO's 30th Management Conference & 9th
Annual Meeting, Paradise Island Hotel, Nassau, Bahamas / contact:
ADAPSO (Association of Data Processing Service Organizations, Inc.),
551 Fifth Ave., New York, N.Y. 10017
Oct. 14-16, 1970: IEEE Systems Science & Cybernetics Conference,
Webster Hall Hotel, Pittsburgh, Pa. / contact: Prof. A. Lavi, Carnegie-Mellon Univ., Pittsburgh, Pa. 15213
Oct. 14·16, 1970: International Conference on Management Informa·
tion Systems, Copenhagen, Denmark / contact: Harald Josefsen,
Program Committee Chmn., The Danish EDP Council, Vesterbrogade
1, DK-1620 Copenhagen V, Denmark
Oct. 14·16, 1970: Conference on Earth Station Technology, London,
England / contact: Helen Kaye, The Institution of Electrical Engineers, Savoy Place, London WC2, England

Oct. 14·17, 1970: International Symposium on Digital Computer Appli·
cations in Engineering Sciences, Technical University of Istanbul,
Turkey / contact: F. A. Akyuz, I.T.U. Hesap Merkezi (Computation
Center), Taskisla 114, Istanbul, Turkey
Oct. 15-16, 1970: 1970 Atlantic Div. of Assoc. for Systems Manage.
ment Eighth Annual Atlantic Systems Conference, New York Hilton,
New York City, N.Y. / contact: Malcolm B. Foster, A.S.C., Box 461,
Pleasantville, N.Y. 10570
Oct. 19·20, 1970: 5th Annual Conference, Digitronics Users Association
(DUA), Ambassador Hotel, Chicago, III. / contact: Executive Secretary, DUA, Box 113, Albertson, L.I., N.Y. 11507
Oct. 19·21, 1970: 11th National Meeting of The Institute of Management Sciences, Los Angeles Hilton Hotel, Los Angeles, Calif. / con~
tael: Gene Saxby, Security Pacific National Bank, P.O. Box 2097
Terminal Annex, Los Angeles, Calif. 90054
Oct. 20, 1970: Division 11 Fall Conference of the Pittsburgh Chapter
of the Data Processing Management Association, Pittsburgh, Pa. /
contact: James J. Dean, P.O. Box 2004, Pittsburgh, PA 15230
Oct. 26·28, 1970: Data Processing Supplies Association, Fall General
Meeting, The Park Sheraton Hotel, 7th Ave., and 56th St., New

COMPUTERS and AUTOMATION for September, 1970

Oct. 26-30, 1970: 12th Annual BEMA Business Equipment Exposition,
New York. Coliseum, New York, N.Y. / contact: Business Equipment
Manufacturers Association (BEMA), 1828 "L" St. NW, Washington,
D.C. 20038
Oct. 27·29, 1970: 12th Annual BEMA Management Conference, American Hotel, New York, N.Y. / contact: Business Equipment Manufacturers Association (BEMA), 1828 "L" St. NW, Washington, D.C. 20038
Oct. 27-30, 1970: Midwest Power Systems Conference and Symposium
(IEEE co-sponsor), Iowa State University, Ames, Iowa / contact:
H. K. Baker, Engineering Extension, 110 Marston Hall, Iowa State
University, Ames, Iowa 50010
Oct. 29-30, 1970: IEEE Joint Engineering Management Conference,
Drake Hotel, Chicago, III. / contact: AilE Headquarters, 345 E. 47th
St., New York, N.Y. 10017
Nov. 10-12, 1970: National Symposium on Criminal Justice Information
and Statistics Systems, Sheraton-Dallas Hotel, Dallas, Texas / contact:
Project SEARCH, 1108 14th St. Fifth Floor, Sacramento, Calif. 95814
Nov. 12-13, 1970: Canadian IEEE Symposium on Communications,
Queen Elii:abeth Hotel, Montrea I, Quebec, Canada / contact: IEEE
Headquarters, Technical Conference Services, 345 E. 47th St., New
York, N.Y. 10017
Nov. 12·13, 1970: CAST '70 Conference (AilE), The Americana Hotel,
Miami Beach, Fla. I contact: Joseph P. Lacusky, American Institute
of Industrial Engineers, Inc., CAST '70, P. O. Box 1081, Miami,
Fla. 33148
Nov. 12-13, 1970: 11th IEEE Symposium on Man·Machine Systems,
Langford Hotel, Winter Park, Fla. / contact: The Institute of Electrical and Electronics Engineers, Inc., 345 East 47th St., New York,
N.Y. 10017
Nov. 16, 1970: ACM Computer Graphics Workshop, Houston, Tex. /
contact: Jackie Potts, ACM, SIGGRAPH, Box 933, Blair Station,
Silver Spring, MD 20910
Nov. 17-19, 1970: Fall Joint Computer Conference, Astro Hall, Houston, Tex. / contact: L. E. Axsom, IBM Scientific Ctr., 6900 Fannin,
Houston, Tex. 77025
Nov. 19·21, 1970: DECUS (Digital Equipment Computer Users Society)
1970 Fall Symposium, Shamrock Hilton, Houston, Texas / contact:
DECUS, Digital Equipment Corp., Maynard, Mass. 01754
Dec. 2·3, 1970: Conference on Display Devices, United Engineering
Ctr., New York, N.Y. / contact: Sam Stone, Gen'l Tel. & Elec.,
208-20 Willets Pt. Blvd., Bayside, N.Y. 11360
Dec. 7-9, 1970: 9th IEEE Symposium on Adaptive Processes: Decision
and Control, Univ. of Texas, Austin, Tex. / contact: Prof. D. G.
Lainiotis, Engineering Science Bldg., 502, Univ. of Texas at Austin,
Austin, Tex. 78712
Dec. 7·9, 1970: 26th Annual National Electronics Conference and
Exhibition (NEC/70), Conrad Hilton Hotel, Chicago, III. / contact:
NEC, Oakbrook Executive Plaza #2, 1211 W. 22nd St., Oak Brook,
III. 60521
Dec. 9·11, 1970: Fourth Conference on Applications of Simulation,
Waldorf-Astoria, New York, N.Y. I contact: Association for Computing Machinery, 1133 Avenue of the Americas, New York, N.Y.
11036
Jan. 31-Feb. 5, 1971: IEEE Winter Power Meeting, Statler Hilton Hotel,
New York, N.Y. / contact: IEEE Headquarters, Technical Conference
Service, 345 E. 47th St., New York, N.Y. 10017

THE MA Y ARTICLE
THE ASSASSINATION OF PRESIDENT 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
_ ~newdevidence 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 s~d cash.) If not satisfactory, returnable i~ 7 days for full refund (if
in salable conditidn)o
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 /.Jay 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 during 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 s~quences. The number of
frames in a movie sequence ranges from about 10 to
abortt 500; and in the count of 510 photographs given
above, the 10 to 500 frames of a single movie sequence are counted just as ~ 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~f ~he·photographs:were~aken-by professional photographers. About 30 of'the photographers
were professionals who worked for newspapers, television .networks ,.:andphotogr,aphic'agencies.

TheWarr.en, Commis'sion .did: not interview .a single
one of the profes:si.onal'.photographers, nor did the
WarrenCommis~ion'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 televi sion 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
"unavail able" to researchers. The Commi ssion 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.

\
",'

\
"
\

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 20:51:56-07:00
Create Date                     : 2009:03:24 20:51:56-07:00
Metadata Date                   : 2009:03:24 20:51:56-07:00
Format                          : application/pdf
Document ID                     : uuid:ae20dc15-3fb7-43ad-aebf-03cb2717a50a
Instance ID                     : uuid:60bfe2a4-97ab-4215-8a75-aa5f6b278c05
Page Layout                     : SinglePage
Page Mode                       : UseNone
Page Count                      : 60

EXIF Metadata provided by EXIF.tools

197009

197009 197009

Navigation menu

Versions of this User Manual:

Views

Navigation